Revised NI 43-101 Technical Report Pre-Feasibility Study ...

NOTE TO READER

An out of date filing titled “NI 43-101 Technical Report Pre-Feasibility Study Marmato Project Colombia” was filed on August 17, 2020. The correct and current technical report titled “Revised NI 43-101 Technical Report Pre-Feasibility Study Marmato Project Colombia” is attached to this filing. The material changes are:

• Statements of mineral reserves and mineral resources have been updated to specify whether such estimates include or exclude the mineral reserve tonnage.

• Figure 17-2 (process flow sheet for the MDZ process plant), which was inadvertently omitted from the original report, is now included in this amended report.

Revised NI 43-101 Technical Report Pre-Feasibility Study Marmato Project Colombia

Effective Date: March 17, 2020 Report Date: September 18, 2020

Report Prepared for

Caldas Gold Corp. 401 Bay Street, Suite 2400

Toronto, Ontario, Canada M5H 2Y4

Report Prepared by

SRK Consulting (U.S.), Inc.

1125 Seventeenth Street, Suite 600

Denver, CO 80202

SRK Project Number: 557200.030

Signed by Qualified Persons:

Ben Parsons, MSc, MAusIMM (CP), Practice Leader/Principal Consultant (Resource Geology)

Eric J. Olin, MSc Metallurgy, MBA, SME-RM, MAusIMM, Principal Consultant (Metallurgy)

Fernando Rodrigues, BS Mining, MBA, MAusIMM, MMSAQP, Practice Leader/Principal Consultant (Mining)

Jeff Osborn, BEng Mining, MMSAQP, Principal Consultant (Mining)

Joanna Poeck, BEng Mining, SME-RM, MMSAQP, Principal Consultant (Mining)

Fredy Henriquez, MS Eng, SME, ISRM, Principal Consultant (Rock Mechanics)

Breese Burnley, P.E., Practice Leader/Principal Engineer (Tailings) Cristian A Pereira Farias, SME-RM, Principal Consultant (Hydrogeology)

David Hoekstra, BS, PE, NCEES, SME-RM, Principal Consultant (Hydrology)

David Bird, PG, SME-RM, Associate Consultant (Geochemistry)

Mark Allan Willow, MSc, CEM, SME-RM, Practice Leader/Principal Consultant (Environmental)

Tommaso Roberto Raponi, P.Eng, Principal Metallurgist (Ausenco)

Reviewed by:

Berkley J. Tracy, MSc Geology, PG, CPG, PGeo, Principal Consultant (Resource Geology)

Tim Olson, BSc Mining, J.D., FAusIMM, Principal Consultant (Mining)

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page ii

MMS/KD Marmato_PFS_NI43-101_557200-030_Rev09.docx August 2020

Table of Contents

1 Summary ....................................................................................................................... 1

1.1 Property Description and Ownership .................................................................................................. 1

1.2 Geology and Mineralization ................................................................................................................ 1

1.3 Status of Exploration, Development and Operations .......................................................................... 3

1.4 Mineral Processing and Metallurgical Testing .................................................................................... 4

1.5 Mineral Resource Estimate ................................................................................................................. 5

1.6 Mineral Reserve Estimate ................................................................................................................. 10

1.7 Mining Methods ................................................................................................................................. 11

1.8 Recovery Methods ............................................................................................................................ 16

1.9 Project Infrastructure ......................................................................................................................... 18

1.9.1 Tailing Management Facilities ............................................................................................... 19

1.10 Environmental Studies and Permitting .............................................................................................. 19

1.10.1 Environmental Studies and Management ............................................................................. 19

1.10.2 Permitting .............................................................................................................................. 20

1.10.3 Social or Community Related Requirements ........................................................................ 21

1.10.4 Community Relations ............................................................................................................ 21

1.10.5 Mine Closure, Remediation, and Reclamation ...................................................................... 22

1.11 Capital and Operating Costs ............................................................................................................. 22

1.11.1 Marmato UZ Capital Costs .................................................................................................... 22

1.11.2 MDZ Capital Costs ................................................................................................................ 23

1.11.3 Marmato Operating Costs ..................................................................................................... 25

1.12 Economic Analysis ............................................................................................................................ 26

1.13 Conclusions and Recommendations ................................................................................................ 28

1.13.1 Property Description and Ownership .................................................................................... 28

1.13.2 Geology and Mineralization ................................................................................................... 28

1.13.3 Status of Exploration, Development and Operations ............................................................ 29

1.13.4 Mineral Processing and Metallurgical Testing....................................................................... 29

1.13.5 Mineral Resource Estimate ................................................................................................... 29

1.13.6 Mining and Reserves ............................................................................................................. 30

1.13.7 Recovery Methods ................................................................................................................ 33

1.13.8 Project Infrastructure ............................................................................................................. 33

1.13.9 Environmental Studies and Permitting .................................................................................. 34

1.13.10 Capital and Operating Costs ............................................................................................. 36

1.13.11 Economic Analysis ............................................................................................................ 37

2 Introduction ................................................................................................................ 39

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page iii


2.1 Terms of Reference and Purpose of the Report ............................................................................... 39

2.2 Qualifications of Consultants (SRK) .................................................................................................. 39

2.3 Details of Inspection .......................................................................................................................... 41

2.4 Sources of Information ...................................................................................................................... 42

2.5 Effective Date .................................................................................................................................... 43

2.6 Units of Measure ............................................................................................................................... 43

3 Reliance on Other Experts ........................................................................................ 44

4 Property Description and Location .......................................................................... 45

4.1 Property Location .............................................................................................................................. 45

4.2 Mineral Titles ..................................................................................................................................... 45

4.2.1 Nature and Extent of Issuer’s Interest ................................................................................... 49

4.3 Royalties, Agreements and Encumbrances ...................................................................................... 49

4.4 Environmental Liabilities and Permitting ........................................................................................... 49

4.4.1 Environmental Liabilities........................................................................................................ 49

4.4.2 Required Permits and Status ................................................................................................ 50

4.5 Other Significant Factors and Risks .................................................................................................. 50

5 Accessibility, Climate, Local Resources, Infrastructure and Physiography ........ 51

5.1 Topography, Elevation and Vegetation ............................................................................................. 51

5.2 Accessibility and Transportation to the Property .............................................................................. 51

5.3 Climate and Length of Operating Season ......................................................................................... 53

5.4 Sufficiency of Surface Rights ............................................................................................................ 53

5.5 Infrastructure Availability and Sources.............................................................................................. 53

5.5.1 Power .................................................................................................................................... 53

5.5.2 Water ..................................................................................................................................... 53

5.5.3 Mining Personnel ................................................................................................................... 53

5.5.4 Potential Tailings Storage Areas ........................................................................................... 54

5.5.5 Potential Waste Disposal Areas ............................................................................................ 54

5.5.6 Potential Processing Plant Sites ........................................................................................... 54

6 History ......................................................................................................................... 55

6.1 Prior Ownership and Ownership Changes ....................................................................................... 55

6.2 Exploration and Development Results of Previous Owners ............................................................. 56

6.3 Historic Mineral Resource and Reserve Estimates .......................................................................... 56

6.4 Historic Production ............................................................................................................................ 58

7 Geological Setting and Mineralization ..................................................................... 60

7.1 Regional Geology .............................................................................................................................. 60

7.2 Local Geology ................................................................................................................................... 62

7.2.1 Graphitic-Sericite Schist (MSG) ............................................................................................ 64

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page iv


7.2.2 Amphibolites (MAB) ............................................................................................................... 64

7.2.3 Serpentinites (MSP) .............................................................................................................. 64

7.2.4 Basalts (VB) .......................................................................................................................... 64

7.2.5 Clastic Sedimentary Rocks (S) ............................................................................................. 64

7.2.6 Marmato Porphyry Stocks (P1 – P5) ..................................................................................... 65

7.2.7 Unconsolidated Quaternary Deposits (QC) ........................................................................... 65

7.3 Property Geology .............................................................................................................................. 65

7.3.1 Structure ................................................................................................................................ 69

7.3.2 Alteration ............................................................................................................................... 71

7.4 Significant Mineralized Zones ........................................................................................................... 74

8 Deposit Type .............................................................................................................. 79

8.1 Mineral Deposit ................................................................................................................................. 79

8.2 Geological Model .............................................................................................................................. 79

9 Exploration ................................................................................................................. 81

9.1 Relevant Exploration Work ............................................................................................................... 81

9.1.1 Topographic Surveys ............................................................................................................ 81

9.1.2 Surface Geochemistry ........................................................................................................... 83

9.1.3 Geophysics ............................................................................................................................ 83

9.1.4 Surface Geological Mapping ................................................................................................. 83

9.1.5 Underground Geological Mapping ........................................................................................ 84

9.2 Sampling Methods and Sample Quality ............................................................................................ 85

9.2.1 Mine Geology - Channel Sampling Procedure ...................................................................... 85

9.2.2 Channel Sampling – Exploration ........................................................................................... 87

9.2.3 SRK Opinion of Quality ......................................................................................................... 89

9.3 Significant Results and Interpretation ............................................................................................... 91

10 Drilling ......................................................................................................................... 93

10.1 Type and Extent ................................................................................................................................ 93

10.2 Procedures ........................................................................................................................................ 95

10.2.1 Core Storage ......................................................................................................................... 97

10.2.2 Collar Surveys Surface .......................................................................................................... 98

10.2.3 Collar Surveys Underground ................................................................................................. 98

10.2.4 Drilling Orientation ................................................................................................................. 98

10.3 Interpretation and Relevant Results ................................................................................................ 101

11 Sample Preparation, Analysis and Security .......................................................... 102

11.1 Security Measures .......................................................................................................................... 102

11.2 Sample Preparation for Analysis ..................................................................................................... 102

11.2.1 Historical Sample Preparation (Pre 2010) ........................................................................... 102

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page v


11.2.2 Sample Preparation Mine Sampling (2010 – 2017) ............................................................ 103

11.2.3 Sample Preparation ............................................................................................................. 105

11.3 Sample Analysis .............................................................................................................................. 106

11.4 Quality Assurance/Quality Control Procedures .............................................................................. 107

11.4.1 Standards ............................................................................................................................ 109

11.4.2 Blanks .................................................................................................................................. 111

11.4.3 Duplicates ............................................................................................................................ 115

11.4.4 Actions/Reassays ................................................................................................................ 118

11.4.5 Check Analysis Results ....................................................................................................... 119

11.5 Opinion on Adequacy ...................................................................................................................... 121

12 Data Verification ....................................................................................................... 122

12.1 Procedures ...................................................................................................................................... 122

12.1.1 Verifications by CGM ........................................................................................................... 122

12.1.2 Verifications by SRK ............................................................................................................ 123

12.2 Limitations ....................................................................................................................................... 125

12.3 Opinion on Data Adequacy ............................................................................................................. 126

13 Mineral Processing and Metallurgical Testing ...................................................... 127

13.1 Metallurgical Program – 2019 ......................................................................................................... 127

13.1.1 Metallurgical Sample Characterization ................................................................................ 127

13.1.2 Mineralogy ........................................................................................................................... 129

13.1.3 Comminution Testwork ........................................................................................................ 129

13.1.4 Whole-Ore Cyanidation ....................................................................................................... 129

13.1.5 Gravity Concentration .......................................................................................................... 131

13.1.6 Cyanidation of Gravity Tailing ............................................................................................. 131

13.1.7 Variability Composites ......................................................................................................... 133

13.1.8 Flotation from Gravity Tailing .............................................................................................. 134

13.1.9 Cyanide Detoxification ......................................................................................................... 135

13.1.10 Solid-Liquid Separation ................................................................................................... 137

13.2 Metallurgical Program – 2020 ......................................................................................................... 139

13.2.1 Metallurgical Sample Location ............................................................................................ 139

13.2.2 Head Analyses .................................................................................................................... 146

13.2.3 Mineralogy ........................................................................................................................... 148

13.2.4 Comminution ....................................................................................................................... 148

13.2.5 Gravity Recoverable Gold (E-GRG) Testwork .................................................................... 151

13.2.6 Gravity Separation Testwork ............................................................................................... 152

13.2.7 Gravity Concentrate Cyanidation ........................................................................................ 154

13.2.8 Gravity Tailing Cyanidation Versus Grind Size ................................................................... 154

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page vi


13.2.9 Cyanidation Versus Cyanide Concentration and Pulp Density ........................................... 156

13.2.10 Cyanidation Versus Preaeration and Air Versus Oxygen Injection ................................ 158

13.2.11 Cyanidation Versus Cyanide Attenuation and Pulp Density ........................................... 160

13.2.12 “Hard Stop” Retention Time Tests .................................................................................. 162

13.2.13 Carbon-In-Leach (CIL) Tests .......................................................................................... 163

13.2.14 Variability Tests ............................................................................................................... 163

13.2.15 CIP Modelling Testwork .................................................................................................. 165

13.2.16 Cyanide Destruction Testwork ........................................................................................ 169

13.2.17 Tailing Thickening ........................................................................................................... 170

13.2.18 Tailings Filtration ............................................................................................................. 172

13.3 Recovery Estimate .......................................................................................................................... 174

13.4 Significant Factors ........................................................................................................................... 175

14 Mineral Resource Estimate ..................................................................................... 177

14.1 Drillhole Database ........................................................................................................................... 178

14.2 Geologic Model ............................................................................................................................... 179

14.2.1 Fault Network ...................................................................................................................... 179

14.2.2 Topographic Wireframes ..................................................................................................... 182

14.2.3 Lithological Wireframes ....................................................................................................... 182

14.2.4 Veins Model ......................................................................................................................... 183

14.2.5 Disseminated Model ............................................................................................................ 185

14.2.6 Splays Model ....................................................................................................................... 186

14.2.7 Porphyry “Pocket” Model ..................................................................................................... 187

14.2.8 MDZ ..................................................................................................................................... 189

14.3 Domains .......................................................................................................................................... 191

14.4 Assay Capping and Compositing .................................................................................................... 193

14.4.1 Outliers ................................................................................................................................ 193

14.4.2 Compositing ........................................................................................................................ 208

14.5 Density ............................................................................................................................................ 211

14.6 Variogram Analysis and Modeling .................................................................................................. 214

14.7 Block Model ..................................................................................................................................... 221

14.8 Estimation Methodology .................................................................................................................. 222

14.8.1 Theoretical Analysis ............................................................................................................ 222

14.8.2 Dynamic Anisotropy ............................................................................................................ 226

14.8.3 Threshold Capping .............................................................................................................. 227

14.8.4 Final Parameters ................................................................................................................. 227

14.9 Model Validation .............................................................................................................................. 230

14.9.1 Visual Comparison .............................................................................................................. 230

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page vii


14.9.2 Comparative Statistics ......................................................................................................... 235

14.9.3 Swath Plots ......................................................................................................................... 240

14.10 Resource Classification .................................................................................................................. 248

14.10.1 Measure Mineral Resources ........................................................................................... 248

14.10.2 Indicated Mineral Resources .......................................................................................... 248

14.10.3 Inferred Mineral Resources ............................................................................................. 249

14.10.4 Final Classification .......................................................................................................... 249

14.11 Depletion ......................................................................................................................................... 250

14.12 Mineral Resource Statement .......................................................................................................... 250

14.13 Comparison to the Previous Estimate ............................................................................................. 253

14.14 Mineral Resource Sensitivity ........................................................................................................... 254

14.15 Relevant Factors ............................................................................................................................. 260

15 Mineral Reserve Estimate ........................................................................................ 261

15.1 Conversion Assumptions, Parameters and Methods ...................................................................... 262

15.1.1 Upper Zone - Dilution .......................................................................................................... 262

15.1.2 Upper Zone - Recovery ....................................................................................................... 263

15.1.3 Upper Zone - Additional Allowance Factors ........................................................................ 264

15.1.4 Upper Zone – Cutoff Grade Calculation .............................................................................. 264

15.1.5 MDZ Mine - Dilution ............................................................................................................. 265

15.1.6 MDZ – Recovery ................................................................................................................. 266

15.1.7 MDZ - Additional Allowance Factors ................................................................................... 266

15.1.8 MDZ – Cutoff Grade Calculation ......................................................................................... 266

15.2 Reserve Estimate ............................................................................................................................ 267

15.3 Relevant Factors ............................................................................................................................. 268

16 Mining Methods ........................................................................................................ 270

16.1 Current Mining Methods .................................................................................................................. 270

16.1.1 Mine Layout ......................................................................................................................... 272

16.1.2 Reconciliation ...................................................................................................................... 274

16.1.3 Dilution................................................................................................................................. 274

16.2 Geotechnical ................................................................................................................................... 275

16.2.1 Geotechnical Data Base ...................................................................................................... 276

16.2.2 Engineering-Geology ........................................................................................................... 277

16.2.3 Stope Stability Assessment ................................................................................................. 279

16.2.4 Dilution................................................................................................................................. 279

16.2.5 Paste Fill Strength Estimation ............................................................................................. 279

16.2.6 PFS Ground Support Requirements ................................................................................... 280

16.2.7 Sill Pillar design ................................................................................................................... 281

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page viii


16.2.8 Critical Infrastructure Stability Assessment ......................................................................... 281

16.2.9 Limitations and Gaps ........................................................................................................... 284

16.2.10 Feasibility Study Recommendations ............................................................................... 285

16.3 Hydrogeology and Mine Dewatering ............................................................................................... 286

16.3.1 Hydrogeological Conditions ................................................................................................ 286

16.3.2 Descriptions of Numerical Groundwater Model .................................................................. 293

16.3.3 Results of Predictions by Groundwater Model .................................................................... 297

16.3.4 Hydrogeological Uncertainties ............................................................................................ 301

16.4 Upper Zone Mining .......................................................................................................................... 302

16.4.1 Stope Optimization .............................................................................................................. 302

16.4.2 Mine Design ........................................................................................................................ 304

16.4.3 Production Schedule ........................................................................................................... 305

16.4.4 Mining Operations ............................................................................................................... 310

16.4.5 Ventilation ............................................................................................................................ 312

16.4.6 Mine Services ...................................................................................................................... 313

16.4.7 Recommendations .............................................................................................................. 315

16.5 MDZ Mining ..................................................................................................................................... 316

16.5.1 Stope Optimization .............................................................................................................. 316

16.5.2 Mine Design ........................................................................................................................ 319

16.5.3 Production Schedule ........................................................................................................... 326

16.5.4 Mining Operations ............................................................................................................... 333

16.5.5 Ventilation ............................................................................................................................ 338

16.5.6 Mine Infrastructure & Services ............................................................................................ 343

16.5.7 Mine Labor .......................................................................................................................... 345

16.5.8 Equipment ........................................................................................................................... 347

16.6 Combined UZ and MDZ Production Schedule ................................................................................ 350

17 Recovery Methods ................................................................................................... 352

17.1 Marmato Process Plant (Current Operations) ................................................................................ 352

17.1.1 Crushing Circuit ................................................................................................................... 355

17.1.2 Grinding and Gravity Concentration Circuit ......................................................................... 355

17.1.3 Flotation and Concentrate Regrind Circuit .......................................................................... 355

17.1.4 Cyanidation and Counter-Current-Decantation (CCD) Circuit ............................................ 355

17.1.5 Merrill-Crowe Circuit and Smelter ....................................................................................... 356

17.1.6 Process Plant Consumables ............................................................................................... 356

17.1.7 Operating Performance ....................................................................................................... 357

17.1.8 Operating Costs .................................................................................................................. 357

17.2 Expansion Plans ............................................................................................................................. 358

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page ix


17.3 MDZ Process Plant ......................................................................................................................... 359

17.3.1 Processing Methods ............................................................................................................ 359

17.3.2 Plant Design and Equipment Characteristics ...................................................................... 361

17.3.3 Process Plant Description ................................................................................................... 362

17.3.4 Primary/Secondary Crushing and Stockpile ........................................................................ 364

17.3.5 Grinding ............................................................................................................................... 365

17.3.6 Gravity Concentration and Intensive Cyanide Leach Circuit ............................................... 366

17.3.7 Leach and Adsorption Circuit .............................................................................................. 366

17.3.8 Carbon Elution and Regeneration Circuit ............................................................................ 367

17.3.9 Electrowinning and Gold Room ........................................................................................... 368

17.3.10 Cyanide Detoxification .................................................................................................... 368

17.3.11 Tailings Thickening and Filtration ................................................................................... 368

17.3.12 Reagents ......................................................................................................................... 370

17.3.13 Services and Utilities ....................................................................................................... 372

17.3.14 Water Supply ................................................................................................................... 372

17.3.15 Operating Costs .............................................................................................................. 372

18 Project Infrastructure............................................................................................... 376

18.1 General Site Access........................................................................................................................ 376

18.2 Marmato Existing UZ Operations Infrastructure ............................................................................. 377

18.2.1 Existing Project Access ....................................................................................................... 377

18.2.2 Existing Project Facilities ..................................................................................................... 378

18.2.3 Energy Supply and Distribution - Existing Marmato Project ............................................... 380

18.2.4 Site Water Supply ................................................................................................................ 382

18.3 MDZ Introduction ............................................................................................................................. 382

18.4 MDZ Process Plant Site Location ................................................................................................... 382

18.4.1 Site Geotechnical ................................................................................................................ 384

18.5 MDZ On-Site Roads and River Crossings ...................................................................................... 384

18.5.1 Site Access Road ................................................................................................................ 384

18.5.2 River Crossing ..................................................................................................................... 384

18.6 MDZ Water Supply .......................................................................................................................... 384

18.6.1 Water Requirements ........................................................................................................... 384

18.6.2 Run-Off Water Collection and Treatment System ............................................................... 385

18.6.3 River Water Collection and Treatment System for MDZ and UZ ........................................ 385

18.7 MDZ Power Supply ......................................................................................................................... 385

18.7.1 Electrical Power Source ...................................................................................................... 385

18.7.2 Electrical Distribution ........................................................................................................... 385

18.7.3 Electrical Rooms ................................................................................................................. 386

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page x


18.7.4 Transformers ....................................................................................................................... 386

18.7.5 Standby/Emergency Power Supply ..................................................................................... 386

18.7.6 Ball and SAG Mill Drives ..................................................................................................... 386

18.7.7 Redundancy ........................................................................................................................ 386

18.8 MDZ Mine Operations Support Facilities ........................................................................................ 387

18.8.1 Mine Administration and Dry Building ................................................................................. 387

18.8.2 General Maintenance Building ............................................................................................ 387

18.8.3 Truck Wash Facility ............................................................................................................. 387

18.8.4 Truck Fuel Facility and Equipment Ready Line................................................................... 387

18.8.5 Explosives Storage .............................................................................................................. 387

18.9 MDZ Process Support Facilities ...................................................................................................... 387

18.9.1 Mill Administration Office and First Aid Facility ................................................................... 387

18.9.2 Laboratory ........................................................................................................................... 387

18.9.3 Warehouse and Storage Yard ............................................................................................. 388

18.9.4 Gatehouse and Weigh-Scale .............................................................................................. 388

18.10 Common Support Facilities ............................................................................................................. 388

18.10.1 Man Camp ....................................................................................................................... 388

18.11 MDZ Support Facilities .................................................................................................................... 388

18.11.1 Communications ............................................................................................................. 388

18.11.2 Wastewater Treatment .................................................................................................... 388

18.11.3 Solid Waste Disposal ...................................................................................................... 389

18.12 MDZ Site Preparation...................................................................................................................... 389

18.12.1 Site Earthwork ................................................................................................................. 389

18.12.2 Site Foundations ............................................................................................................. 389

18.13 MDZ Cemented Paste Backfill Plant ............................................................................................... 389

18.14 Site Water Management ................................................................................................................. 392

18.14.1 Water Supply ................................................................................................................... 392

18.15 Tailings Management Area ............................................................................................................. 394

18.15.1 Existing Tailings Management Facilities ......................................................................... 395

18.15.2 New Tailings Storage Facility Siting Study ..................................................................... 396

18.15.3 New Dry Stack Tailings Storage Facility Design ............................................................. 397

18.15.4 Tailings Risks and Opportunities .................................................................................... 406

18.16 Off-Site Infrastructure and Logistics Requirements ........................................................................ 407

18.16.1 Port .................................................................................................................................. 407

18.16.2 Rail .................................................................................................................................. 407

19 Market Studies and Contracts ................................................................................ 409

19.1 Commodity Price Projections .......................................................................................................... 409

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xi


19.2 Contracts and Status....................................................................................................................... 409

20 Environmental Studies, Permitting and Social or Community Impact ................ 410

20.1 Environmental Studies .................................................................................................................... 410

20.1.1 Environmental Setting ......................................................................................................... 410

20.1.2 Management Procedures and Baseline Studies ................................................................. 411

20.1.3 Geochemistry ...................................................................................................................... 412

20.1.4 Known Environmental Issues .............................................................................................. 412

20.2 Mine Waste Management and Monitoring ...................................................................................... 413

20.2.1 Waste Rock Management ................................................................................................... 413

20.2.2 Tailings Management .......................................................................................................... 413

20.2.3 Site Monitoring .................................................................................................................... 414

20.2.4 Environmental Procedures and Permissions ...................................................................... 415

20.2.5 General Water Management ............................................................................................... 417

20.2.6 Environmental Management Budget ................................................................................... 417

20.3 Project Permitting Requirements .................................................................................................... 417

20.3.1 General Mining Authority ..................................................................................................... 417

20.3.2 Environmental Authority ...................................................................................................... 418

20.3.3 Environmental Regulations and Impact Assessment .......................................................... 419

20.3.4 Water Quality and Water Concessions ............................................................................... 420

20.3.5 Air Quality and Emissions ................................................................................................... 421

20.3.6 Fauna and Flora Protection ................................................................................................. 421

20.3.7 Protection of Riparian Areas and Drainages ....................................................................... 422

20.3.8 Protection of Cultural Heritage or Archaeology ................................................................... 422

20.3.9 Marmato Permitting ............................................................................................................. 422

20.3.10 Performance and Reclamation Bonding ......................................................................... 423

20.4 Social or Community Related Requirements .................................................................................. 424

20.4.1 Social Investment ................................................................................................................ 424

20.4.2 Community Relations .......................................................................................................... 425

20.4.3 Employment ........................................................................................................................ 425

20.4.4 Artisanal and Small-Scale Mining Operations ..................................................................... 425

20.5 Mine Closure, Remediation, and Reclamation ............................................................................... 426

20.5.1 Reclamation and Closure Costs .......................................................................................... 427

21 Capital and Operating Costs ................................................................................... 429

21.1 Capital Cost Estimates .................................................................................................................... 429

21.1.1 Marmato Upper Zone .......................................................................................................... 429

21.1.2 MDZ ..................................................................................................................................... 431

21.2 Operating Cost Estimates ............................................................................................................... 438

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xii


21.3 Basis for Operating Cost Estimates ................................................................................................ 439

21.3.1 Marmato UZ ........................................................................................................................ 439

21.3.2 MDZ ..................................................................................................................................... 440

22 Economic Analysis .................................................................................................. 442

22.1 External Factors .............................................................................................................................. 442

22.2 Production Assumptions ................................................................................................................. 442

22.3 Taxes, Royalties and Other Interests .............................................................................................. 453

22.4 Results ............................................................................................................................................ 453

22.5 Sensitivity Analysis .......................................................................................................................... 458

23 Adjacent Properties ................................................................................................. 459

24 Other Relevant Data and Information ..................................................................... 460

24.1 Project Execution Plan .................................................................................................................... 460

24.1.1 Project Objectives ............................................................................................................... 460

24.1.2 General Project Description ................................................................................................ 460

24.1.3 Site Preparation and Infrastructure ..................................................................................... 461

24.1.4 Underground Mine and Supporting Infrastructure ............................................................... 463

24.1.5 Process Plant ...................................................................................................................... 463

24.1.6 Project Delivery Approach ................................................................................................... 464

24.1.7 Project Team Organization.................................................................................................. 466

24.1.8 Project Execution Supporting Plans .................................................................................... 469

25 Interpretation and Conclusions .............................................................................. 470

25.1 Property Description and Ownership .............................................................................................. 470

25.2 Geology and Mineralization ............................................................................................................ 470

25.3 Status of Exploration, Development and Operations ...................................................................... 470

25.4 Mineral Processing and Metallurgical Testing ................................................................................ 471

25.5 Mineral Resource Estimate ............................................................................................................. 471

25.6 Mining & Reserves .......................................................................................................................... 472

25.7 Recovery Methods .......................................................................................................................... 476

25.8 Project Infrastructure ....................................................................................................................... 476

25.8.1 Water Supply ....................................................................................................................... 476

25.8.2 Tailings Management Facility .............................................................................................. 477

25.9 Environmental Studies and Permitting ............................................................................................ 477

25.10 Capital and Operating Costs ........................................................................................................... 479

25.11 Economic Analysis .......................................................................................................................... 479

26 Recommendations ................................................................................................... 481

26.1 Recommended Work Programs ...................................................................................................... 481

26.1.1 Property Description and Ownership .................................................................................. 481

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xiii


26.1.2 Geology and Mineral Resources ......................................................................................... 481

26.1.3 Mineral Processing and Metallurgical Testing..................................................................... 482

26.1.4 Mining & Reserves .............................................................................................................. 482

26.1.5 Recovery Methods .............................................................................................................. 484

26.1.6 Project Infrastructure ........................................................................................................... 484

26.1.7 Environmental Studies and Permitting ................................................................................ 485

26.1.8 Capital, Operating Costs and Economic Analysis ............................................................... 486

26.2 Recommended Work Program Costs ............................................................................................. 486

27 References ................................................................................................................ 488

28 Glossary .................................................................................................................... 490

28.1 Mineral Resources .......................................................................................................................... 490

28.2 Mineral Reserves ............................................................................................................................ 490

28.3 Definition of Terms .......................................................................................................................... 491

28.4 Abbreviations .................................................................................................................................. 492

List of Tables

Table 1-1: Caldas Mineral Resource(1) Statement with an Effective Date of March 17, 2020 ........................... 9

Table 1-2: Caldas Mineral Reserve Estimate as of March 17, 2020 – SRK Consulting (U.S.), Inc. ................ 11

Table 1-3: 2015 to 2020* Production ................................................................................................................ 12

Table 1-4: Marmato UZ Sustaining Capital (LoM) ............................................................................................ 23

Table 1-5: Marmato UZ Sustaining Capital (2020 to 2026) (US$) ................................................................... 23

Table 1-6: Marmato UZ Sustaining Capital (2027 to 2034) (US$) ................................................................... 23

Table 1-7: MDZ Construction Capital (US$) ..................................................................................................... 24

Table 1-8: MDZ Sustaining Capital (LoM) ........................................................................................................ 24

Table 1-9: MDZ Sustaining Capital (2023 to 2027) (US$) ................................................................................ 25

Table 1-10: MDZ Sustaining Capital (2028 to 2033) ........................................................................................ 25

Table 1-11: UZ Operating Costs Summary ...................................................................................................... 25

Table 1-12: MDZ Operating Costs Summary ................................................................................................... 26

Table 1-13: Marmato Indicative Economic Results .......................................................................................... 27

Table 1-14: LOM All-in Sustaining Cost Breakdown ........................................................................................ 28

Table 1-15: LoM All-in Sustaining Cost Breakdown ......................................................................................... 36

Table 1-16: Marmato Indicative Economic Results .......................................................................................... 37

Table 2-1: Site Visit Participants ....................................................................................................................... 41

Table 6-1: Ownership History at Marmato ........................................................................................................ 55

Table 6-2: SRK Mineral Resource Statement for the Marmato Project, Dated July 31, 2019*, Within Zona Baja** ................................................................................................................................................... 58

Table 6-3: Gold Production from the Municipality of Marmato 2004 to December 2019 ................................. 59

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xiv


Table 10-1: Summary of Drilling Completed by Company ............................................................................... 93

Table 11-1: Summary Of QA/QC Sample Submissions During 2018 Submissions To SGS And ALS Laboratories ....................................................................................................................................... 108

Table 11-2: Summary Of QA/QC Sample Submissions During 2019 Submissions To SGS And ALS Laboratories ....................................................................................................................................... 108

Table 11-3: Summary Of QA/QC Sample Submissions During 2020 Submissions To SGS Laboratory (Up to SMT20-018) ....................................................................................................................................... 109

Table 11-4: Summary of CRM’s Submitted During Routine Assay Submissions .......................................... 109

Table 11-5: Summary Statistics for Field Duplicates (2019-2020) ................................................................. 115

Table 11-6: Summary Statistics for Coarse Duplicates to SGS and ALS Submissions (Au g/t), 2019-2020 117

Table 11-7: Summary Statistics for Coarse Duplicates to SGS and ALS Submissions (Au g/t), 2019-2020 118

Table 11-8: Summary Statistics for 2019 Reassays Program to SGS vs ALS Submissions (Au g/t) ............ 119

Table 12-1: Comparison of Mine Planned Grades (Assayed at Mine Laboratory) Versus Head-Grades ..... 124

Table 13-1: Drillholes and Intervals for MDZ Metallurgical Composites ................................................. 127

Table 13-2: Head Analyses for MDZ and Marmato Test Composites ............................................................ 128

Table 13-3: Comminution Test Results on MDZ and Marmato Test Samples ........................................ 129

Table 13-4: Whole-Ore Cyanidation Test Results on MDZ Test Composite .................................................. 130

Table 13-5: Summary of Gravity Concentration Testwork on MDZ and Marmato Composites (1) ................. 131

Table 13-6: MDZ Master Composite Gravity Tailing Leach Conditions ......................................................... 132

Table 13-7: Gravity Concentration + Gravity Tailing Cyanidation Test Results ............................................. 132

Table 13-8: Summary of Gravity Concentration + Gravity Tailing Cyanidation (Variability Composites) ...... 134

Table 13-9: Summary of Rougher Flotation Tests on Gravity Tailings from MDZ and Marmato Composites ........................................................................................................................................................... 134

Table 13-10: Summary of Flotation Concentrate Cyanidation Test Results .................................................. 135

Table 13-11: Summary of Cyanide Detoxification Testwork on MDZ Composite Leach Residue ........ 136

Table 13-12: Static Thickener Test Conditions .............................................................................................. 137

Table 13-13: Summary of Dynamic Thickener Test Results .......................................................................... 137

Table 13-14: Results of Rheology Testwork on MDZ Thickener Underflow Sample ..................................... 138

Table 13-15: Drill Holes and Intervals Used for the Low Grade MDZ Composite .......................................... 140

Table 13-16: Drill Holes and Intervals Used for the Medium Grade MDZ Composite ................................... 141

Table 13-17: Drill Holes and Intervals Used for the High Grade MDZ Composite ......................................... 143

Table 13-18: Drill Core Holes and Intervals Used for the MDZ Deep Composite .......................................... 144

Table 13-19: Drill Core Holes and Intervals Used for the Transition Composite............................................ 145

Table 13-20: Drill Core Holes and Intervals Used for Crushing (CWI) Testwork ........................................... 146

Table 13-21: Head Analyses for Key Elements .............................................................................................. 147

Table 13-22: Head Analyses and Multi-Element Scan on Each Test Composite .......................................... 147

Table 13-23: Summary of Comminution Test Results .................................................................................... 149

Table 13-24: Summary of SMC Test Results ................................................................................................. 149

Table 13-25: Summary of SPI Tests .............................................................................................................. 150

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xv


Table 13-26: Summary of Bond Ball Mill Work Index (BWI) Tests ................................................................. 150

Table 13-27: Summary of Bond Low Energy Crushing Tests ........................................................................ 151

Table 13-28: Summary of Abrasion Index Determinations ............................................................................. 151

Table 13-29: Summary of E-GRG Test on MDZ Master Composite .............................................................. 151

Table 13-30: Summary of E-GRG Modeling ................................................................................................... 152

Table 13-31: Summary of Gravity Concentration Testwork ........................................................................... 153

Table 13-32: Summary of Intensive Leach Test on Gravity Concentrate ...................................................... 154

Table 13-33: Summary of Cyanide Leach Test Gold Extractions Versus Grind Size .................................... 155

Table 13-34: Summary of Cyanide Leach Test Silver Extractions Versus Grind Size ................................... 156

Table 13-35: Gold Extraction Versus Cyanide Concentration and Slurry Density ......................................... 157

Table 13-36: Summary Cyanidation Tests with Preaeration and Air Versus Oxygen Injection ..................... 159

Table 13-37: Summary of Cyanide Attenuation and Slurry Density Tests ..................................................... 161

Table 13-38: Summary of Hard Stop Leach Retention Time Tests ............................................................... 163

Table 13-39: Summary of CIL Tests Versus Retention Time with Optimized Leach Conditions ................... 163

Table 13-40: Variability Composites – Gold Recovery Under Optimized Conditions .................................... 164

Table 13-41: Variability Composites – Silver Recovery Under Optimized Conditions ................................... 165

Table 13-42: Modeled Design Parameters for a Multi-stage CIP Adsorption Circuit ..................................... 168

Table 13-43: Modeled Gold Concentrations in Solids, Solution and Carbon in a Multi-Stage CIP Circuit .... 169

Table 13-44: Summary of Cyanide Destruction Tests Conducted on Master Composite Leach Residues ... 170

Table 13-45: Summary of High Rate Thickening Test on MDZ Master Composite Leached Tailing ............. 171

Table 13-46: Summary of High Rate Thickening Test on MDZ Transition Composite Leached Tailing ........ 171

Table 13-47: Pressure Filtration Test Results on the Master Composite Tailing Sample .............................. 172

Table 13-48: Pressure Filtration Test Results on the Transition Composite Tailing Sample ......................... 173

Table 13-49: Vacuum Filtration Test Results on the Master Composite Tailing Sample ............................... 173

Table 13-50: Vacuum Filtration Test Results on the Transition Composite Tailing Sample .......................... 173

Table 13-51: Estimated Gold and Silver Recoveries from the MDZ (PFS and PEA Metallurgical Programs) ........................................................................................................................................................... 174

Table 14-1: Summary of Number of Records for Each Exported .csv ........................................................... 178

Table 14-2: Summary of Geological Database Information for Drilling Reported by Company ..................... 178

Table 14-3: Summary of Geological Database Information for Channel Reported by Company .................. 179

Table 14-4: Summary of Leapfrog 0.7 g/t Indictor Grade Shell, ISO Value Sensitivity Study (MDZ Material) ........................................................................................................................................................... 191

Table 14-5: Summary of Domain Coding Used in the 2017 Mineral Resource Estimate .............................. 192

Table 14-6: Summary Raw Sample Statistics Based on Defined Geological Domains (Group) ................... 193

Table 14-7: Summary of Capping Sensitivity – MDZ Domain (Group 5000), Selected Capping Highlighted in Orange ............................................................................................................................................... 201

Table 14-8: Summary of Capping Sensitivity – MDZ Domain (Group 5000), selected capping highlighted in orange ................................................................................................................................................ 205

Table 14-9: Comparison Raw vs Composite Statistics .................................................................................. 207

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xvi


Table 14-10: Comparison Statistics ................................................................................................................ 211

Table 14-11: Summary of Density Statistics by Rock Type and Selected Density ........................................ 212

Table 14-12: Density assigned per rocktype in 2020 Mineral Resources ...................................................... 214

Table 14-13: Summary of Variogram Parameters per Group ........................................................................ 220

Table 14-14: Block Model Prototype (DatamineTM format) ............................................................................. 221

Table 14-15: Summary of Key Fields in Block Model ..................................................................................... 222

Table 14-16: Summary of Datamine Estimates by Search Volume and Validation by Estimation Type (OK, ID, NN)..................................................................................................................................................... 225

Table 14-17: Summary of Domains with Top Capping and Sliding Thresholds for Wider Search Volumes . 227

Table 14-18: Summary of Estimation Search Parameters Used in Estimation .............................................. 229

Table 14-19: Summary of Statistical Validation of Raw, Declustered, OK, ID2 and NN Block Estimates ..... 236

Table 14-20: Summary of CoG Assumptions at Marmato Based on Assumed Costs (Averaged for All Mining Styles) ................................................................................................................................................ 251

Table 14-21: Caldas Mineral Resource(1) Statement with Effective Date of March 17, 2020 ........................ 253

Table 14-22: Grade Tonnage Curve Measured and Indicated - Vein Domains (Group 1000 to 3000) ......... 255

Table 14-23: Grade Tonnage Curve Measured and Indicated - Porphyry Domain (Group 4000) ................. 255

Table 14-24: Grade Tonnage Curve Measured and Indicated - MDZ Domain (Group 5000) ........................ 256

Table 14-25: Grade Tonnage Curve Inferred - Vein Domains (Group 1000 - 3000) ..................................... 256

Table 14-26: Grade Tonnage Curve Inferred - Porphyry Domain (Group 4000) ........................................... 257

Table 14-27: Grade Tonnage Curve Inferred - MDZ Domain (Group 5000) .................................................. 257

Table 15-1: Dilution Assumption ..................................................................................................................... 263

Table 15-2: Mining Extraction/Recovery Assumptions ................................................................................... 263

Table 15-3: Cut-off Grade Parameters for Veins Material .............................................................................. 264

Table 15-4: Cut-off Grade Parameters for Transition Material ....................................................................... 264

Table 15-5: Dilution Assumptions ................................................................................................................... 265

Table 15-6: Additional Ramp Allowance Factors ........................................................................................... 266

Table 15-7: MDZ Underground Cut-off Grade Calculation ............................................................................. 267

Table 15-8: Caldas Mineral Reserve Estimate as of March 17, 2020 – SRK Consulting (U.S.), Inc. ............ 268

Table 16-1: 2015 to 2020* Production ............................................................................................................ 270

Table 16-2: Level Elevations and Description ................................................................................................ 273

Table 16-3: Summary of Structural Sets ........................................................................................................ 278

Table 16-4: Uniaxial Compressive Strength Test Results .............................................................................. 280

Table 16-5: Ground Support Requirements ................................................................................................... 281

Table 16-6: FS Geotechnical Drilling Plan (Tunnel Investigation) .................................................................. 283

Table 16-7: Measured Bedrock Hydraulic Conductivity Values at Depth ....................................................... 287

Table 16-8: Simulated Hydraulic Parameters ................................................................................................. 294

Table 16-9: Predictive Scenarios Evaluated by Groundwater Model ............................................................. 297

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xvii


Table 16-10: Predicted Maximum Mine Inflows and Reduction of Groundwater Discharge to the Rivers and Creeks Under Different Scenarios ..................................................................................................... 301

Table 16-11: Productivity Rates ..................................................................................................................... 305

Table 16-12: Marmato Upper Mine Total Production Schedule ..................................................................... 307

Table 16-13: Marmato Upper Mine Total Development Schedule ................................................................. 308

Table 16-14: Manpower by Department ......................................................................................................... 314

Table 16-15: Marmato Equipment List ........................................................................................................... 315

Table 16-16: Undiluted Stope Optimization Results for Varying Cut-off Grades ........................................... 318

Table 16-17: MDZ Mine Design Summary – by Activity Type ........................................................................ 325

Table 16-18: Productivity Rates ..................................................................................................................... 327

Table 16-19: Schedule Parameters for Underground Mining ......................................................................... 327

Table 16-20: Material Characteristics for Ore and Waste .............................................................................. 327

Table 16-21: Main Ramp Average Development Rate – Long Term Development Openings ...................... 328

Table 16-22: Footwall Access Development Rate – Medium Term Openings* ............................................. 329

Table 16-23: Drift Access Development Rate – Short Term Openings* ........................................................ 329

Table 16-24: Stope Production Rate .............................................................................................................. 330

Table 16-25: MDZ Production Schedule ........................................................................................................ 331

Table 16-26: Truck Hauling Speeds ............................................................................................................... 334

Table 16-27: Backfill Volume Summary – By Type ................................................................................... 336

Table 16-28: Recommended Maximum Air Velocities for Various Airway Types .......................................... 339

Table 16-29: Equipment List and Airflow Requirement .................................................................................. 339

Table 16-30: Auxiliary Ventilation Fan Summary ........................................................................................... 341

Table 16-31: Fan Operating Points* ............................................................................................................... 342

Table 16-32: MDZ Shift Schedule and Rotation ............................................................................................. 345

Table 16-33: MDZ Mining Labor Summary .................................................................................................... 345

Table 16-34: MDZ Mining Labor ..................................................................................................................... 346

Table 16-35: Mine Equipment by Period ........................................................................................................ 348

Table 17-1: Equipment List for Marmato Process Plant ................................................................................. 354

Table 17-2: Marmato Process Plant Consumables ................................................................................... 356

Table 17-3: Summary of Marmato Process Plant Operating Performance and Recovery Estimate ............. 357

Table 17-4: Marmato Process Plant Operating Costs: 2019 - 2020 (Jan-Apr) .............................................. 357

Table 17-5: Summary of Marmato Process Plant Expansion Capex ............................................................. 359

Table 17-6: Process Design Criteria Summary .............................................................................................. 361

Table 17-7: Operating Cost Summary ............................................................................................................ 373

Table 17-8: Operations and Maintenance Manpower Schedule .................................................................... 374

Table 17-9: Light Vehicles and Mobile Equipment Summary ......................................................................... 375

Table 18-1: DSTF Design Criteria .................................................................................................................. 399

Table 18-2: Stormwater Diversion Channel Summary ................................................................................... 403

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xviii


Table 19-1: Marmato Price Assumptions ....................................................................................................... 409

Table 19-2: Marmato Net Smelter Return Terms ........................................................................................... 409

Table 20-1: Water Discharges ........................................................................................................................ 415

Table 20-2: Stationary Emission Sources ...................................................................................................... 415

Table 20-3: Environmental Procedures .......................................................................................................... 416

Table 20-4: Surface Water Concessions ........................................................................................................ 420

Table 21-1: Marmato UZ Sustaining Capital (LoM) ........................................................................................ 429

Table 21-2: Marmato UZ Sustaining Capital (2020 to 2026) (US$) ............................................................... 430

Table 21-3: Marmato UZ Sustaining Capital (2027 to 2034) (US$) ............................................................... 430

Table 21-4: Marmato UZ Capital Development Unit Costs ............................................................................ 431

Table 21-5: Marmato UZ Capital Development Meters (2020 to 2023) ......................................................... 431

Table 21-6: MDZ Construction Capital (US$) ................................................................................................. 432

Table 21-7: MDZ Pre-Production Development Unit Costs ............................................................................ 433

Table 21-8: MDZ Pre-Production Development Meters ................................................................................. 433

Table 21-9: MDZ Processing Plant and Infrastructure Capital ....................................................................... 434

Table 21-10: MDZ Sustaining Capital (LoM) .................................................................................................. 435

Table 21-11: MDZ Sustaining Capital (2023 to 2027) (US$) .......................................................................... 435

Table 21-12: MDZ Sustaining Capital (2028 to 2033) (US$) .......................................................................... 435

Table 21-13: MDZ Development Sustaining Capital Unit Costs .................................................................... 436

Table 21-14: MDZ Development Sustaining Capital Meters (2023 to 2027) (US$) ....................................... 436

Table 21-15: MDZ Development Sustaining Capital Meters (2028 to 2032) .................................................. 436

Table 21-16: MDZ Mining Sustaining Capital (2023 to 2027) (US$) .............................................................. 437

Table 21-17: MDZ Mining Sustaining Capital (2028 to 2032) (US$) .............................................................. 437

Table 21-18: MDZ DSTF Sustaining Capital (2023 to 2027) (US$) ............................................................... 437

Table 21-19: MDZ DSTF Sustaining Capital (2028 to 2032) .......................................................................... 438

Table 21-20: MDZ Rio Sucio Power Line Sustaining Capital (2028 to 2032) ................................................ 438

Table 21-21: Marmato UZ Operating Costs Summary ................................................................................... 438

Table 21-22: Marmato MDZ Operating Costs Summary ................................................................................ 438

Table 21-23: Marmato UZ Operating Development Unit Costs ...................................................................... 439

Table 21-24: Marmato UZ Operating Development Meters (2020 to 2024) ................................................... 439

Table 21-25: Marmato UZ Mineral Processing Operating Costs .................................................................... 440

Table 21-26: Marmato UZ TSF And G&A Operating Costs (2020 to 2026) (US$) ........................................ 440

Table 21-27: Marmato UZ DSTF And G&A Operating Costs (2027 to 2033) (US$) ...................................... 440

Table 21-28: MDZ Mining Operating Costs (2023 to 2027) (US$) ................................................................. 440

Table 21-29: MDZ Mining Operating Costs (2028 to 2033) ........................................................................... 441

Table 21-30: MDZ Mineral Processing Cost ................................................................................................... 441

Table 21-31: MDZ DSTF And G&A Operating Costs (2023 to 2027) ............................................................ 441

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xix


Table 21-32: MDZ DSTF And G&A Operating Costs (2028 to 2033) ............................................................ 441

Table 22-1: Marmato Price Assumptions ....................................................................................................... 442

Table 22-2: Marmato NSR Terms .................................................................................................................. 442

Table 22-3: Marmato Production Summary .................................................................................................... 443

Table 22-4: Marmato Yearly (2020 to 2026) Mine Production Assumptions .................................................. 445

Table 22-5: Marmato Yearly (2027 to 2033) Mine Production Assumptions .................................................. 446

Table 22-6: Marmato Mill Production Assumptions ........................................................................................ 449

Table 22-7: Marmato Mill Production Schedule (2020 - 2026) ....................................................................... 450

Table 22-8: Marmato Mill Production Schedule (2027 - 2033) ....................................................................... 451

Table 22-9: Marmato UZ LoM Cash Flow Metrics .......................................................................................... 454

Table 22-10: MDZ LoM Cash Flow Metrics .................................................................................................... 455

Table 22-11: Marmato Indicative Economic Results (Combined UZ and MDZ) ............................................ 457

Table 22-12: Marmato LoM Annual Production and Revenues ..................................................................... 457

Table 22-13: LoM All-in Sustaining Cost Breakdown ..................................................................................... 458

Table 24-1: Project Work Breakdown Structure – Level 1 and Level 2 .......................................................... 466

Table 24-2: Owner’s Project Team with EPCM Supplements ........................................................................ 468

Table 25-1: LoM All-in Sustaining Cost Breakdown ....................................................................................... 479

Table 25-2: Marmato Indicative Economic Results ........................................................................................ 480

Table 26-1: Summary of Costs for Recommended Work ............................................................................... 487

Table 28-1: Definition of Terms ...................................................................................................................... 491

Table 28-2: Abbreviations ............................................................................................................................... 492

List of Figures Figure 1-1: Cross-Section Showing License Splits at Marmato ......................................................................... 8

Figure 1-2: UZ Production Schedule Colored by Time Period ......................................................................... 14

Figure 1-3: MDZ Mine Production Schedule Colored by Year ......................................................................... 15

Figure 1-4: Combined UZ and MDZ Mining Profile – Tonnes and Grade ........................................................ 16

Figure 1-5: Marmato After-Tax Free Cash Flow, Capital and Metal Production .............................................. 26

Figure 1-6: Marmato Operating Cost Break-Down ........................................................................................... 27

Figure 4-1: Location Map .................................................................................................................................. 45

Figure 4-2: Land Tenure Map(s) ....................................................................................................................... 46

Figure 4-3: Summary of Gap in Licenses Within the Current Operations, with Associated Applications ........ 48

Figure 5-1: Marmato Project, Looking Northwest Towards Cerro El Burro ...................................................... 52

Figure 7-1: Regional Geology Map ................................................................................................................... 61

Figure 7-2: Local Geology Map ........................................................................................................................ 62

Figure 7-3: Regional Geology with Gold Prospects in the Marmato Area ........................................................ 63

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xx


Figure 7-4: Photographs of the Different Porphyritic Intrusive Bodies that Make Up the Porphyry Stock Of Marmato ............................................................................................................................................... 67

Figure 7-5: Property Geology Map ................................................................................................................... 68

Figure 7-6: Cross-Section of the Marmato Gold Deposit Looking NW Showing the Intrusions P1 to P5 ........ 69

Figure 7-7: TCL Interpretation of Vein Orientations at Marmato ...................................................................... 71

Figure 7-8: Types of Alteration Found at Marmato ........................................................................................... 73

Figure 7-9: Structural Features Expected in a North-South Sinistral Riedel Fault System .............................. 75

Figure 7-10: Example of Epithermal Veins as Viewed in the Drilling Core at Marmato ................................... 76

Figure 7-11: Examples from Drill Core of the Different Mineralization Styles .................................................. 77

Figure 7-12: MDZ Mineralization Showing Veinlets Including Visible Gold (Au). BHID MND282-03-17 at a Depth of 1,010 masl, Sample of 1.20 m with 18.06 g/t Au and 2.5 g/t Ag ........................................... 78

Figure 9-1: Development of 3D Topography for the Project Showing LIDAR Survey Points, Shadow Model and 3D View of 1 m Resolution LIDAR Datapoints .................................................................................... 83

Figure 9-2: Example of Level Plan from CGM (Level 20) ................................................................................. 84

Figure 9-3: Channel Sample Marks in Marmato ............................................................................................... 85

Figure 9-4: Underground Workings Survey Using Total Station ...................................................................... 86

Figure 9-5: Sample Collection and Packing ..................................................................................................... 86

Figure 9-6: Distribution of Channel Sampling Along the Vein .......................................................................... 87

Figure 9-7: Channel Sample Cut Using Electrical Saw .................................................................................... 88

Figure 9-8: Identification Ticket and Bags Used to Pack the Channel Sample ................................................ 89

Figure 9-9: 2D Plan View of Sampling Data Versus Vein Interpretations, Showing New Sample Data Highlighted in Red, Versus Plan Section of Veins in Blue (Level 1250 M) ......................................... 92

Figure 10-1: Location Map Showing Drillholes Completed at Marmato by Company ...................................... 94

Figure 10-2: 3D View of Sampling Data, Showing New Exploration Drilling Data Highlighted in Red and Mine Drilling in Purple (Looking North) ......................................................................................................... 95

Figure 10-3: Core Photographs Before and After Making the Respective Cut and Sampling ......................... 96

Figure 10-4: Core Storage Facility at Marmato Constructed in 2010 and Current Status 2019 ...................... 97

Figure 10-5: Plan Showing Primary Drilling Orientation to the South and Southwest Relative to the Main Mineralization Orientation at Depth ..................................................................................................... 99

Figure 10-6: Cross Section (Orientated Looking Northeast), Showing Orientation of Drilling Relative to the Deep Mineralization, and Horizontal Drilling in the Current Operation .............................................. 100

Figure 11-1: Sample Preparation at Mine Laboratory Showing New Equipment (Crusher and Pulverizer) .. 104

Figure 11-2: Sample Preparation Facilities at ACME Laboratories in Medellín ............................................. 105

Figure 11-3: Summary of CRM Submissions to SGS In 2019/2020 Program ............................................... 110

Figure 11-4: Summary of CRM Submissions to ALS In 2019 Program ......................................................... 110

Figure 11-5: Example of Timeline Review Of CRM G914-6 (2019) and G315-2 (2020) Submissions .......... 111

Figure 11-6: SGS and ALS Coarse Blank Submissions 2019 ........................................................................ 112

Figure 11-7: SGS Coarse Blank Submissions 2020 ...................................................................................... 113

Figure 11-8: SGS and ALS Fine Blank Submissions 2019 ............................................................................ 114

Figure 11-9: SGS Fine Blank Submissions 2020 ........................................................................................... 115

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xxi


Figure 11-10: Summary of Field Duplicate 2019-2020 ................................................................................... 116

Figure 11-11: Summary of Coarse Duplicate Submissions to SGS (left) and ALS (right) for 2019-2020 ...... 117

Figure 11-12: Summary of Pulp Duplicate Submissions to SGS (left) and ALS (right), 2019 - 2020 ............ 118

Figure 11-13: Summary of 2019-2020 Reassay (Secondary Laboratory) ..................................................... 119

Figure 11-14: Summary of Check Assays Completed on Pulp Material (Quarterly Checks), Scatter Plot (Left) and Mean vs Relative Difference Plot (Right) ................................................................................... 120

Figure 11-15: Summary of Check Assays Completed on Reject Material (Quarterly Checks), Scatter Plot (Left) and Mean vs Relative Difference Plot (Right) ................................................................................... 120

Figure 12-1: Comparison of Planned Versus Actual Gold Grades at Marmato Mine..................................... 124

Figure 13-1: Drillhole Locations ...................................................................................................................... 128

Figure 13-2: Gold Extraction Versus Retention Time (MDZ Master Comp Gravity Tailings) ......................... 133

Figure 13-3: Yield Stress Versus Thickener Underflow Slurry Density .......................................................... 139

Figure 13-4: Drill Hole Locations Used for Metallurgical Composites ............................................................ 146

Figure 13-5: Gold Extraction Versus Leach Retention Time for Air and Oxygen Injection ............................ 160

Figure 13-6: Gold Extraction Versus Leach Retention Time .......................................................................... 162

Figure 14-1: Fault Network Compared to Mapping on Level 20 (1056 RL) ................................................... 181

Figure 14-2: Cross Section Showing SRK Revised Lithological Model ......................................................... 183

Figure 14-3: Level 20 Geological Mapping Versus Sampling Database and Veins Model, Showing the Level of Information Integrated into the Geological Model .............................................................................. 185

Figure 14-4: Level Plan (1065 RL), Showing Interaction Between Vein and Disseminated Vein Domains ... 186

Figure 14-5: Development of Porphyry Pockets Wireframe Methodology ..................................................... 188

Figure 14-6: Development of MDZ model ...................................................................................................... 190

Figure 14-7: Box Plot Showing Raw Sample Statistics Based on Defined Geological Domains (Group) ..... 192

Figure 14-8: Disintegration Analysis Au (g/t) – Veins, Group 1000 (Vein_N<9000) ...................................... 194

Figure 14-9: Disintegration Analysis Au (g/t) – Veins, Group 1000 (Vein_N>9000) ...................................... 195

Figure 14-10: Disintegration Analysis Au (g/t) – Disseminated Vein, Group 2000 (Vein_N<9000) ............... 196

Figure 14-11: Disintegration Analysis Au (g/t) – Disseminated Vein, Group 2000 (Vein_N>9000) ............... 197

Figure 14-12: Disintegration Analysis Au (g/t) – Splays, (Group 3000) ......................................................... 198

Figure 14-13: Percentile Analysis Au (g/t) – Porphyry Domain, (Group 4000) .............................................. 200

Figure 14-14: Percentile Analysis Au (g/t) – MDZ Domain, (Group 5000) ..................................................... 204

Figure 14-15: Summary Histograms and Cumulative Frequency of Raw Sample Lengths per Domain ....... 209

Figure 14-16: Log Probability Plot of Density Measurements Logged as Vein .............................................. 213

Figure 14-17: Omni Directional Variograms Defined for Au and Ag for Domains Group = 1000 – 4000 ....... 217

Figure 14-18: Group 5000 Au Directional Semi-Variograms .......................................................................... 218

Figure 14-19: Group 5000 Ag Directional Semi-Variograms .......................................................................... 219

Figure 14-20: Group 5000 (MDZ), KNA Analysis Using Snowden Supervisor .............................................. 223

Figure 14-21: Example of Default Search Orientations Used Within the MDZ High-Grade Domain ............. 226

Figure 14-22: Visual Validation of Selected Veins at Marmato ...................................................................... 232

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xxii


Figure 14-23: Visual Validation of Selected Disseminated Veins at Marmato ............................................... 233

Figure 14-24: Section and Level Plan Example of Visual Validation of MDZ (Group 1000) .......................... 234

Figure 14-25: Swath Analysis Group 1000 Au (g/t) ........................................................................................ 241

Figure 14-26:Swath Analysis Group 2000 Au (g/t) ......................................................................................... 242



Figure 14-29: Swath Analysis Group 5000 – INDZONE=0 (LG) Au (g/t) ....................................................... 245

Figure 14-30: Swath Analysis Group 5000 – INDZONE=1 (MG) Au (g/t) ...................................................... 246

Figure 14-31: Swath Analysis Group 5000 – INDZONE=2 (HG) Au (g/t) ....................................................... 247

Figure 14-32: Final Classification for the Marmato Project (Looking Northwest Bearing 305) ...................... 249

Figure 14-33: Final Classification for License #014-89m Marmato Project (Looking Northwest Bearing 305) ........................................................................................................................................................... 250

Figure 14-34: Cross-Section Showing License Splits at Marmato ................................................................. 252

Figure 14-35: Grade Tonnage Curves Showing Sensitivity to Changes in Cut-Off for Measured and Indicated Mineralized Material ........................................................................................................................... 258

Figure 14-36: Grade Tonnage Curves Showing Sensitivity to Changes in Cut-Off for Inferred Mineralized Material .............................................................................................................................................. 259

Figure 15-1: Marmato General Layout ........................................................................................................... 262

Figure 15-2: Veins Dilution ............................................................................................................................. 263

Figure 15-3: UZ Grade/Tonne Curve Based on Au Cut-Off ........................................................................... 265

Figure 15-4: MDZ Grade/Tonne Curve Based on Au Cut-Off ........................................................................ 267

Figure 15-5: License Gap ............................................................................................................................... 269

Figure 16-1: Typical Shrinkage Stoping Diagram ........................................................................................... 271

Figure 16-2: Conventional Cut and Fill Method Diagram ............................................................................... 272

Figure 16-3: Marmato Zona Baja Cross Section Looking NE with Active Levels ........................................... 273

Figure 16-4: Marmato Level 18 with Main Haulage (Mined Out Panels in Cyan) .......................................... 274

Figure 16-5: UZ Planned Dilution ................................................................................................................... 275

Figure 16-6: Location of Geotechnical Drill Holes (As-Builts and MDZ Design Shown) ................................ 276

Figure 16-7: Geotechnical Subdomains ......................................................................................................... 277

Figure 16-8: Structural Domains ..................................................................................................................... 278

Figure 16-9: Conveyor Tunnel Trajectory ....................................................................................................... 283

Figure 16-10: FS Drill Hole Location (Tunnel Investigation) ........................................................................... 284

Figure 16-11: Distribution of Measured Hydraulic Conductivity Values vs. Depth ......................................... 287

Figure 16-12: Estimated Water Table and Direction of Groundwater Flow .................................................... 289

Figure 16-13: Conceptual Hydrogeological Cross-Section ............................................................................ 290

Figure 16-14: Scheme of Current Dewatering System ................................................................................... 291

Figure 16-15: Measured Mine Water Discharge ............................................................................................ 292

Figure 16-16: Model Grid Discretization – Plan View ..................................................................................... 293

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xxiii


Figure 16-17: Modeled Cross-Section ............................................................................................................ 294

Figure 16-18: Simulated Rivers, Creeks and Groundwater Outflows ............................................................ 295

Figure 16-19: Simulated Developments and Stopes for Planned Mine ......................................................... 296

Figure 16-20: Predicted Mine Inflow During Years 2021 to 2032 (Base Case) ............................................. 298

Figure 16-21: Comparison of Total Predicted Dewatering Requirements for Base Case and Sensitivity Scenarios ........................................................................................................................................... 298

Figure 16-22: Predicted Water Table and Direction of Groundwater Flow at End of Mining Shown on West to East Cross-Section through Mine Area ............................................................................................. 299

Figure 16-23: Predicted Drawdown at End of Mining (Base Case, End of 2032) .......................................... 300

Figure 16-24: Stope Optimization results for the Veins (Section looking Northeast) ..................................... 302

Figure 16-25: Stope Optimization Results for the Transition (Section Looking Northeast) ............................ 303

Figure 16-26: Stope Optimization Results for the UZ Colored by Au Grade (Section Looking Northeast) .... 303

Figure 16-27: Transition Mining Method (Magenta is Primary and Cyan is Secondary) ................................ 304

Figure 16-28: Plan View of Transition Development ...................................................................................... 305

Figure 16-29: Production Schedule Colored by Time Period ......................................................................... 310

Figure 16-30: Marmato Hydraulic Backfill System ......................................................................................... 312

Figure 16-31: Level 21 Exhaust Collection Area ............................................................................................ 313

Figure 16-32: New Exhaust System ............................................................................................................... 313

Figure 16-33: Resource Model – AU blocks (g/t) above Mining CoG (Looking North) .................................. 317

Figure 16-34: Undiluted Stope Optimization Results for Varying Cut-off Grades .......................................... 318

Figure 16-35: Stope Cross Section ................................................................................................................ 319

Figure 16-36: Typical Level Section ............................................................................................................... 320

Figure 16-37: MDZ Mine Design (Rotated View Looking Southwest) ............................................................ 321

Figure 16-38: MDZ Main Decline Cross Section ............................................................................................ 322

Figure 16-39: MDZ Mine Design (Rotated View Looking Southwest) ............................................................ 323

Figure 16-40: MDZ Mine Design, Colored by Au Grade ................................................................................ 324

Figure 16-41: MDZ Design with Sur and Ines Faults (Rotated View - Looking Northwest) ........................... 326

Figure 16-42: Mine Production Schedule Colored by Year ............................................................................ 332

Figure 16-43: Haulage Distance – One Way Length ...................................................................................... 335

Figure 16-44: Haulage Cycle Time - Roundtrip .............................................................................................. 336

Figure 16-45: Marmato Project General Ventilation Scheme ......................................................................... 341

Figure 16-46: Estimated Fan Power Demand ................................................................................................ 342

Figure 16-47: Combined UZ and MDZ Mining Profile – Tonnes and Grade .................................................. 350

Figure 16-48: Combined UZ and MDZ Mining Profile - Contained Metal ....................................................... 351

Figure 17-1: Marmato Process Flowsheet.................................................................................................. 353

Figure 17-2: Proposed MDZ Process Flow Sheet .......................................................................................... 360

Figure 17-3: Mill/Leach/Reagents General Arrangement ............................................................................... 363

Figure 17-4: Primary/Secondary Crushing and Stockpile .............................................................................. 364

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xxiv


Figure 17-5: Grinding and Gravity Concentrate and Intensive Leach Area ................................................... 365

Figure 17-6: Pre-Aeration and Leach Tanks .................................................................................................. 366

Figure 17-7: CIP and Detoxification Tanks, Acid Wash and Elution Columns, Regeneration Kiln, and Gold Room Areas ....................................................................................................................................... 367

Figure 17-8: Tailings Thickener and Thickener Underflow Storage Tank Area ............................................. 369

Figure 17-9: Tailings Pressure Filter Plant Area ............................................................................................ 370

Figure 18-1: Marmato Project Location .......................................................................................................... 376

Figure 18-2: Marmato General Access and Major Facilities .......................................................................... 378

Figure 18-3: Marmato Existing Project Site Map ............................................................................................ 379

Figure 18-4: Marmato Electrical System Schematic ...................................................................................... 381

Figure 18-5: Overall Site Plan ........................................................................................................................ 383

Figure 18-6: Paste Plant General Arrangement ............................................................................................. 391

Figure 18-7: Paste Plant 3D Layout ............................................................................................................... 392

Figure 18-8: Makeup and Demand at the Upper Zone Process Plant ........................................................... 393

Figure 18-9: Makeup and Demand at the MDZ Process Plant ...................................................................... 394

Figure 18-10: Proposed Configurations of Cascabels 1 and 2....................................................................... 396

Figure 18-11: Potential DSTF Sites Identified Through Siting Study ............................................................. 397

Figure 18-12: Internal Drainage System ......................................................................................................... 401

Figure 18-13: Pre-Cast Concrete Span Channel Crossing ............................................................................ 402

Figure 18-14: Watersheds for DSTF 1 (Top) and DSTF 2 (Bottom) .............................................................. 404

Figure 18-15: DSTF Haul and Access Roads ................................................................................................ 405

Figure 22-1: Marmato UZ Mine Production Profile ......................................................................................... 447

Figure 22-2: MDZ Mine Production Profile ..................................................................................................... 447

Figure 22-3: Marmato Combined UZ and MDZ Mine Production Profile ....................................................... 448

Figure 22-4: Marmato UZ Processing Production Profile ............................................................................... 452

Figure 22-5: MDZ Processing Production Profile ........................................................................................... 452

Figure 22-6: Marmato Processing Production Profile ..................................................................................... 453

Figure 22-7: Marmato UZ Cash Flow Profile .................................................................................................. 454

Figure 22-8: MDZ Cash Flow Profile .............................................................................................................. 455

Figure 22-9: Marmato After-Tax Free Cash Flow, Capital and Metal Production .......................................... 456

Figure 22-10: Marmato Operating Cost Break-Down (Combined UZ and MDZ) ........................................... 456

Figure 22-11: Marmato NPV Sensitivity ......................................................................................................... 458

Figure 24-1: Project Execution Schedule ....................................................................................................... 465

Figure 24-2: MDZ Project Team ..................................................................................................................... 467

Figure 26-1: 2020 Exploration Plan Showing Phases A through C (Left to Right) ......................................... 482

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page xxv


Appendices Appendix A: Certificates of Qualified Persons

Appendix B: MDZ Tailings Drawings

Appendix C: Economic Model Snapshots

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Project Page 1


1 Summary This report was prepared as a Pre-Feasibility Study (PFS) level Canadian National Instrument 43-101

(NI 43-101) Technical Report (Technical Report) for Caldas Gold Corp. (Caldas Gold) in respect of the

Marmato Project (Marmato Project) owned by Caldas Gold Marmato S.A.S. (CGM or the Company),

an indirect, wholly-owned subsidiary of Caldas Gold, by SRK Consulting (U.S.), Inc. (SRK).

1.1 Property Description and Ownership

The Marmato Project is located between latitudes and longitudes 5°28’24”N and 5°28’55”N, and

75°34’46”W and 75°37’80”W, respectively; with altitudes ranging from approximately 200 to 1,705

meters (m). What has been traditionally termed the Marmato Project was made up of three separate

areas within the historic Marmato mining district named Zona Alta (License #CHG_081), Zona Baja

(License #014-89m) and Echandia (License #RPP_357), of which Zona Baja is 100% owned by CGM

and Zona Alta and Echandia are owned indirectly, through other subsidiaries, by Gran Colombia Gold

Corp. (Gran Colombia). CGM is currently in the process of extending the duration of the Zona Baja

mining contract for which the current 30-year term expires in October 2021.

Notwithstanding the historical designation of the Marmato Project described above, in this report the

“Marmato Project” or “Project” refers to the mining assets (CGM Mining Assets) principally comprising

the existing producing underground gold mine (#014-89m), the existing 1,200 tonnes per day (t/d)

processing plant defined in this report as the Upper Zone, and the area encompassing the Marmato

Deep Zone (MDZ) mineralization, all located within the mining license area referred to as Zona Baja.

The CGM Mining Assets also include two contractual rights:

• One, granted by Minera Croesus, S.A.S. (Croesus), an indirect, wholly owned subsidiary of

Gran Colombia, to mine in the lower portion of the Echandia license (#RPP_357) area

• A second license in the process of being completed, to be granted by Minerales Andinos de

Occidente S.A.S. (MAO), an indirect, wholly owned subsidiary of Gran Colombia, to mine

portions of levels 16 and 17 of Zona Alta (License #CHG_081); this license represents a small

potential upside to add additional material via access from the current mine. This material is

currently excluded from the Mineral Resource Statement and mine plan.

SRK noted within the transfer of licenses from the previous owner a gap between the existing licenses

for Zona Baja (#014-89m) and Echandia (#RPP_357), and CGM applied to the Colombian government

for formal approval to continue mining in the identified gap. SRK has reviewed the application within

the government website and noted that the status is defined as “in progress”, which has been the

reported status since September 30, 2009. SRK understands that at the end of the pre-feasibility study

process (May 2020) the issue was resolved with the government determining that there is no gap and

that the area falls within the license for Zona Baja (#014-89m). As the license gap is no longer an

issue, there may be additional optimization opportunities for the Marmato Project that should be

explored during the next phase of work.

1.2 Geology and Mineralization

The local geology is dominated by porphyritic dacitic and andesitic intrusions, which host the

mineralization at Marmato. The intrusions are characterized by quartz, hornblende, biotite and zoned



plagioclase phenocrysts in a finely crystalline quartz-plagioclase groundmass, with variations in

phenocryst proportion and sizes between intrusions. A total of five different porphyry units have been

identified.

The Marmato gold deposit consists of a structurally controlled epithermal vein system with a mineral

assemblage dominated by pyrite, arsenopyrite, black iron (Fe) rich sphalerite, pyrrhotite, chalcopyrite

and electrum in the Upper Zone (UZ), and a mesothermal veinlet system with a mineral assemblage

dominated by pyrrhotite, chalcopyrite, bismuth minerals and visible gold in the MDZ.

The mineralization in the current mine consists of three distinct phases, a first phase characterized by

the mesothermal vein/veinlet mineralization, which defines the MDZ, followed by an epithermal low

sulfidation style, superimposed by an epithermal intermediate sulfidation phase. Gold‐silver

mineralization is mainly hosted by a pyrite+sphalerite vein to veinlet system fitting in a sinistral

transpressional shearing system, associated with intermediate argillic alteration within the host

porphyryitic rocks. Approximately 92% of the gold/silver-bearing particles are intergrown with sulfides

or occur at sulfide gangue grain boundaries. Current mining in the area is via narrow underground

stoping of the higher-grade vein mineralization.

The MDZ mineralization consists of a network of thin, less than 5 centimeters (cm), sulfide veinlets,

mainly pyrrhotite+chalcopyrite, hosted in weak argillic and deeper potassic alteration which is related

to a previous event and rimmed by a thin sodium-calcitic alteration halo, which is related to the

mineralization. Recent geological reports on MDZ (Sillitoe, 2019) concluded:

• Gold grade distribution in the Zona Baja (MDZ) mineralized orebody is unrelated to the

presence of distinct porphyry phases and is entirely dependent on the intensity of structurally

localized veinlets

• Potassic alteration, represented chiefly by biotite, is progressively better preserved at depth in

the Zona Baja, raising the possibility that early potassic alteration could also be gold bearing,

but further work is required to confirm this theory

• Gold distribution appears to be exclusively a product of veinlet intensity and orientation related

to structural controls during orogenesis. The veinlets responsible for much of the Zona Baja

gold are those containing quartz, pyrrhotite and traces of chalcopyrite and having prominent

albite alteration halos

• The presence of visible gold is also noted in the core and, as expected, relates to increased

assay values when present

Mineralization occurs in parallel, sheeted and anastomosing veins (vein domain), all of which follow a

regional structural control, with minor veins forming splays of the main structures (splays) which often

have limited strike or dip extent. The upper vein domain intersects broader zones of intense veinlet

mineralization (termed porphyry domain in this Technical Report) that is hosted by a lower grade

mineralized porphyry stock. In addition, a discrete, relatively high-grade core (feeder zone) to the main

deeper mineralization termed locally as the MDZ.

The upper portion of the MDZ has been exposed in Level 21 of the existing Caldas mining operations,

while deeper sections have been observed in drillcore, both of which have been confirmed as different

styles of mineralization. The lowest levels of the mine have currently intersected a combination of the

porphyry domain, where the gold is associated with pyrite veinlets, and the MDZ where gold is

associated with pyrrhotite. There is a transition zone existing between the two domains, which is



observed to some extent in the current mine workings with overprinting of the epithermal system on

the MDZ. The vertical extent of the transition is not clearly defined from the current drilling. Currently,

underground mining at the Caldas-operated mine remains focused on the vein structures located in

the central portion (Zona Baja) of the Marmato deposit.

Diamond drilling indicates that the veins typically range between 0.5 and 5 m wide and extend for 250

to 1,000 m along strike and 150 to 750 m down dip. These observations are supported by underground

mining which has confirmed that individual vein structures have good geological continuity and can

extend for 100 to 800 m along strike and 100 to at least 300 m down dip. Between 2017 and 2020,

CGM has worked on updating the quantity of the underground channel sampling captured in the

database, which has increased the information available to model the vein domains.

The broad zones of veinlet mineralization in the porphyry domain was modelled initially by SRK in

2017 and typically varied from 10 to 230 m wide, reaching up to 340 m wide in areas of significant

veinlet accumulation, while extending with good geological continuity for between 200 m and

approximately 950 m along strike and between 100 and 900 m down dip. SRK has updated these

domains during the 2019 geological modelling process using more discrete zones and application of

an indicator grade shell approach using a 0.5 grams per tonne (g/t) gold (Au) cut-off grade (CoG).

At depth within the central portion of the deposit, SRK has noted a zone of elevated grades which has

been referred to as the higher grade MDZ (more than 2 g/t Au). This zone is indicated to be continuous

along strike for approximately 500 m and has a confirmed down dip extent that reaches up to 800 m,

with a thickness that varies between 35 and 150 m. It is possible that the main MDZ mineralization is

bounded within a series of faults but limited drilling at the edges of the deposit make confirmation

difficult to assess at this stage. To avoid the potential for volumetric “blow-outs”, SRK has used the

faults as a hard boundary in the geological domaining process.

1.3 Status of Exploration, Development and Operations

The latest sampling has comprised selective infill drilling targeting the MDZ to a spacing of 50 to 100 m

and additional underground channel sampling within the CGM operated mine, which extends from

Levels 16 to 21.

A total of 1,357 drillholes have been used to inform the 2020 Marmato Mineral Resource Estimate

(MRE) including historic drilling and more recent drilling completed between the 2019 Preliminary

Economic Assessment (PEA) and this PFS. A total of 40 new drillholes from the exploration and mine

developed have been included since the 2019 PEA for a total of 12,555 m of new drilling.

In addition to the drilling information, CGM has captured information from the mine and exploration

channel sampling databases. Limited new sampling has been captured between the 2019 PEA and

the current study; in total, 26,307 channel samples exist in the database for a combined sample length

of 42,328 m. In CGM commissioned a detailed topographic map with 0.5 and 1 m resolution contour

intervals derived from LIDAR imagery, which was supplied to Datamine™ in 2020. The new

topographic map provides a detailed base map for improved accuracy when plotting the results of the

exploration programs, as well as a high-resolution satellite image. All data has been converted and

stored in the Magna Sirgas/Colombia West coordinate system (MSCW).

All samples were prepared, and fire assayed by SGS Laboratories at their facility in Medellin. CGM

has carried out routine Quality Control and Quality Assurance programs (QA/QC) to monitor the quality



during the process. The results of the drilling have validated aspects of the previous interpretation, but

also provided additional information

1.4 Mineral Processing and Metallurgical Testing

Metallurgical programs were conducted by SGS Lakefield (SGS) in 2019 and 2020 to evaluate the

processing requirements for the MDZ. The 2019 metallurgical program was conducted as part of the

2019 PEA that was prepared for the Project, and the 2020 metallurgical program was conducted to

support the current PFS. The 2020 metallurgical program was conducted to further define the process

parameters and design criteria for the selected flowsheet that includes gravity concentration followed

by cyanidation of the gravity tailings. The test program included gravity concentration, gravity

recoverable gold (E-GRG determination) cyanide leach optimization and carbon-in-pulp (CIP)

modelling, cyanide destruction (CND), solid/liquid separation and environmental testwork. The

optimization and metallurgical design tests were all completed using the MDZ master composite. Once

the optimized flowsheet had been selected, the variability test samples were tested under these

optimized gravity/cyanidation conditions.

Key findings from the 2020 metallurgical program include the following:

• The PFS metallurgical program was conducted on an MDZ master composite and on

variability composites representing low, medium and high grade MDZ ore, transition zone and

the MDZ deep zone.

• Native gold was by far the predominant gold carrier, and the majority (more than 99%) of the

gold particles occurred within mineral structures that would be readily accessible by leaching

solutions. Gold particles were not often in direct contact with sulfides, yet very commonly

pyrrhotite, chalcopyrite, and bismuth minerals were found in close vicinity to the gold

mineralization

• The metallurgical program optimized process parameters required to recover gold and silver

values from MDZ ore using a process flowsheet that includes gravity concentration followed

by cyanidation of the gravity tailing.

• Comminution tests were conducted on the MDZ master composite, MDZ deep zone

composite, three MDZ sub-composites (low grade, medium grade and high grade) and on the

Marmato mine composite. The comminution tests included SAG Mill Comminution (SMC),

SAG Mill Power Index (SPI) and Bond ball mill work index (BWI) tests. In addition, Bond Low

Impact Crushing work index (CWI) and abrasion (AI) tests were conducted on selected ½ HQ

drill core pieces.

o The results of the SMC (A x b) values ranged from 23 to 29, indicating the ore is hard with

respect to impact breakage.

o The BWI values for the MDZ composites range from 17.7 kilowatt hour per tonne (kWh/t)

to 19.8 kWh/t, which places them in the hard range of hardness.

• E-GRG testwork and modeling indicate that about 40% of the gold contained in the MDZ ore

can recovered into a gravity concentrate. Gold contained in the gravity tailing could be

recovered in a standard CIP cyanidation leach circuit.

• An intensive cyanide leach test on the gravity concentrate demonstrated that 99.7% of the

contained gold and 87.9% of the contained silver could be extracted from the gravity

concentrate without regrinding.



• Based on the results of the PFS metallurgical program, overall gold recovery (gravity

concentration + gravity tailing cyanidation) is estimated at 95% and overall silver recovery is

estimated at 51%. This is very similar to the results from the PEA metallurgical program in

which gold recovery was estimated at 95% and silver recovery was 47%. There is little

difference in reported gold recoveries for the master and variability composites, and gold

recovery appears to be independent of ore grade over the range tested.

• Cyanide destruction tests demonstrated that weak acid dissociable cyanide (CNWAD) could be

reduced to less than 10 milligrams per liter (mg/L) with the SO2/air process. However, CNWAD

levels would further attenuate to less than 1 mg/L with time.

• Pressure filtration will be required to dewater thickened tailings in order to achieve less than

15% moisture content required for disposal in a dry stack tailings facility (DSTF).

1.5 Mineral Resource Estimate

The Mineral Resource model presented herein represents an updated resource evaluation prepared

for the Marmato Project. The resource estimation methodology involved the following procedures:

• Database compilation and verification

• Construction of wireframe models for the fault networks and centerlines of mining development

per vein

• Definition of resource domains

• Data conditioning (compositing and capping) for statistical and geostatistical analysis

• Variography

• Block modelling and grade interpolation

• Resource classification and validation

• Assessment of “reasonable prospects for economic extraction” and selection of appropriate

reporting cut-off grades (CoGs)

• Preparation of the Mineral Resource Statement

The resource evaluation work was completed by Mr. Benjamin Parsons, MAusIMM (CP#222568), with

assistance from Mr. Giovanny Ortiz, FAusIMM (#304612). The effective date of the Mineral Resource

Statement is March 17, 2020, which is the last date assays were provided to SRK.

The mineral resource estimation (MRE) process was a collaborative effort between SRK and CGM

staff. CGM provided SRK with an exploration database with flags of the main veins as interpreted by

CGM. In addition to the database, CGM has also supplied a geological interpretation comprising

preliminary three dimensional (3D) digital files (DXF) of the areas investigated by core drilling for each

of the main veins.

SRK imported the geological information into Seequent Leapfrog® Geo (Leapfrog®) to complete the

geological model. Leapfrog® has been selected due to the ability to rapidly create accurate geological

interpretations, which interact with a series of geological conditions and data types.

SRK has produced block models using Datamine™ Studio RM Software (Datamine™). The procedure

involved the import from Leapfrog™Geo of wireframe models for the fault networks, veins, definition

of resource domains (e.g. high-grade sub-domains), data conditioning (compositing and capping) for

statistical analysis, geostatistical analysis, variography, block modelling and grade interpolation

followed by validation.



Grade estimation for the veins has been based on block dimensions of 5 m by 10 m by 10 m. Sub-

blocking to 0.5 m by 1 m by 1 m has been allowed to reflect the narrow nature of the geological model.

The block size reflects the relatively close-spaced underground channel sampling and spacing within

veins compared to the wider drilling spacing, with the narrower block size used in the MDZ at depth to

reflect the proposed geometry of the mineralization (i.e. steeply dipping feeder zone).

SRK reviewed and updated the geostatistical properties of the domains. Gold grades have been

interpolated using nested three-pass estimates within Datamine™, using an Ordinary Kriging (OK)

routine. SRK has also run Inverse Distance Weighted (IDW2) and Nearest Neighbor (NN) estimates

for validation purposes.

The search ellipses follow the typical orientation of the mineralized structures and where appropriate,

were aligned along the mineralized veins, as detailed below:

• Dynamic searches were used for the vein mineralization domains. Within these domains, the

true dip and true dip direction has been calculated on a block by block basis

• In comparison, given the relatively short strike and dip of the splay, SRK has elected to use

an average dip and strike for each structure

• For the porphyry domain, SRK has generated a default dip and dip-direction to orientate the

search volume along the main regional trend

• For the MDZ, a single dip and strike has been used with the search ranges orientated along

the main dip and strike of the domain

• All contacts between the veins have been treated as hard boundaries for domaining with only

coded samples from any given vein used in the estimation of that domain

• Statistical characteristics such as search volume used, kriging variance, and number of

samples used in an estimate, were also computed and stored in each individual block for

descriptive evaluations

SRK has validated the block model using a combination of visual checks, statistical comparison of

composite grades to all three estimation methods and via swath plot analysis. SRK considered the

estimates to be representative of the underlying data.

Block model quantities and grade estimates for the Marmato Project were classified according to the

CIM Definition Standards for Mineral Resources and Reserves (CIM, 2014). SRK developed a

classification strategy which considers the confidence in the geological continuity of the mineralized

structures, the quality and quantity of exploration data supporting the estimates and the geostatistical

confidence in the tonnage and grade estimates. Data quality, drillhole spacing and the interpreted

continuity of grades controlled by the veins have allowed SRK to classify portions of the veins in the

Measured, Indicated and Inferred Mineral Resource categories.

Measured: Measured Resources are limited to vein material within the current levels being mined by

CGM and estimated within the first search volume, which required a minimum of five composites and

a maximum of 20 composites. These areas are considered to have strong geological knowledge as

they have been traced both down-dip and along strike via mapping, plus underground channel

samplings provided sufficient data populations to define internal grade variability.

Indicated: SRK has delineated Indicated Mineral Resources using two methods split by the material

types:



• Veins/Disseminated/Splays: Primarily between Level 16 to 21 currently in operation. Indicated

Mineral Resources have been given at the following approximate data spacing, as a function

of the confidence in the grade estimates and modelled variogram ranges. SRK has expanded

the limits of the Indicated resources to also cover areas within the licensed portion of Echandia

where:

o Spacing of 50 m by 50 m (XY) existed from the nearest drillhole

o Multiple holes were enabled to be used during the estimation process

o Support from both diamond drilling and channel sampling was present

• MDZ: Based primarily on 2018/2019 drilling with the following conditions:

o 50 by 50 m (XY) drillhole spacing (defined by a distance buffer of 25 m from drilling of

underground [UG] levels)

o Enabled multiple holes to be used during the estimation process

o Search volume less than 2 (i.e. volumes 1 and 2)

o Additional caution has been paid when classifying the dip extensions on the series of holes

drilled to the northeast as limited information is known up and down dip from the current

drilling

Inferred: In general, Inferred Mineral Resources have been limited to within areas of reasonable grade

estimate quality and sufficient geological confidence, and are extended no further than 150 m from

peripheral drilling on the basis of modelled variogram ranges.

SRK has defined the proportions of Mineral Resource to have potential for economic extraction for the

Mineral Resource based on different CoGs relating to the mineralization style (i.e. vein versus

porphyry) and potential differences in selective underground mining methods.

During the site inspection, SRK noted and discussed with the mine geologists that some mining has

been attempted within the porphyry “pockets”. SRK considers this to have uncertainty as no detailed

survey of mining volumes in the porphyry pockets is available. Based on the level of uncertainty, SRK

has downgraded areas identified as having potential historical mining to Inferred.

To assign the final classification, the mathematical criteria as defined has then been applied to the

block model, which is subsequently digitized on 50 m sections (across strike), with the final wireframe

based on interpretation of polylines in Leapfrog™ to smooth changes in interpretation between

sections.

To determine the potential for economic extraction, SRK used the following key assumptions for the

costing but notes that the deposit has variable mining costs depending on the mining types resulting

in a range of CoGs. A metallurgical recovery of 95% Au has been assumed for the MDZ and 90% for

the veins and porphyry material based on the current performance of the operating plant. Mining and

processing costs have been defined from aspects of the current study and historical production. The

initial cut-off is based on the mining of the veins using the current mining processes and assumed

costs, with a second method (longhole) defined for mining the MDZ and potentially areas of wider

porphyry mineralization in the upper levels.

SRK has reported the tonnage and grades associated with the current mine and the MDZ project,

which are the assets owned indirectly by CGM. As such, the Mineral Resource includes all material

within the #014-89m license and a sub-portion of the #RPP_357 (Echandia) below an elevation of

1,300 m, which can be accessed from the existing operation through an agreement with Gran

Colombia. SRK has also included the proportion of Mineral Resources currently under application



(Application #KIU-11401) within the Mineral Resources, but these have been excluded from the

Mineral Reserves as the timing on granting this license remained uncertain at the date of this report

(however, this license was recently confirmed as approved by the government; so will therefore be

included in future technical studies).

The proposed mining plan is predicated on splitting the above Mineral Resources into three styles of

mineralization within three distinct areas. These areas are referred to as the UZ (existing mine levels

16 through 21), the Transitional Zone (which includes mining of MDZ material to an elevation of 950

m) and the MDZ project (which includes all material below the 950 m elevation).

The three styles of mineralization are based on the key geological types defined in the Mineral

Resources of veins, porphyry, and MDZ. Therefore, the estimation domains for the Mineral Resource

Statement have been grouped into veins, porphyry and MDZ mineralization. The veins account for the

veins, halos and splay material and have used a 1.9 g/t Au cut-off. The porphyry material also uses a

cut-off of 1.9 g/t Au. As the potential mining method will require further investigation, the MDZ material

has used a lower cut-off of 1.3 g/t Au to account for the larger bulk mining methods involved.

SRK highlights that all Mineral Resources within #CHG_081 (yellow and orange) and the upper areas

of #RPP_357 (above 1,300 m) as highlighted in Figure 1-1 in light blue have not been reported and

are excluded from the current Mineral Resource statement herein for CGM because any Mineral

Resources that may occur in these areas have not been transferred from Gran Colombia to CGM.

Table 1-1 shows the Mineral Resource Statement for the Project, with an effective date of March 17,

2020.

Source: SRK, 2020

Figure 1-1: Cross-Section Showing License Splits at Marmato

Licence #014-89m

Licence #CHG-081 (above 1300)

RPP #357 -(above 1300)

RPP #357 -(below 1300)

Licence #CHG-081 (below 1300)

Area under applicationNo # KIU-11401Note: Included in Mineral Resources but excluded from Mineral Reserves

Upper Mine(above 950)

MDZ Projectbelow 950)

Transition Zone MDZ(above 950)



Table 1-1: Caldas Mineral Resource(1) Statement with an Effective Date of March 17, 2020

Caldas Marmato Project - Effective Date March 17, 2020, Basis for MRE and PFS (CGM including RPP_357 less than 1,300 m)(1)

Category Quantity (Mt) Grade (g/t) Metal (kozs)

Au Ag Au Ag

Upper Mine (2)

Measured 2.1 5.65 27.0 387 1,853

Veins (5) 2.1 5.6 27.0 387 1,853

Porphyry (5) 0.0 0.0 0.0 0 0

Indicated 9.2 4.45 18.7 1,320 5,545

Veins 7.2 5.0 21.1 1,156 4,862

Porphyry 2.1 2.5 10.3 165 682

Measured and Indicated 11.4 4.67 20.2 1,707 7,397

Veins 9.3 5.2 22.4 1,543 6,715

Porphyry 2.1 2.5 10.3 165 682

Inferred 4.5 3.70 15.5 532 2,224

Veins 2.7 4.4 17.9 386 1,574

Porphyry 1.7 2.6 11.7 145 650

Transition Zone (3) (6)

Measured 0.0 0.0 0.0 0 0

Indicated 3.4 2.68 7.2 294 785

Measured and Indicated 3.4 2.68 7.2 294 785

Inferred 0.0 1.95 3.7 2 3

MDZ (4) (6)

Measured 0.0 0.0 0.0 0 0

Indicated 24.7 2.63 3.6 2,085 2,870


Inferred 21.9 2.32 2.1 1,639 1,506

Combined

Measured 2.1 5.6 27.0 387 1,853

Indicated 37.3 3.1 7.7 3,699 9,200


Inferred 26.4 2.6 4.4 2,172 3,733 (1) Mineral Resources are reported inclusive of the Mineral Reserve. Mineral resources are not mineral reserves and do not have demonstrated economic viability. All figures are rounded to reflect the relative accuracy of the estimate. All composites have been capped where appropriate. The Mineral Resources were estimated by Benjamin Parsons, MSc, MAusIMM #222568 of SRK, a Qualified Person pursuant to NI 43-101. (2) Upper Mine is defined as the current operating mines from levels 16 through 21 using existing mining methodology (cut and fill). (3) “Transition Zone” is defined as mining of MDZ above an elevation of 950 m (accessed from the current operations) using a modified longhole stoping method. (4) MDZ is defined as mining of MDZ below an elevation of 950 m using longhole open stope mining methods. (5) Porphyry and vein mineral resources are reported at a CoG of 1.9 g/t. CoGs are based on a price of US$1,500/oz Au and gold recoveries of 90% for underground resources without considering revenues from other metals. (6) MDZ mineral resources are reported at a CoG of 1.3 g/t. CoGs are based on a price of US$1,500/oz Au and gold recoveries of 95% for underground resources without considering revenues from other metals. Source: SRK, 2020

The 2020 Mineral Resource represents a number of changes in the defined Mineral Resource

compared to the 2019 PEA Mineral Resources, due to the following key factors:

• Infill drilling within the MDZ areas has increased the confidence in the estimates significantly

from the Inferred to Indicated category.

• Minor reduction in the vein domains as a result of additional depletion accounted for between

the PEA and PFS models, plus changes in the geological interpretation of veins and

disseminated material.



SRK highlights that the current MDZ Mineralization represents a notable change in the style of

mineralization and considerations for mining methods at the Project and has maintained the use of a

high-grade core to the mineralization at depth.

The main changes in the Mineral Resource Statement since the previous estimate can be defined on

the combined Mineral Resource as follows:

• Increase in the Indicated MDZ material from 6.4 million tonnes (Mt) at 2.6 g/t Au, for 537

thousand ounces (koz), to 28.1 Mt at 2.6 g/t Au, for 2,379 koz, which is an increase of

1,842 koz within the MDZ. This is reflected in a reduction in the Inferred from 41.2 Mt at 2.1 g/t

for 2,812 koz to 22 Mt at 2.3 g/t for 1,640 koz, which is a reduction of 1,172 koz.

• Increase in the proportion of Measured and Indicated material within the vein domain from

9.2 Mt at an average grade of 4.6 g/t to 9.3 Mt at an average grade of 5.2 g/t Au, which is an

increase of 180 Koz or 13.2%.

• Reduction in the proportion of Inferred material within the veins from 3.3 Mt at 4.4 g/t Au for

466 koz, to 2.7 Mt at 4.4 g/t Au for 386 koz, which represents a difference of 80 koz.

• Minor increase in proportion for Indicated of porphyry (pockets) material of 25 koz.

• Increase in the Inferred portion of the porphyry material from 0.3 Mt at 3.1 g/t Au for 34 koz,

to 1.7 Mt at 2.6 g/t Au for 145 koz.

1.6 Mineral Reserve Estimate

The mine is currently developed to the 1,000 m elevation. A transition is occurring from narrow vein

mineralization to large porphyry mineralized areas (gold associated with pyrrhotite veinlets).

Mineralization is generally vertical with vein widths ranging from more than 1 m to several m. Porphyry

mineralized areas also have a vertical mineralization trend and can be up to approximately 100 m in

width. For this PFS, there are three different mining methods, separated into three distinct zones.

• The first zone is the mineralized vein material between 950 m elevation and 1,300 m elevation,

referred to as the Veins. This is the existing mine where conventional cut and fill stope methods

will continue to be used.

• The second zone is the wider porphyry material between 950 m elevation and 1,050 m

elevation, referred to as the Transition Area. A modified longhole stoping method will be used

in this area.

• The third zone is the porphyry material below 950 m elevation, referred to as MDZ. There is a

10m sill pillar left in-situ between the MDZ and the bottom of the Transition Area. The MDZ

material will be mined using a longhole stoping method. The MDZ area is currently not

developed.

The first two zones (Veins and Transition) are considered the Upper Mine, and the material is

processed in the existing processing facility. Material mined from the third zone (MDZ) will be sent to

a new processing facility to be constructed.

Mineral Reserves were classified using the 2014 CIM Definition Standards. Indicated Mineral

Resources were converted to Probable Mineral Reserves by applying the appropriate modifying

factors, as described herein, to potential mining shapes created during the mine design process. In

the same manner, Measured Mineral Resources were converted to Proven Mineral Reserves.



A 3D design has been created representing the planned reserve mining areas. Dilution and recovery

have been included in the estimate, specific to each mining method. The underground mine design

process resulted in 19.7 Mt at an average grade of 3.19 g/t Au and 6.87 g/t Ag. Table 1-2 presents the

Mineral Reserve statement as of March 17, 2020.

Table 1-2: Caldas Mineral Reserve Estimate as of March 17, 2020 – SRK Consulting (U.S.), Inc.

Underground Mineral Reserves Cut-Off (1): 1.61 to 2.23 g/t

Area Category Tonnes

(kt) Au

(g/t) Ag

(g/t) Contained Au

(koz) Contained Ag

(koz)

Veins(2)

Proven 762 5.01 21.80 123 534

Probable 3,049 4.20 16.85 412 1,652

Veins Total 3,812 4.37 17.84 535 2,186

Transition(3)

Proven 40 7.63 28.16 10 36

Probable 1,293 3.43 7.92 143 329

Transition Total

1,333 3.56 8.52 152 365

MDZ(4)

Proven - - - - -

Probable 14,556 2.85 3.84 1,333 1,799

MDZ Total 14,556 2.85 3.84 1,333 1,799

Caldas Total

Proven 802 5.14 22.12 133 570

Probable 18,898 3.11 6.22 1,888 3,780

Total 19,700 3.19 6.87 2,021 4,350

Source: SRK, 2020 Notes: All figures are rounded to reflect the relative accuracy of the estimates. Totals may not sum due to rounding. Mineral Reserves have been stated on the basis of a mine design, mine plan, and economic model. Mineral Resources are reported inclusive of the Mineral Reserve. (1): Veins reserves are reported using a CoG of 2.23 g/t Au. The Veins CoG calculation assumes a US$1,400/oz Au price, 85% Au metallurgical recovery, US$49.45/t mining cost, US$13.63/t G&A cost, US$12.24/t processing cost, and US$8.96/t royalties. Transition reserves are reported using a CoG of 1.91 g/t Au. The Transition CoG calculation assumes a US$1,400/oz Au price, 95% Au metallurgical recovery, US$46/t mining cost, US$13.63/t G&A cost, US$12.24/t processing cost, and US$8.96/t royalties. MDZ reserves are reported using a CoG of 1.61 g/t Au. The MDZ CoG calculation assumes a US$1,400/oz Au price, 95% metallurgical recovery, US$42/t mining cost, US$14/t processing cost, US$6.75/t production taxes, US$3/t G&A cost, and US$3/t tailings cost. Note that costs/prices used here may be somewhat different than those in the final economic model. This is due to the need to make assumptions early on for mine planning prior to finalizing other items and using long-term forecasts for the life-of-mine plan. (2): The Veins area is currently mined using cut-and-fill methods. Mining dilution ranging from 20% - 55%, averaging 26%, is included in the reserves using a zero grade for dilution. A mining recovery of 90% is applied to stopes. The Veins Mineral Reserves were estimated by Fernando Rodrigues, BS Mining, MBA, MMSAQP #01405, MAusIMM #304726 of SRK, a Qualified Person. (3): The Transition area will be mined using a modified longhole stoping method. A mining dilution of 7% is included in the reserves using a zero grade for dilution. A mining recovery of 90% is applied to stopes. The Transition Mineral Reserves were estimated by Fernando Rodrigues, BS Mining, MBA, MMSAQP #01405, MAusIMM #304726 of SRK, a Qualified Person. (4): The MDZ portion of the Project will be mined by longhole open stoping mining methods. Mining dilution (internal and external) is included in the reserve. Stope dilution is 8%, and a portion of the stope dilution is applied using grade values based on average surrounding block information. A mining recovery of 92.5% is applied to stopes. The MDZ Mineral Reserves were estimated by Joanna Poeck, BEng Mining, SME-RM, MMSAQP #01387QP, a Qualified Person.

1.7 Mining Methods

Marmato has been in operation in various forms since the mid-1500s. Mineros Nacionales (MN) was

awarded the contract for the concessions in 1989. The Project was originally developed as a 300 t/d

underground mine in 1997 and has expanded through the years to the existing 1,200 t/d operation.

Table 1-3 shows the production from 2015 to May 2020.



Table 1-3: 2015 to 2020* Production

o Year o Unit 2015 2016 2017 2018 2019 2020*

o Ore Processed o t 303,279 341,309 365,119 338,902 370,245 119,069

o Au Grade o g/t 2.79 2.56 2.48 2.67 2.49 2.47

o Au Recovered o oz 23,954 23,449 25,163 24,909 25,750 8,318

*January through May of 2020 Source: CGM, 2020

Historically, shrinkage stoping was used to mine the Veins material as well as a caving method where

poor ground conditions were encountered. Currently, a conventional cut and fill (CaF) mining method

is used. Blasted material is either transferred down to Level 18 via ore passes or is transferred up via

the incline shaft (apiques) hoist, loaded into rail carts and hauled to the mill.

In the Transition area, a modified longhole stoping method will be used. The stope size is 15 m wide

by 15 m high with varying length of up to 26 m. These stopes are mined in a primary-secondary

sequence with paste backfill for the primary stopes and unconsolidated waste rockfill for the secondary

stopes. Where waste rock is unavailable, hydraulic sand fill will be used to fill the secondary stope.

Blasted material in the Transition area is also transported up to Level 18 via apiques and hauled to the

mill via rail carts.

The MDZ material is mined using a longhole stoping method with stope sizes that are 10 m wide by

30 m high, with varying lengths of up to 30 m. The MDZ area is currently not developed. The main

access will be a decline, hosting a conveyor from the plant area to the underground crusher area. A

dedicated ventilation drift will serve as secondary egress from the mine. Ventilation infrastructure

development underground was designed to support the mining method and was sized based on mining

equipment and production rate requirements. Trucks will dump into a surge bin at an elevation of 790

m. Material will go through the surge bin into the crusher and then be conveyed out of the mine.

Geotechnical

SRK and the Marmato exploration team collaborated on a geotechnical investigation program for the

MDZ from June 26, 2018 to March 4, 2020. The program was designed to characterize subsurface

geotechnical conditions to assist in the development of a PFS mine design. Based on the observed

ground conditions, SRK considers that the geotechnical investigation fulfills the industry standards to

support stope design and ground support requirements at a PFS project level. For a PFS project level,

SRK considers the proposed PFS mine design acceptable. The proposed stope designs, sill pillar

design, back filling specifications and ground support specification must be considered as PFS level

only and should not be implemented before an FS level investigation is conducted. Full geotechnical

investigation is described in the Marmato Geotechnical PFS Study (SRK, 2020).

Hydrogeology

The mine area is located in the hydrogeological regional area of Magdalena Cauca, specifically in the

Cauca River catchment (Caldas Department). The region is comprised of igneous and metamorphic

rocks with limited groundwater storage capacity and hydraulic conductivity (IDEAM, 2013). The

porphyry units represent the main hydrogeological units in the mine area, with a low hydraulic

conductivity and limited groundwater storage capacity. Groundwater flow is compartmentalized within

structural blocks with limited hydraulic communication across fault boundaries due to fault gouge,

weathering, or an offset of geological units (Knight Piésold, 2012).



Previous field campaigns were performed by Knight Piesold (KP) in 2011 and 2012 (Knight Piésold,

2012) and currently by SRK starting in early 2020, primarily consisted of packer isolated interval

testing, monitoring well and Vibrating Wire Piezometer (VWP) installations in underground coreholes

or locations distal to the mine area.

The zone of enhanced hydraulic conductivity values at depths of 600 to 800 m below the ground

surface corresponds to fractured zones associated with Fault 2 and Fault 1-3 in the mine area.

Measured water levels show elevations from 661 to 2,022 m Magna Sirgas/Colombia West coordinate

system (EPSG 3115) (MSCW), following the topography at 100 m depth in most of the locations

outside the mine area. A depressurization zone was detected in the underground piezometers where

the water levels have a horizontal trend. The shape or extent of the depressurization zone is currently

unknown. On a regional scale, the groundwater flows west to east, following the topographical gradient

to the Cauca River, located at 692 m elevation, which represents the main discharge for the

hydrogeological system.

KP developed 172 packer tests, three underground piezometers and 11 piezometers at the surface

(Knight Piésold, 2012). In the 2020 field campaign, 70 packer tests and two multi-level VWP

installations were performed. As a result, the geometrical mean of hydraulic conductivity values ranges

from 1.1 by 10-3 meters per day (m/d) to 4 by 10-2 m/d in the porphyry units depending on the depth

intervals. The shallow zone (less than 200 m depth) corresponds to saprolite and more permeable

bedrock and the deep zone (more than 850 m depth) has less permeable conditions. However, it is

apparent that high-permeability zones (hydraulic conductivity greater than 0.1 m/d, which may be

associated with Fault 2 and Fault 1-3, were encountered in the vicinity of the planned mine at depths

of 600 to 800 m below ground surface (bgs), or at an elevation of 700-900 m MSCW.

Mine Dewatering

The measured monthly average total dewatering rate in the Marmato mine is 37 liters per second (L/s),

varying from 26.8 L/s to 46.4 L/s. Strong seasonal trends were not observed, however a decrease of

approximately 20 L/s can be observed in the last 12 months. A major structure zone with significant

water flow (7 to 8 L/s) was detected at levels 17 and 21 to the north of the Criminal Fault.

The dewatering rate is a combination of groundwater inflows and water content in the backfill material

(50% of water). According to Marmato operational personnel, the contribution of the backfill material

is 7 to 14 L/s, depending on the number of hydraulic backfill equipment units in operation. Therefore,

the average fresh groundwater inflow into the mine could vary from 23 to 30 L/s.

SRK developed a preliminary 3D numerical groundwater flow model using MODFLOW-USG code,

based on available climatic, geological and hydrogeological data. The majority of the predicted inflow

to the MDZ planned mine (up to 78 L/s with a possible range from 56 to 159 L/s) is expected from the

upper levels above 730 m where elevated hydraulic conductivity values of the bedrock groundwater

system were measured. Mine inflow to the lower planned mine below 730 m is predicted to be lower

(15 L/s with an upper limit of 34 L/s) due to reduced measured hydraulic conductivity with depth.

Total maximum discharge into the entire mine complex, including flow to existing mine levels, is

predicted to be up to 111 L/s with a possible range from 89 to 168 L/s.

The mine is 2.5 km to the west of the Cauca River with a proposed bottom of 212 m below the river

stage (or 480 m MSCW). There is a risk of surface-water inflow through the riverbed sediments and



fractured bedrock when hydraulic gradient will be reversed by mine dewatering. Structural features

similar to those detected to the north of the Criminal Fault could connect mine developments with the

river. In SRK’s opinion, this represents a medium risk for the Project. Further hydrogeological

investigations of this area are required to evaluate potential significant increments in groundwater

inflow.

Production Schedule

The production and development schedules were completed using iGantt software from Minemax. The

production schedule is based on the rate assumptions either from current mining practices or

developed from first principles.

The UZ production schedule targets a total ore production of 1,500 t/d or 525,000 tonnes per year (t/y)

(based on 350 days per year) to the mill. A gradual ramp up of 1,100 t/d (385,000 t/y) in 2020, 1,250 t/d

(437,500 t/y) in 2021, 1,400 t/d (490,000 t/y) in 2022 and 1,500 t/d in 2023. The Transition Zone

accounts for 400 t/d while the rest comes from the Veins. Life of Mine (LoM) for the Veins is 12 years

for a total production of 3.81 Mt at 4.37 g/t Au. LoM for the Transition Zone is 11 years for a total

production of 1.33 Mt at 3.56 g/t Au.

Combined UZ production is 5.14 Mt at 4.16 g/t Au. Figure 1-2 shows the UZ production schedule

colored by time period. Note that there is also a 2 Mt/y permit limit of moved material, which limits the

production of the UZ.

Source: SRK, 2020

Figure 1-2: UZ Production Schedule Colored by Time Period

The MDZ mining schedule is based on 365 days/year seven days/week, with three 8 hour shifts each

day. Actual operational mining days are 360. For simplicity the schedule has been completed

assuming 365 with pro-rated productivity rates. A production rate of 4,000 t/d (1.46 Mt/yr) was targeted



with ramp-up to full production as quickly as possible. The schedule timeframe is quarterly for four

years and annually for the remainder of the mine life.

Decline activities begin in October 2021 with initial mine development through Q4 2023. Stoping begins

in Q4 of 2024, with a one year ramp up period until the mine and plant are operating at full capacity.

Figure 1-3 shows the mine production schedule colored by year.

Source: SRK, 2020

Figure 1-3: MDZ Mine Production Schedule Colored by Year

Figure 1-4 summarizes the combined UZ and MDZ schedules. This combined schedule is used in the

economic model results shown in section 22.



Source: SRK, 2020

Figure 1-4: Combined UZ and MDZ Mining Profile – Tonnes and Grade

Mining of the Veins in the UZ is with handheld pneumatic equipment (jacklegs and stopers) for

development and production. Blasted material is mucked using slushers, microscoops and skid steer

loaders into rail carts and hauled out to the mill.

The Transition Zone utilizes jumbo drills for lateral development. The same jumbo drills are used for

ore mining with a longhole drill attachment. Blasted material is loaded by 4 t load haul dumps (LHD)

to 10 t trucks (or to the orepass) and is then transferred to rail carts and transported out of the mine

via the apiques.

The UZ mine (Veins and Transition) is a producing mine and all infrastructure is already established.

The MDZ mine will utilize jumbo drills for lateral development and down-the-hole drills for vertical

development and production stoping. Mechanical bolters will be used for ground support. The mine

will operate a fleet of 45 t haul trucks being loaded by 17 t LHDs. The ore will be fed through a grizzly

with rock breaker into an underground crusher and conveyor system to the surface. The mine will have

full infrastructure underground, including; ventilation, cemented paste backfill booster pump and

distribution system, dewatering pumping system, electrical substation and distribution system, fuel

storage, warehousing, explosives storage, communications system, and maintenance shops. The

MDZ mine will have a staff of approximately 429 people at the peak of production. Owner mining has

been assumed for steady state with contractor mining development early in the mine life.

1.8 Recovery Methods

CGM operates a 1,200 t/d process plant to recover gold and silver values from material produced from

current Marmato mining operations in the UZ and plans to expand this facility to 1,500 t/d capacity in



2021. In addition, CGM is evaluating the development of the MDZ, which is below the current mining

operations and the construction of a new 4,000 t/d plant to process material solely from the MDZ.

Marmato Process Plant

The Marmato process plant flowsheet incorporates unit operations that are standard to the industry

and includes:

• Three-stage crushing

• Closed circuit ball mill grinding

• Gravity concentration

• Flotation

• Flotation and gravity concentrate regrind

• Cyanidation of the flotation and gravity concentrates

• Counter-current-decantation

• Merrill-Crowe zinc precipitation

• Smelting of precipitates to produce final doré product

During the period from 2013 to 2020 (Jan to May) ore processed through the Marmato plant has

increased from 274,191 to 370,245 tonnes per year (t/y) while grades have declined slightly from 2.90

g/t Au in 2013 to 2.49 g/t Au in 2019 and silver grades have ranged from 12.36 to 9.13 g/t Ag. Overall

gold recovery has ranged from 83.7 to 88.9% and has averaged about 87.1% during the period 2019

to 2020 (Jan to May). Silver recovery has ranged from 33 to 41.1% and has averaged 33.2% during

the period 2019 to 2020 (Jan to May). Annual gold production has increased from 22,566 ounces in

2013 to 25,750 ounces in 2019.

MDZ Process Plant

The MDZ process plant was designed by Ausenco and is based on the 2020 metallurgical program

conducted by SGS Lakefield, Ausenco’s industry experience and input from equipment suppliers. The

process plant is designed to process ore at a rate of 1,460,000 dry t/y (4,000 dry t/d) based on a 92%

plant availability and includes unit operations that are well proven and standard to the industry,

including:

• Crushing/Grinding


• Cyanide leaching of the gravity tailings

• Carbon-in-pulp (CIP) gold adsorption

• Desorption/Electrowinning/Refining

• Cyanide detoxification

• Tailings thickening and filtration

The MDZ process plant will be located North-East of the town of Marmato, Colombia. Access to the

plant will be via the plant roads off National Route 25. The primary crusher will be located underground,

and the secondary crusher positioned at the surface near the entrance to the mine portal. The crushed

ore stockpile will be east of the main process plant. The main plant will be outdoors and will include

the grinding, gravity recovery, leach/CIP tanks, reagent, elution/carbon regeneration, cyanide

detoxification and tailings thickening circuits. The electrowinning and refining area will be located in a



separate building. Plant tailings will be thickened and pumped either to the mine backfill plant or to the

tailing filter plant, located next to the main plant. Filtered tailings will be hauled and stored in a DSTF.

1.9 Project Infrastructure

The existing Marmato Project has a mature and functioning infrastructure system including all the

necessary facilities and supporting utilities to produce at the planned production levels. The current

facilities include a security checkpoint that provides access to the office and administrative office area.

The facilities also include employee motorcycle parking, meeting area, cafeteria, multiple shops and

warehouses, a camp with cafeteria, exercise and sports field, equipment storage yards, compressor

station, welding shop, a 500 kilowatt (kW) backup generator, processing plant, underground mine,

explosives storage a short distance from the mine that is managed by the military, main power

substation and distribution powerlines with motor control centers at key loads. The site has three

portals that access the mine workings. Water Supply for the existing Marmato Project is provided by

mine dewatering and water reclaimed from the DSTF, Additional water supply from the Cauca River

to supplement the existing plant water availability during the dry season is planned to be in-place

before the MDZ project startup.

The MDZ project infrastructure will be developed on a separate greenfield site approximately 3

kilometers (km) north-east of the existing site by road. The new site will require new access roads off

the existing El Llano access to a new processing facility, camp area, and mine portal with access to

the MDZ.

The new infrastructure will include an additional transmission line from the 115 KV Salamina substation

to the new MDZ substation with local MDZ distribution to the mine substation and processing facility.

Surface facilities will include the mine portal, truck shop, processing facility, fuel storage and fuel

distribution system, paste backfill plant, shotcrete plant, processing plant facilities, a tailings filter plant,

a new water supply plant near the Cauca River, a new camp, offices, and a small temporary run of

mine (RoM) stockpile. Explosives storage is planned to be offsite.

The site will have a crushing area with a surge stockpile feeding the main processing plant. Support

facilities will include warehouses, shops, offices, a camp, administrative office, change house and

laydown yards. The camp and administrative facilities will be located at a separate location

approximately 300 m to the south from the processing plant. Parking will be provided near the entrance

to the MDZ site and at the camp location with a security gate for restricted access that will be

constructed at the entrance to the facility.

Water supply for the MDZ project will be supplied by mine dewatering, recycled water from the tailings

filter press and runoff and seepage collected from the DSTF, as well as from supplemental water

drawn from the Cauca River as needed.

The area already supports a significant mining population and skilled labor will be available from the

region.



1.9.1 Tailing Management Facilities

SRK completed a study of potential options for DSTF siting in the vicinity of the existing Cascabel

tailings storage facility and the proposed portal and plant location. Factors considered in the siting

study included topography, permitting requirements for stream crossings, property ownership and

acquisition potential, and municipality boundaries. Also, as part of the study, SRK developed

conceptual designs for seven potential DSTF locations. From that analysis, only three locations, sites

1, 2 and 6, were identified within the area of study as potentially feasible for:

• Providing the capacity required through mine life

• Achieving global stability in the steep terrain in the site vicinity

Due to property access difficulties and travel restrictions because of the COVID-19 pandemic, SRK

and CGM were unable to complete a geotechnical investigation at any of the sites. All conclusions and

costs presented in this study related to DSTF design and operation are therefore based on necessary

assumptions that will require investigation and confirmation in the next phase of study. Where input

assumptions were required, SRK has attempted to use conservative inputs to arrive at a reasonable

but conservative estimate of costs, risks and potential opportunities associated with DSTF siting,

construction and operation.

Based on the results of an SRK trade-off study (ToS) evaluating major cost items for Sites 1, 2 and 6,

CGM indicated a preference to evaluate the feasibility of developing DSTF 2 and then DSTF 1 to

achieve the desired tailings storage capacity through the currently predicted mine life. The combination

of DSTF 2 and DSTF 1 provides sufficient capacity based on current projections. DSTF 6 provides

sufficient capacity on its own for the currently predicted mine life, although the distance to the plant

provides some additional planning and access complexities.

Operation of the current Cascabel 1 and 2 DSTFs is required to provide enough capacity and time to

begin phased construction of DSTF 2 to provide for uninterrupted tailings storage. A review of available

design and stability analyses of the Cascabel 1 and 2 configurations indicates they have not been

designed or evaluated in accordance with internationally accepted standards of practice. Engineering

consultants from Dynami recently completed a stability review of both the existing and expanded

Cascabel 1 and 2 designs and concluded there is not enough information currently available to

establish the current or future stability of the facility. Dynami recommended extensive characterization.

To achieve the timeline currently presented in the PFS, CGM has committed to immediate

implementation of Dynami’s recommendations and subsequent design and mitigation aimed at

ensuring the facility’s compliance with internationally accepted standards of practice. SRK

recommends that CGM identify other options for filtered tailings storage that may provide additional

interim storage capacity in the event Cascabel 1 and 2 cannot be shown to be stable to internationally

accepted standards.

1.10 Environmental Studies and Permitting

1.10.1 Environmental Studies and Management

The existing Marmato Project predates the regulatory requirements to prepare an environmental

impact assessment (EIA) as part of the permitting process. Instead, the operations were authorized

through the approval of an Environmental Management Plan (Planes de Manejo Ambiental or PMA).

The original PMA for Marmato was approved by the regional environmental authority (Corporación



Autónoma Regional del Caldas or Corpocaldas) on October 29, 2001 under Resolution 0496, File No.

616. The site-specific PMA covers environmental studies and required management procedures and

practices. In addition, baseline data collection programs were initiated in 2019 to gather relevant and

appropriate site information with respect to both the existing Marmato Project and the proposed MDZ

expansion. The data was compiled and reported in Capítulo 20: Caracterización Ambiental y Social

del Proyecto, Caldas Gold Marmato S.A.S., Título Minero #014 – 89m (May 2020). The assessment

of potential impacts associated with the MDZ expansion project can only begin in earnest once the

PFS mine plan has been finalized, at which point, CGM will initiate engagement with Corpocaldas

(anticipated in Q1 2021).

SRK directed a sampling and analytical program to generate environmental geochemistry data for

tailings and waste rock for the existing operations and MDZ expansion project. Data from SRK’s

metallurgical program indicates that tailings will be discharged with a neutral to alkaline supernatant.

However, the tailings solids will be potentially acid generating (PAG) with the potential to eventually

exceed the alkaline supernatant and produce acidic drainage in the longer term. Detoxified cyanide

tailings are anticipated to have elevated concentrations of arsenic, sulfate, and total dissolved solids

in potential leachates. Testing on paste backfill tailings suggest that the material will be acid-

neutralizing in the short term, but in the long term, the material could become acidic. A waste rock

geochemical characterization program is in progress. An analytical program completed in 2012, in

support of the defunct open pit mine design, indicated that a significant fraction of waste rock could be

potentially acid generating. Effective management of both tailings and waste rock will be a critical issue

for success of the project.

Water balance modeling indicates the project is net positive and will continue to discharge excess

mine dewatering flows during some periods of the project. Infrequent discharges from facility surface

water management controls are also predicted. Based on water quality predictions and existing

infrastructure at the mine, additional water treatment facilities are not included in this study. However,

water treatment may be required dependent upon the outcome of ongoing geochemical studies.

SRK is not aware of any known environmental issues that could materially impact CGM’s ability to

extract the mineral resources or mineral reserves at the Marmato project. While there will be some

challenges associated with land acquisition and surface water control during operations, the Marmato

project has not had, nor does it currently have, any legal restrictions which affect access, title, mining

rights, or capacity to perform work on the property. Likewise, in regard to environmental compliance,

the operation is covered by the PMA and associated environmental permits, which further reduces

environmental risks.

1.10.2 Permitting

The Marmato Project is authorized under a number of resolutions issued by Corpocaldas in the name

of CGM’s predecessor, Mineros Nacionales S.A.S. These include, among others:

• Environmental Management Plan or PMA (Resolution No. 496)

• Various water concessions

• Discharge permits (Resolutions 270 modified by 254)

• Emissions permit (Resolution 270)

CGM is currently in the process of modifying the PMA to include a second DSTF area (Cascabel 2).

To this end, CGM has presented the impact assessment and technical documentation for this



modification to Corpocaldas for review. Corpocaldas has evaluated the request and is waiting for the

Ministry of the Interior to certify the presence, or not, of ethnic communities in the area of the new

facility in order to determine the need for prior consultations, before issuing its final decision. Once

Corpocaldas authorizes the Cascabel II modification, a new modification request will be submitted for

the phased construction of a DSTF2 to accommodate the UZ tailings during the development of the

full MDZ project and tailings capacity.

The PMA will require a major modification to allow for the proposed MDZ expansion project, which

envisions an increase in production in a second processing plant to be constructed. During

construction, Channel Occupancy Permits will need to be obtained for the new tailings site, the process

plant site, and the site of the underground portal (bocamina). Likewise, a Forest Exploitation Permit

will be needed for areas of proposed surface disturbance with trees.

The final environmental impact assessment deliverable includes the application for all the

environmental permits that will be required for the construction and operation phases of the project.

Once the EIA is officially delivered to Corpocaldas, the review process can begin based on the agreed-

upon terms of reference. CGM estimates that a minimum of six months will be required to review the

complete application and issuance of the Environmental License by Corpocaldas for the MDZ

expansion of the Marmato project. However, this process has been delayed as a result of the COVID-

19 pandemic and CGM does not anticipate fully reengaging Corpocaldas with the submittal of the EIA

in Q1 of 2021. The current timeline envisioned for the permitting of the project should be considered

to be aggressive and that permitting timeline expectations should be reviewed as the process begins.

In accordance with the terms and conditions of the PMA, CGM maintains an Environmental Insurance

Policy for the current operation. That policy is renewed annually with Corpocaldas as the beneficiary.

This policy is intended to cover the entire Marmato operations and all aspects of environmental

compliance. According to CGM, the current amount covered by the policy is COL$302,835,000

(USD$91,768). This amount will be reviewed and adjusted during the modification process of the PMA

for the MDZ expansion project.

1.10.3 Social or Community Related Requirements

The 2001 PMA for Marmato specifically requires the management of the social component of the

Project. Caldas is required to maintain records on all community activities (including number of

participants, topics, duration, etc.), which is to be turned over to Corpocaldas every six months as part

of the ongoing monitoring programs. As part of the social management and monitoring program, CGM

has developed a social investment model which seeks to promote the development of communities in

the area of influence, with the purpose of contributing to the consolidation of society and fostering

economic development (Economic Development), guaranteeing the care and respect for the

environment (Environmental Development) and supporting and participating in actions aimed at

improving the quality of life and well-being of its inhabitants (Social Development and Promotion of

Solidarity Actions).

1.10.4 Community Relations

Between 2014 and 2018, CGM developed and implemented a social engagement program at Marmato

specifically designed to focus on the well-being of the community and care for the environment. These

initiatives are incorporated in the Community Relations Plan (Plan de Relaciones con la Comunidad).



The Marmato Project currently operates with 152 administrative employees, 1,090 operating workers,

and 54 apprentice workers, most of whom are from the municipalities surrounding the project. With

the MDZ expansion, CGM anticipates hiring approximately 900 temporary workers during construction

and around 550 permanent employees as part of the new operations.

1.10.5 Mine Closure, Remediation, and Reclamation

Article 209 of Law 685 of 2001 requires that the concession holder, upon termination of the agreement,

shall undertake the necessary environmental measures for the proper reclamation and closure of the

mining operation. To ensure that these activities are carried out, the Environmental Insurance Policy

shall remain in effect for three years from the date of termination of the contract. While a formal closure

plan is not legally required at this stage of the operation, currently there is a closure plan for Marmato,

Plan de Cierre y Abandono de Mina La Maruja – Gran Colombia Gold Marmato S.A.S. (May 2019)

which discusses basic reclamation and closure actions including aspects of temporary, progressive,

and final closure. Reclamation and closure costs for the current operation provided in the closure plan

are based on percentages of costs to build the facilities. SRK did not independently calculate or

validate this estimate however, it is within keeping of other moderate-sized underground mining

operations in South America. The reclamation and closure cost for the existing mine plan is estimated

to be COL$20,128,000,000 (US$6.1 million based on exchange rate of 3,300 to 1). A requirement for

long-term post-closure water treatment, if deemed necessary, could increase this estimate.

Using first principles and the Nevada-developed Standardized Reclamation Cost Estimator, local

equipment and labor rates, and based on limited PFS engineering design information and drawings

for the MDZ expansion project, an additional cost of US$3.1 million was included in the technical

economic model to account for the increase in production anticipated for the new operations and the

construction of a new plant and tailings storage facilities. These are actual reclamation activity cost

estimates rather than percentages of construction costs. SRK strongly recommends that a more

detailed and thorough calculation of closure costs be prepared for the next level of study, looking at

both the existing facilities and planned expansion. Again, long-term post closure water treatment

requirements, if necessary, could significantly increase this estimate. This too should be more closely

examined during the next study phase.

1.11 Capital and Operating Costs

1.11.1 Marmato UZ Capital Costs

The Marmato UZ is a currently operating underground mine. The estimate of capital expenditures

(capex) includes expansion capex to increase the mineral processing capacity and sustaining capex

to maintain the equipment and all supporting infrastructure necessary to continue operations until the

end of the projected production schedule. The estimate conforms to Class 4 guidelines for a PFS level

estimate with a ±25% accuracy according to the Association for the Advancement of Cost Engineering

International (AACE International). The capital cost estimate is presented in Q2 2020 US Dollars

(US$). The estimate includes processing, maintenance, general and administration (G&A) and

accommodations costs.

The sustaining capital cost estimates developed for the UZ includes the costs associated with the

engineering, procurement, construction and commissioning. The cost estimate is based on budgetary

estimates prepared by Marmato and reviewed by SRK. The estimate indicates that the Project requires



sustaining capital of US$54.8 million to support the projected production schedule throughout the LoM.

Table 1-4 summarizes the LoM sustaining capital estimate, Table 1-5 and Table 1-6 present the same

estimate by year.

Table 1-4: Marmato UZ Sustaining Capital (LoM)

Description LoM (US$)

Upper Zone Infill Drilling 11,847,000

Upper Zone Development 6,396,225

Upper Zone Mine Sustaining 10,049,860

Upper Zone Plant Expansion 11,626,000

Upper Zone Plant Expansion Contingency 2,906,500

Upper Zone Plant Sustaining 3,600,000

Upper Zone Dewatering 2,275,706

Closure Costs 6,100,000

Total $54,801,292

Source: CGM/SRK, 2020

Table 1-5: Marmato UZ Sustaining Capital (2020 to 2026) (US$)

Description 2020 2021 2022 2023 2024 2025 2026

Infill Drilling 2,200,000 2,200,000 2,200,000 2,200,000 2,200,000 121,000 121,000

Development 1,187,325 2,998,025 1,986,625 224,250 - - -

Mine Sustaining 2,127,399 1,777,862 1,852,400 3,858,800 - 154,000 279,400

Plant Expansion 5,035,000 3,511,000 1,210,000 440,000 1,430,000 - -

Plant Expansion Contingency

1,258,750 877,750 302,500 110,000 357,500 - -

Plant Sustaining 300,000 300,000 300,000 300,000 300,000 300,000 300,000

Dewatering 135,000 713,569 1,427,137 - - - -

Closure Costs - - - - - - -

Total $12,243,474 $12,378,206 $9,278,662 $7,133,050 $4,287,500 $575,000 $700,400



Description 2027 2028 2029 2030 2031 2032

Infill Drilling 121,000 121,000 121,000 121,000 121,000 -

Development - - - - - -

Mine Sustaining - - - - - -

Plant Expansion - - - - - -

Plant Expansion Contingency - - - - - -

Plant Sustaining 300,000 300,000 300,000 300,000 300,000 -

Dewatering - - - - - -

Closure Costs - - - - - 6,100,000

Total $421,000 $421,000 $421,000 $421,000 $421,000 $6,100,000


1.11.2 MDZ Capital Costs

The MDZ is a lower part of the deposit that is undeveloped. Before CGM can exploit this part of the

deposit it will have to expand the existing operation. The expansion is planned to be executed between

the years of 2021 and 2023.

The capital cost estimates prepared for the expansion into the MDZ area also include estimates for

Engineering, Procurement and Construction Management (EPCM) and the Owner’s cost to manage

it. The cost estimate is based on cost models prepared by SRK and Ausenco with site specific inputs

from CGM. The estimate indicates that the expansion will require an investment of US$269.4 million,



this includes an estimated capital of US$237.2 million plus 13.6% contingency of US$32.2 million.

Table 1-7 summarizes the expansion capital estimate.

Table 1-7: MDZ Construction Capital (US$)

Description LoM 2020 2021 2022 2023

Development 19,719,753 - 2,279,534 10,343,401 7,096,818

Mining Equipment Purchases 52,430,929 - 16,868,012 15,295,229 20,267,688

Mining Services 11,589,225 - 1,288,744 6,206,511 4,093,970

Infrastructure 33,201,830 - 16,600,915 16,600,915 -

Process Plant 42,371,769 - 21,185,884 21,185,884 -

DSTF 19,660,473 - 17,212,986 1,279,528 1,167,958

Temporary Power Line 272,727 - 272,727 - -

Mining EPCM 9,276,559 - 2,883,922 4,999,126 1,393,512

Mining Owner's 15,721,708 - 3,978,018 7,881,638 3,862,053

Infrastructure + Plant EPCM 10,484,229 - 5,242,114 5,242,114 -

Infrastructure + Plant Owner's 13,602,581 1,087,625 4,663,472 5,298,567 2,552,917

Infrastructure + Plant Other Indirect 8,860,555 - 4,430,278 4,430,278 -

Sub-Total 237,192,337 1,087,625 96,906,605 98,763,190 40,434,916

Mining Contingency 15,091,967 - 2,508,648 5,950,365 6,632,954

Plant + Infrastructure Contingency 14,237,757 - 7,118,879 7,118,879 -

DSTF Contingency 2,871,944 - 2,581,948 191,929 98,067

Total Contingencies (13.6%) 32,201,668 - 12,209,474 13,261,173 6,731,021

Total $269,394,005 $1,087,625 $109,116,079 $112,024,363 $47,165,937

Source: CGM/Ausenco/SRK, 2020

The MDZ will require sustaining capital to maintain the equipment and all supporting infrastructure

necessary to continue operations until the end of its projected production schedule. The sustaining

capital cost estimate developed for this mining area includes the costs associated with the engineering,

procurement, construction and commissioning. The cost estimate is based on PFS designs and cost

models prepared by SRK with site specific inputs from CGM. The estimates indicate that the Project

requires sustaining capital of US$131.3 million to support the projected production schedule through

the LoM. Table 1-8 summarizes the LoM sustaining capital estimate and Table 1-9 and Table 1-10

present the same estimate by year.

Table 1-8: MDZ Sustaining Capital (LoM)


Drilling -

Development 34,285,846

Mine Equipment Purchases 17,166,844

Mine Equipment Rebuilds 26,862,004

Mining Owner's Cost 5,892,624

Mining Contingency 14,671,389

DSTF Sustaining 23,806,666

115kV Power Line 5,614,521


Total $131,299,895




Table 1-9: MDZ Sustaining Capital (2023 to 2027) (US$)

Description 2023 2024 2025 2026 2027

Drilling - - - - -

Development 2,735,635 4,986,400 3,834,632 2,168,451 3,433,459

Mine Equipment Purchases 6,646,459 3,972,308 - - 1,186,305

Mine Equipment Rebuilds - 1,162,732 2,300,471 4,985,557 4,601,468

Mining Services - - - - -

Mining Owner's Cost 1,689,596 943,372 402,871 227,366 487,479

Mining Contingency 1,322,664 1,704,601 1,307,595 1,476,275 1,799,109

DSTF Sustaining 6,817,007 21,934 64,320 15,054 13,150,673

115kV Power Line 280,726 561,452 561,452 561,452 561,452

Closure Costs - - - - -

Total $19,492,087 $13,352,799 $8,471,341 $9,434,155 $25,219,945


The sustaining capital cost estimate to support the 115kV power line was in fact estimated as a total

cost of US$3.24 million. This cost estimate was converted to a loan payment program that considers

a 10 year payment schedule and an 11.5% yearly interest rate. Each individual payment is calculated

to be approximately US$561,452.

Table 1-10: MDZ Sustaining Capital (2028 to 2033)

Description 2028 2029 2030 2031 2032 2033

Drilling - - - - - -

Development 6,412,653 3,918,836 3,897,980 2,467,849 429,949 -

Mine Equipment Purchases 208,000 4,232,979 920,793 - - -

Mine Equipment Rebuilds 681,459 4,291,695 2,278,851 6,399,725 160,047 -

Mining Services - - - - - -

Mining Owner's Cost 454,151 875,725 507,741 258,506 45,817 -

Mining Contingency 1,540,853 2,184,960 1,382,954 1,825,216 127,163 -

DSTF Sustaining 166,510 2,714,184 502,892 166,510 187,582 -

115kV Power Line 561,452 561,452 561,452 561,452 561,452 280,726

Closure Costs - - - - - 3,000,000

Total $10,025,079 $18,779,831 $10,052,664 $11,679,257 $1,512,011 $3,280,726


1.11.3 Marmato Operating Costs

SRK, Ausenco and CGM prepared the estimate of operating costs for the PFS production schedule.

Marmato UZ LoM cost estimate is presented in Table 1-11 and MDZ LoM cost estimate is presented

in Table 1-12.

Table 1-11: UZ Operating Costs Summary

Description LoM (US$/t-Ore) LoM (US$000’s)

Mining 48.45 249,251

Process 12.07 62,082

G&A 13.82 71,086

Total Operating $74.33 $382,419

Source: CGM/SRK/Ausenco, 2020



Table 1-12: MDZ Operating Costs Summary


Mining 35.19 512,288

Process 13.68 199,113

G&A 8.23 119,771



1.12 Economic Analysis

The valuation results of the Marmato Project indicate that is has an after-tax IRR of 19.5% and an

after-tax Net Present Value (NPV) of approximately US$256.1 million, based on a 5% discount rate

and gold and silver prices of US$1,400/oz and US$17.00/oz respectively. The cash flow profile also

shows a shorter payback for the investment when comparing to a stand-alone MDZ operation, to the

combine operations present a payback within the year of 2026, while a stand-alone MDZ operations

would present a payback in the year of 2027. The operation is projected to have negative cash flows

between the years 2020 and 2023, when the MDZ is installed, with payback for the expansion expected

by 2026. The annual free cash flow profile of the Project is presented in Figure 1-5.

Source: SRK, 2020

Figure 1-5: Marmato After-Tax Free Cash Flow, Capital and Metal Production

Indicative economic results are presented in Table 1-13. The Project can be considered a gold

operation with a sub-product of silver, where gold represents 99% of the total projected revenue and

silver the remaining 1%. The underground mining cost is the heaviest burden on the operation

representing 62% of the operating cost, as presented in Figure 1-6.



Source: SRK, 2020

Figure 1-6: Marmato Operating Cost Break-Down

Table 1-13: Marmato Indicative Economic Results

LoM Cash Flow (Unfinanced)

Total Revenue USD 2,625,861,238

Mining Cost USD (761,539,531)

Processing Cost USD (270,396,073)

G&A Cost USD (190,857,579)

Total Opex USD (1,222,793,183)

Operating Margin USD 1,403,068,055

Operating Margin Ratio % 53%

Taxes Paid USD (210,374,619)

Free Cashflow (before initial capital) USD 760,268,116

Before Tax

Free Cash Flow USD 701,248,730

NPV @ 5% USD 396,654,830

NPV @ 8% USD 279,571,263

NPV @ 10% USD 219,652,793

IRR % 26%

After Tax


NPV @ 5% USD 256,075,253

NPV @ 8% USD 167,009,205

NPV @ 10% USD 121,855,455

IRR % 19.5%

Payback Year 2026

Source: SRK, 2020



The estimated All-in Sustaining Costs (AISC), including sustaining capital, is US$880/Au-oz.

Table 1-14 presents the breakdown of the Marmato AISC.

Table 1-14: LOM All-in Sustaining Cost Breakdown

LOM All-in Sustaining Cost Breakdown

Mining US$/Au-oz 408

Processing US$/Au-oz 145

G&A US$/Au-oz 102

Refining US$/Au-oz 6

Royalty US$/Au-oz 130

Sustaining Capital US$/Au-oz 102

Silver Credit US$/Au-oz (14)

AISC US$/Au-oz 880

SRK’s standard Cash Cost reporting methodology for NI 43-101 reports includes smelting/refining costs; whereas CGM’s basis of reporting treats these costs as a reduction of realized gold price (the refinery discounts the selling price by a factor to cover these charges) and excludes them from its reported “total cash cost per ounce”. Source: SRK, 2020

1.13 Conclusions and Recommendations

1.13.1 Property Description and Ownership

SRK noted within the transfer of licenses from the previous owner there is a gap between the existing

licenses for #014-84M and RPP-357. This ground was under application from CGM with the Colombian

government for formal approval to continue mining. SRK reviewed the application within the

government website and noted that the status is defined as “in progress”, which has been the status

since September 30, 2009. The Company has taken steps to get the approval finalized. It is SRK’s

understanding that at the time of writing CGM has received notification (May 2020) to continue mining

in this area and that under the new Colombia mining license coding, the government does not consider

the gap to be present. SRK has not completed sufficient work to confirm this but would highlight that

it should be resolved and enable additional material to be used in mine plans for future studies.

In 2017 CGM began the process and submitted to the government the application for the license

extension to the current operation and future exploration for license #014-89, with the original license

currently held to October 2021. The process is expected to be completed in Q4 2020.

1.13.2 Geology and Mineralization

SRK produced an updated 3D geological model for the Marmato deposit as part of the current study.

SRK considers this to have increased the confidence in the spatial location of the various geological

units. CGM geologists as part of the on-going exploration continue to develop the geological

knowledge on the project and have supplied additional fault information which should be integrated

into further lithological models. SRK does not consider these faults to have a material impact on the

current mineral resource estimate but notes that it may impact future underground infrastructure (such

as a decline).



1.13.3 Status of Exploration, Development and Operations

The databases comprise a combination of historical and recent diamond core and underground

channel samples. In total, there are some 1,317 diamond drillholes for a combined length of

266,390 m; plus 24,824 individual underground channel samples, inclusive of current mine sampling

contained in the databases

SRK is of the opinion that the exploration and assay data is sufficiently reliable to support evaluation

and classification of Mineral Resources in accordance with generally accepted CIM Estimation of

Mineral Resource and Mineral Reserve Best Practices Guidelines (2014).

SRK notes that CGM exploration continues at the project throughout 2020 and SRK has reviewed the

2020/2021 drilling plan. The drilling is targeting mineralization in the hanging wall of the current

estimate which is referred to by CGM as the New Zone, which may impact on current mining

infrastructure if further mineralization is located, which may require modifications to the current mine

design. SRK therefore recommends that the geological model and mineral resource should be updated

to reflect the new drilling upon completion as the impact of these in future models may impact the

design prior to construction

1.13.4 Mineral Processing and Metallurgical Testing

Native gold is the predominant gold carrier and over 99% of the gold particles occurred within mineral

structures that would be readily accessible by leaching solutions.

The PFS metallurgical program optimized process parameters required to recover gold and silver

values from MDZ ore using a process flowsheet that includes gravity concentration followed by

cyanidation of the gravity tailing.

Comminution tests demonstrated that the MDZ ore is classified as hard with regard to impact breakage

and grinding characteristics.

Overall gold recovery is estimated at 95% and overall silver recovery is estimated at 51%. There is

little difference in reported gold recoveries for the master and variability composites and gold recovery

appears to be independent of ore grade over the range tested.

Cyanide destruction tests demonstrated that weak acid dissociable cyanide (CNWAD) could be reduced

to less than 10 mg/L with the SO2/air process. However, CNWAD levels will further attenuate to less

than 1 mg/L with time.

Pressure filtration will be required to dewater thickened tailings in order to achieve less than 15%

moisture content required for disposal in a DSTF.

1.13.5 Mineral Resource Estimate

The resource evaluation work was completed by Mr. Benjamin Parsons, MAusIMM (CP#222568). The

effective date of the Mineral Resource Statement is March 17, 2020, which is the last date assays and

the surveyed depletion outlines were provided to SRK.

SRK has produced block models using Datamine™. The procedure involved import from

Leapfrog™Geo of wireframe models for the fault networks, veins, definition of resource domains (high-

grade sub-domains), data conditioning (compositing and capping) for statistical analysis, geostatistical

analysis, variography, block modelling and grade interpolation followed by validation. Grade estimation



for the veins has been based on block dimensions of 5 m by 10 m by 10 m for the Porphyry and MDZ

units. Sub-blocking to 0.5 m by 1 m by 1 m has been allowed to reflect the narrow nature of the

geological model. The block size reflects the relatively close-spaced underground channel sampling

and spacing within veins compared to the wider drilling spacing, with the narrower block size used in

the MDZ at depth to reflect the proposed geometry of the mineralization (steeply dipping).

SRK is of the opinion that the MRE has been conducted in a manner consistent with industry best

practices and that the data and information supporting the stated mineral resources is sufficient for

declaration of Measured, Indicated and Inferred classifications of resources. SRK considers the veins

(including splays) and the MDZ to be of sufficient confidence for use in a mining study but recommends

further work on the short scale variability within the porphyry be completed to confirm the current

interpretation within areas of the existing mining infrastructure prior to use in any mining studies.

1.13.6 Mining and Reserves

UZ Mine Design

CaF is the current mining method used for the Veins and is appropriate for the deposit geometry. A

modified longhole stoping method will be used for the Transition zone to take advantage of the bulk

characteristics of the deposit.

Stope optimizations were run using a minimum CoG of 2.23 g/t Au for the Veins and 1.91 g/t Au for

the Transition zone.

Access to the Veins is already established. Primary haulage is on level 18 and material from levels

above is transferred down via existing ore passes. Material below level 18 is transported up via an

incline or via the apiques. The main production apique is at level 22, a secondary production apique

is at level 20 and will extend down to level 22.

The Transition zone is accessed via level 21 and level 22. A ramp will also connect the two levels as

a secondary egress and ventilation exhaust.

Tonnage and grades presented in the reserve include dilution and recovery. Productivities are based

on the current mine productivities

A quarterly/yearly production schedule was generated using iGantt software. The schedule targeted

1,500 t/d with a gradual ramp up to meet the upgraded mill capacity. There is also a 2 Mt/y permit limit

of moved material, which limits the production of the UZ.

MDZ Mine Design

Longhole stoping is an appropriate mining method for the deposit geometry. Stopes are sized to be

large enough to take advantage of bulk mining methods, yet small enough to maintain stability and

minimize dilution.

Optimizations were run using various CoG to identify higher grade mining areas and understand the

sensitivity of the deposit to CoG. Results show large quantities of lower grade material where a small

increase/decrease in CoG has a material impact on the quantity of economic material available for

design. A minimum CoG of 1.61 g/t Au was used for design/reserve. Higher grade stopes based on

3.5 g/t stope optimization results were designed as a first pass, with the lower grade stopes added as

separate stopes. This allowed for scheduling of higher grade stopes first.



The MDZ is accessed through a decline drift with conveyor. Tonnage and grades presented in the

reserve include dilution and recovery and are benchmarked to other similar operations. Productivities

were generated from first principles with inputs from mining contractors, blasting suppliers, and

equipment vendors where appropriate. The productivities were also benchmarked to similar

operations. Equipment used in this study is standard equipment used world-wide with only standard

package/automation features.


4,000 t/d.

Geotechnical

The geotechnical investigation, laboratory tests and design are suitable for a PFS project level design.

The proposed design parameters are acceptable for a PFS study only.

Empirical charts suggest that the side walls are located in unsupported transition zones, which could

require some spot ground support for potential wedge formations depending on discontinuity

persistence/continuity.

SRK used the Bieniawski, 1993, empirical chart to estimate the open stope stand-up time. A 10 m

span stope can likely be open for one to six months without ground support.

Dilution was estimated using the empirical Clark and Pakalnis (1997) method. The thickness of

external dilution is estimated as Equivalent Linear Overbreak/Slough (ELOS). The ELOS charts

indicate that significant dilution is unlikely due to the good rock mass quality (RMQ). Wall damage

would likely be associated with blasting overbreak. SRK considerers it relevant to conduct a blasting

study during the FS to evaluate the degree of overbreak.

To estimate the backfill strength requirements, SRK applied the Mitchell et al, 1982 analytic solution

which suggests that a backfill uniaxial compressive strength (UCS) of 1 megapascals (MPa) will be

adequate to maintain backfill stability and prevent backfill from sloughing into the open stope.

Negligible wall sloughing is anticipated.

Hydrogeology

The 3D groundwater flow model for the Marmato project was developed, reasonably calibrated to

available measured water level and groundwater flow data, and used to make predictive simulations

of:

• Passive inflow to the existing and planned deep underground mines

• Propagation of drawdown during proposed dewatering during mining

• Changes in groundwater discharge to rivers and creeks during mining

The model predicts that:

• The majority of inflow to the planned mine (up to 78 L/s with a possible range from 56 to

159 L/s) is expected from the upper levels above 730 m, where elevated hydraulic conductivity

values of bedrock groundwater system were measured.

• Mine inflow to the MDZ planned mine below 730 m is predicted to be lower (15 L/s with upper

limit of 34 L/s) due to reduced measured hydraulic conductivity with depth.

• The total maximum planned mine discharge is predicted to be up to 88 L/s, with a possible

range from 61 to 167 L/s.



• Total maximum discharge into the entire mine complex, including flow to existing mine levels,

is predicted to be up to 111 L/s, with a possible range from 89 to 168 L/s.

• Major sources of mine inflow are depletion of groundwater storage and capturing of

groundwater discharge to surface water bodies (i.e., streams). The model does not predict

reversing of hydraulic gradient between the mine area and the Cauca River and does not

predict inflow to the mine from the river. However, further investigation of the structures and

their hydrogeological role are needed to verify this conclusion.

• Lowering of the water table in the mine area of up to 140 m and drawdown propagation of up

to 2 km away from the mine, assuming a 10-m drawdown extent

In SRK’s opinion, the completed predictions are conservative, given the following:

• The model is based on extrapolation of the measured hydraulic conductivity values in mine

area for entire model domain, including topographic highs areas outside of the mine area,

where measured water levels are high and hydraulic conductivity values are most likely lower

than in the mine area.

• The model uses high recharge from precipitation to calibrate the model to measured water

levels, combined with geomean hydraulic conductivity values in discrete depth intervals that

are derived from measured hydraulic conductivity values in the mine area.

• The model uses calibrated conductance values that reproduce measured inflow to the existing,

relatively shallow mine for simulation of groundwater inflow to the deep underground

developments of the planned mine.

• The model simulates no restriction of groundwater inflow to the backfilled stopes for Base

Case and Maximum Inflow scenarios.

The completed analysis of available hydrogeological data and numerical groundwater modeling

indicate that several uncertainties remain in understanding of the hydrogeology, including

hydrogeological role of the faults, hydraulic properties of bedrock outside of the mine area, recharge

estimates, spatial and vertical distribution of groundwater inflow to the current mine, water table

elevation, and water level changes due to passive mine dewatering and seasonal changes in

precipitation.

To reduce these uncertainties, SRK recommends completing the following additional hydrogeological

investigations/analyses for the FS:

• Structural analysis of the geological features and faults outside of the mining area, with

emphasis on potential connection to the Cauca River

• Detailed water balance and estimate of recharge from precipitation

• Detailed groundwater inflow mapping in existing developments

• Evaluation of the role of backfilling in reduction of groundwater inflow to the mine

• Improvement of mine discharge measurements at each level of the existing mine

• Re-survey existing monitoring locations, with emphasis on ground and collar elevations

• Installation of groundwater level monitoring network outside of mine area and along the river

valley, including hydrogeological testing during construction of monitoring wells

• Detailed water level measurements to observe:

o Drawdown propagation as result of mine dewatering

o Seasonal variation as result of precipitation



• Additional large-scale hydraulic testing to identify zone of enhanced permeability related to

Fault 2 (in areas where planned conveyor decline and egress ramp plan to intersect this fault

at multiple locations/elevations) and Fault 1-3 (intersects planned stopes in multiple

elevations). In addition, test the S. Ines Fault (intersects the planned stopes in the upper levels

and part of the egress ramp)

• Drilling and hydraulic testing of pilot holes in places where ventilation declines are planned

• Updates to the developed numerical groundwater model based on above items to improve its

predictability:

o Better calibration of the model to water levels for future pore pressure predictions

o Re-evaluation of pumping design based on updated inflow predictions

o Evaluation of flow-through hydrogeological conditions during post-mining

• Groundwater chemistry sampling

1.13.7 Recovery Methods

An ore processing plant has been designed to process MDZ ore at the rate of 4,000 t/d using

conventional processes that are standard to the industry including: primary and secondary crushing,

SAG/ball mill grinding, gravity concentration, agitated cyanide leaching, carbon-in-pulp (CIP), gold

elution, electrowinning and smelting to produce a final doré product.

1.13.8 Project Infrastructure

The existing infrastructure for the UZ operations is established and meets the project requirements.

The addition of the water supply pumping system from the Cauca River will address potential water

sourcing issues during drought seasons.

The new MDZ infrastructure includes the required access, power supply, water supply, tailings storage,

and support facilities to support the production of 4,000 t/d from the new plant and mine.

A full understanding of the mine water and DSTF water requirements and runoff will allow for

optimization of the site runoff pond and water treatment capacities.

Tailings Management Facility

SRK advanced the conceptual designs of DSTF 2 and DSTF 1 to a level sufficient for cost estimating.

The designs include consideration of the following specific elements:

• Subgrade preparation include topsoil salvaging, removal of unsuitable material and excavation

of stability benches and embankment keys

• Construction of rockfill starter embankments using a combination of imported and on-site

borrow

• Construction of underdrain network and underdrain flow management

• Construction of seepage collection drains on dry stack benches and seepage management

systems

• Construction of stormwater diversion and control channels

• Management of contact stormwater on dry stack top deck and return to process

• Access and haul roads between plant and DSTF 2 and DSTF 1

• Temporary storage area for filtered tailings

• Temporary holding pond for non-filtered tailings



• Topsoil and unsuitable soil stockpile area with underdrainage system

Currently identified risks and opportunities with respect to the costs developed for the PFS have been

identified in relation to the following:

• The inability to characterize the foundation conditions beneath the conceptual DSTF

footprints.

• Ongoing geochemical characterization of both waste rock and ore/tailings indicating some of

the waste rock and tailings may be acid generating and therefore require special management

considerations.

• Immediate characterization and analysis of Cascabels 1 and 2 to demonstrate compliance

with internationally accepted standards of practice and provide for tailings management

through commissioning of a new DSTF.

• More extensive testing of tailings to confirm tailings geotechnical characteristics and cement

addition requirements.

• Stormwater maintenance requirements at both DSTF 1 and DSTF 2 constitute higher costs

through operations and closure than is currently allowed for in the PFS costs.

1.13.9 Environmental Studies and Permitting

The following interpretations and conclusions have been drawn with respect to the currently available

information provided for the Marmato Project:

• Environmental Studies: Baseline studies have been completed or are currently underway

with respect to the existing facilities (additional tailings storage capacity request) and MDZ

proposed expansion. These resource studies will be used for impact analysis and the

development of mitigation actions and environmental management planning.

• Environmental and Social Management: Environmental and social issues are currently

managed in accordance with the approved PMA and will likely need to be updated and/or

modified for the proposed MDZ expansion project.

• Monitoring: Routine monitoring is currently conducted on seven domestic wastewater

discharges and three non-domestic (industrial) wastewater discharges. Air quality emissions

from the metallurgical laboratory and smelter are also monitored for: particulate matter (PM),

sulphur dioxide (SO2) nitrogen oxides (NOX) and lead (Pb). The tailings are infrequently

monitored for hazard classification purposes through a Corrosive, Reactive, Explosive, Toxic,

Inflammable, Pathogen [biological] (CRETIP) program. The results of the monitoring are

provided to Corpocaldas. This monitoring program will require significant modification to

include the facilities for the proposed MDZ expansion project, and to bring it up to international

best practice standards.

• Geochemistry: Acid-generating sulfide minerals identified in the deposit include pyrite,

arsenopyrite, iron-bearing sphalerite, pyrrhotite, and chalcopyrite (SRK, 2017). Samples of

groundwater discharging into the underground are predominantly acidic. The underground

water samples contain elevated metal(loid) concentrations. While the tailings will be

discharged with a neutral to alkaline supernatant, the tailings themselves will be potentially

acid generating (PAG) with the potential to eventually overwhelm the alkaline supernatant and

produce acid drainage in the long term. A waste rock analytical program completed in 2012 in



support of an open mine design indicated that a significant fraction of waste rock could be

potentially acid generating (KP, 2012).

• Permitting: Operations are permitted through the posting of an Environmental Management

Plan (PMA) and secondary permits for use of water abstraction, forest use, air emissions,

discharges and river course (channel) construction. The PMA for the current operations was

originally approved in 2001. Minor modification of the PMA (including and environmental

impact analysis) is currently underway as part of the request for additional tailings storage

areas. Major modification of the PMA will be required for the MDZ expansion project.

• Stakeholder Engagement: Company has conducted extensive stakeholder identification and

analysis programs and has set stakeholder engagement objectives and goals to develop

communications plans with government, community, media and small miners but the company

does not currently have a formal stakeholder engagement plan.

• Closure Costs: The reclamation and closure cost estimate provided for the current operations

is approximately US$6.1 million, though there is considerable uncertainty surrounding the

basis for this estimate. An additional US$3.1 million is estimated for the MDZ expansion

facilities (assuming concurrent taili8ngs reclamation), for a total of US$9.2 million. A

requirement for long-term post-closure water treatment, if any, could significantly increase this

estimate.

There do not appear to be any other known environmental issues that could materially impact CGM’s

ability to conduct mining and milling activities at the site. Preliminary mitigation strategies have been

developed to reduce environmental impacts to meet regulatory requirements and the conditions of the

PMA.

Recommendations

Environmental Studies and Permitting

The following recommendations are made with respect to environmental, permitting and social issues

regarding the Marmato Project:

Prepare a more detailed site-wide closure plan for the existing Marmato facilities, including building

plans and equipment inventories) from which a more accurate final closure cost estimate can be

developed.

Continue work on groundwater hydrogeology and surface water to better define the risk associated

with potential groundwater contamination and underground dewatering impacts. A detailed evaluation,

including a groundwater model, could provide information that would assist in forecasts of post-closure

mine water discharge and possible long-term water treatment requirements. Such an investigation

could also provide vital information on underground geotechnical stability, both during operations and

post closure.

Characterization work should be completed on artisanal tailings and waste rock to understand their

Acid Rock Drainage Medal Leaching (ARDML) potential and devise a long-term management plan.

A comprehensive baseline surface and groundwater sampling program will be important to establish

the baseline condition and try to quantify the contributions from artisanal or pre-mining conditions,

especially with respect to mercury from artisanal mining.



Substantial financial resources and technical specialist support will be required to implement the

environmental monitoring and mitigation measures likely to be presented in the updated PMA for the

expansion project.

1.13.10 Capital and Operating Costs

Marmato UZ is a currently operating underground mine, the estimate of capital includes some

expansion capex to increase the mineral processing capacity and sustaining capital to maintain the

equipment and all supporting infrastructure necessary to continue operations until the end of the

projected production schedule. The estimate prepared for this study indicates that the Project requires

a sustaining capital of US$54.8 million to support the projected production schedule throughout the

LoM.



the years of 2021 and 2023. The cost estimate indicates that the expansion will require an investment

of US$269.4 million, this includes an estimated capital of US$237.2 million plus 13.6% contingency of

US$32.2 million.

Ausenco prepared a detailed cost estimate for the MDZ mineral processing facility and other mine

infrastructure but did not prepare an annual expenditure schedule for this capital.

SRK, Ausenco and CGM prepared the estimate of operating costs for the PFS’s production schedule.

The estimated operating cost for the Marmato UZ is US$76.12/t-ore and for the MDZ is US$57.10/t-

ore

The estimated AISC, including sustaining capital, is US$880/Au-oz. Table 1-15 presents the

breakdown of the Marmato AISC.

Table 1-15: LoM All-in Sustaining Cost Breakdown

LoM All-in Sustaining Cost Breakdown

Mining USD/Au-oz 408

Processing USD/Au-oz 145

G&A USD/Au-oz 102

Refining USD/Au-oz 6

Royalty USD/Au-oz 130

Sustaining Capital USD/Au-oz 102

Silver Credit USD/Au-oz (14)

AISC USD/Au-oz 880


The following recommendations are made with respect to capital and operating costs of the Marmato

Project:

• Prepare first principles estimate of capital and operating costs with enough accuracy to

support future studies of the project, including:

o Prepare cash flow model based on shorter periods of production

o Prepare an expenditure curve for MDZ Mineral Processing and Site Infrastructure

construction costs



o Further detail the site-specific operating cost data and cost models to include fixed and

variable nature of costs and detail the cost models to include breakdown by area and

function

o Improve cost models to include currencies used to estimate each cost and prepare

sensitivity to currencies variability

1.13.11 Economic Analysis


after-tax NPV of approximately US$256.1 million, based on a 5% discount rate and gold and silver

prices of US$1,400/oz and US$17.00/oz respectively. The cash flow profile also shows a shorter

payback for the investment required for the MDZ, bringing it back about a year to 2026. The operation

is projected to have negative cash flows between the years 2020 and 2023, when the MDZ is installed,

with payback for the expansion expected by 2026. LoM is projected to end in 2033 resulting in a total

production of 1.87 Moz of gold and 1.57 Moz of silver in the form of doré bars containing both precious

metals. Indicative economic results are presented in Table 1-16.






G&A Cost USD (190,857,579)






Before Tax


NPV @ 5% USD 396,654,830

NPV @ 8% USD 279,571,263

NPV @ 10% USD 219,652,793

IRR % 26%

After Tax


NPV @ 5% USD 256,075,253

NPV @ 8% USD 167,009,205

NPV @ 10% USD 121,855,455

IRR % 19.5%

Payback Year 2026

Source: SRK, 2020

The Project is a gold operation with a sub-product of silver, where gold represents 99% of the total

projected revenue and silver the remaining 1%. The underground mining cost is the heaviest burden

on the operation representing 62% of the operating cost, while processing costs represent 22% and

G&A costs the remaining 16%.



The following recommendations are made with respect to the economic evaluation of the Marmato

Project:

• The schedule prepared for Marmato UZ doesn’t fully utilize its mineral processing capacity for

several years of the life of mine. Investigate the possibility to expand the total mine movement

permit to allow Marmato UZ to process its run of mine using its plant at full capacity, as this

will very likely improve the overall project economics.



2 Introduction

2.1 Terms of Reference and Purpose of the Report

This report was prepared as a PFS Level Canadian National Instrument 43-101 (NI 43-101) Technical

Report (Technical Report) disclosing the findings for Caldas Gold Corp. (Caldas Gold), which indirectly

holds all of the shares of CGM, by SRK Consulting (U.S.), Inc. (SRK) on the Marmato Project, located

in Colombia. The Project consists of the current Marmato operating mine and the MDZ.

The quality of information, conclusions, and estimates contained herein is consistent with the level of

effort involved in SRK’s services, based on: i) information available at the time of preparation, ii) data

supplied by outside sources, and iii) the assumptions, conditions, and qualifications set forth in this

report. This report is intended for use by CGM subject to the terms and conditions of its contract with

SRK and relevant securities legislation. The contract permits CGM to file this report as a Technical

Report with Canadian securities regulatory authorities pursuant to NI 43-101, Standards of Disclosure

for Mineral Projects. Except for the purposes legislated under provincial securities law, any other uses

of this report by any third party is at that party’s sole risk. The responsibility for this disclosure remains

with CGM. The user of this document should ensure that this is the most recent Technical Report for

the property as it is not valid if a new Technical Report has been issued.

This report provides Mineral Resource and Mineral Reserve estimates, and a classification of

resources and reserves prepared in accordance with the Canadian Institute of Mining, Metallurgy and

Petroleum Standards on Mineral Resources and Reserves: Definitions and Guidelines, May 10, 2014

(CIM, 2014).

2.2 Qualifications of Consultants (SRK)

The Consultants preparing this technical report are specialists in the fields of geology, exploration,

Mineral Resource and Mineral Reserve estimation and classification, underground mining,

geotechnical, environmental, permitting, metallurgical testing, mineral processing, processing design,

capital and operating cost estimation, and mineral economics.

None of the Consultants or any associates employed in the preparation of this report has any beneficial

interest in CGM. The Consultants are not insiders, associates, or affiliates of CGM. The results of this

Technical Report are not dependent upon any prior agreements concerning the conclusions to be

reached, nor are there any undisclosed understandings concerning any future business dealings

between CGM and the Consultants. The Consultants are being paid a fee for their work in accordance

with normal professional consulting practice.

The following individuals, by virtue of their education, experience and professional association, are

considered Qualified Persons (QP) as defined in the NI 43-101 standard, for this report, and are

members in good standing of appropriate professional institutions. QP certificates of authors are

provided in Appendix A. All QP’s stated below are independent of the Company. The QP’s are

responsible for specific sections as follows:

• Ben Parsons, Principal Consultant (Resource Geologist) is the QP responsible for data

verification, preparation of the geological model and the mineral resource estimate. Sections

2 through 12 (except 4.4), 14, 23 and portions of Sections 1, 24, 25 and 26 summarized

therefrom, of this Technical Report.



• Eric Olin, Principal Consultant (Metallurgy) is the QP responsible for Metallurgy Sections 13,

17.1 17.2 and the Upper Zone processing portion of Section 21 and portions of Sections 1,

24, 25 and 26 summarized therefrom, of this Technical Report.

• Robert Raponi, Principal Metallurgist at Ausenco Engineering Canada Inc., is the QP

responsible for the MDZ process plant and infrastructure engineering and related portions of

Sections 17.3, 18.3 through 18.12, and the MDZ processing and infrastructure portions of

21.1.2, 21.3.2 and portions of Sections 1, 24, 25 and 26 summarized therefrom, of this

Technical Report.

• Fernando Rodrigues, Principal Consultant (Mining Engineering) is the QP responsible for

Upper Zone Mining and Economics and related portions of Section 15.1.1 through 15.1.4, and

the portions of Sections 15.2 and 15.3 pertaining to the Upper Zone, and Sections 16.1, 16.4,

portions of 16.6 pertaining to the Upper Zone, 19 and 22, and portions of Sections 1, 24, 25

and 26 summarized therefrom, of this Technical Report.

• Jeff Osborn, Principal Consultant (Mining Engineering) is the QP responsible for Infrastructure

and Cost Estimation Sections 18.1, 18.2, 18.13,18.16, and 21 (excluding processing and

tailings portions of Section 21), and portions of Sections 1, 24, 25 and 26 summarized


• Joanna Poeck, Principal Consultant (Mining Engineering) is the QP responsible for the

opening statement in Section 15 and portions of Section 15.1.5 through 15.1.8, and the

portions of Sections 15.2 and 15.3 pertaining to the MDZ, and Section 16.5, portions of 16.6

pertaining to the MDZ and portions of Sections 1, 24, 25 and 26 summarized therefrom, of this

Technical Report.

• Fredy Henriquez, Principal Consultant (Geotechnical Engineering) is the QP responsible for

Geotechnical Section 16.2 and portions of Sections 1, 24, 25 and 26 summarized therefrom,

of this Technical Report.

• Breese Burnley, Principal Consultant (Geotechnical Engineering) is the QP responsible for

Tailings Section 18.15, and the tailings portions of Section 21, and portions of Sections 1, 24,

25 and 26 summarized therefrom, of this Technical Report.

• Cristian Pereira, Senior Consultant (Hydrogeology) is the QP responsible for Hydrogeology

Section 16.3, and portions of Sections 1, 24, 25 and 26 summarized therefrom, of this

Technical Report.

• David Hoekstra, Principal Consultant (Water Resource Engineering) is the QP responsible for

Section 18.14, Hydrology Section 20.2.5, and portions of Sections 1, 24, 25 and 26

summarized therefrom, of this Technical Report.

• David Bird, Associate Principal Consultant (Geochemistry) is the QP responsible for

Geochemistry Section 20.1.3, and portions of Sections 1, 24, 25 and 26 summarized


• Mark Willow, Principal Consultant (Environmental) is the QP responsible for Section 4.4,

Environmental Section 20 (except section 20.1.3), and portions of Sections 1, 24, 25 and 26

summarized therefrom, of this Technical Report.



2.3 Details of Inspection

Table 2-1 presents a site visit summary. SRK was given full access to relevant data requested, and

conducted discussions with junior and senior CGM personnel regarding procedures and

interpretations.

Table 2-1: Site Visit Participants

Personnel Company Expertise Date(s) of Visit Details of Inspection

Ben Parsons SRK Mineral Resources

June 11 to June 13, 2019 August 17, 2017 and March 12 to March 14, 2012

Underground Site visit levels 18 – 21, review latest drilling intersections, Underground Site visit levels 17 – 20, review latest drilling intersections, Underground Site visit, review latest drilling intersections.

Eric Olin SRK Metallurgy December 17 and 18, 2019

Reviewed Marmato process operations and site locations for the MDZ process plant

Jeff Osborn SRK Mining/Infrastructure

July 16 to July 18, 2019 August 22 and 23, 2017

Surface Facilities and New MDZ location Underground and Surface Facilities including DSTF as well as core shack area

Fernando Rodrigues

SRK Mining/Reserves August 22 and 23, 2017

Underground and Surface Facilities including DSTF as well as core shack area

Fredy Henriquez SRK Geotechnical

January 8 and 11, 2020 July 16 to July 18, 2019

Underground Mine core shack area and the Tunnel proposed portal location

Mark Willow SRK Environmental December 1, 2016 Environmental Impact review

Cristian Pereira SRK Hydrogeology August 12 to August 13, 2019

Hydrogeology review



Personnel Company Expertise Date(s) of Visit Details of Inspection

Giovanny Ortiz SRK Geology/Mineral Resources

September 30 to October 2, 2019 February 10 to February 21, 2020

Review Veins model with CGM team Review Underground Sampling Process and Mapping procedures

Breese Burnley SRK Tailings January 28, 2020

Existing tailings facilities, Site 6 and Site 2 proposed for MDZ.

Source: SRK, 2020

2.4 Sources of Information

SRK’s opinion contained herein is based on information provided to SRK by CGM throughout the

course of its investigations. SRK has relied upon the work of other consultants in the project areas in

support of this Technical Report.

The Consultants used their experience to determine if the information from previous reports was

suitable for inclusion in this technical report and adjusted information that required amending. This

report includes technical information, which required subsequent calculations to derive subtotals, totals

and weighted averages. Such calculations inherently involve a degree of rounding and consequently

introduce a margin of error. Where these occur, the Consultants do not consider them to be material.

This report is based in part on internal Company technical reports, previous technical studies, maps,

published government reports, Company letters and memoranda, and public information as cited

throughout this report and listed in the References Section 27.

SRK has been supplied with numerous technical reports and historical technical files. SRK’s report is

based upon:

• Numerous technical review meetings held at CGM’s offices in Medellín, Colombia

• Discussions with directors, employees and consultants of the Company

• Data collected by the Company from historical exploration on the Project

• Access to key personnel within the Company, for discussion and enquiry

• A review of data collection procedures and protocols, including the methodologies applied in

determining assays and measurements

• Gran Colombia Gold Marmato S.A.S. for the site-specific closure plan and cost estimate

presented in Plan de Cierre y Abandono de Mina La Maruja (May 2019)



• Site Environmental Manager, Ing. Adrián Quintero Jiménez, and corporate Environmental

Manager, Erwin Wolff Carreño, for information on permits, monitoring programs and data, and

the environmental management budget estimate

• Knight Piésold (2012) for information on the geochemistry of the deposit

• Marmato’s exploration team provided geotechnical core logging and laboratory tests results;

• Geology and major faults were provided by Marmato’s exploration team.

• Existing reports provided to SRK, as follows:

o NI 43-101 Mineral Resource Estimate on the Marmato Project, Colombia, June 21, 2012

o NI 43-101 Mineral Resource Estimate on the Marmato Project, Colombia, June 16, 2017

o NI 43-101 Technical Report Preliminary Economic Assessment Marmato Project

Colombia, February 6, 2020

o Geochronology, Geochemistry and Magmatic-Hydrothermal Oxide Characterization of the

Marmato Gold Deposit, Colombia,

o Lead isotopic compositions of the gold mineralization of Marmato, Colombia:

Characterization of the transition domain in epithermal - porphyry systems

o Further Geological Observations on The Lower Zone Gold Deposit at Marmato, Colombia,

Richard H Sillitoe, July 2019

o Marmato Structural Geology Review (Memorandum), SRK Consulting (Canada) Inc,

March 11, 2020

• Data files provided by the Company to SRK as follows:

o Topographic grid data in digital format

o Drillhole database, including collar, survey, geology, and assay

o QA/QC data including details on duplicates, blanks and certified reference material (CRM)

o DXF files, including geological interpretation, vein domain digitized 2D section

interpretations, stope outlines and mined depletions

2.5 Effective Date

The effective date of this report is March 17, 2020.

2.6 Units of Measure

The metric system has been used throughout this report. Tonnes are metric of 1,000 kg, or 2,204.6 lb.

All currency is in U.S. dollars (US$) unless otherwise stated.



3 Reliance on Other Experts The Consultant’s opinion contained herein is based on information provided to the Consultants by

CGM throughout the course of the investigations.

SRK has not performed an independent verification of land title and tenure as summarized in Section 4

of this report. SRK did not verify the legality of any underlying agreement(s) that may exist concerning

the permits or other agreement(s) between third parties but have relied on the Company and its legal

advisor for land title issues. SRK has been supplied with a Legal Opinion by Dentons Cardenas and

Cardenas entitled “Legal_Opinion_Caldas_Finance_Corp”, which summarized the findings of their

review of CGM’s land title and tenure, upon which Dentons Cardenas and Cardenas have agreed SRK

can rely on for this disclosure.

These items have not been independently reviewed by SRK and SRK did not seek an independent

legal opinion of these items.

.



4 Property Description and Location

4.1 Property Location

The Marmato Project is located in the Municipality of Marmato, Department of Caldas, Republic of

Colombia and is approximately 125 km due south of the city of Medellín, the capital of the Department

of Antioquia (Figure 4-1).

The property sits between latitudes and longitudes 5°28’24”N and 5°28’55”N, and 75°34’46”W and

75°37’80”W, respectively; with altitudes ranging from approximately 200 to 1,705 meters (m). The

Project can be accessed from Medellín via paved roads on the Medellín to Cali highway (Route 25)

which forms part of the Pan America Highway.

Source: SRK, 2012

Figure 4-1: Location Map

4.2 Mineral Titles

The Marmato project area has historically been divided into three main zones with numerous license

boundaries defined within. What has traditionally been termed the Marmato project was made up of

three separate concessions (Figure 4-2), named Zona Alta (#CHG_081), Zona Baja (#014-89m) and

Echandia (#RPP_357), of which Zona Baja is 100% owned by CGM and Zona Alta and Echandia are

owned indirectly, through other subsidiaries, by Gran Colombia.



CGM currently holds the Zona Baja license (#014-89m) and has rights to continue mining the

neighboring Echandia license (#RPP_357) in which the Company conducts mining operations, and is

in the midst of completing a license to be granted by MAO, an indirect, wholly owned subsidiary of

Gran Colombia, to mine Levels 16 and 17 of Zona Alta (License #CHG_081). These are also referred

to as the CGM Mining Assets in this report.

Source: Mineros Nacionales, 2010

Figure 4-2: Land Tenure Map(s)

The horizontal division of mining rights at Marmato is unique in Colombia and was created in 1946 by

Law 66 to enable mining title contracts to be defined by horizontal mine levels. This is defined as an

Aporte Minero Mine (Mining Contribution 1017 for precious metals), which was granted in 1981. The

top of the Zona Baja is defined in Contract #014-89 with Mineros Nacionales S.A. (Mineros Nacionales)

and coincides with the road and varies from 1,207 m to 1,298.3 m in elevation.

The Zona Baja license lies below the Marmato Zona Alta property and is adjacent to Echandia. Zona

Baja extends east to the River Cauca. The license is bounded vertically by the Zona Alta and Cerro El

Burro in Marmato, but in the other parts it continues to surface. The license continues vertically to

depth in all parts.

The Zona Baja contract was owned by Mineros Nacionales, a private Colombian corporation which

was owned 94.5% by Mineros S.A. (Mineros), a Colombian corporation whose shares are traded on

the Colombian stock exchange (BVC – Bolsa de Valores de Colombia). The remaining 5.5% of

Mineros Nacionales was owned by a number of private and juridical persons. The contract registration

number is 014-89m and the mining title registration number is GAFL-11. It covers a surface area of

952.5830 hectares (ha). The Zona Baja contract was awarded to Mineros Nacionales, since renamed



CGM, in October 1991 and is valid for 30 years until October 2021. In October 2017, CGM commenced

the process to renew the contract for another 30-year term.

On February 15, 2010, Medoro Resources Ltd. (Medoro) acquired all of the issued and outstanding

ordinary shares of Mineros Nacionales S.A. from Mineros S.A., for total cash consideration of

US$35 million. With this acquisition, Medoro acquired 100% of Mineros Nacionales' interests in the

Zona Baja concession (the Zona Baja property). Medoro merged into Gran Colombia in 2011.

The Echandia property lies to the north east of the Zona Alta limit, and extends to depth. The Echandia

license has contract number RPP_357 and mining title registry number EDMN-01. The

Reconocimiento de Propiedad Privada (RPP) type of contract translates as Recognition of Private

Property. RPPs were created by Law 20 in 1969. The law respected prior mining and land rights and

required that proof be submitted of mining. Echandia is an old freehold property dating from the 19th

century. The RPP titles grant surface and subsurface rights in perpetuity.

Exploitation is required in order to maintain the validity of an RPP license. Mining on a relatively small

scale is being maintained in the area of contract number RPP 357 and the Company has an operating

contract permitting it to mine underground in this area.

Effective June 10, 2011, Gran Colombia completed a merger with Medoro and the combined company

continued under the name Gran Colombia. As a result, Gran Colombia acquired 100% of Medoro’s

interest in the Marmato project, including the Company’s license in Zona Baja.

On February 24, 2020, Caldas Gold completed its reverse takeover transaction (RTO Transaction)

with Caldas Finance Corp., an indirect wholly-owned subsidiary of Gran Colombia, pursuant to which

Caldas Gold, until then known as Bluenose Gold Corp., acquired the CGM Mining Assets through the

acquisition of all of the issued and outstanding shares of Gran Colombia’s newly incorporated, indirect

wholly-owned subsidiary, Caldas Finance Corp., which holds all of the issued and outstanding shares

of Caldas Gold Colombia Inc., a Panamanian company. Caldas Gold Colombia Inc. holds all of the

issued and outstanding shares of CGM, which in turn, holds all of the CGM Mining Assets included in

the RTO Transaction. The CGM Mining Assets principally comprise the existing producing

underground gold mine (#014-89m), the existing 1,200 t/d processing plant and the area

encompassing the MDZ mineralization, all located within the mining license area referred to as Zona

Baja. The CGM Mining Assets also include two contractual rights:

• One, granted by Croesus, an indirect, wholly owned subsidiary of Gran Colombia, to mine in

the lower portion of the Echandia license (#RPP_357) area

• Another, in the process of being completed, to be granted by MAO, an indirect, wholly owned

subsidiary of Gran Colombia, to mine Levels 16 and 17 of Zona Alta (License #CHG_081)

The CGM Mining Assets comprise the Marmato Project that is the subject of this report.

The purchase price for the CGM Mining Assets was CAD$57,500,200, satisfied through the issuance

to Gran Colombia (through an affiliate) of 28,750,100 common shares of Caldas Gold, CGM’s ultimate

parent, having a deemed price of CAD$2.00 per common share.

SRK noted within the transfer of licenses from the previous owner, there is a gap between the existing

licenses for #014-84M and RPP-357. This ground was under application from the CGM with the

Colombia government for formal approval to continue mining. SRK reviewed the application within the

government website and noted that the status is defined as “in progress”, which has been the status



since September 30, 2009. The Company has been taking steps to get the approval finalized. It is

SRK’s understanding that at the time of writing CGM has received notification (May 2020) to continue

mining in this area and that under the new Colombia mining license coding, the government does not

consider the gap to be present. SRK has not completed sufficient work to confirm this but would

highlight that it should be resolved and enable additional material to be used in mine plans for future

studies.

The exclusion zone (gap) was considered in the PFS where CGM ownership was not secured at the

beginning of the PFS work within the Mineral Resources. As the PFS was nearing completion CGM

informed SRK that the gap was no longer an issue, however a re-design to include the gap area was

not completed. For the reserves stated here, UZ development mining does go through the gap area.

A summary of the location of the area of concern is shown in Figure 4-3. It is expected that this will not

limit the current mining operation. SRK estimates within this area on a global basis for the Mineral

Resources approximately <5% of the Measured and Indicated Mineral Resources and approximately

3% of the Inferred material.

Level 1,050 m, area in green is under application but has been historically mined. Source: SRK, 2019

Figure 4-3: Summary of Gap in Licenses Within the Current Operations, with Associated Applications

Area under Application

No# KIU-11401 Original Application: 9/30/2009Status: Under Review RPP_357

014_89m



4.2.1 Nature and Extent of Issuer’s Interest

In regard to surface rights at the Project, CGM compiled a GIS database of surface rights ownership

within a 6 km radius of Marmato. Each of the properties was reviewed to determine discrepancies

between legal descriptions and actual ownership. The mining law allows for expropriation of land if

negotiations among subsurface and surface owners are unsuccessful.

As part of the mining licenses, CGM’s continuing obligations to maintain its mining concession in good

standing are:

• Conducting mining activities in the concession without interruptions. Any suspensions of

mining activities for more than six continuous months (save for force majeure) have to be

authorized by the Mining Authority

• Payment of the economic considerations based on production as set forth in the concession

contract (royalties, production taxes, and other economic consideration)

• Timely compliance of legal or contractual reporting obligations before the Mining Authority,

such as the annual and semi-annual filing of basic mining formats

• Compliance with relevant environmental laws and regulation, which includes obtaining and

complying with any permits that are required to carry out the corresponding mining activities

(such as water concession permits, discharge permits, Environmental Management Plans,

etc.)

• Maintaining insurance policies as required under the concession contract (civil liability policy,

contractual compliance policy, and compliance of labor obligations)

4.3 Royalties, Agreements and Encumbrances

In 1991, CGM (formerly Mineros Nacionales S.A.S.) entered into an agreement with Ecominas (a State

Industrial and Commercial Organization) for the exploration and exploitation of Mining Title No.014-

89M. The mentioned title was previously granted by the Colombian State to Ecominas. It was agreed

by the parties that CGM would pay a royalty to Ecominas (now referred to as Agencia Nacional de

Mineria) equal to 6% on gold revenue and 8% on silver revenue as economic compensation.

Additionally, CGM is bound by law to pay the Colombian State a 4% royalty.

CGM also pays a royalty of 4% on gold and silver revenue to an associated company owned by Gran

Colombia, Minera Croesus S.A.S. (Croesus), in respect of production sourced from the neighboring

Echandia mining title (#RPP_357) owned by Croesus. This royalty obligation remains in place.

4.4 Environmental Liabilities and Permitting

The main environmental details for the Project are covered in Section 19.1 of this report

4.4.1 Environmental Liabilities

The existing Marmato project is authorized through the approval of an Environmental Management

Plan (Planes de Manejo Ambiental or PMA). The PMA for Marmato was approved by the regional

environmental authority, Corpocaldas, on October 29, 2001 under Resolution 0496, File No. 616. The

PMA, and its requisite environmental management procedures and practices, amount to

approximately US$482,000 annually. This amount is likely to increase with the MDZ expansion project,

as additional monitoring and management will be required.



The 2001 mining code requires the concession holder to obtain an Insurance Policy to guarantee

compliance with mining and environmental obligations which must be approved by the relevant

authority, renewed annually, and remain in effect during the life of the Project and for three years from

the date of termination of the concession contract. According to CGM, the current amount covered by

the policy is COL$302,835,000 (USD$91,768). This amount will be reviewed and adjusted during the

modification process of the PMA for the MDZ expansion project.

According to the Code, the concession holder is liable for environmental remediation and other

liabilities based on actions and/or omissions occurring after the date of the concession contract, even

if the actions or omissions occurred at a time when a third-party was the owner of the concession title.

The owner is not responsible for environmental liabilities which occurred before the concession

contract, from historical activities, or from those which result from non-regulated mining activity, as has

occurred on and around the Marmato Project site.

Current liabilities at the site are generally associated with the reclamation and closure of the mining

facilities and tailings disposal areas. Given the extensive impacts associated with artisanal mining in

the area, a clear delineation between possible environmental liabilities attributable to CGM and those

from unregulated mining activities is not possible; however, CGM has been making a concerted effort

to deal with legacy environmental issues in order to better make that separation. The social issues

related to mining in Colombia, especially the interactions between mining companies and artisanal

operators could continue for CGM employees and the neighboring communities.

4.4.2 Required Permits and Status

Discussion related to mining in Colombia, the Mining and Environmental Codes, as well as the permits

and authorizations necessary for mineral exploration and exploitation is provided in Section 20.3.

4.5 Other Significant Factors and Risks

There are no legal restrictions that affect access, title or right or ability to perform work on the property

with the exception of the pending license approval.



5 Accessibility, Climate, Local Resources, Infrastructure and Physiography

5.1 Topography, Elevation and Vegetation

The Marmato project is located in an area of steep mountainous terrain with a relief of approximately

1,600 m (Figure 5-1). The Project is bound to the east by the Cauca River at 600 m elevation, and

otherwise surrounded by the peaks of the nearby mountains that reach up to 2,200 m elevation and

are commonly incised through landslide activity. Despite the abrupt relief, the landscape is in general

well vegetated and supports crop cultivation and livestock. The dominant land use in the area of

Marmato is cattle grazing, coffee, sugar cane, citrus fruit, bananas, and mining. The Middle Cauca

region, where Marmato is located, was occupied for two thousand years before the Spanish conquest

by farmers, potters, gold miners and goldsmiths of the Quimbaya culture (500 BC to 1600 AD).

The ecological zones defined on the Holdridge Life Zone climatic classification system are zoned by

elevation (Municipio de Marmato, 2004; Correa, 2006; Cia Minera de Caldas, 2008):

• Premontane (subtropical) wet forest transitional to tropical moist forest and dry forest; defined

as temperatures >24°C, annual rainfall of 1,500 to 2,800 mm, and elevation of 700 to 1,000

m. This area includes the Cauca River valley and the lower part of El Llano town.

• Premontane (subtropical) wet forest defined as temperatures of 18°C to 24°C, rainfall of 2,000

mm to 4,000 mm, and elevation of 1,000 to 1,900 m. The main areas of mining and exploration

are in this zone.

• Lower montane (warm temperate) wet forest defined as temperatures of 12°C to 18°C, rainfall

of 2,000 to 4,000 mm, and elevation of 1,900 m to 2,900 m.

Much of the original forest cover has been cleared for agriculture and grazing, especially at lower

elevations. Land is used for cattle grazing, coffee, sugar cane, citrus fruit, bananas, and mining in

Marmato.

5.2 Accessibility and Transportation to the Property

The Project is in the Municipality of Marmato in Caldas. The concessions of the Marmato Project are

located on the eastern side of the Western Cordillera (Cordillera Occidental) of Colombia on the west

side of the Cauca River.

Marmato is 200 km east of the Pacific Ocean and 300 km south of the Caribbean Sea and Atlantic

Ocean. The nearest port is Buenaventura on the Pacific Ocean (320 km by the Pan American Highway

to the south west).

The property is a three-hour drive from Medellín, via the Medellín to Cali highway which is part of the

Pan American Highway, National Route 25. The route from Medellín is via Itaguí (7 km), Caldas (12

km), Alto de Minas (13 km), Santa Barbara (27 km), La Pintada (26 km), La Guaracha del Rayo (32

km), and then a turn onto a secondary road to an 8 km long partially asphalted road to Marmato. There

is an international airport located in Medellín with flights to the USA, Panamá, Venezuela, Spain and

Peru, and a national airport in Manizales with flights to Medellín and Bogotá.



Source: SRK, 2012

Figure 5-1: Marmato Project, Looking Northwest Towards Cerro El Burro



5.3 Climate and Length of Operating Season

Climate at the site is typical of the equatorial zone, with the region falling within the Köppen

classification zone of Am, described as moist tropical climates with high temperatures year-round and

short dry seasons in a monsoon cycle. Average annual precipitation was estimated as 1,889

millimeters per year (mm/y) (Knight Piésold, 2012) with two drier periods around January and July and

wetter periods around April-May and October-November. Temperatures are warm year-round, with

maximum temperatures ranging from 28.7 degrees Celsius (°C) to 31.6°C and minimum temperatures

in the range of 17.4°C to 18.7°C (Knight Piésold, 2012). Relative humidity at the site is typically in the

70 to 80% range. The climate allows year-round operations.

5.4 Sufficiency of Surface Rights

Refer to Section 4.2 of this report

5.5 Infrastructure Availability and Sources

5.5.1 Power

The power supply is well established for the operating Marmato UZ operations. Power is available

through the Colombian power supplier Central Hidroeléctrica de Caldas (CHEC), a subsidiary of

Empresas Públicas de Medellín (EPM) through existing local substations. Substantial transmission

capacity is available in the region around the Project, with energy provided over the transmission

system by the third largest electricity producer in Colombia, ISAGEN.

Major electrical power will be required at the new MDZ plant site as all process facilities and major

infrastructure buildings are located there. Electrical power to the MDZ plant is planned to be supplied

by Central Hidroeléctrica de Caldas S.A. (CHEC) from the 115 KV Salamina substation located 15 km

away.

Site power will be obtained from a 115 KV HV line that will be provided by the local power authority up

to the MDZ plant outdoor substation

5.5.2 Water

External water supplies are available from both groundwater and surface water sources. Dewatering

for the underground mine is currently being utilized as a water supply for the existing operations, and

withdrawals from the nearby Cascabel River have been utilized as well. Additional dewatering flows

are expected to be produced as a result of dewatering the MDZ Project. Water is also available from

the nearby Cauca River.

5.5.3 Mining Personnel

The region has currently and historically a strong mining presence with around 1,400 people working

at the current Marmato Project and a substantial number of artisanal miners in the area close to the

mine. Skilled personnel should be available from the local miners as well as supplemented from the

other nearby areas to support the workforce as needed.



Field personnel for the exploration program have been employed from the towns of Marmato and El

Llano and neighboring municipalities. In the long term, personnel currently working on the large

number of small-scale mines and from the surrounding region would be able to supply the basic

workforce for any future mining operation.

5.5.4 Potential Tailings Storage Areas

The existing processing plant has an active storage known as the Cascabel site. Three additional sites

were identified as potential tailings storage areas, DSTF 1, DSTF 2 and DSTF 6. DSTF 1 and 2 were

advanced to PFS level and were included in the PFS. DSTF 6 was identified as potential future

expansion.

5.5.5 Potential Waste Disposal Areas

Waste rock for the existing project is being used underground as backfill. The new project waste rock

will be used for construction purposes on the plant and tailings storage facilities (DSTF). Excess waste

rock will be placed in secondary stopes as backfill if not needed for other purposes.

5.5.6 Potential Processing Plant Sites

The existing project processing plant is established and operating at its current location. A viable site

has been identified for the new MDZ processing facility. The site is an undeveloped location

approximately 3 km east by road of the ore body and existing plant on a naturally occurring plateau.



6 History Colombian gold production between 1514 and 1934 has been estimated at 49 million ounces (Moz)

which makes Colombia number one in South America with 38% of the total historical production

(Emmons, 1937). Two-thirds of the gold production was from placer deposits. Subsequent Colombian

production is estimated at 30 million ounces (Moz) by the Banco de la Republica (Shaw, 2000), which

gives Colombia a total recorded historical gold production of approximately 80 Moz. 75% of this

production came from the Departments of Antioquia and Caldas, with the Marmato Project located

near the border between the two departments.

6.1 Prior Ownership and Ownership Changes

Marmato is one of the most important historical gold properties in Colombia and lies in the heart of the

main historical gold producing region (dating back to 500 BC). The location name is derived from

“marmato” or “marmaja”, an old Spanish term for pyrite. The property has a long and complex

ownership history, summarized in Table 6-1.

Table 6-1: Ownership History at Marmato

Date Ownership History

1525 Colonization of Colombia and first references to Marmato

1634 First larger scale workings begin; and first gold mill

1798 Silver mines located at Echandia, with two near surface veins exploited

1819 to 1925 Various English companies mine gold at Marmato

1925 to 1938 Mines were expropriated and initially remained closed, then later leased to contractors

1946 Marmato was divided into two zones (law 66), Alta (Upper) and Baja (Lower)

1981 to 2004 Marmato becomes part of the Aporte Minero scheme and was managed by a succession of state mining companies

1984 to 1985 Minera Phelps Dodge de Colombia S.A. (Minera Phelps Dodge) explores the Zone Baja of Marmato

1991 Contract for the Zona Baja is awarded to Mineros Nacionales in October 1991 for a period of 30 years by the state entity Empresa Colombiana de Minas (Ecominas); the contract is now administered by Agencia Nacional de Mineria (National Mining Agency or ANM)

1996 to 2000

Conquistador Mines Ltd. (Conquistador), a Vancouver listed junior company (now called Orsa Ventures Corp), explored the Project through its Colombian subsidiary Corona Goldfields S.A. (Corona Goldfields). Conquistador had an option to explore the Zona Baja over 4 years and to acquire 50.1% of Mineros Nacionales (it bought 13.15% which it later sold in 2001), and acquired several mines in the Zona Alta.

1995 to 1997 Gran Colombia Resources Inc. (unrelated to GCM and now defunct) carried out exploration at Echandia and Chaburquia properties on the northern portion of the Marmato System

2005 - 2008 Minera de Caldas began exploration of Marmato and surrounding areas with the aim of identifying bulk mineable targets of low grade gold and silver. Colombia Goldfields Limited (CGL), began acquisition of property within Zona Alta, plus completed 46,000 m of drilling

2009 - 2010 Medoro purchased CGL (Zona Alta), Colombia Gold (Echandia) and Mineros Nacionales (Zona Baja), which consolidated the three primary gold properties at Marmato

2011 - 2019 Gran Colombia and Medoro Resources Ltd merged to create the largest underground gold and silver producer in Colombia, under the name of Gran Colombia Gold Corp.

2020 CGM acquires the current operating mine and the Marmato Deeps Project via the RTO Transaction (see section 4.2)

Source: SRK, 2020



6.2 Exploration and Development Results of Previous Owners

Modern exploration at Marmato began in the mid-1980s and has continued into the present day under

various entities. The exploration and development results of prior owners are listed below:

1984 to 1985: Minera Phelps Dodge explored the Zona Baja of Marmato, with the objective of defining

a 300 t/d underground operation. It completed surface and underground sampling and drilled seven

underground core holes and defined a Proven reserve of 102,900 t at 7.83 g/t gold and 24 g/t silver

and a total reserve (Proven, Probable and Possible) of 754,600 t at the same grade.

1993: Mineros Nacionales began mining the Maruja mine via a 300 t/d underground operation under

contract (No. 041-89M). Mineros S.A. acquired 51.75% of Mineros Nacionales and upgraded the mine

and mill. Mineros S.A. subsequently increased ownership of Mineros Nacionales to 94.5%. Further

exploration was completed through the 1990s with 24 underground core holes drilled and three reverse

circulation (RC) holes drilled. The plant was expanded to a capacity of 800 t/d.

1996 to 2000: Conquistador drilled 44 holes (14,873 m), 30 from surface (11,496 m) and 14

underground diamond holes (3,377 m), plus 1,147 channel samples totaling 2,847 m from surface

trenches and underground cross-cuts. Conquistador also commissioned MRDI to complete a resource

estimate and scoping study in 1998 but carried out no further work on the Project due to the expiration

of the option contract.

1995 to 1997: Gran Colombia Resources Inc. conducted soil surveys, surface magnetic and

geophysical surveys, channel samples (La Negra, La Felicia and La Palma adits) and completed 75

diamond drillholes (surface and underground) totaling 15,000 m. A scoping study was completed by

Geosystems International, Denver, in 1997 which concluded that there was not sufficient grade

continuity for a bulk-tonnage resource and mining operation, and no further work was carried out.

2005: Minera de Caldas began exploration of Marmato and surrounding areas with the aim of

identifying bulk mineable targets of low grade gold and silver. CGL carried out underground sampling,

surveying and mapping, preliminary metallurgical test work and diamond drilling to define a mineral

resource. CGL carried out 46,000 m of drilling in 2007 and 2008.

2010: Medoro commenced infill drilling of the project via surface and underground diamond drilling

with a view of producing a pre-feasibility study in 2011.

2011 to 2017: CGM completed further infill drilling from surface and underground locations, plus

channel sampling of existing cross-cuts.

2017 to 2020: CGM exploration has focused drilling on defining and infilling the MDZ, plus on-going

exploration within the current mining operations and cross-cuts on levels 20 and 21.

6.3 Historic Mineral Resource and Reserve Estimates

A number of different MREs have been completed on the property during the history of the project.

Between 2010 and 2019, SRK has produced several MREs for the Project. The most recent Mineral

Resource Statement for the Project has an effective date of July 31, 2019, which is the last date assays

were provided to SRK.


Leapfrog™Geo of wireframe models for the fault networks, veins, definition of resource domains (e.g.

high-grade sub-domains), data conditioning (compositing and capping) for statistical analysis,



geostatistical analysis, variography, block modelling and grade interpolation followed by validation.

Grade estimation for the veins has been based on block dimensions of 5 m by 5 m by 5 m for the

Porphyry and MDZ units. Sub-blocking to 0.5 m by 1 m by 1 m has been allowed to reflect the narrow

nature of the geological model. The block size reflects the relatively close-spaced underground

channel sampling and spacing within veins compared to the wider drilling spacing, with the narrower

block size used in the MDZ at depth to reflect the proposed geometry of the mineralization (i.e. steeply

dipping feeder zone).

SRK reviewed and updated the geostatistical properties of the domains. Gold grades have been

interpolated using nested three-pass estimates within Datamine™, using an OK routine. SRK has also

run IDW2 and NN estimates for validation purposes.


CIM Definition Standards for Mineral Resources and Reserves (CIM, 2014). SRK developed a

classification strategy which considers the confidence in the geological continuity of the mineralized

structures, the quality and quantity of exploration data supporting the estimates, and the geostatistical

confidence in the tonnage and grade estimates. Data quality, drillhole spacing and the interpreted

continuity of grades controlled by the veins have allowed SRK to classify portions of the veins in the

Measured, Indicated and Inferred Mineral Resource categories.



Table 6-2: SRK Mineral Resource Statement for the Marmato Project, Dated July 31, 2019*, Within Zona Baja**

Category Quantity

Grade Metal

Au Ag Au Ag

Mt gpt gpt 000'oz 000'oz

Underground Vein***

Measured 2.1 4.9 23.2 325 1,543

Indicated 7.2 4.5 18.1 1,038 4,168


Inferred 3.3 4.4 14.7 466 1,577

Underground Porphyry***

Measured

Indicated 1.6 2.7 10.1 140 527


Inferred 0.3 3.1 9.6 34 107

Underground Deeps****

Measured

Indicated 6.4 2.6 4.7 537 978


Inferred 41.2 2.1 2.7 2,812 3,609

Underground Combined

Measured 2.1 4.9 23.2 325 1,543

Indicated 15.2 3.5 11.6 1,714 5,674


Inferred 44.9 2.3 3.7 3,312 5,293

* Mineral resources are not mineral reserves and do not have demonstrated economic viability. All figures are rounded to reflect the relative accuracy of the estimate. All composites have been capped where appropriate. ** Zona Baja defined as all material within License #014-89m and the portion of #RPP_357 below 1,340 masl pursuant to a current mining contract with Gran Colombia. *** Porphyry and Vein mineral resources are reported at a CoG of 1.9 g/t. CoGs based on a price of US$1,500 per ounce of gold, suitable benchmarked technical and economic parameters and gold recoveries of 95 percent for underground resources, without considering revenues from other metal. **** Deeps mineral resources are reported at a CoG of 1.3 g/t. CoGs based on a price of US$1,500 per ounce of gold, suitable benchmarked technical and economic parameters and gold recoveries of 95 percent for underground resources, without considering revenues from other metal.

The Mineral Resources quoted in Table 6-2 are no longer deemed current and should not be relied

on, and has been updated with the current Mineral Resource estimate defined in Section 14 of this

report.

No historical Mineral Reserves have been quoted for the Project.

6.4 Historic Production

Production has occurred from the Marmato property since pre-colonial times, but there are no

published historical records of the actual gold and silver production for all periods since mining

commenced, however sporadic records for different periods have been noted.

To give an indication of the current mining activity at the deposit SRK has reproduced a summary

(Table 6-3) of the total produced gold and silver at Marmato on an annual basis between 2004 and

2019. The figures also represent only the official declared gold recovered and does not include illegal

mining which persists at Marmato even to present times.



Table 6-3: Gold Production from the Municipality of Marmato 2004 to December 2019

Year Ore Tonnes (t) Grade Au (g/t) Au Produced (oz)

2004 186,330 3.60 21,583

2005 231,540 3.30 24,541

2006 262,517 3.10 26,171

2007 300,756 3.22 31,127

2008 254,474 2.95 24,138

2009 250,638 3.51 24,372

2010 252,136 3.39 23,318

2011 250,553 3.19 22,715

2012 268,137 2.85 21,717

2013 274,190 2.90 22,566

2014 295,023 2.85 24,116

2015 303,279 2.79 23,954

2016 341,308 2.55 23,447

2017 366,485 2.46 25,162

2018 340,052 2.67 24,951

2019 370,494 2.48 25,750

Source: CGM, 2020



7 Geological Setting and Mineralization

7.1 Regional Geology

The Colombian Andes are part of the Northern Andean Block which includes the Northern Volcanic

Zone of the Andes (Gansser, 1973; Shagam, 1975). They are formed of three N to NNE trending

mountain ranges, the Western, Central and Eastern Cordilleras, separated by two major intermontane

basins, the Cauca-Patía Depression and the Magdalena Depression, which represent terrane

boundaries. The Colombian Andes have a complex history of volcanism, subduction, accretion and

faulting, represented by the juxtaposition of metamorphic, igneous and sedimentary rocks of various

ages from the Precambrian to the present (Aspden et al., 1987; Restrepo and Toussaint, 1988). Cediel

et al. (2011) have defined nine principal tectonic terranes in Colombia which are:

• Guyana shield

• Maracaibo sub-plate

• Central continental sub-plate

• Pacific terranes

• Caribbean terranes

• Choco-Panama arc

• Guajira terrane

• Caribbean Plate

• Nazca Plate

Marmato is located on the eastern side of the Western Cordillera which is separated from the Central

Cordillera by the River Cauca. It lies within the Romeral terrane which is bounded by the Cauca Fault

on the west side and the Romeral Fault to the east and is part of the Pacific terranes realm. The recent

tectonic setting of the Colombian Andes is characterized by the subduction of young (less than 20

mega annum [Ma]) oceanic crust beneath relatively thin continental crust (less than 40 km; Cediel and

Caceres, 2000; Cediel et al., 2003). The Benioff zone is located at around 140 to 200 km depth below

the volcanic belt of the Colombian Andes which has slightly migrated to the east during the last 10 Ma

(Pennington, 1981; Vargas & Mann, 2013).

The Marmato stock is part of the Miocene magmatism characterized by calc-alkalic subvolcanic

intrusions and volcanic rocks of the Combia Formation. The Miocene magmatism cross-cuts the units

of the Romeral terrain, the plutonic units of the Albian and early Cenozoic, and the siliciclastic

sequences of the Amagá Formation (Cáceres et al. 2003; Tassinari et al, 2008). Miocene gold related

magmatism in Colombia has been well-recognized in the Western and Central cordilleras associated

with stocks (Sillitoe et al., 1982; Toussaint and Restrepo, 1988; Lodder et. al, 2010; Lesage et al.,

2013). In addition, late Miocene-Pliocene magmatism with gold mineralization has also been

recognized in the Santander Massif in the northern part of the Eastern Cordillera (Mantilla et al., 2009).

The regional geology is shown in Figure 7-1.



Source: Modified from the Geological Map of Colombia, 1:1 million scale, Colombian Geological Survey, 2015

Figure 7-1: Regional Geology Map



7.2 Local Geology

The Marmato gold deposit is hosted by the porphyritic andesitic to dacitic Marmato stock which is

18 km long and 3 to 6 km wide and is elongated north to south (Calle et al., 1984). It intrudes the

Arquía Complex and Amagá Formation on the east side in the Cauca Valley and the Combia Formation

on the west side. The Marmato gold deposit is hosted in a multiphase porphyry suite, the Marmato

Porphyry Suite, which is about 3 km long by 1.6 to 2.5 km wide and is located near the southern end

of the larger Marmato stock. Five main porphyry pulses have been identified in the Marmato Porphyry

Suite by cross-cutting relationships in core logging and named P1 to P5 from oldest to youngest,

respectively. The ages of the intrusions have been reported recently between 6.58 ± 0.07 Ma to 5.74

± 0.14 Ma by U-Pb LA-ICP-MS of zircon (Caldas, 2016, dating carried out by the Brasilia University

Isotope Geochronology Laboratory, Brasil). The Aguas Claras Porphyry Suite is located 3 km

southwest of the Marmato Porphyry Suite and also has five porphyry pulses identified from cross-

cutting relationships in core logging named AP1 to AP5 from oldest to youngest. Two intrusions of the

Marmato Porphyry Suite, P3 and P5, cross-cut the Aguas Claras Porphyry Suite as dikes. There is no

previous dating of the Aguas Claras Porphyry Suite. The local geology map is presented in Figure 7-2.

Source: Caldas, 2017

Figure 7-2: Local Geology Map



Source: Modified from Caldas, 2018. Original Source in Figure

Figure 7-3: Regional Geology with Gold Prospects in the Marmato Area



The main rock types within the area are summarized below in Sections 7.2.1 to 7.2.7.

7.2.1 Graphitic-Sericite Schist (MSG)

They constitute the metamorphic units of the local area and appear as dark rocks, with schist texture,

demonstrating evidence of a ductile deformation crenulation cleavage, with predominant orientation in

NNW-SSE, and steeply dipping. Evidence exists of several metamorphic events. This lithology is found

as suspended ceilings towards the central and northwestern area of the title, as well as large xenoliths

(more than 20 m in diameter) within the bodies of the main porphyry, their best outcrops are observed

in the upper areas of the El Burro hills and the Echandía area.

7.2.2 Amphibolites (MAB)

This lithology is not present within the title in any significance. The metabasites of the Arquía Complex

correspond to green schists and amphibolites; they correspond to dark green to light green rocks,

interspersed with small bands of quartz-muscovitic schists with graphite. Petrographically they are

represented by quartz-chloritic schists with muscovite, chloritic schists, actinolite/horblende schist with

chlorite, quartz actinolytic schists, with a mineralogy composed of actinolite (24 to 45%), chlorite (8 to

35%), horblende (less than 5%), zoisite/clinozoisite (3 to 8%), quartz (5 to 37%), plagioclase (4 to 6%),

muscovite (1 to 4%). The most common accessory minerals are calcite, titanite, hematite, magnetite

and ilmenite; the characteristic textures are nematoblastic (actinolite), lepidoblastic (chlorite) and

porphyroblastic (epidote) (Moreno et al, 2012).

Amphibolites are dark green to light green rocks, with compositional banding of plagioclase and

horblende, occasionally with garnet, the modal composition is given by horblende (42 to 46%),

plagioclase (25 to 28%), garnet (15 to 17%), quartz (9 to 14%); the characteristic textures are

nematoblastic (horblende) and porphyroblastic (garnet).

7.2.3 Serpentinites (MSP)

This lithology is not present in the mining title area. The mafic and ultramafic rocks are represented by

an elongated strip in the North-South direction, corresponding to Gabros of fine to medium grain, dark

greenish-gray color, with a mottled appearance, exhibiting an incipient fluid texture, near the fault

zones it presents genesis textures being difficult to differentiate the primary foliation from the dynamics.

The essential components are plagioclase, horblende, pyroxene, sericite, clinozoisite, zoisite, chlorite,

calcite, epidote, biotite, sphene, apatite, magnetite-ilmenite. These rocks are associated with the

remains of an Ophiolithic Complex of presumably Pre-Jurassic age and that was located during the

Cretaceous (Gonzales, et. Al, 1982).

7.2.4 Basalts (VB)

They correspond to massive crystalline rocks, of aphanitic texture, dark (melanocratic) coloration,

slightly magnetic, composed of plagioclase, pyroxenes and amphiboles which are generally

chloritized. Its outcrops are located towards the southeast of the title on the Cascabel and Aguas

Claras ravines and is interpreted as roof pendants associated with the intrusion of the porphyritic stock.

7.2.5 Clastic Sedimentary Rocks (S)

The remnants of a clastic sedimentary sequence made up of quartz-lithic sandstones, from cream to

brown tones, with a medium to coarse grain size, friable, are observed in subtabular layers of varying



thicknesses, which are observed interstratified with clay. The units are characterized by presenting a

high state of weathering and have a preferred orientation of the stratification is N20-30W/30-50NE.

7.2.6 Marmato Porphyry Stocks (P1 – P5)

The porphyritic andesitic to dacitic Marmato stock which is 18 km long and 3 to 6 km wide and is

elongated north to south (Calle et al., 1984). It intrudes the Arquía Complex and Amagá Formation on

the east side in the Cauca Valley, and the Combia Formation on the west side.

The ages of intrusion, alteration, and mineralization are Late Miocene. The Marmato Stock has been

dated with 6.3+-0.7 Ma (K/Ar) porphyritic dacite (Sillitoe, 1982); 7.1+-0.2 Ma (K/Ar) in porphyritic

andesite biotite (Rossetti & Colombo, 1999); 6.7+-0.06 Ma (Ar/Ar) in porphyritic andesite biotite

(Vinasco, 2001); 5.6+-0.6 Ma (K/Ar) for sericitized plagioclase in porphytic dacite from the MDZ (R &

Tassinari, 2003), the latter is interpreted as the age of deposit formation being slightly younger than

the intrusions. Likewise, Santacruz (2016) finds two different ages and exposes the idea of two

magmatic events, the first one dated at 6.58-6.3 Ma (U/Pb in zircon) and a second event dated at 5.7

Ma (U/Pb in zircon)

The Marmato gold deposit is hosted in a multiphase porphyry suite, the Marmato Porphyry Suite, which

is about 3 km long by 1.6 to 2.5 km wide and is located near the southern end of the larger Marmato

stock. Five main porphyry pulses have been identified in the Marmato Porphyry Suite by cross-cutting

relationships in core logging and named P1 to P5 from oldest to youngest, respectively. The ages of

the intrusions have been reported recently between 6.58±0.07 Ma to 5.74±0.14 Ma by U-Pb LA-ICP-

MS of zircon (Caldas, 2016, dating carried out by the Brasilia University Isotope Geochronology

Laboratory, Brasil).

The Aguas Claras Porphyry Suite is located 3 km southwest of the Marmato Porphyry Suite and also

has five porphyry pulses identified from cross-cutting relationships in core logging named AP1 to AP5

from oldest to youngest. Two intrusions of the Marmato Porphyry Suite, P3 and P5, cross-cut the

Aguas Claras Porphyry Suite as dikes.

7.2.7 Unconsolidated Quaternary Deposits (QC)

Unconsolidated quaternary deposits are present within the mining areas, which include old landslide

bodies made up of alluvial gravels and the waste material extracted by artisanal miners that are

collected on the slopes of Cerro El Burro over the municipality of Marmato. There are also small

deposits of poorly calibrated gravels, with angular to sub-rounded clasts whose sizes vary from

pebbles to blocks, generally supported matrices, which do observe discordant outcropping lithologies

in the main stream gorges in the area.

7.3 Property Geology

The Marmato gold deposit consists of a structurally-controlled epithermal vein system with a mineral

assemblage dominated by pyrite, arsenopyrite, black Fe-rich sphalerite (the type locality for

“marmatite”, Boussingault, 1830), pyrrhotite, chalcopyrite and electrum in the UZ and a mesothermal

veinlet system with a mineral assemblage dominated by pyrrhotite, chalcopyrite, bismuth minerals and

visible gold in the MDZ.

Dacitic and andesitic intrusions at Marmato are characterized by quartz, hornblende, biotite and zoned

plagioclase phenocrysts in a finely crystalline quartz-plagioclase groundmass, with variations in



phenocryst proportion and sizes between intrusions. A total of five different porphyry units have been

identified which are summarized below:

Dacitic Porphyry (P1)

It is a compact, solid, granular rock characterized by the presence of large quartz phenocrystals

(5 millimeter [mm] diameter), hornblende, biotite and plagioclase phenocrystals zoned with a fine

granular matrix of quartz and plagioclase, with variations in sizes and percentages. The P1 intrusion

is the main Dacitic body of the Marmato Porphyritic Suite and therefore represents most of the geology

of the title, it is overlain by quartz-sericitic-graphite shales of the Arquía Complex and contains small

intrusions of the andesite P4 as well as the other porphyritic intrusive bodies.

Andesitic Porphyry (P4)

The P4 intrusion corresponds to an igneous, slightly equigranular, massive igneous rock, of grayish

hue, medium grain size, characterized by the scarce or almost null presence of quartz, accompanied

by abundant phenocrystals of coarse plagioclase with biotite, horblende and magnetite, the

amphiboles are generally chloritized. They represent the second body in terms of importance within

the suite since it also acts as a host to the mineralization.

An Andesite P4 forms a stock on the NW side of Cerro Los Novios and extends NE through Echandía;

It has a NE orientation with dimensions of approximately 1,600 m length and 750 m width; towards the

southeast there are numerous P4 dikes with a NW and E-W tendency.

Porphyritic Dacite (P2)

It appears as a smaller intrusion than the P1 and P4 intrusions and is characterized by being a solid,

hypocrystalline rock, with a medium grain size (finer than P1), light gray coloration, mineralogically

constituted by small sub-rounded quartz crystals, accompanied by mildly zoned and epidotic

plagioclase phenocrystals, biotite’s and euhedral and generally chloritized amphiboles. This intrusive

body has not been observed on the surface but exclusively logged in drill core.

Porphyritic Andesite (P3)

Porphyritic andesite dikes with a NW and EW tendency and 400 m in length have been observed in

contact with P1 in the Cascabel gorge and other occurrences in Echandía. The P3 units is

characterized by the presence of slightly zoned plagioclase euhedral megacrysts (15 mm long), with

small subhedral crystals of plagioclase, biotite, hornblende, and to a lesser extent quartz and

magnetite (Figure 7-4 photos of porphyry). At the drilling core level, it has been observed cutting the

P1 and P2 bodies.

Porphyritic Dacite (P5)

Porphyritic Dacite Dikes P5 have been observed cutting the P1 porphyry within the Cascabel gorge

and selected drilling cores. P5 is characterized by massive, uneven, hypocrystalline texture, with small

euhedral crystals of quartz and elongated plagioclase phenocrystals of up to 10 mm in length, with

additional, biotite and horblende crystals present. Mineralization is absent in this lithology and there is

dating that gave an age of 5.7 Ma (U/Pb in zircon) (Santacruz, 2016).




Figure 7-4: Photographs of the Different Porphyritic Intrusive Bodies that Make Up the Porphyry Stock Of Marmato

The P1 Intrusion is a main dacitic porphyry stock in the Marmato Porphyry Suite and is characterized

by large β quartz phenocrysts more than 7 mm. It is cross-cut by intrusion P2 which corresponds to a

porphyry dacite intrusion with fewer and smaller phenocrysts. Intrusion P3 forms dikes of andesitic

porphyry with plagioclase megacrysts more than 10 mm, and cross-cuts intrusions P1 and P2 (Figure

7-5). Intrusion P4 is an andesitic porphyry stock which cross-cuts P1, P2 and P3, and is characterized

by smaller plagioclase phenocrysts. The youngest porphyry P5 is dacitic and forms dikes cross-cutting

P1. It is characterized by large quartz phenocrysts and elongate plagioclase phenocrysts.

Mineralization is hosted mainly by stocks P1 to P4, while is absent in P5.



Source: SRK, 2020

Figure 7-5: Property Geology Map



Source: CGM, 2019 The deposit outcrops on El Burro Hill and Echandía Hill

Figure 7-6: Cross-Section of the Marmato Gold Deposit Looking NW Showing the Intrusions P1 to P5

7.3.1 Structure

The dominant NW and E-W trends of the veins are interpreted to be due to regional tectonic forces

and may have formed as tension fractures related to NW-SE compression and sinistral strike-slip

movement on the N-S trending Cauca and Romeral Faults which lie on either side of the deposit.

In April 2010, the Company commissioned Telluris Consulting Ltd (TCL) to complete a review of the

local and regional geology to define a structural-hydrothermal model for the Marmato deposit. TCL

defined the Marmato deposit as a series of N-NW to E-W trending steep to moderately dipping, gold-



bearing, sulfide-rich veins hosted in a north-south trending late Miocene porphyry complex. TCL noted

that the porphyry complex was emplaced in folded and thrusted Paleozoic and Mesozoic metamorphic

and sedimentary sequences adjacent to the eastern margin of the broadly N-S trending Cauca-

Romeral terrane accompanied by east northeast to northwest-southeast compression. This resulted

in N-S trending thrust and transpressional structures along with steep NW and NE conjugate fault

zones.

Within the relatively young intrusive rocks of the Marmato deposit there are principally two deformation

stages recognized from the TCL study:

• Syn-mineralization W-NW-E-SE compression that reactivated some of the basement

structures as well as generating a range of second order shear and extensional structures

along N-NW to E-W trends as well as N-NE trending thrusts

• Continued post-mineralization compression into the late-Pliocene, (approximately 2 Ma) that

resulted in uplift due to renewed thrusting along the main terrane boundaries forming thrust-

bounded intermontane basins such as the Cauca-Patia depression

Within the Marmato area, there are four principal trends of mineralized structures:

• NW trending steep to sub-vertical faults/fractures (140° to 150°N)

• W-NW trending steep to moderately inclined structures (110° to 120°N)

• E-W trending structures (100° to 090°N) that tend to have moderate to relatively low-angle

dips

• E-NE to NE-trending structures (065° to 080°N) that show a range of dips

In addition to these ore-bearing structures, there is a set of N-NE trending structures of varying dips

that appear to represent different components of a reverse/thrust fault system. Both the W-NW and E-

W veins tend to splay from the main NW structures which is consistent with extensional and Riedel

shear components to a sinistral shear system. TCL reported that kinematic indicators show that

mineralization accompanied a phase of W-NW to E-SE orientated compression (Figure 7-7). The

north-northeast trending reverse faults and conjugate fractures reflect this compression component.



Source: TCL, 2010

Figure 7-7: TCL Interpretation of Vein Orientations at Marmato

Post-mineral failures of preferential trend N80E to E-W/60NE are observed in the margins of some

veins and veins with alteration to soft white clay gouge with ground pyrite registered as fault gouge,

(FLG). In some places, there is coarse euhedral pyrite in the clay gouge. Brittle fault gaps without clay

gouge (BXF) are also observed. In the mining works of the MDZ of CGM, it is observed that the N30-

40W trend veins rotate counterclockwise to the northwest at N50-60W. They have competent wall

rocks and do not require maintenance by the miners: while the veins that have an east-west tendency,

are faulted in the backing rock with soft clay gouge and require support from the miners.

Within the MDZ, mineralization is hosted within veinlets. These veinlets develop in tension fractures

that are arranged parallel to the main stress tensor which, as indicated before, is of the WNW-ESE

trend. Within a Riedel-type analysis, these are called "T" type fractures, do not show rotation or elapse,

and their deformation mechanism is exclusively traction opening perpendicular to compression σ1.

The veinlets on occasions have intersected the drilling core at high angles indicating the steep dip of

the mineralization.

After the mesothermal mineralization was placed in the "T" veinlets, these veinlets were reactivated

and affected by subsequent epithermal events. They can be observed in several drillholes as the

veinlets served to delimit the intensity of the alteration or to separate between type of alterations.

7.3.2 Alteration

Two stages of pervasive alteration have been recognized, early propylitic and later intermediate

argillic. These affect all types of porphyryitic rocks, although alteration is weak in P5. The propylitic

alteration is characterized by epidote replacement of plagioclase cores, albite replacement of

plagioclase rims and matrix, chlorite replacement of mafics, with disseminated pyrite and pyrrhotite,



and varies in intensity from veinlet-halo to pervasive. Calcite partially replaces plagioclase where

propylitic alteration is weakly developed. Cross-cutting relationships show evidence for multiple events

of propylitic alteration related to each phase of intrusion.

Intermediate argillic alteration overprints the propylitic alteration and varies in intensity from

vein/veinlet-halo to pervasive, associated to the intermediate sulfidation mineralization style and

replaces epidote, chlorite and albite. There is a strong but generally narrow halo of white to green illite

or sericite alteration related to veins and veinlets of the mesothermal mineralization event which grades

outwards to pervasive illite, with smectite in distal parts. The main disseminated sulfide is pyrite,

although pyrrhotite and iron-rich sphalerite also occur, which to some extent formed the basis for the

previous model domains.

Additionally, weak and patchy potassic alteration, represented chiefly by biotite occurs at depth in the

MDZ. Progressively better preservation of early potassic alteration at depth may indicate the possibility

of early gold-bearing phases.



o Propylitic

o Sodium-Calcite

o Potassic

o Argilic

Source: CGM, 2020

Figure 7-8: Types of Alteration Found at Marmato

Chloritized Biotite and Hornblende

EpidoticPlagioclase



7.4 Significant Mineralized Zones

Gold mineralization occurs in veins and veinlets with dominant NW and W-NW trends. The deposit

mainly comprises sulfide-rich veinlets and veins composed of minor quartz, carbonate, pyrite,

arsenopyrite, Fe-rich sphalerite (i.e. marmatite), pyrrhotite, chalcopyrite and electrum in the epithermal

Upper Zone, quartz, pyrrhotite, chalcopyrite, bismuth sulfide, telluride minerals and free gold in the

mesothermal MDZ. Pervasive early propylitic alteration is over-printed principally by phyllic and

intermediate argillic alteration related to the gold mineralized veins of low to intermediate sulfidation

epithermal type, with weak and patchy potassic (biotite) alteration at depth.

The Marmato deposit lacks known surface epithermal features, such as lithocaps, sinters and

crustifome-banded streaks. The veins can be found from the surface (1,700 meters above sea level

[masl]) and there is evidence of their continuity up to the 900 masl, on course there is a continuity that

goes from 50 m for secondary veins or splays to 600 m for the main veins.

The current significant mineralized zones in title #014-89m, are located in the NW sector of the title,

covering an approximate area of 30 Ha, with a distribution towards depth, approximately from elevation

1,200 m to 600 m, emphasizing that gold mineralization remains open at depth.

The mineralization in the current mine consists of three distinct phases, a first phase characterized by

the mesothermal vein/veinlet mineralization which defines the MDZ, followed by an epithermal low to

(subsequent) intermediate sulfidation style generally found in the UZ (SRK, 2019).

TCL recognized two principal deformation stages within the Marmato stock (TCL, 2010):

• Syn-mineralization W-NW to E-SE compression that reactivated some basement structures

as well as generating a range of second-order shear and extensional structures along N-NW

to W trends, as well as N-NE trending thrust faults.

• Continued post-mineralization compression into the late-Pliocene, (approximately 2 Ma) that

resulted in uplift due to renewed thrusting along the main terrane boundaries, forming thrust

bounded intermontane basins such as the Cauca-Patia depression.

TCL outlined four principal trends of auriferous structures within the Marmato area:

• NW trending steep to sub-vertical faults/fractures.

• W-NW trending steep to moderately inclined structures.

• W trending structures that tend to have moderate to relatively low angle dips.

• E-NE to NE trending structures that show a range of dips.

TCL reported that kinematic indicators show that gold mineralization accompanied a phase of W-NW-

E-SE orientated compression. The N-NE trending reverse faults and conjugate fractures reflect this

compression component. Within this tectonic framework the E-W faults should be predominantly

dextral strike-slip and the W-NW faults should be predominantly sinistral strike-slip (Figure 7-9). CGM

interprets the rotation of some of these structures to be the result of rotation during progressive

compressional deformation event, however CGM also noted that there are pre-gold mineralization and

post-gold mineralization phases of fault movement on a number of faults and veins (J. Ceballos, 2019,

pers. comm.)



Source: CGM, 2018 (based on sources listed in text)

Figure 7-9: Structural Features Expected in a North-South Sinistral Riedel Fault System

The epithermal mineralization occurs in parallel, sheeted and anastomosing veins, all of which follow

a regional structural control, with minor veins forming splays of the main structures which often have

limited strike or dip extent (Figure 7-10). The upper vein domain intersects broader zones of intense

veinlet mineralization that is hosted by a lower grade auriferous porphyry stock (which are termed

locally as “Porphyry Pockets” or “Porphyry” mineralization.




Figure 7-10: Example of Epithermal Veins as Viewed in the Drilling Core at Marmato

Gold mineralization occurs in veins and veinlets with dominant NW and W-NW trends. The deposit

mainly comprises sulfide-rich veinlets and veins composed of minor quartz, carbonate, pyrite,

arsenopyrite, Fe-rich sphalerite (i.e. marmatite), pyrrhotite, chalcopyrite and electrum in the epithermal

UZ, and quartz, pyrrhotite, chalcopyrite, bismuth sulfide and telluride minerals and free gold in the

mesothermal MDZ. Pervasive early propylitic alteration is over-printed principally by phyllic and

intermediate argillic alteration related to the gold mineralized veins of low to intermediate sulfidation

epithermal type, with weak and patchy potassic (biotite) alteration at depth (Figure 7-11).

Further detail on the length and size of the veins is defined in Section 14.2 of this report.



Low to intermediate sulfidation epithermal style mineralization

Mesothermal style mineralization Source: CGM, 2017

Figure 7-11: Examples from Drill Core of the Different Mineralization Styles



In addition, to the mineralization within the current mine (Levels 16 to 21), CGM has identified through

exploration a deeper mesothermal mineralization referred to as the MDZ. The deeper mineralization

is hosted primarily within the P1 porphyry unit and forms fine veinlet mineralization with a discrete,

relatively high-grade gold core to the main MDZ has been identified by CGM and delimited by SRK

with more than1.7 g/t Au iso-shell (SRK, 2019).

The MDZ type corresponds to several sets of N60-70W/90 veinlets that run parallel to each other as

sheeted veins, these are controlled by the main trend stress tensor WNW-ESE, as explained in the

chapters of regional and local structural geology. For this reason, these veinlets are considered to

represent stress fractures at a very early stage in the porphyritic intrusion, which may represent a

greater extension of this mineralization to that currently explored, modeled and estimated.

The MDZ gold mineralization consists of a network of thin, less than 5 cm thick sulfide veinlets, mainly

consisting of pyrrhotite+chalcopyrite, and typically rimmed by a thin sodium-calcitic alteration halo, all

hosted in weak argillic and deeper potassic alteration related to a pre-gold mineralization event (SRK,

2019).

The Mesothermal veinlets within the MDZ mineralization follow a standard pattern, presenting a

predominantly NW orientation between 40 to 62°, with steep dips (between 70 to 84°). Minor variations

in the trend from NNW to E-W have also been noted but are less common and within borehole MT-IU-

009, another family of mesothermal veins with an N10W tendency and a dip of 47° is identified.

Source: SRK, 2020

Figure 7-12: MDZ Mineralization Showing Veinlets Including Visible Gold (Au). BHID MND282-03-17 at a Depth of 1,010 masl, Sample of 1.20 m with 18.06 g/t Au and 2.5 g/t Ag



8 Deposit Type

8.1 Mineral Deposit

The alteration and mineralization in the UZ at Marmato evolved through early stage, higher-

temperature propylitic alteration to later, lower-temperature intermediate argillic alteration, with most

of the gold and silver being deposited in the later stage.

The gold-silver and base metal association in the UZ at Marmato is typical of the intermediate

sulfidation epithermal type. The veins lack distinctive epithermal textures and the mineralization has a

relatively high depth and temperature of formation, which straddles the deep epithermal to

mesothermal transition as defined by the original classification of Lindgren (1922) and by estimates of

formation temperature of 300°C (Heald et al., 1987). The Marmato deposit lacks known shallow and

surface epithermal features such as lithocaps, sinters and crustiform banded veins.

Mineralization is interpreted to be genetically related to the host porphyritic rocks, as shown by the

inter-mineral timing of the porphyry phases cross-cutting earlier stages of propylitic alteration, the late-

mineral timing of the final dacite P5, and miarolitic cavities lined with propylitic-stage minerals. The

veins and veinlets are structurally controlled and did not form a multi-directional porphyry stockwork

or breccia related to hydro-brecciation. In this model, the host stocks might be considered as late-

mineral intrusions with respect to a postulated porphyry gold-copper-molybdenum centers.

The upper portion of the MDZ has been exposed in Level 21 of the existing CGM mining operations

and is referred to as the Transitional Zone, while deeper sections have been observed in drillcore,

both of which have been confirmed as separate styles of mineralization. The lowest levels of the mine

have currently intersected a combination of the porphyry domain, where the gold is associated with

pyrite veinlets and the MDZ where gold is associated with pyrrhotite. Gold grade distribution in the

MDZ orebody is unrelated to the presence of distinct porphyry phases and is entirely dependent on

the intensity of structurally localized veinlets. Sillitoe (2019) concluded, that the only geological

parameter than can be used to constrain the grade model is veinlet intensity, although the presence

of visible native gold also acts as a useful grade indicator.

8.2 Geological Model

As part of the updated Mineral Resource, SRK initially focused on the creation of a lithological model

(i.e., one encompassing the major geological features inclusive of the current veins being mined). The

lithological database provided to SRK contained 64 separate logging codes, which has been refined

to 14 logging codes by SRK. The main geological features and units modelled by SRK were:

• Major Fault Network

• Porphyry (P1 – P5)

• Meta Schist

• Intrusive

• Volcanic

• Breccia

• Veins

• MDZ



In comparison to the PEA lithological model, SRK made additional definition of the units within the P4

and P5 dikes. The other key change is the definition of the breccia units which crosscut the MDZ and

may need consideration for geotechnical criteria prior to mining.



9 Exploration

9.1 Relevant Exploration Work

9.1.1 Topographic Surveys

The Company commissioned a detailed topographic map with 0.5 and 1 m resolution contour intervals

derived from LIDAR imagery, which was supplied to Datamine™ in 2020. The new topographic map

provides a detailed base map for improved accuracy when plotting the results of the exploration

programs, as well as a high-resolution satellite image.

For exploration work and also for other infrastructure and mining operation works, the topographic

information has been extracted from different remote sensors (satellite, radar) seeking to optimize the

quality (resolution) of the information.

In early 2007, a high-resolution satellite image of Ikonos and a detailed topographic map with contour

lines every 2 m were obtained. This map provided a detailed basis that served to improve the accuracy

of the drilling programs and their results. The topography was converted to a solid model in Vulcan™

to limit the grade estimate to the surface. This model has been supplied to SRK by the Company.

In 2008, the image of Ikonos and the topographic map were expanded, which were joined to the

original map of 2007 to leave a seamless final product on which the infrastructure for the Marmato

Project could be located, such as rock waste storage areas, tailings and exploration.

In 2019, the ISATECH company carried out a geodetic control work with LIDAR technology producing

a detailed topographic map with 0.5 and 1 m resolution contour intervals, which was supplied to

Datamine™ in 2020. The new topographic map provides a detailed base map for improved accuracy

when plotting the results of the exploration programs, as well as a high-resolution satellite image. In

2019, in order to ensure that the project's coordinate system was adjusted to the national geodetic

network in the MAGNA SIRGAS system, in this work 10 points of control and 9 materialized GPS

points, the area in which the work was carried out is presented in Figure 9-1.

Due to the project having gone through several stages of development and with differences in the grid

systems being used between the mine and exploration (Arenas, UTM), the Company undertook a

study to validate the geometric transformation procedures that were developed between the

planimetric and altimetric reference systems. In addition to this study, it was sought to establish a

mechanism that would allow all areas of the company that generate geographic information to have a

guideline to work in the same coordinate system according to the national standard (defined by IGAC),

for this the company Geosoluciones DAJ was hired in February 2020. The study involved the validation

of the topographic survey carried out by ISATECH in the Magna Sirgas reference frame, Gauss-Krüger

projection, west origin, by tracking and processing data from the Global Navigation Satellite System

(GNSS) of the nine materialized GPS points (Figure 9-1). The study also included a surface-

underground mooring traverse of the La Maruja mine (six levels) so that the underground database

could be tied to the MAGNA SIRGAS reference grid. Using the information, the required geometric

coordinate transformation, between the ARENA and MAGNA SIRGAS reference framework has been

established as well as the use of the UTM (Universal Transversal Mercator) coordinate system. CGM

exploration geologist then adjusted the database and geological information into the standardized

system and supplied SRK with the converted database on March 17, 2020.



Source: SRK, 2020

Figure 9-1: Development of 3D Topography for the Project Showing LIDAR Survey Points, Shadow Model and 3D View of 1 m Resolution LIDAR Datapoints

9.1.2 Surface Geochemistry

CGM collected 1,880 rock chip samples and 700 soil samples on surface in Echandia, for a total of

2,580 samples. The geochemical samples identified anomalies coincident with low magnetic

anomalies covering an area of about 800 m by 1,100 m in size.

9.1.3 Geophysics

During 2007 and 2008, a helicopter survey which included both magnetic and radiometrics was

completed.

9.1.4 Surface Geological Mapping

Geological mapping at 1:1,000 scale has been carried out on surface, although outcrop exposures are

limited away from the steep face of Cerro El Burro above Marmato.



9.1.5 Underground Geological Mapping

CGM supplied AutoCAD drawings of mine level plans and sections for all veins currently being mined

(Levels 16 through 21). SRK has been supplied with this information for the current update and utilized

the information during the construction of the geological model. The level plans and information have

a degree of time lag as they are not updated on a routine basis (every six months) but based on the

current production levels at CGM. It is not anticipated that any changes will have a significant impact

on the MRE. SRK has used these underground level maps (Level 16 through Level 21) as the basis

for the current interpretation of the veins, which has been supplemented with information from mining

where available. An example of a level plan is shown in Figure 9-2.

Source: CGM, 2020

Figure 9-2: Example of Level Plan from CGM (Level 20)



9.2 Sampling Methods and Sample Quality

Two stages of exploration have been completed at Zona Alta (#CHG_081 license). The initial phase

used basic hammer and chisel techniques to cut channels 2 m long and 5 cm wide by 1 cm deep. Due

to the relatively poor quality of sampling, many of these samples were repeated during a second stage

of exploration using a hand-held core saw.

Where access was not possible due to poor ground conditions, the original hand-cut samples have

been retained in the database but are considered of lower quality for classification purposes. The Zona

Alta area is currently mined by multiple small-scale mining operations and the Company does not have

input into the sampling processes used by the miners, nor is an active database of any sampling

compiled within these areas. SRK highlights that these sample locations are within the Zona Alta

CHG_081 license which are not included in the current Mineral Resources and are deemed to have

no material impact on the current estimate.

The Company completes routine grade control sampling using channel sampling within the current

mining operation. The process is completed by both mine geologist as part of the routine grade control

process and exploration geologist for verification. Differences exist between the two procedures, as

summarized below.

9.2.1 Mine Geology - Channel Sampling Procedure

The CGM mine geologist collect channels samples as part of the routine grade control process within

the current mine. The following is description of the sampling procedure:

• Once the underground advance faces are cleaned, the geologist completes a brief mapping

and description rock face, including samples locations, lithology, alteration, structures and

mineralization (intensity, styles). The information is handwritten and stored in paper format.

• The geologists then mark the limits of the samples on faces using spray paint. The limits are

defined based on lithological and mineralization contacts, including intensity and style of

mineralization. The minimum and maximum length used are 0.5 m and 1.0 m. The channel is

marked perpendicular to the mineralized structures (Figure 9-3).

• The geologist locates the samples using tape and compass from the closer surveyed control

point. The survey of the underground workings is done using total station (Figure 9-4)

Source: SRK, 2020

Figure 9-3: Channel Sample Marks in Marmato



Source: SRK, 2020

Figure 9-4: Underground Workings Survey Using Total Station

The geology technicians clean the surface and collect rock chips using chisel and hammer with the

aim to construct a channel of 10 cm width and 2 to 5 cm depth. The rock chips are collected in a plastic

pan and then packed into plastic bags which are labelled and then closed with tape or ties. The

samples are identified using metal numbering plates. Figure 9-5 shows the helpers collecting the

samples and the packed sample. The final weight of each sample varies from 0.5 to 1 kilogram (kg).

Source: SRK, 2020

Figure 9-5: Sample Collection and Packing



The samples collected during the day are delivered to a company geologist, who reviews the samples

and personally delivers the samples to the onsite laboratory to provide a chain of custody. SRK noted

that internal quality controls are not included in the sample stream by the geologist and SRK has

recommended that insertion of samples should be completed to follow generally accepted best

practice. In the absence of quality control information SRK has relied upon reconciliation of the planned

versus head-grade from the grade control systems to determine if the performance of the channel

sampling is reasonable. A study of the planned versus head-grades for 2006 to 2019 (discuss further

in Section 12.1.2) shows the differences in the grades range between -10% to +8% on an annual basis

but the overall performance is in the order of 98% during this period, which SRK considers reasonable.

While the study add confidence to the mine sampling it should not replace the need for an industry

standard QA/QC protocol in future sampling.

The samples are collected approximately every 2 m along the vein according to the advance of the

underground working (Figure 9-6). The geologists try to maintain constant the distance from the floor

to the sample channel.

Source: CGM, 2019

Figure 9-6: Distribution of Channel Sampling Along the Vein

9.2.2 Channel Sampling – Exploration

The CGM exploration staff conduct channel sampling of the underground workings which have been

focused on crosscuts. The procedure differs to the one used by the mine geology department, with

more focus on the sample size quality. The following is the description of the procedure:

2m

2m

2m

2m

muestra 1

muestra 2

muestra 3

muestra 4

muestra 5



• The surface of the underground tunnel/crosscut is cleaned with water to expose fresh rock.

The geologist initially completed a geological description of the channel, with the samples

were logged using the same logging codes as utilized in the diamond drilling procedures

including the description of:

o Lithology (color, texture, grain size)

o Alteration (type, intensity and mineralogy)

o Mineralization (styles, intensity, mineralogy, quantification)

o Structures (description, Dip-Dip/Dip, size, width, mineralogy, counting)

• The geologist locates the start point of the channel using tape and compass from a reference

point previously surveyed with total station

• A photograph of each sample is taken and registered with the sampling information

• The geologist defines the interval length of the sample (minimum 0.5 m and maximum 2 m),

perpendicular to the mineralization trend and respecting the changes in lithology, alteration

and mineralization marking the limits of the samples with aerosol paint. The quality controls

are defined (Blanks, duplicates, reference materials) and their tickets and bags are marked

• The geology technicians delimit the channel using two parallel line guide marks separated

7 cm apart. Two cuts are made using a diamond saw following the parallel horizontal lines, at

an approximately depth of 3 cm along the line guides. To facilitate the collection of the sample,

a vertical cut is made with the diamond saw perpendicularly to the parallel line guides every 5

to 10 cm (Figure 9-7).

Source: CGM, 2019

Figure 9-7: Channel Sample Cut Using Electrical Saw

• The sample width and depth were designed to give a sample weight similar to a split HQ drill

core sample.



• The surface is cleaned with water and a metallic brush and then the collection of the sample

is completed using hammer and chisel directly into the bag (canvas or plastic) where the

identification ticket had been introduced.

Source: CGM, 2019

Figure 9-8: Identification Ticket and Bags Used to Pack the Channel Sample

• Approximately 10 kg per 2 m of sample is collected.

• In selected samples a second channel is constructed above or below to collect a field

duplicate.

• Once the channel sample is taken a photograph is taken for record, which can also be used

to assess sampling quality.

• The geologist in charge of the QA/QC reviews the samples, photographs and inserts quality

controls as defined by the Company’s internal protocols which includes; duplicates, standards

and blanks.

• The collected samples are returned to surface by the Company geologist and packed in

batches of maximum 150 samples (minimum 20) including the controls for submission to the

laboratory. The bags with the samples are sent to an external commercial laboratory using

Company vehicles with a person who is responsible for the delivery of the shipment, and a

chain of custody maintained.

9.2.3 SRK Opinion of Quality

SRK reviewed the sampling locations of a large continuous exploration cross-cut during an

underground site visit and is satisfied that the sampling procedures used are in line with industry best

practice and no evidence of selective sampling of higher grade vein material was evident.



In an underground operation, SRK recommends the use of routine sampling compared to selective

sampling as preferred to ensure the best confidence for the geological models, as it is just as important

to know where low grade exists for mine planning requirements.

Mr. Giovanny Ortiz of SRK completed a review of the mine geology grade control sampling in

September 2019 and February 2020. While SRK noted that the current procedures produced

reasonable samples, improvements in the procedure were recommended which included:

• By definition, sampling must be equi-probable and for this type of vein mineralization,

systematic sampling is used, taking samples every 2 m in channels perpendicular to the

direction of the vein.

• In each advance, sampling should be completed using consistent sampling procedure to limit

any potential sampling bias. Sample locations should be standardized and perpendicular to

the structure where possible. Sample lengths should be maintained where possible with

avoidance of multiple samples being taken in areas of potential high grades.

• The defined channels used for sample selection should be homogeneous in width and depth

in order to take all the components of the mineralization, avoiding the collection of a greater

quantity of softer material that may generate bias in the results of the sample towards the soft

material.

• The use of a diamond saw to make two cuts that delimit the channel as used by exploration is

preferred to manual sampling as it produces a more homogeneous sample, although the loss

of fine material must be avoided.

• The delimitation of the samples according to the geological contacts must always be respected

and the technicians must adhere with the sampling contacts set by the geologists.

• The weight of the sample should be a reflection of sample length, with the laboratory

procedures adjusted should additional splitting be required for sample sizes with greater

weight.

• To collect the rock sample, it is recommended to use a container that allows for cleaning to

avoid cross contamination. Direct packaging can also be done in the plastic bag, if it is possible

to capture the entire sample without sample loss.

• SRK recommends regular training of sampling assistants to keep protocols on track and to

raise awareness of the importance to CGM of good sampling methodology.

• In order to guarantee the quality of the results, SRK recommends that the mine geology

department implement a control program for quality of all chemical sample preparation and

analysis processes in the internal laboratory.

All CGM verification sample points were surveyed using either total station of theodolite. A total of

4,285 samples (over 6,699 m) have been taken over 1,431 channels. SRK has integrated these

channels into the database and treated them as horizontal drillholes, with samples cut to sufficient size

to relate to that of a diamond drillhole. SRK considers this approach to be acceptable and has used

this data in producing the resource estimates presented here.



9.3 Significant Results and Interpretation

SRK has reviewed the sampling methods and sample quality for the Marmato Project and is satisfied

that the results are representative of the geological units seen and that acceptable minimal biases

have been identified. SRK has accepted the results from the channel sampling program as presented

for the definition of geological model and Mineral Resources at the Marmato Project.

In the updated database, SRK notes some channel samples from the mine have been limited to the

vein only and are not supported by other channels. In cases where this occurs SRK has selected

samples for use in the estimation process during the geological modelling stage. Using the

interpretation of veins and disseminated material surrounding the veins SRK has been able to account

for the vein sampling spatially. In areas with isolated grades or a lack of continuity, SRK assigns

geological coding to limit or remove the impact on the estimation process.

The impact of additional short samples that have been logged typically with the lithology “P1”, within

the larger indicator-based grade shells could potentially result in the over-estimation of grade and

therefore require further restriction. SRK tested a number of scenarios and has made additional

adjustments for the estimation procedures within the porphyry sampling by applying filters on the

information used during geological modelling and the estimation process. The adjustments have

resulted in a reduction of both tonnage and contained metal within this zone. SRK has discussed these

issues with the CGM geology team and will work on a method to improve the modelling of these

domains in future estimates.

SRK has reviewed the methods employed by the Company during the underground sampling of Zona

Baja which showed clearly marked sampling intervals and associated check sampling. It is SRK’s view

that the sampling intervals and density of samples are adequate for the definition of a compliant MRE.

SRK recommends the Company continue with the current underground sampling program on the lower

levels of the CGM mine as per the current exploration program.

There has been limited increase in the underground channel database between the PEA and PFS

geological models. A total of 272 new channels exist in the database all taken by the CGM geological

department. The total cumulative length of the channels is 796 m, but contains a combination of the

routine channel sampling in the veins which accounts for 262 channels ranging from 0.3 to 5.2 m, for

a total combined sampling length of 569.4 m. In addition to the routine vein channels, the mine has

conducted a series of channels (following exploration protocols), within the transitional areas of the

MDZ on level 21. These channel samples are taken from cross-cuts across the width of the

mineralization and range in length from 11.7 to 35.9 m, for a cumulative length of 226.6 m.

The results of the latest drilling have been compared to the geological mapping information to confirm

location of the veins, faults and potential transitional material with the MDZ (Figure 9-9).



Source: SRK, 2020

Figure 9-9: 2D Plan View of Sampling Data Versus Vein Interpretations, Showing New Sample Data Highlighted in Red, Versus Plan Section of Veins in Blue (Level 1250 M)



10 Drilling For the purpose of the Mineral Resources, the combined CGM and Gran Colombia database from all

licenses has been used inclusive of drilling and sampling information from Zona Alta (#CHG_081), and

Echandia (#RPP_357).

10.1 Type and Extent

The drillings performed vary in depth, inclination and direction depending on the objectives plotted in

the different drilling campaigns. The Company initially targeted a regular spacing of drilling of between

150 and 200 m for new mineralization targets, which are latter infilled to 50 to 75 m.

Drillholes, where regularly spaced, are inclined -60 and -75° predominantly to the southwest, with

occasional scissor holes towards the northeast. Fan drilling has been utilized both at surface and from

underground, which are also typically orientated towards the southwest, with a small number of less

extensive fans orientated towards the northeast.

In title #014-89m, different diamond drilling campaigns have been carried out with core recovery in

diameters from HQ to AW, the latter in rapid mineralization verification procedures adjacent to mining

areas; for HQ and NQ diameters are more specifically related to exploration-focused perforations.

A technical report completed by SRK on September 4, 2011, titled “A NI 43-101 Mineral Resource

Estimate on the Marmato Project, Colombia”, provides in-depth detail on the historic drilling programs

completed from surface.

Table 10-1: Summary of Drilling Completed by Company

Company Series Number of Holes Total (m)

Zona Alta

Compañía Minera de Caldas CMdC 205 46,377.8

Minerales Andinos de Occidente MAdO 146 45,095.0

Zona Alta Subtotal 351 91,472.9

Echandia

Colombia Gold CGD 20 5,933.4

Gran Colombia Resources CGD-GCL 75 11,184.7


Mineros Nacionales MNL 6 768.4

Echandia Subtotal 189 55,474.4

Zona Baja

Mineros Nacionales CNQ 47 14,873.0

Mineros Nacionales CNQ-MNL 25 1,803.4

Minera Phelps Dodge CNQ-PDG 6 696.0


Mineros Nacionales MNL 574 47,869.8

Gran Colombia Exploration CALDAS 57 27,788.6

Zona Baja Subtotal 817 131,997.9

Grand Total 1357 278,945.2

Source: SRK, 2020



Source: SRK, 2020

Figure 10-1: Location Map Showing Drillholes Completed at Marmato by Company



Source: SRK, 2020

Figure 10-2: 3D View of Sampling Data, Showing New Exploration Drilling Data Highlighted in Red and Mine Drilling in Purple (Looking North)

10.2 Procedures

All surface hole collars have been surveyed using a Differential Global Positioning System (DGPS)

and have been surveyed to a high degree of confidence in terms of the XY location. The Z locations

have been adjusted to the topography. Underground drilling collars have been surveyed by the mines

survey department and verified against existing development.

Drilling has been completed by various drilling contractors during the history of the project. While

different drilling contractors have been used over time the Company ensured that:

• A geologist is assigned to each drilling machine

• A trained technician was assigned to measure recovery and RQD measurements

• The transport of the drilling witnesses from the machine to the area of core shed is completed

by a Company employee

• Logging, core cutting and sampling is supervised by suitably trained employees

• Prior to sending samples to the laboratory, all sample bags and number strings were checked

for continuity and sample bag integrity

• All diamond cores were photographed as a routine documentation of samples



• Drill core was stored in locally constructed wooden core boxes with painted labels on the end

of each core box detailing box number, drillhole number and sampling intervals

Source: SRK, 2020

Figure 10-3: Core Photographs Before and After Making the Respective Cut and Sampling



Once the initial transportation and documentation of the hole has been completed at the core shed,

the following processes are completed:

• Conduct a quick log of the geology (i.e. the rapid review of lithological contacts and the most

important information even with approximate depths based on the drill plugs)

• Documentation and measurement of recovery

• Complete a geotechnical log including lithology, recovery (%), RQD, fractures, Jn conditions

and weathering

• Complete a geological log including lithology, mineralization, alteration

• Definition sampling intervals (Minimum 50 cm; Maximum 200 cm)

• Mark the cutting line for core cutting

• Insertion of appropriate quality control standards

• Selection of samples, for petrographic, density and metallurgical testwork

10.2.1 Core Storage

CGM constructed a core storage facility at Marmato during 2010 (Figure 10-4), which acts as the main

exploration facility with logging and offices setup also located at the facility. The core facility has

approximate dimensions of 70 m wide and 90 m long; the capacity of the facility for the storage of the

drilling cores is approximately 350,000 m including the coarse and fine rejects of the samples sent to

the laboratory for geochemical analysis.

SRK has visited the core storage facility during multiple site visits and found the facility to be organized

and clean, with sufficient space for the ongoing exploration.

Source: SRK, 2019

Figure 10-4: Core Storage Facility at Marmato Constructed in 2010 and Current Status 2019



To confirm that no bias has been introduced, a test program was completed whereby samples of the

cuttings were taken from the core saw tray and sludge samples were taken from the sumps used to

settle out fine solids from the drill water. The settling tanks were installed as part of the environmental

management plan for drilling. The core intervals to which these samples correspond were recorded so

that the cuttings and sludge sample grades can be compared with the average grade for the interval.

10.2.2 Collar Surveys Surface

All surface drill hole collars have been surveyed using a DGPS and have been surveyed to a high

degree of confidence in terms of the XY location. Data has been provided to SRK in digital format

using UTM grid coordinates.

Drillhole collar elevations have been adjusted for errors based on projections on a digital terrain model

(dtm) surveys based on the Ikonos satellite imagery, which gives contour levels every 2 m. It is SRK’s

view that even given the extreme topography found at Marmato that the current procedures site the

collar locations with a sufficient degree of confidence.

10.2.3 Collar Surveys Underground

The collar for underground drillholes are defined using a differential GPS and total station measures.

The geologist adjusts the coordinates as required based on conditions underground. SRK notes that

in some cases the drilling is assumed to have occurred from a single location within an underground

drilling station but notes in reality there are likely adjustments for orientation and setup which have

occurred but not been accounted for in the final database. SRK does not consider these adjustments

to have a material impact on the estimation and modelling process.

Inside the mine, the topographic survey is used to verify the accuracy of the result of the calculations;

the adjustment of the points is performed by the Mine surveyors and the results of the measurements

are verified by a GIS Specialist.

10.2.4 Drilling Orientation

Down-hole directional surveys were conducted using a GyroSmart digital gyro tool, manufactured by

Flexit Navigation A.B. (Flexit) and Imego A.B. (Imego) of Sweden, which was purchased from Ingetrol.

Prior to the purchase of this instrument in 2007, down-hole directional surveys were conducted using

a Flexit Multishot tool supplied by Terramundo.

During the initial exploration, drillholes were regularly spaced and orientated -60 and -75°

predominantly to the SW, with occasional scissor holes towards the NE. More recently with the focus

on the MDZ, the Company has used fan drilling from underground adits, which are also typically

orientated towards the SW, with a smaller number of less extensive fans orientated towards the NE.

Drillholes have been drilled from four purpose-built underground drilling stations with two contractor

rigs being used to date. Three of the drilling stations are located on Level 20 with a single station

established on Level 21. Drillholes have been drilled in a fan pattern and dips ranging from -20 to -75°

predominantly to the southwest (ranging from 151º to 255º). Holes targeting the upper portions of the

MDZ have shallower angles while the deeper targeted holes longer (more than 400 m) and steeper

(Figure 10-5 and Figure 10-6).



In addition to the drilling completed by the CGM exploration team (MT-IU series), the current operating

mine has also completed routine exploration drilling ahead of mining. The routine exploration is

typically completed using horizontal drilling from the existing drives to aid in the mapping and

delineation of the known veins prior to mining.

Source: SRK, 2020

Figure 10-5: Plan Showing Primary Drilling Orientation to the South and Southwest Relative to the Main Mineralization Orientation at Depth

LegendVeinsFaultsHistorical DrillingNew Drilling (PEA and PFS)

Primary orientation of deep mineralization



Source: SRK, 2019

Figure 10-6: Cross Section (Orientated Looking Northeast), Showing Orientation of Drilling Relative to the Deep Mineralization, and Horizontal Drilling in the Current Operation

LegendVeinsFaultsHistorical DrillingNew Drilling



10.3 Interpretation and Relevant Results

The updated drilling database indicates that the veins typically range between 0.5 and 5 m wide and

extend for 250 to 1,000 m along strike, and 150 to 750 m down dip. This is supported by underground

mapping and mining which has confirmed that individual vein structures have good geological

continuity and can extend for 100 to 800 m along strike and 100 to at least 350 m down dip. Between

2017 and 2020, CGM has worked on updating the quantity of the underground channel sampling

captured in the database, which has increased the information available to model the vein domains in

greater detail, which has be integrated with detailed level mapping of the veins.

The broad zones of veinlet mineralization in the porphyry domain modelled initially by SRK in 2017

typically varied from 10 to 230 m wide, reaching up to 340 m wide in areas of significant veinlet

accumulation, while extending with good geological continuity for between 200 m and approximately

950 m along strike and between 100 and 900 m down dip. SRK has updated these domains during

the 2019 geological modelling process using more discrete zones and application of an indicator grade

shell approach using a 0.5 g/t Au CoG.

At depth within the central portion of the deposit, SRK has noted a zone of elevated grades which has

been referred to as the higher grade MDZ (more than 2 g/t Au). This zone is indicated to be continuous

along strike for approximately 500 m and has a confirmed down dip extent that reaches up to 800 m,

with a thickness that varies between 35 and 150 m. It is possible that the main MDZ mineralization is

bounded within a series of faults but limited drilling at the edges of the deposit make confirmation

difficult to assess at this stage. SRK notes that the sampling lengths are not perpendicular to the

defined steeply dipping structures but notes that access to ideal drilling locations is limited to current

underground drilling stations. Drilling of the deeper portions of the deposit will not produce ideal

intersections from the current stations, without considerations of directional drilling. The Company is

currently investigating the option to use directional drilling in future drilling programs.

In the QP’s opinion the drilling completed to date has produced reasonable intersection angles to

define the mineralization in three dimensions. In the case of intersections in the latest round of drilling

between the PEA and PFS, which identified additional mineralization in the hangingwall, SRK has

limited the spatial extents of the geological interpretations until further infill drilling establish continuity.

SRK considers the hangingwall mineralization to represent upside to the current model and additional

exploration is recommended prior to mining to ensure underground infrastructure is designed outside

of the mineralization.



11 Sample Preparation, Analysis and Security

11.1 Security Measures

The Chain of Custody procedures for sample security were set up for the Company by Dr. Stewart

Redwood in December 2005 (with the latest update in August 2009). During the initial exploration

(2005 to 2006), sample numbers were created in the field based on a combination of the sample

location, sample type and sample point, with descriptions of each sample noted in a field book and

later transcribed. While providing useful information, the decision was taken in 2006 to change to a

sequential numbering system based on preprinted sample tickets. In both cases the sample numbers

have been transcribed on the sample bags to avoid errors (e.g. lost tickets).

At the drill rig, the drilling contractors are responsible for removing the core from the core barrel (using

manual methods) and placing the core in prepared core trays (3 m length). The core is initially cleaned

to remove drilling additives, but attempts are made to ensure fine material is not lost. Once completed,

the core tray is closed with a wooden lid, hammered shut, and CGM geologists or technicians take

possession. The drill core is then transported to the core shed for selection of sampling intervals and

initial sample preparation. On receipt at the core shed, CGM geologists and technicians follow the

logging and sampling procedures laid out in Section 11.2. Once completed and the half core has been

photographed, the core boxes are again sealed and then transported to the onsite core storage facility.

The core storage facility is within a secure area with a single access gate controlled by a 24-hour

security guard.

In preparation for shipment, samples were packed into nylon rice sacks with approximately five

samples per rice sack. The shipments were accompanied with the laboratory submittal forms and were

transported to Medellín. Samples were accumulated at sample dispatch (in the case of historical holes

this was a warehouse in Medellín) until a hole was completed. Drillholes were only submitted in their

entirety once sampling was completed. The samples were transported by CGM employees to the

preparation facilities. Upon reception at the sample preparation facility, the laboratory company

checked that the samples received matched the work order and signed that it had accepted the

samples.

Once the sample preparation was completed, the laboratory dispatched the sample pulps by courier

to selected overseas laboratories. The laboratories were instructed to retain excess sample pulps after

analysis which can be used in the event that check analyses are requested by CGM.

The coarse sample rejects and sample pulps from the preparation facilities in Medellín were picked up

by CGM technicians during routine sample shipments to the preparation facilities. The coarse rejects

and pulps were returned to the CGM core shed at Marmato for long-term storage.

11.2 Sample Preparation for Analysis

11.2.1 Historical Sample Preparation (Pre 2010)

Prior to the opening of the Inspectorate and SGS sample preparation laboratories in Medellín in

August 2006 and November 2007, respectively, there were no internationally certified sample

preparation laboratories for mineral exploration in Colombia. At the start of exploration work by CGM,

all samples had to be sent to other countries to be prepared and analyzed. Entire rock samples were

sent by air to Inspectorate in Sparks, Nevada (ISO 9001:2000 and ISO 9002:1994 certified) for



preparation and analysis, or to ALS Chemex in Quito, Ecuador (ISO 9001:2000 and ISO 17025:2005

certified) for preparation, with analysis by ALS Chemex in Lima. ALS Chemex in Reno, Nevada was

also used for some check analyses.

The Sparks analytical laboratory was used until late 2007; however, considerable QA/QC problems

were experienced during 2007 as well as long delays in turnaround time and from late 2007 to 2010

the Lima analytical laboratory was used. The analyses from Inspectorate’s Sparks laboratory which

failed QA/QC were repeated. Other samples analyzed initially at Inspectorate’s Sparks laboratory were

re-analyzed at Inspectorate’s Lima, Peru laboratory. Only sample batches that passed QA/QC were

accepted and stored in the final database.

The secondary laboratory used was SGS (ISO 9001 certified) at a sample preparation facility in

Medellín and at their analytical laboratory run by SGS del Perú S.A.C., El Callao, Lima. Inspectorate

is used as the laboratory for check on any SGS submissions and replicate assays of samples analyzed

initially at SGS del Peru S.A.C in Callao, Lima, Peru.

The sample preparation at the Inspectorate laboratory in Medellín consisted of drying the entire sample

and crushing it to more than 70% passing -10 mesh by jaw crusher and roll mill. This was later changed

to more than 85% passing -10 mesh using a TM Terminator Jaw Crusher. A split of 250 to 500 grams

(g) was then obtained using a Jones splitter and was pulverized to more than 80% passing -150 mesh

with Labtech LM2 pulverizing ring mill. Tested barren silica sand was used as a clean wash between

each sample in pulverization.

The sample preparation procedures at the SGS laboratory in Medellín and SGS Colombia S.A. facility

in Barranquilla, comprised drying the sample, crushing the entire sample in two stages to -6 mm and

-2 mm by jaw crusher (more than 95% passing), riffle splitting the sample to 250 to 500 g, and

pulverizing the split to more than 95% passing -140 mesh in 800 cubic centimeters (cm3) chrome steel

bowls in a Labtech LM2 pulverizing ring mill (preparation code 321).

The sample preparation method at the Inspectorate laboratory in Sparks, Nevada was to dry and crush

the entire sample to more than 85% passing -10 mesh by TM Terminator Jaw Crusher, spilt 250 g to

300 g using a Jones splitter and pulverize this to more than 90% passing -150 mesh with a Labtech

LM2 pulverizing ring mill. Tested barren silica sand was used as a clean wash between each sample

in pulverization (rock chip 0 to 10 pound (lb) method).

The sample preparation procedure at the ALS Chemex laboratory in Quito was to log the sample into

the tracking system, weigh, dry, crush the entire sample to more than 70% passing 2 mm, split off up

to 1.5 kg and then pulverize the split to more than 85% passing 75 microns (code PREP-32).

11.2.2 Sample Preparation Mine Sampling (2010 – 2017)

The sample preparation method at the internal CGM mine laboratory comprised of drying the sample,

crushing the entire sample to -5 mm by jaw crusher (more than 95% passing), riffle splitting a sub-

sample of 200 to 300 g and pulverizing the sub-sample in a disc mill to more than 80% passing -200

mesh. SRK visited the facility during the 2017 site inspection and noted new equipment had been

purchased and was not currently in use at the time of visit (Figure 11-1). The new equipment was

consistent with those used at the third-party commercial laboratory. One issue noted during the site

inspection is the stacking of sample trays (full) prior to pulverizing, which SRK does not consider to be



best practice as this could result in cross contamination of samples. Ideally sample should be stacked

on individual trays on a trolley as shown in Figure 11-1.

Source: SRK, 2017

Figure 11-1: Sample Preparation at Mine Laboratory Showing New Equipment (Crusher and Pulverizer)

Since January 2010, the primary laboratory used for the exploration samples in the drill and

underground sampling programs was ACME Laboratories for sample preparation in Medellín, and

analytical laboratories in Sparks, Nevada, USA and Lima, Peru. The 2011 drill program utilized the

ACME sample preparation laboratory in Medellín and the ACME assay laboratory in Santiago, Chile.

In addition, the SGS laboratory in Lima, Peru was used as a check laboratory.

SRK visited the ACME sample preparation facilities on November 4, 2010. The sample preparation

method at ACME, Medellín was to dry the sample in large controlled and crush the entire sample to

more than 85% passing -10 mesh by TM Terminator Jaw Crusher (Figure 11-2).

The sample is then spilt to 250 to 300 g using a Jones splitter and pulverized to more than 90% passing

-150 (75 µm) mesh with a Labtech LM2 pulverizing ring mill.

Tested barren silica sand was used as a clean wash between each sample in the crushing and

pulverization stages (rock chip 0 to 10 lb method).



Showing: (a) Terminator Jaw Crusher (b) Jones Riffle Splitter (c) LM2 Mill (d) Final Bar-Coded Sample Pulp Source: SRK, 2010

Figure 11-2: Sample Preparation Facilities at ACME Laboratories in Medellín

11.2.3 Sample Preparation

Since 2017, CGM has used SGS Laboratories in Medellin is the primary laboratory for both sample

preparation and analyzing all exploration drilling core samples. All CGM mine drilling has undergone

preparation and analysis at ALS Laboratory, to ensure sample quality. CGM has incorporated routine

check analysis on each laboratory with secondary assays at ALS for the SGS submissions and vice-

versa.

(a) (b)

(c)

(d)



Samples sent to SGS were prepared using method PRP93, which involved drying in oven at 100°C

followed by jaw crushing. The sample was crushed to 90% passing -10 mesh size. The crusher was

cleaned with compressed air between samples. A 250 g split, using a Jones splitter, was pulverized to

95% passing -140 mesh using a ring and puck pulverizer.

11.3 Sample Analysis

The ACME laboratory in Santiago analyzed the samples (from the 2010 to 2012 drill programs) for

gold by fire assay (FA) with atomic absorption spectrophotometer (AAS) finish. Samples over 10 g/t

Au were assayed by FA with gravimetric finish. Silver was assayed by aqua regia digestion and AAS

finish. Silver samples above 100 g/t were assayed by FA with gravimetric finish.

The historical samples by Conquistador (one of the previous owners) have been assayed by Barringer

for gold by FA with atomic absorption (AA) finish and checks by gravimetric finish for some high grade

samples. Silver was determined by acid digestion with AA finish.

A detailed description of the sample analytical procedure undertaken for the 2011 SRK MRE (January

2011) and is provided, given the incorporation of these samples in to the current estimate:

The Inspectorate laboratory in Lima analyzed the samples for gold by FA with an AA finish (detection

limits 0.005 parts per million [ppm] to 3 ppm, method FA/AAS). Silver was analyzed by aqua regia

digestion and AA finish (method AA, detection limits 0.2 to 200 ppm). Over-limit gold assays (above

3,000 parts per billion [ppb] or 3 ppm) were repeated by FA (1 assay ton, 29.2 g) with gravimetric finish

(method Au FA/GRAV). Samples above a 200 g/t silver upper limit of detection were repeated by FA

(1 assay ton, 29.2 g) with gravimetric finish (method Ag FA/GRAV). Samples were analyzed for

multiple elements by aqua regia digestion and inductively coupled plasma (ICP) finish (32 Element

ICP Package for Ag, Al*, As, Ba*, Bi, Ca*, Cd, Co, Cr*, Cu, Fe, Hg, K*, La*, Mg*, Mn, Mo, Na*, Ni, P,

Pb, S*, Sb*, Se, Sn*, Sr*, Te*, Ti, Tl*, V, W, Zn). Inspectorate states that for elements marked * the

digestion is partial in aqua regia in most silicate matrices and the analysis is partial. Over-limit zinc and

lead analyses (more than 10,000 ppm) were rerun by aqua regia digestion and AA. Multi-element

analyses were not carried out on the final batches of samples.

The Inspectorate laboratory in Sparks, Nevada analyzed samples for gold and silver by FA with an AA

finish for gold (detection limits 2 ppb to 3,000 ppb) and AA finish for silver (detection limits 0.1 ppm to

200 ppm) (method Au, Ag FA/AA/AAS). Over-limit gold assays (above 3,000 ppb or 3 g/t) were

repeated by FA with gravimetric finish (method Au FA/GRAV). Samples above a 200 ppm silver upper

limit of detection were repeated by FA with gravimetric finish (method Ag FA/GRAV). Samples were

analyzed for multi-elements by aqua regia digestion and ICP finish (30 Element ICP Package for Ag,

Al*, As, B*, Ba*, Bi, Ca*, Cd, Co, Cr*, Cu, Fe, Hg, K*, La*, Mg*, Mn, Mo, Na, Ni, P, Pb, Sb*, Se, Sr*,

Ti, Tl*, V, W, Zn). Inspectorate states that for elements marked * the digestion is partial in aqua regia

in most silicate matrices and the analysis is partial. Over-limit zinc and lead analyses (more than

10,000 ppm) were rerun by aqua regia digestion and AA.

SGS del Perú S.A.C. analyzed samples for gold by FA (30 g sample) with an AA finish (code FAA313;

detection limits 0.005 ppm to 10 ppm), and for silver with an aqua regia digestion and an AAS finish

(code AAS12CP), or three acid digestion with AAS finish (code AAS42C); detection limits in both are

0.3 ppm to 500 ppm). Multi-element geochemical analyses were done by two different methods. One

method (ICM40B) uses a four acid digestion and both ICP-AES and ICP-MS for 50 elements (Ag, Al,

As, Ba*, Be*, Bi, Ca*, Cd, Ce, Cr*, Co, Cs, Cu, Fe*, Ga*, Ge*, Hf*, In*, K*, La*, Li*, Lu*, Mg*, Mn*, Mo,



Na*, Nb*, Ni*, P*, Pb, Rb*, S*, Sb, Sc*, Se, Sn*, Sr*, Ta*, Tb, Te*, Th*, Ti*, Tl*, U*, V*, W*, Y*, Yb*,

Zn*, Zr*), elements marked * the digestion is partial.

The second method (ICP12B) uses a two acid (HNO3 and HCl) digestion and both ICP-AES and ICP-

MS for 38 elements (Ag, Al, As, Ba*, Be*, Bi, Ca*, Cd, Co, Cr*, Cu, Fe*, Ga*, Hg, K*, La*, Mg*, Mn*,

Mo, Na*, Nb*, Ni*, P*, Pb, S*, Sb, Sc*, Sn*, Sr*, Ti*, Tl*, V*, W*, Y*, Zn*, Zr*), elements marked * the

digestion is partial. SGS indicates that the analysis is partial for elements marked * and depends on

the mineralogy. Over limit gold values were repeated by FA with a gravimetric finish (method FAG303)

and a lower limit of detection of 0.02 g/t. Silver grades above 100 ppm and zinc grades above 1% were

repeated by four acid digestion and AA (method AAS41B). Gold and silver for some samples was by

FA with gravimetric finish on 30 g (method FAG323 with lower limit of detection of 0.03 g/t Au and 0.03

g/t Ag).

Since 2017, Caldas has used SGS Laboratories in Medellin as the primary laboratory for both sample

preparation and analyzing all exploration drilling core samples. Samples have been analyzed at each

laboratory for gold and silver by FA. At SGS, the assaying using an Au 30 g AAS (method FAA313).

All CGM mine drilling have undergone preparation and analysis at ALS Laboratory, to ensure sample

quality. CGM has incorporated routine check analysis on each laboratory with secondary assays at

ALS for the SGS submissions and vice-versa.

The CGM channel samples have been assayed at an onsite internal mine laboratory for Au and Ag by

FA with gravimetric finish. SRK reviewed the laboratory and noted some areas of improvement relating

to the state of the equipment. The mine has recently purchased new sample preparation equipment,

which should result in improved assay quality. SRK recommends CGM complete routine check

analysis between the mine laboratory and an independent commercial laboratory in Medellín. SRK

recommends that all exploration samples are kept clear of the mine laboratory to avoid any potential

contamination.

11.4 Quality Assurance/Quality Control Procedures

SRK completed a detailed review of the QA/QC procedures and results as part of the 2012 MRE.

Limited drilling was completed between 2012 and 2017, and since 2018 to early 2020 the drilling was

increased to explore the mineralization at depth. The results are summarized in the current report.

The routine QA/QC program at Marmato comprises certified standard reference materials (CRM),

quartz blanks, preparation duplicates (PD), coarse duplicates (CD), field duplicates (FD) and check

and replicate assays. The CRM, quartz sand blanks and duplicate samples make up the portion of the

QA/QC program which provides ongoing monitoring of the geochemical laboratories. The check assay

and replicate assay samples are submitted at longer time intervals (less frequently) and provide a

secondary control on the accuracy of the geochemical data.

Sampling protocols suggest the following submission rates:

• For the CRM, five random numbers are generated between 1 and 100

• For the FD samples, two random numbers are generated between 1 and 100

• For the PD samples, two random numbers are generated between 1 and 100

• In contrast, the blanks are inserted at points within the sample stream where, based on the

geology, the geologist believes that there is a high likelihood of significant mineralization, and

therefore potential for contamination



CGM employed a database administrator for the QA/QC program at Marmato during this period. SRK

held discussions with the database administrator during the site visit to review how the data was

captured.

SRK has been supplied with a complete QA/QC assay database for the Project in separate excel files

which summarize the submissions in 2019 and 2020 (up to borehole SMT20-018). Within the 2018 to

2020 exploration period when the majority of the drilling has been completed by CGM, a total of 946

certified standards, 742 blanks and 1,096 duplicates, representing approximately 14% of total routine

sample submissions for the CGM drilling programs at Marmato to SGS have been completed.

Additionally, CGM has a total of 133 certified standards, 90 blanks and 87 duplicates, representing

approximately 18% of the total sample routine submissions for the CGM drilling programs at Marmato

to ALS. In addition to the duplicates a series of re-assays and check assays have been completed

using alternative laboratories (SGS vs ALS). In 2018, a total of 190 assays along with the associated

QA/QC were re-assayed from SGS submissions and 42 re-assays from ALS submissions. In 2019, a

total of 981 and 77 re-assays have been selected from SGS and ALS submissions respectively. In

2020, a total of 105 re-assays have been selected to date from SGS. SRK considers the level of

QA/QC submissions to be of acceptable levels for the current stage of the Project.

A summary of the breakdown per sample type and laboratory are shown in Table 11-1 (2018),

Table 11-2 (2019) and Table 11-3 (2020).

Table 11-1: Summary Of QA/QC Sample Submissions During 2018 Submissions To SGS And ALS Laboratories

Marmato Exploration

Original Shipments

Sent Shipments Received Analysis QA/QC

Laboratory Shipments Samples Shipments Samples Standard Blanks Duplicates

SGS 29 4530 29 4530 224 178 275

ALS 10 909 10 909 46 36 36

Re-Analysis Shipments



SGS 6 203 6 203 12 0 1

ALS 3 46 3 46 3 0 1

Source: CGM, 2018

Table 11-2: Summary Of QA/QC Sample Submissions During 2019 Submissions To SGS And ALS Laboratories

Marmato Exploration

Original Shipments



SGS 105 14,295 105 14,295 661 514 747

ALS 11 804 11 804 87 54 51




SGS 18 1,086 18 1,086 58 0 47

ALS 6 86 6 86 6 1 2

Source: CGM, 2020



Table 11-3: Summary Of QA/QC Sample Submissions During 2020 Submissions To SGS Laboratory (Up to SMT20-018)

Marmato Exploration

Original Shipments



SGS 12 1225 12 1225 61 50 74




SGS 4 105 1 15 58 0 47

Source: CGM, 2020

11.4.1 Standards

In the 2019 to 2020 program, 946 Certified Reference Material (Standards) were submitted during

routine submissions to SGS and 133 with ALS submissions.

A summary of the standard codes is shown in Table 11-4. In the submissions to SGS, SRK concluded

the majority of standards had a greater number of overestimations than underestimations. The

discrepancies noted are likely due to occasional laboratory issues, however this has not resulted in a

material bias overall.

SRK has reviewed the CRM results of the 2019 to 2020 program results provided by CGM and is

satisfied with the results and the failures management procedure, which give sufficient confidence in

the assays for these to be used to derive a MRE. CGM has utilized CRM from Geostats Pty Ltd.,

Rocklabs, and OREAS. In 2018-20, 24 different CRM’s were inserted into the sample stream.

Table 11-4: Summary of CRM’s Submitted During Routine Assay Submissions

STM_NAME Std Value SD1Low SD1High SD2Low SD2High SD3Low SD3High

G310-6 0.65 0.61 0.69 0.57 0.73 0.53 0.77

G312-4 5.3 5.08 5.52 4.86 5.74 4.64 5.96

G313-1 1 0.95 1.05 0.9 1.1 0.85 1.15

G313-2 2.04 1.97 2.11 1.9 2.18 1.83 2.25

G314-1 0.75 0.71 0.79 0.67 0.83 0.63 0.87

G314-5 5.29 5.12 5.46 4.95 5.63 4.78 5.8

G315-2 0.98 0.94 1.02 0.9 1.06 0.86 1.1

G914-6 3.21 3.09 3.33 2.97 3.45 2.85 3.57

G914-9 16.77 16.29 17.25 15.81 17.73 15.33 18.21

G915-5 17.95 17.09 18.81 16.23 19.67 15.37 20.53

G915-6 0.67 0.63 0.71 0.59 0.75 0.55 0.79

OREAS-15Pc 1.61 1.571 1.649 1.532 1.688 1.493 1.727

OREAS-60P 2.6 2.56 2.64 2.52 2.68 2.48 2.72

OREAS-62Pa 9.64 9.56 9.72 9.48 9.8 9.4 9.88

OREAS-67A 2.238 2.142 2.334 2.046 2.43 1.95 2.526

SH35 1.323 1.279 1.367 1.235 1.411 1.191 1.455

SH82 1.333 1.306 1.36 1.279 1.387 1.252 1.414

SJ80 2.656 2.599 2.713 2.542 2.77 2.485 2.827

SK94 3.899 3.815 3.983 3.731 4.067 3.647 4.151

SL76 5.96 5.768 6.152 5.576 6.344 5.384 6.536

SN91 8.679 8.485 8.873 8.291 9.067 8.097 9.261

SP73 18.17 17.75 18.59 17.33 19.01 16.91 19.43

SQ88 39.723 38.776 40.67 37.829 41.617 36.882 42.564

OREAS 62c 8.79 8.58 8.79 8.37 9.21 8.16 9.42

Source: SRK, 2020



Source: SRK, 2020

Figure 11-3: Summary of CRM Submissions to SGS In 2019/2020 Program

Source: SRK, 2020

Figure 11-4: Summary of CRM Submissions to ALS In 2019 Program



Source: CGM, 2020

Figure 11-5: Example of Timeline Review Of CRM G914-6 (2019) and G315-2 (2020) Submissions

11.4.2 Blanks

Blanks are inserted at points within the sample stream where, based on the geology, the geologist

believes that there is a high likelihood of significant or high-grade mineralization and therefore potential

for contamination. Coarse and Fine blank samples are submitted (812 in total) to both SGS and ALS



between 2018 and early 2020 (Figure 11-6 to Figure 11-9) the results have been reviewed to check

for any potential evidence of contamination. SRK comments that no evidence has been noted with

limited samples reporting above a value of 0.025 g/t Au.

Source: SRK, 2020

Figure 11-6: SGS and ALS Coarse Blank Submissions 2019



Source: SRK, 2020

Figure 11-7: SGS Coarse Blank Submissions 2020



Source: SRK, 2020

Figure 11-8: SGS and ALS Fine Blank Submissions 2019



Source: SRK, 2020

Figure 11-9: SGS Fine Blank Submissions 2020

11.4.3 Duplicates

The Company has submitted three different types of duplicates during the routine sample submissions.

The three different types have been defined as field duplicates, coarse duplicates and pulp duplicates.

The field duplicates have only been submitted with the exploration drilling submitted to SGS with no

duplicates in the mine drilling programs to ALS. A total of 281 field duplicates have been analyzed

between 2019 and early 2020. The difference in the mean grades from the two data populations is 6%

higher in the duplicate dataset, which returned 1.25 g/t and 1.18 g/t Au respectively. A summary of the

results is shown in Table 11-5 and Figure 11-10. Many samples are outside of 30% of acceptability,

reflecting the heterogeneity of the mineralization but suggest as well that it is appropriate to review the

sampling procedures and carry out some additional training of the geology helpers.

Table 11-5: Summary Statistics for Field Duplicates (2019-2020)

Original Au (g/t) Duplicate Au (g/t)

Mean 1.18 1.25

Standard Deviation 1.79 2.48

Sample Variance 3.21 6.14

Range 10.84 21.46

Minimum 0.006 0.006

Maximum 10.85 21.47

Count 271 271

Source: SRK, 2020



Source: SRK, 2020

Figure 11-10: Summary of Field Duplicate 2019-2020

The coarse duplicates have been submitted in both exploration and mine drilling programs submitted

to SGS and ALS. A total of 274 coarse duplicates have been analyzed at SGS and 24 at ALS between

2019 and 2020. The difference in the mean grades from the two data populations is 5.6% higher in the

duplicate dataset, which returned 1.48 g/t and 1.58 g/t Au respectively at SGS. This is skewed by two

high-grade samples, which once removed reduces the difference to 1% in the mean grades. The

difference in the mean grades from the two data populations is 3.4% lower in the duplicate dataset,

which returned 2.39 g/t and 2.31 g/t Au respectively at ALS. A summary of the results is shown in

Table 11-6 and Figure 11-11. The correlation coefficient in each case is greater than R2> 0.9, which

indicates a strong correlation between the original and duplicate assays. Only a few samples are

outside of the acceptability range of 20% in the two set of samples.



Table 11-6: Summary Statistics for Coarse Duplicates to SGS and ALS Submissions (Au g/t), 2019-2020

SGS Submissions ALS Submissions

Original Au (g/t) Duplicate Au (g/t) Original Au (g/t) Duplicate Au (g/t)

Mean 1.48 1.56 2.39 2.31

Standard Deviation 3.54 3.95 2.43 2.18

Sample Variance 12.56 15.58 5.92 4.75

Range 47.60 45.72 9.23 7.83

Minimum 0.003 0.003 0.036 0.081

Maximum 47.60 45.72 9.27 7.91

Count 274 274 24 24

Source: SRK, 2019

Source: SRK, 2020

Figure 11-11: Summary of Coarse Duplicate Submissions to SGS (left) and ALS (right) for 2019-2020

The pulp duplicates have been submitted in both exploration and mine drilling programs submitted to

SGS and ALS. A total of 276 pulp duplicates have been analyzed at SGS and 27 at ALS between

2019 and 2020. The difference in the mean grades from the two data populations is 0.3% lower in the

duplicate dataset, which returned 1.332 g/t and 1.329 g/t Au respectively at SGS. The difference in the

mean grades from the two data populations is 5% lower in the duplicate dataset, which returned 1.39

g/t and 1.32 g/t Au respectively at ALS. A summary of the results is shown in Table 11-7 and Figure

11-12. It is observed that a high percentage of sample duplicates in ALS are not accomplishing the

acceptability range of 10%, that is an aspect to review with the laboratory.



Table 11-7: Summary Statistics for Coarse Duplicates to SGS and ALS Submissions (Au g/t), 2019-2020

SGS Submissions ALS Submissions

Original Au (g/t) Duplicate Au (g/t) Original Au (g/t) Duplicate Au (g/t)

Mean 1.332 1.329 1.39 1.32

Standard Deviation 2.59 2.50 2.19 2.00

Sample Variance 6.70 6.24 4.79 4.00

Range 25.55 22.51 9.52 9.13

Minimum 0.003 0.003 0.087 0.068

Maximum 25.55 22.51 9.61 9.20

Count 276 276 27 27

Source: SRK, 2020

Source: SRK, 2020

Figure 11-12: Summary of Pulp Duplicate Submissions to SGS (left) and ALS (right), 2019 - 2020

11.4.4 Actions/Reassays

The Company continued the reassay programs on the 2019 and 2020 submissions, respectively.

These represent resubmission of samples to a secondary laboratory (SGS to ALS and vice-versa).

From 2019 to 2020, a total of 1,076 sample pulps were reanalyzed (including reference materials)

which returned similar values across all grade ranges.

The difference in the mean grades from the two data populations is 0.2 % higher in the reassay dataset,

which returned 1.809 g/t and 1.812 g/t Au respectively in the original and reassays. The correlation

coefficient was R2>0.95, which is considered an indication of no bias during this period at SGS. A

number of points are outside of the 10% acceptance range that is an aspect to review with the lab. A

summary of the results are shown in Table 11-8 and Figure 11-13.



Table 11-8: Summary Statistics for 2019 Reassays Program to SGS vs ALS Submissions (Au g/t)

SGS Submissions (Primary)

Original Au (g/t) Duplicate Au (g/t)

Mean 1.809 1.812

Standard Deviation 3.00 3.14

Sample Variance 8.98 9.88

Range 45.13 47.33

Minimum 0.003 0.003

Maximum 45.13 47.33

Count 1,076 1,076

Source: SRK, 2020

Source: SRK, 2020

Figure 11-13: Summary of 2019-2020 Reassay (Secondary Laboratory)

11.4.5 Check Analysis Results

In addition to the re-assay program CGM also completed a check analysis on pulps and reject material

on a quarterly basis. From 2019 to 2020, a total of 249 sample pulp and rejects were reanalyzed from

SGS samples which returned similar values across all grade ranges.

The difference in the mean grades from check pulps is 4.4% lower in the check dataset, which returned

3.33 g/t and 3.19 g/t Au respectively in the original and check assays. Both results return very similar

high-grades (44.67 g/t Au versus 45.8 g/t Au), and the correlation coefficient was R2>0.71, which

shows slight bias towards the original pulps results. A summary of the results is shown in Figure 11-14.

It is noted that a number of points are outside of the 10% acceptance range, which is an aspect to

review with the laboratory. After completing this analysis SRK does not consider it to be material during

this time period.



In comparison the difference in the mean grades from rejects is 7.9% lower in the check dataset, which

returned 3.33 g/t and 3.07 g/t Au respectively in the original and check assays. SRK notes that the

correlation coefficient is improved within the reject check analysis with R2>0.86, showing a slight bias

towards the original pulp values. A summary of the results is shown in Figure 11-15.

It is noted that for the pulps and rejects that a number of points are outside of the 10% and 20%

acceptance ranges, which is an aspect to review with the laboratory. After completing this analysis,

SRK does not consider it to be material during this time period.

Source: Gran Colombia, 2020

Figure 11-14: Summary of Check Assays Completed on Pulp Material (Quarterly Checks), Scatter Plot (Left) and Mean vs Relative Difference Plot (Right)

Source: SRK, 2020

Figure 11-15: Summary of Check Assays Completed on Reject Material (Quarterly Checks), Scatter Plot (Left) and Mean vs Relative Difference Plot (Right)



11.5 Opinion on Adequacy

It is the opinion of the QP that the frequency of QA/QC sample inserted in the 2018 and 2020 campaign

is at an acceptable rate as stipulated in the Company’s internal guidelines (approximately 14%).

In general, it is the opinion of SRK that the results of the QA/QC analysis display a reasonably good

correlation to the original assays and are acceptable for use in defining compliant MRE.

In the opinion of the QP, the sampling preparation, security and analytical procedures used by CGM

are consistent with generally accepted industry best practices and are therefore adequate.



12 Data Verification

12.1 Procedures

12.1.1 Verifications by CGM

CGM has completed a number of verification sampling programs during the history of the Marmato

Project. The work completed has ensured sample integrity and allowed SRK to have confidence to

use the combined historical and CGM data as supplied by the Company. The work completed by CGM

can be sub-divided into the validation and verification of the on-going exploration drilling programs,

and the validation of underground channel sampling from the operating mine.

CGM employs a Database & GIS Manager who is responsible for tracking the samples through the

laboratory. The Sample Order Form is given to the Database Manager. A Microsoft® Excel

spreadsheet is used to track Company reference number, lab order number, date of delivery to lab,

date of receipt of assays by email, date of receipt of certificate and date of receipt of invoice.

The Database & GIS Manager is responsible for receiving the assay results and importing these into

the database. This is the only person with authority to do this in order to maintain integrity and quality

control of the database.

On receipt of each batch of assays for the exploration drilling, the QA/QC samples are checked to

accept or reject the batch. If there is a problem the Chief Geologist is notified and he requests that the

laboratory identify and solve the problem, if possible, or carry out re-analysis, as necessary. If re-assay

is required, either the whole batch or the sample tray between the good QC samples on either side is

re-analyzed. Microsoft® Excel or Access spreadsheets and graphs are used to check QC results and

update these with each batch so that the whole program is monitored progressively.

The laboratory also carries out its own internal QA/QC samples and the results for these are requested

and monitored on an ongoing basis by CGM.

On-going validation included a detailed survey of historical collar positions using a DGPS which

highlighted a number of minor discrepancies (typically less than 10 m). Based on the new survey data,

the database has been updated accordingly and the interpretations adjusted to the new drillhole

positions.

The on-going validation of the underground channel sampling database has been a considerable task,

which has required capture of the sampling information from the mines operating long-sections into a

3D database. The program continued between 2017 and 2019, which has resulted in an increase of

approximately 100% in the size of the channel sample database. To complete the task, CGM has

completed surveys of the existing development to ensure accuracy of the placement of sampling on

the main levels. CGM geologists have then created collar, survey and assay database for each sample

relative to the strike of the deposit using the assumption that sampling has been completed

perpendicular to the vein, as per the mines sampling procedure.

In 2020, CGM commissioned a separate study with WL Engineering (WLI) to review the Marmato

project for internal requirements as part of a license extension application to the Colombian National

Mining Association (NMA). As part of the study WLI completed a detailed validation of the data used

to generate the geological model and mineral resource. These studies included:



• Visiting Competent Person, Peter Bergsneider Serrano

• Visit by a Competent Person, Mauricio Castañeda Gómez

• Spatial verification of the geographic reference system for the topographic surface, and the

Marmato Project Database; Magna Sirgas, Origin West

• Spatial agreement between the surface collars and the surface topography of the Marmato

Project

• Validation of duplicate coordinates and/or anomalous height values outside the topographic

surface or underground work

• Verification of anomalous deviations from the survey table

• Consistency between the gold and silver data reported in the test table versus the laboratory

certificates

• Search for duplicate or abnormal records in the various tables of the Database, and QA/QC

• Verification of the reported values of gold and silver in the modeling compounds, and their

correspondence in the resource model

• Verification of the structural data reported in the various mineralization styles of the Marmato

Project

• Review and verification of available geotechnical data

• Visual verification of modeled solids versus database

12.1.2 Verifications by SRK

In accordance with NI 43-101 guidelines, Mr. Ben Parsons of SRK most recently visited the Marmato

Project on June 11, 2019. The main purpose of the site visit was to:

• Witness the extent of the exploration work completed to date

• Complete an underground site inspection to understand the changes in the geological settings

and possible exposure of the MDZ style mineralization

• Inspect core logging and sample storage facilities

• Discuss geological interpretation and inspect drill core

• Assess logistical aspects and other constraints relating to the exploration property

• Review data for the assay database

• Hold discussions with personnel involved in the current and historical exploration activities

SRK did not complete an independent visit to the SGS Laboratory facility during the recent site visit

but visited the facility previously during the November 2011 site visit also completed by Mr. Ben

Parsons.

In 2019, SRK undertook a number of site visits by specialized geological staff to review both the

structural model controls and implementation for the PFS model, and to witness the sampling

procedures for the mine. The site visits were completed by Mr. Blair Hrabi for the structural review and

by Mr. Giovanny Ortiz for the sampling and mapping review.

SRK has been working with the exploration team since 2017, when data has been captured from the

mine to generate a detailed geological model. For the most recent iteration of the database, in addition

to the site inspection, SRK has completed a series of technical meetings with CGM geologists to review

the on-going capture of the underground channel sampling program and integration into the database.



SRK reviewed the capture and geo-referencing of the underground development with existing

geological maps for each of the mining levels.

SRK highlighted a lack of QA/QC in the operating mine channel sampling program. In the absence of

quality control information, SRK has relied upon reconciliation of the planned versus head-grade from

the grade control systems to determine if the performance of the channel sampling is reasonable. The

mine current run and annual study of the planned grade (based on channel samples and drilling

assayed at the mine laboratory) versus the reported grade (head-grade). A study of the planned versus

head-grades for 2006 to 2019 shows the differences in the grades range between -10.7% to + 8.6%

on an annual basis (Table 12-1) but the overall performance is in the order of 2.3% during this period,

when weighted for tonnage, which SRK considers reasonable, but notes a minor high-bias in results

since, which should be monitored via regular QA/QC.

Table 12-1: Comparison of Mine Planned Grades (Assayed at Mine Laboratory) Versus Head-Grades

Year Planned Grade Au (g/t) Head Grade Au (g/t) Comparison (%)

2006 3.59 3.64 -1.4%

2007 3.51 3.32 5.7%

2008 3.24 3.44 -5.8%

2009 3.24 3.51 -7.7%

2010 3.34 3.39 -1.5%

2011 3.31 3.19 3.8%

2012 3.02 2.84 6.2%

2013 2.94 2.83 3.7%

2014 2.91 2.85 2.0%

2015 2.95 2.79 5.8%

2016 2.83 2.56 10.7%

2017 2.69 2.48 8.6%

2018 2.60 2.67 -3.0%

2019 2.63 2.49 5.4%

Source: CGM, 2020

Source: CGM, 2020

Figure 12-1: Comparison of Planned Versus Actual Gold Grades at Marmato Mine

3.59

3.51

3.24 3.24

3.343.31

3.02

2.94 2.912.95

2.83

2.69

2.602.63

3.64

3.32

3.443.51

3.39

3.19

2.84 2.83 2.852.79

2.56

2.48

2.67

2.49

-15.0%

-10.0%

-5.0%

0.0%

5.0%

10.0%

15.0%

2.00

2.20

2.40

2.60

2.80

3.00

3.20

3.40

3.60

3.80

4.00

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

20

17

20

18

20

19

Diff

ere

nce (

%)

Gra

de A

u (

g/t)

Year

Comparison of Planned versus Actual Gold Grades (Au g/t)

Planned Grade Au (g/t) Head Grade Au (g/t) Comparison (%)



While the study adds confidence to the mine sampling, it should not replace the need for an industry

standard QA/QC protocol in future sampling.

SRK has undertaken basic validation for all tabulated data in both Leapfrog® and Datamine.

Additionally, in order to independently verify the information incorporated within the latest drill program,

SRK has:

• Completed a review of selected drill core for selected holes, to confirm both geological and

assay values stored in the database show a reasonable representation of the Project

• Verified the digital database against the original issued assay certificates

• Visited underground workings to check the continuity of vein and veinlet mineralization at

depth, including a site inspection to levels 19 through 21 to understand changes in the

mineralization styles

• Verified the quality of geological and sampling information and developed an interpretation of

gold grade distributions appropriate to use in the resource model

• Reviewed the QA/QC database for the recent drilling and channel sampling programs

12.2 Limitations

SRK has reviewed the data acquired for the Project and held a number of technical meetings at the

Company office in Medellín to review the progress on the data validation. The efforts should remain

ongoing and a lack of definition in portions of the 3D survey of the mines has limited the ability to

accurately place all the samples in their “true” location. SRK notes that the information for the raise

sampling shows the most significant variations from SRK geological interpretation using mapping

between levels.

SRK has highlighted to CGM that in the validation phase, there still remains a large number of data

points which contain significant mineralization that require constraining which lie outside of the revised

2019 vein interpretations. SRK noted a number of areas where, based on short channel samples, the

geological model would likely result in overstating the tonnage if left unconstrained. Additionally, where

these occur, the grade in any subsequent estimate will overstate grade locally with possible vein or

veinlet material being incorrectly projected into the lower grade porphyry style domains.

The current structure of the database and naming convention for the underground channel sampling

results in some limitations on generating an automated process to update the geological model.

Isolated sampling of veins without surrounding samples can led to overstating the tonnage when using

Leapfrog® and therefore caution has been required to review intersections on a case by case basis.

Additionally, in a number of cases long cross-cut drift sampling has been logged as individual

Hole_ID’s, so restrictions based on length cannot be applied.

Regarding the significant rise in the amount of channel sampling, some of which has not been captured

in the veins or splay model, it is the opinion of SRK that there is potential for over-estimation. SRK has

noted this risk and therefore applied filters to the database to minimize these risks in the porphyry

domain estimates. SRK considers this approach to be conservative, and it has resulted in a reduction

in the contained metal within the domains if estimated with no filter; however, these estimates provide

a more reasonable base case for classification. Further work is recommended between SRK and CGM

to improve the database structure and geological modelling of this domain, prior to any consideration

for use in a mining study.



SRK completed a series of test models, which removed the influence of any of these samples from

the database, using filters on logging codes and channel length, which is considered a more

conservative approach by discounting the influence of short channel samples. The resultant

interpretation contained approximately half the volume of the optimistic scenario for the porphyry

mineralization. Due to the uncertainty in the impact of these domains, SRK has therefore excluded this

material from the current mining assessment. If a solution can be found in the short term to improve

the confidence in the geological continuity, this may represent upside potential. SRK recommends

follow-up work from CGM which includes mapping and verification of the presence of the porphyry-

style mineralization and additional sampling (drilling if required), prior to inclusion in the mine plan.

12.3 Opinion on Data Adequacy

Based on the validation work completed by SRK, the database has been accepted as provided by

CGM’s Resource Geologist. SRK is satisfied with the quality of assays returned from the laboratory

for the latest drilling program and that there is no evidence of bias within the current database which

would materially impact on the estimate.

While there are areas for potential improvement, SRK is of the opinion that the exploration and assay

data is sufficiently reliable to support evaluation and classification of Mineral Resources in accordance

with generally accepted CIM Estimation of Mineral Resources and Reserves: Definitions and

Guidelines (CIM, 2014).



13 Mineral Processing and Metallurgical Testing Metallurgical programs were conducted by SGS Lakefield (SGS) in 2019 and 2020 to evaluate

processing requirements for the MDZ. The 2019 metallurgical program was conducted as part of the

2019 PEA that was prepared for the Project and the 2020 metallurgical program was conducted to

support the current PFS. For ease of reference, the results of the 2019 metallurgical program are

briefly summarized in this section along with the results of the 2020 metallurgical program.

13.1 Metallurgical Program – 2019

The results of the 2019 metallurgical program are fully documented in SGS’s report, “The Recovery of

Gold from Marmato Deposit Samples” dated April 16, 2019. The metallurgical program included

comminution testwork, mineralogical studies and an evaluation of several different flowsheet options

including:

• Whole-ore cyanidation

• Gravity concentration followed by cyanidation of the gravity tailings

• Gravity concentration followed by gold flotation from the gravity tailing and cyanidation of the

flotation concentrate

13.1.1 Metallurgical Sample Characterization

The test program was conducted on test samples prepared from drill core from the East, West and

Central MDZ and a Master MDZ composite was formulated on a weighted basis from the East, West

and Central MDZ samples. In addition, a composite representing the current Marmato material was

also tested. The drill holes and intervals used to formulate the test composites are shown in Table

13-1 and the location of each drillhole is shown in Figure 13-1. Head analyses for each of the test

composites are shown in Table 13-2.

Table 13-1: Drillholes and Intervals for MDZ Metallurgical Composites

Location Drill Hole From (m) To (m)

East Zone

MT-IU-001 138.3 148

MT-IU-004 152.4 182.2

MT-1445 562 589

Central Zone

MT-IU-002 146 179.3

MT-IU-003 209.4 256.5

MT-IU-006 252.5 292.8

MT-1498 334 386

West Zone

MT-IU-007 159.4 173.7

MT-IU-008 251 266.6

MT-IU-011 129.3 144.6 Source: SGS Lakefield, 2019



Source: SRK, 2019

Figure 13-1: Drillhole Locations

Table 13-2: Head Analyses for MDZ and Marmato Test Composites

Element MDZ West MDZ Center MDZ East MDZ Master Marmato Comp

Au (S.M.) g/t 1.54 2.69 2.65 2.32 5.48

Au (Calc.) g/t 1.3 2.61 1.8 2.36 4.83

Ag g/t 0.9 3.9 6.7 4.2 19.6

S % 1.22 2.04 2.2 1.95 10.5

Te g/t <4 <4 <4 <4 <4

Hg g/t 0.3 <0.3 <0.3 <0.3 <0.3 Source: SGS, 2019 Au (S. M.) = average of screened metallic assay Au (Calc) = average from testwork



13.1.2 Mineralogy

A mineralogical evaluation was conducted on a single sample from the MDZ Center Zone by Terra

Mineralogical Services Inc. Key findings included the following:

• Native gold was by far the predominant gold carrier

• The majority (more than 99%) of the gold particles occurred in locations that would be readily

accessible by leaching solutions

• The gold grains were predominately associated with silicate gangue minerals. Gold particles

were not often in direct contact with sulfides, yet very commonly pyrrhotite, chalcopyrite, and

bismuth minerals were found in close vicinity to the gold mineralization

• The average grain size of the gold particle was very fine (<6 μm), however, a small amount of

coarse gold particles also identified.

13.1.3 Comminution Testwork

Comminution testwork included SMC (SAG mill comminution), BWI (Bond ball mill work index) and AI

(Abrasion index) index determinations and the results are shown in Table 13-3. The SMC tests were

conducted on the East, West and Center MDZ composites and the reported Axb values ranged from

28 to 31 and averaged 29, indicating that the material is very hard with respect to SAG mill impact

grinding. The BWI tests were conducted on all test composites using a 150 mesh (105 µm) closing

screen, the MDZ composites ranged from 19 to 20.7 kWh/t and averaged 19.8 kWh/t, indicating that

the MDZ material is very hard with respect to ball mill grinding. The AI tests on the MDZ composites

ranged from 0.626 to 0.731 indicating that the samples were very abrasive and high liner and grinding

media wear rates can be expected.

Table 13-3: Comminution Test Results on MDZ and Marmato Test Samples

Sample Relative Density JK Parameters

BWI (kWh/t) AI (g) A x b ta SCSE

Marmato Mine Comp - - - - 15.7 0.199

MDZ Comp - - - - 20.7 0.652

Center Zone Comp 2.65 31 0.30 11.0 19.0 0.704

East Zone Comp 2.64 28 0.27 11.6 20.3 0.626

West Zone Comp 2.69 28 0.27 11.7 19.0 0.731

Average 2.66 29 0.28 11.4 19.0 0.582

Source: SGS, 2019

13.1.4 Whole-Ore Cyanidation

Two whole-ore cyanidation tests were completed on the MDZ Master composite. The tests were

conducted at a grind size of 80% passing (P80) 60 µm with a maintained cyanide concentration of 1 g/L

sodium cyanide (NaCN) and evaluated the impact of pre-aeration and dissolved oxygen concentration

on gold extraction and leach kinetics. The results of these tests are summarized in Table 13-4. These

tests showed that without pre-aeration and only air injection to maintain the dissolved oxygen

concentration during leaching at 5 to 8 mg/L, 98% of the gold could be extracted after 72 hours of

leaching with sodium cyanide consumption reported at 1.83 kilograms per tonne (kg/t). However, with

inclusion of pre-aeration for two hours and oxygen injection sufficient to maintain the dissolved oxygen

concentration during leaching at 20 mg/L, sodium cyanide consumption was reduced to 0.66 kg/t and

leach kinetics were significantly increased with gold leaching complete after 24 to 48 hours.



Table 13-4: Whole-Ore Cyanidation Test Results on MDZ Test Composite

Grind Size (P80 µm)

Aeration Conditions Reagent Cons. kg/t of

CN Feed Au Extraction (%) Au Residue (g/t) Au Head (g/t)

Pre-air Leach NaCN CaO 8 h 24 h 48 h 72 h A B Avg. Calc. Direct

56 n/a Air, ~5-8 ppm 1.83 1.52 40 70.4 88.7 98.3 0.04 0.05 0.05 2.65 2.32

61 2 h O2 O2, ~20 ppm 0.66 1.29 93.5 95.6 ~99 98.2 0.04 0.04 0.04 2.26

Source: SGS, 2019



13.1.5 Gravity Concentration

Gravity concentration tests were conducted on the MDZ Master, West Zone, Center Zone, East Zone

and Marmato composites at grind sizes ranging from P80 70 to 223 µm with a Knelson MD-3 centrifugal

gravity concentrator followed by upgrading on a Mozley table. The results of these tests are shown in

Table 13-5. This testwork demonstrated that the MDZ material (including West, Center and East Zone

composites) is highly amenable to gravity concentration with gold recoveries ranging from 50.6 to 69%

and silver recoveries ranging from 14.6 to 24.7% into gravity concentrates containing 0.05 to 0.16

weight percent (wt%) and 1,575 to 2,494 g/t Au and 182 to 1,432 g/t Ag.

Table 13-5: Summary of Gravity Concentration Testwork on MDZ and Marmato Composites (1)

Test No.

Composite

Grind Size

Assay Head Calc. Head Gravity Concentrate Distribution

(%)

P80

µm Au

(g/t) Ag

(g/t) Au

(g/t) Ag

(g/t) Mass

(%) Au (g/t

Ag (g/t)

Au Ag

G-1 MDZ 223 2.32 4.2 2.31 3.9 0.07 1,815 1,006 55.1 17.9

G-3 MDZ 223 2.32 4.2 2.27 4.5 0.05 2,494 1,432 50.6 14.6

G-9 MDZ 112 2.32 4.2 2.51 4.0 0.10 1,575 921 63.7 23.6

G-6 West Zone 88 1.54 0.9 1.3 1.2 0.16 531 182 66.1 24.7

G-7 Center Zone 94 2.69 3.9 2.61 4.3 0.12 1,498 752 69.0 21.0

G-8 East Zone 99 2.65 6.7 1.8 8.1 0.11 831 1,149 51.7 15.9

G-4 Marmato 70 5.48 19.6 4.67 18.8 0.17 1,586 1,278 57.9 11.6

G4R Marmato 78 5.48 19.6 4.98 18.5 0.13 2,192 1,668 56.9 11.6

Note: (1) Marmato test G-2 not shown due to high variance between assay and calculated head Source: SGS Lakefield, 2019

13.1.6 Cyanidation of Gravity Tailing

MDZ Master Composite

Cyanidation tests were conducted on the gravity tailings from the MDZ Master Composite. The leach

conditions are shown in Table 13-6 and the test results are summarized in Table 13-7. Tests were

conducted over a range of grind sizes and cyanide concentrations, both with and without pre-aeration

and oxygen injection. These tests demonstrated that overall gold extractions

(gravity concentration + gravity tailing cyanidation) of about 97 to 98% could be achieved. A grind size

of about P80 100 µm appeared optimum with a cyanide concentration of 0.5 g/L NaCN. Preaeration

appears to be beneficial in reducing cyanide consumption. Gold extraction versus leach retention time

is shown in Figure 13-2.



Table 13-6: MDZ Master Composite Gravity Tailing Leach Conditions

CN Test No.

Feed Grind P80, µm

Aeration Conditions Lead Nitrate (100 g/t)

Cyanide (g/L

NaCN)

Reagent Addition

(kg/t)

Reagent Cons. (kg/t)

Pre-Air Leach NaCN CaO NaCN CaO

3 G-1 223 n/a Air, ~5-8

ppm N 1 1.65 1.09 0.83 1.09

4 G-1 70 n/a Air, ~5-8

ppm N 1 2.27 1.09 1.55 1.09

12 G-3 138 4 h, O2 O2, ~20

ppm N 1 1.23 1.22 0.24 1.18

13 G-3 96 4 h, O2 O2, ~20

ppm N 1 1.24 1.27 0.22 1.23

14 G-3 76 4 h, O2 O2, ~20

ppm N 1 1.24 1.32 0.25 1.30

15 G-3 80 2 h, O2 O2, ~20

ppm Y 1 1.24 1.20 0.28 1.17

21 G-3 99 4 h, O2 O2, ~20

ppm N 0.75 0.75 1.10 0.19 1.09

22 G-3 98 4 h, O2 O2, ~20

ppm N 0.50 0.50 1.11 0.16 1.08

23 G-3 97 4 h, O2 O2, ~20

ppm Y 0.50 0.50 1.13 0.14 1.11

Source: SGS Lakefield, 2019

Table 13-7: Gravity Concentration + Gravity Tailing Cyanidation Test Results

CN Test No.

Feed

Au Extraction/Recovery, % Residue (Au g/t) Calc. Head

(Au g/t) CN (Unit)

Grav Grav +

CN 8 h 24 h 48 h 72 h A B Avg.

3 G-1 58.6 79.9 85.8 55.1 93.6 0.12 0.18 0.15 1.06

4 G-1 88.2 95.6 55.1 98.0 0.07 0.06 0.07 1.48

12 G-3 87.3 93.8 93.2 93.3 50.6 96.7 0.08 0.08 0.08 1.19

13 G-3 89.8 96.1 94.7 95.1 50.6 97.6 0.06 0.05 0.06 1.13

14 G-3 87.8 95.6 95.2 95.0 50.6 97.5 0.05 0.06 0.06 1.10

15 G-3 90.7 95.1 97.0 95.9 50.6 98.0 0.04 0.05 0.05 1.08

21 G-3 80.3 90.1 92.6 93.1 50.6 96.6 0.08 0.07 0.08 1.08

22 G-3 83.1 93.3 92.9 93.4 50.6 96.7 0.07 0.07 0.07 1.06

23 G-3 91.5 95.9 95.2 95.4 50.6 97.7 0.05 0.06 0.06 1.21





Figure 13-2: Gold Extraction Versus Retention Time (MDZ Master Comp Gravity Tailings)

13.1.7 Variability Composites

The individual East, Center and West Zone composites and the Marmato composite were subjected

to gravity concentration followed by cyanidation of the gravity tailings using the following test

conditions:

• Grind size: ~ P80 of 100 μm

• Preaeration: 4 hours with oxygen

• Dissolved O2: 20 mg/L

• NaCN: 1.0 g/L (maintained)

• Retention Time: 48 hours

• Slurry Density: 50% solids

The results of these tests are summarized in Table 13-8, which show that overall gold recoveries for

the West, Center and East Zone composites were very similar to the results obtained from the MDZ

Master composite and ranged from 96.7 to 97.9% with cyanide consumption ranging from 0.19 to 0.34

kg/t. Overall gold recovery from the Marmato composite was about 92% with cyanide consumption at

about 0.50 kg/t.



Table 13-8: Summary of Gravity Concentration + Gravity Tailing Cyanidation (Variability Composites)

Gravity Test Cyanidation Test Composite Grind Size Au Distribution (%)

P80 µm Gravity Grav. + Cyan.

G-6 CN-18 West Zone 88 66.1 97.3

G-7 CN-19 Center Zone 94 69.0 97.9

G-8 CN-20 East Zone 99 51.7 96.7

G-4 CN-16 Marmato 70 57.9 92.0

G4R CN-16R Marmato 78 56.9 91.8


13.1.8 Flotation from Gravity Tailing

Rougher flotation tests were conducted on gravity tailings from the MDZ Master and Marmato

composites using flotation conditions provided by CGM. All tests were conducted at natural pH with

20 minutes of retention time and used potassium amyl xanthate (PAX) and MX5160 as the collectors,

copper sulfate as a sulfide mineral activator and Dowfroth 250 as the frother. The results of selected

tests are shown in Table 13-9. Overall gold recoveries (gravity concentration + rougher flotation) of

96% to 97% were reported for the MDZ Master composite and 97.4% for the Marmato composite.

Rougher flotation concentrate grades produced from the MDZ Master composite ranged from about

10 to 13 g/t Au. The rougher flotation concentrate produced from the Marmato composite contained

43.6 g/t Au. Although generally high overall gold recoveries were reported, it should be noted that the

gold grade of the final flotation tailing produced from the MDZ Master composite was significantly

higher than the cyanidation leach residues (0.09 g/t Au versus 0.06 g/t Au).

Table 13-9: Summary of Rougher Flotation Tests on Gravity Tailings from MDZ and Marmato Composites

Composite Test Grind P80 µm Au Recovery (%) Au Grade (g/t)

Grav Flot (unit) Flot + Grav Flot Conc Flot Tail

MDZ G1/F2 74 55.1 93.4 97.0 9.6 0.08

MDZ G1/F4 63 55.1 91.9 96.4 13.1 0.09

MDZ G1/F5 63 55.1 90.8 95.9 12.1 0.10

Average 67 55.1 92.0 96.4 11.6 0.09

Marmato G-2/F3 210 11.5 97.1 97.4 43.6 0.39


Cyanidation of Flotation Concentrates

Cyanidation tests were conducted on the rougher flotation concentrates that had been reground to

about 22 µm. Cyanidation tests were conducted at 1 g/L NaCN for 48 hours and the results are

summarized in Table 13-10. These tests demonstrated that about 98% of the gold contained in the

flotation concentrates could be extracted by cyanidation. It is important to note that 98% gold extraction

from the flotation concentrate implies an overall gold recovery of about 95% to 96% from a gravity +

flotation + cyanidation flowsheet. This is about 2% lower gold recovery than by the gravity + gravity

tailing cyanidation flowsheet.



Table 13-10: Summary of Flotation Concentrate Cyanidation Test Results

CN Test No.

Sample Feed

Au Extraction, % CN (Unit)

Au Residue (g/t) Au Calc. Head (g/t)

8 h 24 h 48 h A B Avg.

1 MDZ G-1/F-1 95.4 97.1 98.1 0.22 0.20 0.21 11.1

2 MDZ G-1/F-2 93.3 95.4 98.2 0.17 0.17 0.17 9.57

10 MDZ G-1/F-4 93.0 95.1 97.2 0.38 0.36 0.37 13.1

11 MDZ G-1/F-5 94.2 95.9 98.1 0.24 0.23 0.24 12.1

7 Marmato G-2/F-3 56.6 92.4 98.0 0.87 0.84 0.86 43.6

Source: SGS, 2019

13.1.9 Cyanide Detoxification

The cyanidation leach residue produced from the MDZ Master composite under optimized leach

conditions was subjected to cyanide detoxification testing using the industry-standard SO2/Air process

to reduce the weak acid dissociable cyanide (CNWAD) to less than 10 mg/L. The main parameters

adjusted during the testwork were sodium metabisulphite and copper addition rates. The results of the

detoxification testwork are shown in Table 13-11. The initial leach residue contained 151 mg/L CNWAD,

which was subsequently reduced to 8.95 CNWAD (Test 1-7) with the addition of 8.05 g SO2/g CNWAD

and 0.22 g Cu/g CNWAD. This testwork established that the following operating conditions will achieve

a discharge CNWAD concentration of <10 mg/L.

• Slurry density: 50% solids (w/w)

• SO2 addition: 8 g SO2 /g CNWAD

• Cu addition: 0.22 g Cu /g CNWAD

• pH: 8.5 (with lime added as needed (~0.5 kg/t)

• Retention time: 90 minutes



Table 13-11: Summary of Cyanide Detoxification Testwork on MDZ Composite Leach Residue

Test

Test Duration

Reten. Time

Product (Solution Phase) Reagent Addition

Min Min pH

CNT CNWAD by Cu Fe g/g CN WAD g/L Feed Pulp kg/t Solids

mg/L Ana. Lab

mg/L

Picric Acid mg/L

Aged Picric Acid mg/L

mg/L mg/L SO2

Equiv. Lime Cu (1)

SO2 Equiv.

Lime Cu (1) SO2

Equiv. Lime Cu (1)

Feed (CN-24) … 10.6 150 151 … … 26.6 2.56 … … … … … … … … …

Batch Test

270 8.5 … … 1.12 … … … 11.1 9.71 0.070 1.23 1.08 0.007 1.67 1.46 0.011 CND 1

270

Continuous Tests 8.5 … … 55.6 … … … 6.98 0.54 0.000 0.78 0.061 0.000 1.05 0.081 0.000

1-1 94 60

1-2 90 55 8.7 5.7 <0.1 14.7 1.14* 0.70 1.7 6.12 0.11 0.070 0.68 0.013 0.007 0.92 0.02 0.011

1-3 90 60 8.5 … … 50.2 … 7.95 4.84 0.070 0.88 0.55 0.007 1.20 0.73 0.011

1-4 60 55 8.5 … … 30.5 … … … 7.31 3.60 0.130 0.81 0.41 0.015 1.10 0.54 0.020

1-5 115 59 8.5 2.5 <0.1 9.88 2.51** 4.3 1.2 7.57 2.19 0.220 0.88 0.26 0.025 1.14 0.33 0.033

1-6 120 59 8.5 … … 18.3 … … … 7.37 2.38 0.350 0.86 0.28 0.041 1.11 0.36 0.053

1-7 100 89 8.5 … … 8.95 … … … 8.05 3.62 0.220 0.94 0.43 0.026 1.21 0.55 0.033

1-8 170 91 8.5 <0.1 <0.1 6.45 … 14.7 <0.1 10.7 4.66 0.240 1.19 0.53 0.026 1.61 0.70 0.036

…No sample submitted for assays (1) Cu added using CuSO4, 5H2O, SO2 added using sodium metabisulphite * CND1-2 aged five-day sample ** CND1-5 aged one day sample Bold red values indicate a key parameter that has changed Source: SGS Lakefield, 2019



13.1.10 Solid-Liquid Separation

Thickening and rheological studies were conducted on a cyanidation leach residue at a P80 105 µm

grind size that had adjusted to pH 8.5 with lime to simulate the detoxified slurry pH.

Flocculant Screening

Flocculant screening tests identified BASF Magnafloc 10, which is a very high molecular weight,

slightly anionic polyacrylamide flocculant, as a suitable flocculant for this application at an application

rate of 40 g/t. Both static and dynamic thickening testwork were conducted with this flocculant.

Static Thickening

Preliminary static settling tests were performed in two-liter graduated cylinders which were fixed with

rotating picket-style rakes. Static settling test results were used to determine the starting conditions for

subsequent dynamic thickening tests. The selected conditions based on these tests are summarized

in Table 13-12, and indicated a specific thickener settling area of 0.11 square meters per tonne per

day (m2/[t/day]) with an underflow density of 62% solids and an overflow containing 61 mg/L of total

suspended solids (TSS).

Table 13-12: Static Thickener Test Conditions

o Sample ID o Flocculant Dose

(g/t) o Feed

%w/w o U/F

%w/w o Unit Area

m2/(t/day) o ISR

m3/m2/day o Supernatant

Clarity o TSS

mg/L

o MDZ Comp o 40 o 8 o 62 o 0.11 o 833 o Hazy o 61


Dynamic Thickening

Dynamic thickening testwork was initiated with a 50 g/t dosage of BASF Magnafloc 10 flocculant at a

feedwell slurry density of 8% (w/w) solids. The dynamic thickening test responded very differently to

the static thickening test under these conditions with a very turbid overflow with total suspended solids

(TSS) measured at 450 mg/L. In order to improve the overflow clarity, BASF Magnafloc 1687 coagulant

was applied to the diluted thickener feed prior to flocculant dosing. A series of additional tests

established a dosage of Magnafloc 1687 at 15 g/t followed by a dosage of 25 g/t of Magnafloc 10 as

optimal. The result of dynamic thickener tests conducted over a range of unit settling areas (m2/[t/d])

are summarized in Table 13-13.

Table 13-13: Summary of Dynamic Thickener Test Results

1687 Dosage (g/t)

10 Dosage (g/t)

Unit Area m2/(t/d)

Solids Loading (t/m2/h)

Net Rise Rate (m3/m2/d)

Underflow %w/w solids

Overflow TSS (mg/L)

Residence Time (h)

U/F Yield Stress (Pa)

15 25 0.13 0.32 84.2 64.1 54 1.36 60

15 25 0.11 0.38 99.5 62.9 57 1.15 49

15 25 0.09 0.46 121.7 61.7 43 0.94 52

15 25 0.07 0.60 156.4 59.5 58 0.73 38

Underflow extended for 30 minutes: 65.1 93

Note: Bed height was maintained around 160 mm Source: SGS Lakefield, 2019



Rheology on Thickener Underflow

The results of rheology testwork on the thickener underflow are summarized in Table 13-14. A Critical

Solids Density (CSD) of approximately 61% solids was established, which corresponds to

approximately 20 pascals (Pa) on the unsheared yield test and 13 Pa on the sheared yield test. As

shown in Figure 13-3, CSD is the solids density at which a small increase of the solids density causes

a significant decrease of the flowability of the slurry. The CSD value is also predictive of the maximum

underflow solids density achievable in a commercial thickener.

Table 13-14: Results of Rheology Testwork on MDZ Thickener Underflow Sample

Test Code

Solids %w/w

Unsheared Sample Sheared Sample

Observations Ɣ ƮyВ ɳP Ɣ ƮyВ ɳP

Range, 1/s

Pa mPa.s Range,

1/s Pa mPa.s

CSD= ~61% solids, corresponding to ~20 Pa unsheared and 13 Pa sheared yield stress.

T1 65.8 Plug Flow 73 -- 200-400 26 50 Thixotropic

T2 64.2 200-400 50 3.9 200-400 20 36 Thixotropic

T3 62.1 200-400 28 15 200-400 15 24 Thixotropic

T4 60.3 200-400 17 16 200-400 12 17 Thixotropic

T5 58.3 200-400 12 15 200-400 8.9 15 Minor Thixotropic, minor settling

T6 54.2 200-400 5.1 13 200-400 4.8 13 Minor settling

T7 50.3 200-400 2.3 10 200-400 2.1 15 Moderate settling, dilatant after 425 1/s

Notes:

• The MDZ Comp underflow samples contained 15 g/t BASF Magnafloc 1687 coagulant and 25 g/t BASF Magnafloc 10 flocculant.

• The values are based on data produced by the unsheared and sheared slurry sample.

• Variable shearing was produced in the 0 to 600 s-1 range, increasing and decreasing (up and down curves).

• Constant shearing was produced by subjecting the slurry sample to a constant rotation at 300 1/s for 180 seconds.

• Bingham Plastic parameters: yield stress (ƮyВ) and plastic viscosity (ɳP) values, for the specified Ɣ range.

• Ɣ – Shear rate range at which the rheological parameters were calculated. Source: SGS Lakefield, 2019



Source: SGS, 2019

Figure 13-3: Yield Stress Versus Thickener Underflow Slurry Density

13.2 Metallurgical Program – 2020

The 2020 metallurgical program was conducted to further define the process parameters and design

criteria for the selected flowsheet that includes gravity concentration followed by cyanidation of the

gravity tailing. The test program included gravity concentration, gravity recoverable gold (E-GRG

determination) cyanide leach optimization and carbon-in-pulp (CIP) modelling. Cyanide destruction

(CND), solid/liquid separation, and environmental testwork was also completed. The optimization and

metallurgical design tests were all completed using the Master Composite. Once the optimized

flowsheet had been selected, the variability test samples were tested under these optimized

gravity/cyanidation conditions. The results of this program are fully documented in SGS’s report, “The

Recovery of Gold from Marmato MDZ Deposit Samples”, June 18, 2020.

13.2.1 Metallurgical Sample Location

The 2020 metallurgical program was conducted on an MDZ master composite and on variability

composites representing low, medium and high grade MDZ ore, transition zone and the MDZ Deep.

In addition, an ore sample from the existing Marmato mine was tested. The MDZ master and variability

composites were formulated from selected drill core holes and intervals. The master composite was

prepared from the low, medium and high grade MDZ variability composites on a weighted basis to

represent the average grade of the MDZ. In addition, selected core intervals from five different drill

holes were prepared for crushability testwork. The selected drill holes and core intervals used to



formulate each of the MDZ test composites are shown in Tables 13-15 through 13-20. Figure 13-4

shows the location of the selected drill holes and intervals.

Table 13-15: Drill Holes and Intervals Used for the Low Grade MDZ Composite

Hole Sample ID From (m) To (m) Metallurgy ID Weight (Kg)

MT-1445

D97563 546.0 548.0 001 1.90

D97564 548.0 550.0 002 1.90

D97565 550.0 552.0 003 1.90

D97566 552.0 554.0 004 1.90

D97568 554.0 556.0 005 2.00

D97569 556.0 558.0 006 2.00

D97570 558.0 560.0 007 2.00

MT-IU-021

D119582 169.4 170.6 008 2.04

D119583 170.6 171.7 009 1.78

D119584 171.7 172.9 010 1.90

D119585 172.9 174.0 011 2.08

D119586 174.0 175.2 012 2.14

D119587 175.2 176.4 013 2.08

D119588 176.4 177.7 014 1.80

D119589 177.7 179.2 015 2.08

D119591 179.2 180.7 016 2.52

MT-IU-006

D115349 184.2 185.2 017 2.04

D115350 185.2 185.9 018 1.00

D115351 185.9 187.4 019 2.96

D115352 187.4 188.9 020 3.24

D115353 188.9 189.9 021 2.08

D115355 189.9 191.2 022 2.72

D115356 191.2 192.6 023 2.64

D115357 192.6 193.3 024 1.50

D115358 193.3 194.8 025 1.68

D115359 194.8 196.3 026 1.64

D115360 196.3 197.3 027 1.14

D115361 197.3 198.3 028 1.18

D115362 198.3 199.1 029 0.54

D115365 199.1 200.6 030 1.50

D115366 200.6 202.1 031 1.72

D115367 202.1 203.8 032 1.52

D115368 203.8 204.8 033 0.62

D115370 204.8 205.6 034 0.74

D115371 205.6 207.1 035 1.56

D115372 207.1 208.6 036 1.50

MT-IU-015

D117573 294.0 295.2 037 1.76

D117574 295.2 296.0 038 1.28

D117575 296.0 297.0 039 1.56

D117577 297.0 298.2 040 1.50

D117579 298.2 299.4 041 1.60

D117580 299.4 300.9 042 2.34

D117581 300.9 302.4 043 2.34

D117582 302.4 303.9 044 2.58

D117583 303.9 305.4 045 2.52

D117584 305.4 306.9 046 2.36

Source: SRK, 2020



Table 13-16: Drill Holes and Intervals Used for the Medium Grade MDZ Composite


MT-1445

D97541 513.0 515.0 001 1.90

D97542 515.0 517.0 002 2.00

D97543 517.0 519.0 003 1.90

D97544 519.0 521.0 004 1.90

D97546 521.0 522.5 005 1.50

D97547 522.5 524.0 006 1.60

D97548 524.0 525.5 007 1.40

D97549 525.5 526.7 008 1.10

D97550 526.7 528.0 009 1.00

MT-1499-A

D105992 506.0 507.3 010 1.00

D105993 507.3 509.0 011 1.40

D105994 509.0 511.0 012 1.56

D105996 511.0 513.0 013 1.50

D105997 513.0 515.0 014 1.64

D105998 515.0 517.0 015 1.52

D105999 517.0 519.0 016 1.80

D106000 519.0 521.0 017 1.56

D107877 521.0 523.0 018 1.60

D107878 523.0 525.0 019 1.84

D107879 525.0 527.0 020 1.78

D107880 527.0 529.0 021 1.76

D107881 529.0 531.0 022 1.70

MT-1455-A

D93860 585.7 587.0 023 1.26

D93861 587.0 589.0 024 1.92

D93862 589.0 591.0 025 2.00

D93863 591.0 593.0 026 1.72

D93864 593.0 595.0 027 1.92

D93865 595.0 597.0 028 2.06

D93867 597.0 598.5 029 1.42

D93868 598.5 600.0 030 1.50

D93869 600.0 602.0 031 1.80

D93870 602.0 604.0 032 2.06

D93871 604.0 605.0 033 1.06

D93872 605.0 607.0 034 2.06

D93873 607.0 609.0 035 2.10

D93874 609.0 610.8 036 1.70

D93875 610.8 612.0 037 1.12

D93876 612.0 613.6 038 1.26

D93877 613.6 614.5 039 1.00

D93879 614.5 616.0 040 1.56

MT-IU-024

D120855 498.7 500.2 041 1.20

D120856 500.2 501.7 042 1.50

D120857 501.7 503.2 043 1.40

D120859 503.2 504.3 044 1.10

D120862 504.3 505.5 045 0.82

D120864 505.5 506.9 046 1.42

D120865 506.9 508.4 047 1.00

D120866 508.4 510.1 048 1.90

D120867 510.1 511.9 049 1.96

D120868 511.9 513.7 050 1.60

D120869 513.7 515.5 051 1.30

D120870 515.5 517.3 052 1.56

D120871 517.3 519.1 053 1.84

D120872 519.1 520.9 054 1.64

D120873 520.9 522.4 055 1.54




D120874 522.4 523.9 056 1.20

D120875 523.9 525.4 057 1.36

D120876 525.4 526.7 058 1.10

D120878 526.7 528.1 059 1.26

D120879 528.1 529.6 060 1.42

D120880 529.6 531.0 061 1.12

D120881 531.0 531.5 062 0.60

D120882 531.5 532.0 063 0.60

D120884 532.0 532.7 064 0.50

D120885 532.7 534.2 065 1.80

D120886 534.2 535.7 066 1.34

D120887 535.7 537.2 067 1.66

D120888 537.2 539.0 068 1.84

D120889 539.0 540.7 069 1.82

D120890 540.7 542.4 070 1.86

D120740 268.1 269.6 071 2.82

D120741 269.6 271.0 072 2.68

D120742 271.0 272.5 073 2.88

D120743 272.5 273.6 074 2.34

D120744 273.6 274.7 075 2.22

D120745 274.7 275.7 076 2.08

MT-IU-017

D120746 275.7 276.7 077 2.04

D120747 276.7 277.7 078 1.76

D120748 277.7 279.0 079 2.62

D120749 279.0 280.3 080 2.62

D120750 280.3 281.6 081 2.54

D120751 281.6 282.9 082 2.76

D118372 241.8 243.2 083 2.48

D118373 243.2 244.6 084 1.34

D118374 244.6 246.0 085 2.60

D118375 246.0 247.3 086 2.22

D118376 247.3 248.8 087 2.46

D118377 248.8 250.2 088 2.68

D118378 250.2 251.6 089 2.20

D118379 251.6 253.0 090 2.50

D118381 253.0 254.4 091 2.48

D118382 254.4 255.7 092 1.92

D118383 255.7 257.0 093 2.50

D118384 257.0 258.3 094 2.38

D118385 258.3 259.6 095 2.54

D118386 259.6 260.8 096 1.74

MT-IU-013

D116944 203.7 205.0 097 2.56

D116945 205.0 206.0 098 1.94

D116946 206.0 207.0 099 1.96

D116947 207.0 207.5 100 0.80

D116948 207.5 208.2 101 1.64

D116949 208.2 209.0 102 1.66

D116950 209.0 209.8 103 1.38

D116951 209.8 210.6 104 1.66

D116952 210.6 211.7 105 1.86

D116954 211.7 212.9 106 2.54

D116955 212.9 213.9 107 1.86

MT-IU-018

D118757 228.5 229.7 108 1.96

D118758 229.7 231.1 109 1.84

D118759 231.1 232.4 110 2.12

D118760 232.4 233.8 111 2.40




D118761 233.8 235.2 112 2.38

D118762 235.2 236.4 113 2.62

D118764 236.4 237.6 114 1.20

D118765 237.6 239.0 115 2.34

D118766 239.0 240.5 116 2.38

MT-IU-015

D117389 254.6 256.0 117 2.14

D117390 256.0 257.5 118 2.32

D117391 257.5 258.5 119 1.42

D117392 258.5 260.0 120 1.76

D117393 260.0 260.7 121 1.04

D117395 260.7 262.2 122 2.52

D117398 262.2 263.7 123 2.18

D117399 263.7 264.4 124 1.06

D117400 264.4 265.9 125 1.66

D117401 265.9 266.9 126 2.28

D117402 266.9 267.8 127 1.66

D117403 267.8 269.2 128 2.22

D117404 269.2 270.0 129 1.26

MT-IU-004

D115136 182.2 183.3 130 1.82

D115137 183.3 184.3 131 1.84

D115139 184.3 185.2 132 2.22

D115140 185.2 186.4 133 1.76

D115141 186.4 187.4 134 1.68

Source: SRK, 2020

Table 13-17: Drill Holes and Intervals Used for the High Grade MDZ Composite

Hole Sample ID From (m) To (m) Metallurgy ID Weight (kg)

MT-1499-A

D105948 428 430 001 2.00

D105949 430 432 002 2.02

D105950 432 434 003 2.00

D105952 434 436 004 2.04

D105953 436 438 005 1.72

D105954 438 440 006 1.90

D105955 440 442 007 2.10

D105956 442 444 008 2.10

D105957 444 446 009 1.90

D105959 446 448 010 2.00

D105960 448 450 011 1.72

D105961 450 452 012 2.12

D105893 335 337 013 1.86

D105894 337 339 014 1.90

D105896 339 341 015 1.84

D105897 341 343 016 1.92

D105898 343 345 017 2.00

D105899 345 347 018 1.80

D105900 347 349 019 1.94

D105901 349 351 020 1.92

D105903 351 353 021 1.62

D105904 353 355 022 1.84

D105905 355 357 023 1.82

D105906 357 359 024 1.80

D105907 359 361 025 1.90

D105908 361 363 026 1.92

D105914 371 373 027 2.02

D105915 373 374 028 1.18




D105916 374 376 029 2.18

D105917 376 378 030 2.16

MT-1390

D83285 181.25 183.25 031 3.40

D83286 183.25 185.25 032 3.40

D83287 185.25 186.1 033 1.54

D83288 186.1 187.55 034 2.28

D83289 187.55 189.55 035 3.42

D83291 189.55 190.45 036 1.56

D83292 190.45 192.45 037 3.40

D83293 192.45 194.45 038 3.18

D83294 194.45 196.25 039 3.14

D83296 196.25 196.9 040 1.24

D83297 196.9 197.5 041 0.80

Source: SRK, 2020

Table 13-18: Drill Core Holes and Intervals Used for the MDZ Deep Composite


MT-IU-024

D120892 544.2 546.0 001 1.52

D120893 546.0 547.7 002 1.10

D120894 547.7 549.6 003 1.22

D120896 549.6 551.4 004 1.24

D120897 551.4 553.1 005 1.60

D120898 553.1 554.8 006 1.58

D120900 554.8 556.6 007 1.92

D120901 556.6 558.4 008 1.84

D120902 558.4 560.2 009 1.50

MT-IU-026

D122079 603.4 605.3 062 2.02

D122080 605.3 607.2 063 1.72

D122081 607.2 609.1 064 1.82

D122082 609.1 611.0 065 2.00

D122083 611.0 612.9 066 1.98

D122084 612.9 614.8 067 1.66

D122085 614.8 616.4 068 1.50

D122086 616.4 617.6 069 1.14

D122087 617.6 618.8 070 1.10

MT-IU-27

D122780 584.2 585.7 106 1.62

D122781 585.7 587.1 107 1.50

D122782 587.1 588.7 108 1.56

D122783 588.7 590.2 109 1.62

D122784 590.2 591.8 110 1.72

D122785 591.8 593.6 111 1.96

D122786 593.6 595.5 112 1.96

D122787 595.5 597.4 113 1.78

D122788 597.4 599.3 114 2.04

D122789 599.3 601.3 115 2.22

D122790 601.3 603.2 116 2.02

MT-IU-031

D124720 610.2 612.1 141 2.12

D124721 612.1 613.9 142 2.04

D124722 613.9 615.8 143 2.04

D124723 615.8 617.6 144 2.04

D124724 617.6 619.5 145 0.20

D124726 619.5 621.3 146 2.26

D124727 621.3 623.0 147 1.96

Source: SRK, 2020



Table 13-19: Drill Core Holes and Intervals Used for the Transition Composite


MT-IU-035

D125952 185.3 186.3 006 2.08

D125953 186.3 187.5 007 2.20

D125954 187.5 189.5 008 3.66

D125955 189.5 190.7 009 2.12

D125957 190.7 191.9 010 2.22

D125960 191.9 193.5 011 2.98

D125961 193.5 194.4 012 1.46

D125963 194.4 196.0 013 2.88

MT-IU-001

D101771 237.3 238.0 028 0.62

D101772 238.0 239.7 029 1.56

D101773 239.7 241.0 030 1.12

D101774 241.0 243.0 031 1.52

D101775 243.0 244.0 032 0.98

D101776 244.0 244.8 033 0.68

D101778 244.8 245.3 034 0.64

D101779 245.3 245.9 035 0.46

D101780 245.9 247.8 036 1.70

D101781 247.8 248.8 037 0.82

D101782 248.8 250.6 038 1.54

D101783 250.6 252.0 039 1.44

D101784 252.0 254.0 040 1.66

D101785 254.0 255.3 041 1.34

D101787 255.3 257.0 042 1.42

MT-IU-05

D109789 206.2 207.7 043 1.98

D109790 207.7 208.9 044 2.14

D109791 208.9 210.0 045 1.98

D109792 210.0 211.1 046 1.89

D109793 211.1 212.2 047 1.80

D109795 212.2 213.0 048 1.24

D109796 213.0 214.0 049 1.34

D109797 214.0 215.4 050 2.26

D109798 215.4 216.8 051 2.16

D109799 216.8 218.2 052 2.32

D109800 218.2 219.5 053 2.08

D109801 219.5 220.7 054 1.36

D109802 220.7 221.9 055 2.38

D109803 221.9 223.0 056 1.79

D109807 223.0 224.0 057 1.76

D109808 224.0 224.7 058 1.36

D109809 224.7 226.3 059 2.36

MT-IU-013

D116922 188.0 188.8 100 1.58

D116923 188.8 189.8 101 1.78

D116924 189.8 190.8 102 2.14

D116925 190.8 191.8 103 1.78

D116926 191.8 192.9 104 2.22

D116927 192.9 194.0 105 2.08

D116928 194.0 195.0 106 2.20

D116932 195.0 196.1 107 1.56

D116934 196.1 197.0 108 1.76

Source: SRK, 2020



Table 13-20: Drill Core Holes and Intervals Used for Crushing (CWI) Testwork

Hole From (m) To (m) Weight (Kg)

MT-IU-006 182.6 183.3 3.28

250.2 251.5 2.84

MT-IU-015 289.5 241.0 5.44

MT-IU-017 237.6 239.0 5.00

282.7 283.9 4.30

MT-IU-018 241.8 243.1 4.92

Source: SRK, 2019

Source: SRK, 2019

Figure 13-4: Drill Hole Locations Used for Metallurgical Composites

13.2.2 Head Analyses

Head analyses for each of the MDZ metallurgical composites are shown in Table 13-21. Direct and

calculated head analyses for both gold and silver are provided. Calculated gold and silver analyses

are based on the average of all relevant tests and are considered a better indication of actual grades

due to the test sample size and number of tests conducted. The calculated gold and silver grades for

the master composite were 2.99 g/t Au and 3.3 g/t Ag. Calculated gold grades for the variability

composites ranged from 1.8 to 4.52 g/t Au and calculated silver grades ranged from 1.1 to 5.6 g/t Ag.

Calculated gold and silver grades for the Marmato mine composite were 3.15 g/t Au and 10.1 g/t Ag.

Cyanide soluble copper, organic carbon, mercury and arsenic were low in all test composites and will

not present any problem during processing. Total sulfur and sulfide sulfur analyses show that sulfur

occurs primarily as sulfide sulfur. Sulfide sulfur in the MDZ composites ranged from 0.82 to 1.54% S⁼.



Sulfide sulfur in the Marmato mine composite was reported at 8.62% S⁼. A multi-element ICP scan of

each test composite is provided in Table 13-22 and shows that there are no elements in the test

composites that would present special processing challenges.

Table 13-21: Head Analyses for Key Elements

Composite

Direct Assay

Calculated(1) S

(%) S⁼

(%) Cu CN

Sol (%) Corg

Hg (ppm)

As (ppm) Au

(g/t) Ag

(g/t) Au

(g/t) Ag

(g/t)

Master 3.73 2.9 2.99 3.3 1.22 1.15 0.004 <0.05 <0.3 <60

Low Grade 1.95 2.8 1.80 3.1 1.12 1.09 0.002 <0.05 <0.3 <60

Medium Grade

3.27 2.4 2.58 2.9 1.32 1.21 0.003 <0.05 <0.3 <60

High Grade 3.58 3.6 3.99 3.8 0.89 0.82 0.004 <0.05 <0.3 <60

Transition 2.50 3.7 2.82 5.6 1.74 1.54 0.004 <0.05 <0.3 <30

Deep 4.70 0.8 4.52 1.1 1.34 1.22 0.002 <0.05 <0.3 <30

Marmato 4.18 10.3 3.15 10.1 9.27 8.62 0.005 <0.05 <0.3 <100

Note: (1) Calculated: Average of all relevant tests Source: SGS, 2020

Table 13-22: Head Analyses and Multi-Element Scan on Each Test Composite

o Element Master Comp

Low Grade Comp

Med Grade Comp

High Grade Comp

Transition Comp

Deep Comp

Marmato Comp

o Au (S.M.), g/t

3.74 1.97 3.62 3.53 2.67 5.33 3.13

o Au 1, g/t 5.05 2.05 3.01 3.22 2.60 5.23 2.87

o Au 2, g/t 2.93 1.93 2.58 3.69 2.89 4.18 3.38

o Au 3, g/t 3.21 1.87 4.23 3.84 2.01 4.68 6.30

o Au Avg, g/t 3.73 1.95 3.27 3.58 2.50 4.70 4.18

o Au Calc., g/t

2.99 1.80 2.58 3.99 2.82 4.52 3.15

o Ag 1, g/t 2.7 2.9 2.4 4.2 4.0 0.9 10.9

o Ag 2, g/t 3.6 3 2.6 3.7 3.5 0.6 10.2

o Ag 3, g/t 2.5 2.5 2.1 3 3.7 0.8 9.9

o Ag Avg, g/t 2.9 2.8 2.4 3.6 3.7 0.8 10.3

o Ag Calc, g/t 3.3 3.1 2.9 3.8 5.6 1.1 10.1

o AuCN, g/t 3.2 2 2.7 4 1 1.2 1.4

o Cu NaCN, %

0.004 0.002 0.003 0.004 0.003 0.002 0.005

o S, % 1.22 1.12 1.32 0.89 1.74 1.34 9.27

o S=, % 1.15 1.09 1.21 0.82 1.54 1.22 8.62

o SO4, % <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.2

o S(o), % <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 <0.05

o CT, % 0.12 0.11 0.14 0.09 0.17 0.15 0.58

o C(g), % <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 <0.05

o TOC, % <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 0.14

o CO3, % 0.8 0.75 0.84 0.55 0.88 0.84 2.31

o Hg, g/t <0.3 <0.3 <0.3 <0.3 <0.3 <0.3 <0.3

o SG 2.74 2.75 2.75 2.73 2.71 2.70 2.93

o AI, g/t 79,500 80,200 77,200 78,800 75,700 80,600 61,100

o As, g/t <60 <60 <60 <60 <30 <30 <100

o Ba, g/t 1,240 1,330 1,200 1,210 1,240 1,110 766

o Be, g/t 1.09 1.06 1.03 1.1 1.06 1.16 0.67

o Bi, g/t <30 <30 <30 <30 <20 <20 <20

o Ca, g/t 17,800 17,000 17,200 17,100 12,800 16,700 16,700

o Cd, g/t <2 <2 <2 <2 <2 <2 30

o Co, g/t <20 <20 <20 <20 <8 <8 13



o Element Master Comp

Low Grade Comp

Med Grade Comp

High Grade Comp

Transition Comp

Deep Comp

Marmato Comp

o Cr, g/t 24 31 22 16 65 71 94

o Cu, g/t 317 233 251 387 249 210 218

o Fe, g/t 36,800 35,400 36,000 32,400 43,300 34,000 107,000

o K, g/t 24,600 26,800 23,100 24,400 26,300 17,400 32,600

o Li, g/t 8 9 8 8 <20 <20 <5

o Mg, g/t 8,310 8,240 7,980 8,240 8,180 8,070 6,190

o Mn, g/t 226 236 209 245 266 101 856

o Mo, g/t <20 <20 <20 <20 <5 <5 <6

o Na, g/t 32,300 30,900 31,300 34,100 27,100 38,100 8,320

o Ni, g/t <20 <20 <20 <20 <20 <20 <20

o P, g/t 792 815 783 807 848 812 662

o Pb, g/t <20 <20 <20 <20 <20 <20 128

o Sb, g/t <20 <20 <20 <20 <20 <20 <10

o Se, g/t <30 <30 <30 <30 <30 <30 <30

o Sn, g/t <30 <30 <30 <30 <40 <40 <20

o Sr, g/t 625 621 584 647 543 589 210

o Ti, g/t 2,450 2,500 2,400 2,500 2,580 2,460 1,840

o Tl, g/t <30 <30 <30 <30 <30 <30 <30

o U, g/t <20 <20 <20 <20 <30 <30 <20

o V, g/t 65 65 65 66 61 61 47

o Y, g/t 7.5 7.2 7.2 7.9 6.5 7.6 3.7

o Zn, g/t 40 45 35 44 51 21 1,300

Au (S.M) = Screened Metallics Au and Ag (calc) = Calculated head from test program Source: SGS Metallurgical Report, 2020

13.2.3 Mineralogy

Mineralogical studies were performed by Terra Mineralogical Services Inc (Terra) on the Master

composite and three variability composites from the MDZ. The results of this mineralogical

investigation are presented in Terra’s report, “Determination of Gold Deportment in Four Master

Composite Samples from the Top of the Marmato Deep Zone”, November 8, 2016. Key findings were

similar to the 2019 investigation and include:

• Native gold is the predominant gold carrier.

• The great majority of gold grains occur in locations that would be readily accessible by leaching

solutions (more than 98%).

• Gold grains are predominantly associated with silicate gangue minerals. Gold particles are not

often in direct contact with sulfides, yet very commonly pyrrhotite, chalcopyrite and bismuth

minerals are found in close vicinity to the gold mineralization.

• Pyrrhotite, subordinate pyrite, and minor chalcopyrite are the predominant sulfide minerals

occurring alongside the gold, silver, tellurium, and bismuth mineralization in the quartz/ silicate

veinlets.

• The average grain size of the gold particles identified was very fine grained (less than 6

microns), however, a small population of coarse gold particles was also identified.

13.2.4 Comminution

Comminution tests were conducted on the MDZ master composite, MDZ deep zone composite, three

MDZ sub-composites (low grade, medium grade and high grade) and at the Marmato mine composite.

The comminution tests included SAG Mill Comminution (SMC), SAG Mill Power Index (SPI) and Bond



ball mill work index (BWI) tests. In addition, Bond low impact crushing work index (CWI) and abrasion

(AI) tests were conducted on selected 1/2 HQ drill core pieces. The results of these tests are

summarized in Table 13-23.

Table 13-23: Summary of Comminution Test Results

o Sample Name Relative Density

JK Parameters(1) SPI®

(Min) CWI

(kWh/t) BWI

(kWh/t) AI

(g) A x

b ta(2) SCSE

o Low Grade Comp 2.65 24.5 0.24 12.4 137 - 18.9 -

o Med Grade Comp 2.65 25.3 0.25 12.2 140 - 17.7 -

o High Grade Comp 2.65 22.8 0.22 12.9 154 - 18.6 -

o Master Comp 2.63 24.0 0.24 12.5 146 - 18.7

o Transition Comp 2.66 27.8 0.27 11.7 146 - 21.1 -

o Mine (Marmato) Comp 3.12 141 1.17 6.3 23.2 - 12.4 -

o Deep Zone Comp 2.67 28.8 0.28 11.5 178 - 19.8 -

o CWI #1 - - - - - 7.9 - 0.470

o CWI #2 - - - - - 13.9 - 0.644

o CWI #3 - - - - - 10.5 - 0.582

o CWI #4 - - - - - 12.2 - 0.527

o CWI #5 - - - - - 9.2 - 0.642

Note: (1) JK Parameters are the result of the SMC test procedure (2) The ta value reported as part of the SMC procedure is an estimate Source: SGS Metallurgical Report, 2020

SMC Test

The results of the SMC tests are summarized in Table 13-24. The samples (excluding the Marmato

composite) were characterized as hard with respect to resistance to impact breakage, with A x b values

ranging from 23 to 29. The Marmato composite was much softer with A x b value of 141 (lower A x b

values indicate harder material). The samples were also characterized as hard with respect to

resistance to abrasion breakage, with an average ta value of 0.27 (excluding the Marmato composite).

Table 13-24: Summary of SMC Test Results

Sample Name

A b A x b Hardness Percentile

ta(1) DWI

(kWh/m3) Mia

(kWh/t)

Mih

(kWh/t) Mic

(kWh/t) SCSE

(kWh/t) Relative Density

o Low Grade Comp

90.8 0.27 24.5 95 0.24 10.6 28.7 23.3 12 12.4 2.65

o Med Grade Comp

93.8 0.27 25.3 94 0.25 10.4 28.2 22.8 11.8 12.2 2.65

o High Grade Comp

99 0.23 22.8 97 0.22 11.5 30.6 25.2 13 12.9 2.65

o Master Comp

100 0.24 24.0 96 0.24 11.1 30 24.5 12.7 12.5 2.63

o Transition Comp

86.8 0.32 27.8 90 0.27 9.5 26.1 20.7 10.7 11.7 2.66

o Mine (Marmato) Comp

60.7 2.33 141 6 1.17 2.2 6.9 4.1 2.1 6.3 3.12

o Deep Zone Comp

92.9 0.31 28.8 87 0.28 9.4 25.8 20.4 10.6 11/5 2.67

(1) The ta value reported as part of the SMC procedures is an estimate Source: SGS Metallurgical Report, 2020

SAG Power Index (SPI) Test

The results of the SPI tests are summarized in Table 13-25 along with CEET Crusher Index (Ci)

measurements. The SPI was used to measure the hardness of the ore while the Ci was used to predict

the SAG feed size distribution of the ore. The SPI ranged from 137 to 178 minutes, except for the



Marmato composite, which had an SPI value of 23.2 minutes. The Ci ranged from 1.7 to 3.1 for the

composites (excluding the Marmato composite).

Table 13-25: Summary of SPI Tests

o Sample Name SGS ID CEET Crusher Index (Ci) SPI (Minute) Hardness Percentile

o Low Grade Comp 1-20597 2.2 137 82

o Med Grade Comp 1-20598 3.2 140 83

o High Grade Comp 1-20599 3.1 154 87

o Master Comp 1-20600 2.5 146 84

o Transition Comp 1-20906 3.0 146 85

o Mine (Marmato) Comp 1-20907 27.7 23 9

o Deep Zone Comp 1-20908 1.7 178 91

Source: SGS metallurgical report 2020

Bond Ball Mill Grindability Test

The results of Bond ball mill work index (BWI) tests using a 120 mesh (125 µm) closing screen are

summarized in Table 13-26. The BWI values for the MDZ composites range from 17.7 kWh/t to 19.8

kWh/t, which places them in the hard range of hardness. The Marmato mine ore BWI value was much

lower at 12.4 kWh/t.

Table 13-26: Summary of Bond Ball Mill Work Index (BWI) Tests

o Sample Name Mesh of

Grind F80

(µm) P80

(µm) Gram per

Revolution Work Index

(kWh/t) Hardness Percentile

o Low Grade Comp

120 2,566 96 1.05 18.9 88

o Med Grade Comp

120 2,525 98 1.15 17.7 82

o High Grade Comp

120 2,450 99 1.10 18.6 87

o Master Comp 120 2,426 98 1.09 18.7 87

o Transition Comp 120 2,552 98 0.93 21.2 95

o Mine (Marmato) Comp

120 2,313 98 1.81 12.4 29

o Deep Zone Comp

120 2,586 99 1.01 19.8 92

Source: SGS Metallurgical Report, 2020

Crushing Work Index and Abrasion Index Tests

Bond low impact crushing work index (CWI) and Bond Abrasion Index (AI) tests were conducted on

drill core pieces selected to provide spatial representivity through the MDZ deposit. The results of CWI

tests are summarized in Table 13-27 and show that the average CWI was 10.7 kWh/t and the average

SGS hardness percentile was 56 (medium range of hardness). The results of Bond Abrasion (AI) tests

are summarized in Table 13-28. The AI’s ranged from 0.470 to 0.644, which would classify the MDZ

ore as abrasive and will result in relatively high wear rates for liners and grinding media.



Table 13-27: Summary of Bond Low Energy Crushing Tests

Sample Name

Number of Specimens

Average CWI (kWh/t)

Hardness Percentile

Min CWI (kWh/t)

Max CWI (kWh/t)

Std Dev (kWh/t)

Relative Density

CWI #1 9 7.9 37 2.7 13.4 3.4 2.68

CWI #2 20 13.9 74 6.2 22.5 3.9 2.66

CWI #3 16 10.5 53 5.7 17.6 3.3 2.65

CWI #4 15 12.2 66 6.4 26.4 4.8 2.62

CWI #5 13 9.2 47 3.5 16.3 3.4 2.67 Source: SGS Metallurgical Report, 2020

Table 13-28: Summary of Abrasion Index Determinations

Sample Name AI (g) Percentile of Abrasivity

CWI Rocks 1 0.470 78






13.2.5 Gravity Recoverable Gold (E-GRG) Testwork

The MDZ master composite was submitted for an extended gravity recoverable gold (E-GRG) test.

The three stage gravity test was completed at the SGS facility in Lakefield, Ontario and the results

were forwarded to FLSmidth (Knelson) for analysis and modelling. The E-GRG test involved sequential

gravity separation tests at successively finer grinds (P80 659, 257 and 98 µm) and the results are

shown in Table 13-29. An E-GRG value of 78.1% was determined for the MDZ master composite.

Table 13-29: Summary of E-GRG Test on MDZ Master Composite

Stage Grind Size (P80 µm) Mass (%) Au (g/t) Au Dist. (%)

Conc. Tail Conc. Tail Conc. Tail

1 659 0.39 1.03 319.2 2.12 39.1 0.7

2 257 0.37 1.19 202.4 1.57 23.3 0.6

3 98 0.40 96.6 125.9 0.68 15.7 20.6

Total 1.16 98.82 215.6 0.71 78.1 21.9

Note: Calc. Head: 3.21 g/t Au Source: SGS Metallurgical Report, 2020

The E-GRG value determined by SGS was used by FLSmidth (Knelson) to model gold recovery in the

MDZ process under the following conditions:

• Plant feed: 116 tph

• Circulating load: 300%

• Grind size (P80 µm): 105

• Conc. cycle time (min): 40

Table 13-30 provides a summary of the modeling results under two scenarios. The first scenario

included processing 45% of the cyclone underflow with a single Knelson concentrator (model QS40)

followed by leaching in an Acacia intensive leach reactor (model CS2000) which would result in an

estimated recovery of 51% of the E-GRG (78.1% per SGS) and result in about 40% overall gold

recovery to the gravity circuit. The second scenario included processing 90% of the ball mill discharge

to two Knelson concentrators (model QS48) followed by leaching in an Acacia intensive leach reactor



(model CS4000) which would result in an estimated recovery of 67% of the E-GRG and result in about

52% overall gold recovery to the gravity circuit. Estimated gold recoveries include Acacia leach

recoveries and therefore represent gold recovery to doré.

Table 13-30: Summary of E-GRG Modeling

Concentrator % Circ.

Load Treated

Tonnes Treated

GRG (1) Recovery (%)

Total Au Recovery (%)

Upgrade Method

Acacia Size

QS40 45 225 51.0 39.7 Acacia CS2000

2 x QS48 90 600 67.3 52.4 Acacia CS4000

Note: (1) Value represents percentage of 78.1% GRG determined by SGS Source: FLSmidth, 2019

13.2.6 Gravity Separation Testwork

Gravity separation testwork was conducted using a Knelson MD-3 laboratory concentrator operated

under standard laboratory conditions. The Knelson gravity concentrate was upgraded on a Mozley

Laboratory Mineral Separator (model C-800) targeting recovery of 0.05 to 0.1 wt% into the final Mozley

concentrate. The Mozley tailing was recombined with the Knelson tailing and used for downstream

cyanidation testwork. The grind size of about P80 212 µm was used for the first test (G1) on the MDZ

Master composite. A target grind size of P80 105 µm was used for the remaining gravity separation

tests. The results of all gravity separation tests are summarized in Table 13-31. Gold recovery from

the Master composite averaged 58.5%. The percent mass pull to the gravity concentrate ranged from

about 0.07 to 0.09% in these tests. Silver recovery ranged from about 16 to 21%. Gravity gold

recoveries for the variability test composites ranged from about 49% to 82%. Gravity silver recoveries

ranged from about 9% to 34%. These results demonstrate that a gravity separation circuit should be

considered in the overall process flowsheet and will serve to significantly reduce carbon handling in

the downstream CIP or CIL circuit.



Table 13-31: Summary of Gravity Concentration Testwork




13.2.7 Gravity Concentrate Cyanidation

An intensive cyanide leach (ICN) test was conducted on the gravity concentrate produced from a 30

kg gravity concentration test without regrinding and used standard ICN test conditions that included:

• Cyanide concentration: 20 g/L NaCN

• LeachAid: 100 kg/t conc (0.12 kg/t ore)

• Retention Time: 48 hours

• pH: 11 to 11.5

The results of this test are presented in Table 13-32 and show that 99.7% of gold and 87.9% of the

silver contained in the gravity concentrate were extracted. Sodium cyanide consumption was 26.4 kg/t

concentrate (0.032 kg/t ore). LeachAid addition was equivalent to 0.12 kg/t ore which has not been

optimized.

Table 13-32: Summary of Intensive Leach Test on Gravity Concentrate

CN Test No.

Reagent Cons. Kg/t of CN Feed

Extraction % Residue (g/t) Calc. Head (g/t)

Au Ag

NaCN CaO 24 h 48 h 24 h 48 h Au Ag Au Ag

56 26.4 ~0 94.4 99.7 76.4 87.9 4.53 73.5 1,615 609


13.2.8 Gravity Tailing Cyanidation Versus Grind Size

Whole-ore and gravity tailing cyanidation tests versus grind size were conducted on the MDZ Master

composite. Five of the tests were conducted on whole-ore sample and five were conducted on the G-1

gravity tailing. The grind size P80 targets ranged from about 212 to 53 µm. Test conditions for this

series included:

• Grind size target P80’s: 212, 150, 100, 75, and 53 µm

• Pulp Density: 45% solids (w/w)

• No pre-aeration

• Dissolved oxygen: 7 to 8 mg/L (air sparged)

• Pulp pH: 10.5 to 11 (maintained with lime)

• Cyanide Concentration: 0.5 g/L NaCN (maintained)

• Retention Time: 48 hours (with kinetic subsampling)

The results of gold extraction versus grind size is shown in Table 13-33 and silver extraction test results

are shown in Table 13-34. Whole-ore leach results indicate that gold extractions of about 91 to 95%

could be achieved over the grind size range tested. Gold extractions from the gravity tailing increased

from 89.5 to 95.4% with decreasing grind size. Overall (gravity + cyanidation) gold recoveries

increased from about 96 to 98%.

Cyanidation test results on the gravity tailing indicated that there was a clear linear relationship

between grind size and residue grade. An engineering review of these test results was completed and

a grind size of P80 105 µm was selected and used for all remaining cyanidation tests. There was no

clear relationship versus grind size for the whole-ore leach tests, likely due to the presence of small

amounts of coarse free gold. The test results indicated that an additional 0.1 g/t to 0.15 g/t gold will be

recovered with gravity concentration included in the flowsheet. Overall silver recovery was about 61%

at the target grind size.



Table 13-33: Summary of Cyanide Leach Test Gold Extractions Versus Grind Size

CN Test No.

Feed Size P80, µm

Reagent Cons. kg/t of CN Feed

Au Extraction % Gravity

% Au O’All % Au

Au Residue (g/t) Au Head (g/t)

NaCN CaO 2 h 6 h 12 h 18 h 24 h 30 h 48 h A B Avg Calc. Grav +

CN Direct

Whole Ore Tests

1 216 0.77 0.78 14.2 28.5 44.2 57.0 67.3 73.9 91.2 … 91.2 … … 0.27* 3.09 … 3.74

2 155 0.94 0.77 8.9 26.2 45.1 60.1 71.6 79.9 95.2 … 95.2 … … 0.16* 3.42

3 108 1.01 0.80 5.4 23.3 43.2 59.4 71.2 78.3 93.7 … 93.7 … … 0.20* 3.17

4 72 1.10 0.87 3.5 20.2 39.1 55.6 67.2 75.7 94.0 … 94.0 … … 0.23* 3.88

5 49 1.20 0.92 2.3 19.3 36.1 50.7 62.1 71.3 93.7 … 93.7 … … 0.23* 2.61

Gravity Tailing Tests (G-1)

6 211 0.63 0.88 19.7 40.5 60.4 70.1 76.8 81.2 89.5 58.8 95.7 0.15 0.11 0.13 1.23 3.02

7 148 0.70 0.75 18.2 41.9 64.9 75.1 83.1 87.1 92.4 96.9 0.10 0.10 0.10 1.32

8 107 0.87 0.80 10.9 37.2 62.8 74.7 84.1 89.2 94.3 97.7 0.07 0.08 0.08 1.31

9 79 0.99 0.85 8.1 37.6 61.6 73.4 85.8 90.4 95.8 98.3 0.05 0.05 0.05 1.20

10 52 1.01 0.85 5.6 30.8 54.2 64.5 76.3 84.9 95.4 98.1 0.06 0.06 0.06 1.31

*Average of 4 or 14 cuts, depending on test (all 30 g FA to extinction) Source: SGS Metallurgical Report, 2020



Table 13-34: Summary of Cyanide Leach Test Silver Extractions Versus Grind Size

CN Test No.

Ag Extraction (%) Gravity

% Ag O’All %

Ag

Ag Residue

(g/t) Avg Head (g/t)

2 h 6 h 12 h 18 h 24 h 30 h 48 h Calc Grav +

CN Direct

Whole Ore Tests

1 19 30.4 37.4 44.6 46.5 50.4 57.0 … 57.0 1.3 3 … 2.9

2 12.2 29.4 38.6 42.7 46.1 48.2 55.9 … 55.9 1.4 3.2

3 6.3 28.5 38.7 43.9 46.8 49.7 58.6 … 58.6 1.2 2.9

4 3.9 26.4 36.2 43.5 47.7 51.5 58.8 … 58.8 1.3 3.2

5 … 24 33.0 38.7 40.8 46.7 56.8 … 56.8 1.4 3.2

Gravity Tailing Tests (G-1)

6 25.5 36.3 41.7 46.5 47.0 48.3 50.9 16.5 59.0 1.7 3.5 3.8

7 23.1 36.3 42.9 46.0 48.2 50.0 52.7 60.5 1.7 3.6

8 14.8 36.2 44.9 47.8 49.6 51.3 54.4 61.9 1.6 3.5

9 8.7 34.9 42.2 47.0 48.4 50.4 53.4 61.1 1.7 3.7

10 4.8 31.4 38.8 41.9 43.6 47.1 50.7 58.8 2.0 4.1


13.2.9 Cyanidation Versus Cyanide Concentration and Pulp Density

A series of cyanidation tests were conducted to evaluate cyanide concentration over the range from

0.25 to 1 g/L NaCN (maintained) and leach slurry densities over the range from 45 to 55% solids (w/w).

The results of this test series are shown in Table 13-35. Gold extraction and leach residue grade were

independent of cyanide concentration above 0.5 g/L NaCN. At 0.5 g/L NaCN, overall gold extraction

(gravity + cyanidation) as reported at 97.4% and cyanide consumption was reported at 0.88 kg/t.

Cyanidation tests versus slurry density demonstrated that a slurry density of 50% solids (w/w) was

optimum. Above 50% solids gold extraction decreased significantly, most likely due to the increased

viscosity of the slurry.



Table 13-35: Gold Extraction Versus Cyanide Concentration and Slurry Density

CN Test No.

Feed Size P80, µm

NaCN (g/L)

% Solids (w/w)

Reagent Cons. kg/t of

CN Feed Au Extraction % Gravity

% Au O’All % Au


NaCN CaO 2 h 6 h 12 h 24 h 30 h 48 h A B Avg Calc Grav+CN Direct

11 105 0.25 45 0.45 0.97 8.7 26.0 49.8 71.0 78.4 89.8

58.8

95.8 0.14 0.12 0.13 1.27

3.02 3.74

12 105 0.5 45 0.88 0.87 9.4 36.8 64.7 87.4 90.3 93.6 97.4 0.09 0.08 0.09 1.32

13 106 0.75 45 1.17 0.84 14.8 44.1 71.9 88.4 91.1 93.9 97.5 0.08 0.08 0.08 1.31

14 108 1 45 1.41 0.78 11.1 49.7 76.9 91.5 94.2 94.2 97.6 0.07 0.08 0.08 1.29

15 106 0.5 50 0.79 0.88 8.9 37.5 62.4 84.0 85.7 92.6 97.0 0.09 0.09 0.09 1.21

16 108 0.5 55 0.88 0.81 8.7 30.7 55.3 75.5 82.0 88.8 95.4 0.16 0.13 0.15 1.29




13.2.10 Cyanidation Versus Preaeration and Air Versus Oxygen Injection

Cyanidation tests on the gravity tailing were conducted to evaluate the impact of pre-aeration and the

use of air versus oxygen injection. During tests with air injection, dissolved oxygen levels were reported

at about 7 to 9 mg/L while dissolved oxygen levels during tests with oxygen injection were reported at

about 21 to 25 mg/L. The results of these tests are presented in Table 13-36. Gold extractions were

about 95% and residue grades were 0.06 g/t Au for all tests. Overall gold extractions (gravity +

cyanidation) were consistently close to 98%. The impact of oxygen on cyanide and lime consumptions

was significant for each pre-aeration time that was tested. The cyanide consumptions in tests

conducted with oxygen injection were approximately half as much as those tests conducted with air

injection. Lime consumptions were about 25% less. This test series demonstrated that pre-aeration

and oxygen injection resulted in lower cyanide consumption and significantly reduced leach retention

time. Gold extraction versus leach retention time for each test is shown in Figure 13-5.



Table 13-36: Summary Cyanidation Tests with Preaeration and Air Versus Oxygen Injection

CN Test No.

Feed Size

P80,µm

Pre-Air

h

CN Air or O2


Au Extraction (%)

Gravity % Au

O’All %

Au


NaCN CaO 2 h 4 h 8 h 24 h 30 h 48 h A B Avg Calc. Grav

+ CN

Direct

17 103 8 Air 0.40 1.18 42.0 60.0 75.8 91.5 94.3 94.4

58.8

97.7 0.09 0.05 0.07 1.26

3.02 3.74

18 104 8 O2 0.19 0.78 54.8 72.0 83.2 94.4 94.8 95.2 98.0 0.06 0.06 0.06 1.26

19 105 4 Air 0.35 1.04 37.5 55.9 74.1 92.4 93.7 95.5 98.1 0.06 0.05 0.06 1.21

20 107 4 O2 0.16 0.75 60.8 75.6 85.3 95.8 97.4 95.0 97.9 0.06 0.06 0.06 1.20

21 106 2 Air 0.37 0.94 33.4 52.3 71.0 92.6 94.6 95.0 97.9 0.06 0.06 0.06 1.21

22 104 2 O2 0.15 0.70 57.8 72.4 85.8 95.1 94.7 95.1 98.0 0.06 0.06 0.06 1.23





Figure 13-5: Gold Extraction Versus Leach Retention Time for Air and Oxygen Injection

13.2.11 Cyanidation Versus Cyanide Attenuation and Pulp Density

A series of leach tests were conducted at slurry densities of 45, 50 and 55% solids at an initial cyanide

concentration of 0.5 mg/L NaCN which was allowed to attenuate to 0.2 mg/L. Tests included pre-

aeration (4 hours) and oxygen injection to maintain dissolve oxygen levels at about 20 to 25 mg/L. The

results of these tests were similar and are summarized in Table 13-37. Gold extraction ranged from

94.7 to 95.6% and overall gold extractions (gravity + cyanidation) were about 98% for all tests. Gold

residue grades were 0.05 to 0.07 g/t. Allowing cyanide to attenuate throughout the test significantly

reduced sodium cyanide consumption to 0.13 to 0.15 kg/t. The leach kinetic results are shown in Figure

13-6 and demonstrate that gold extraction was complete after about 24 hours of leaching.

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35 40 45 50

Au E

xtr

action,

%

Retention Time, h

CN-8, No PA, Air

CN-12, No PA, Air

CN-17, 8 h PA, Air

CN-18, 8 h PA, O2

CN-19, 4 h PA, Air

CN-20, 4 h PA, O2

CN-21, 2 h PA, Air

CN-22, 2 h PA, O2



Table 13-37: Summary of Cyanide Attenuation and Slurry Density Tests

CN Test No.

Feed Size P80, µm

% Solids (w/w)


Au Extraction (%) Gravity

% Au O’All % Au


NaCN CaO 2 h 4 h 8 h 24 h 30 h 48 h A B Avg Calc Grav +

CN Direct

23 107 45 0.13 0.76 67.0 76.1 86.1 95.4 96.5 95.6

58.8

98.2 0.05 0.06 0.06 1.25

3.02 3.74 24 104 50 0.15 0.76 66.9 76.6 85.7 95.4 95.8 95.6 98.2 0.05 0.05 0.05 1.25

25 103 55 0.15 0.74 63.8 73.0 83.4 94.5 95.0 94.7 97.8 0.07 0.06 0.07 1.24


SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Page 162



Figure 13-6: Gold Extraction Versus Leach Retention Time

13.2.12 “Hard Stop” Retention Time Tests

“Hard stop” retention time tests were conducted at 18, 24, 30 and 36 hours. All tests were conducted

under the following conditions:

• Grind size P80: 105 µm

• Pulp density: 50% solids (w/w)

• Pre-aeration: 4 hours

• O2 injection: 20 to 25 mg/L dissolved O2


• Cyanide Conc: 0.5 g/L NaCN (allowed to attenuate to 0.2 g/L)

The results of these tests are shown in Table 13-38 and confirm that with the use of oxygen a retention

time of 24 hours is sufficient to achieve maximum gold extraction. Gold extractions ranged from about

93 to 94% with leach residues ranging from 0.07 to 0.08 g/t Au. Overall gold extraction (gravity +

cyanidation) was 97.7% after 24 hours and sodium cyanide consumption was reported at 0.08 kg/t.

An optimized leach retention time of 24 hours was selected based on these test results.

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35 40 45 50

Au E

xtr

action,

%

Retention Time, h

CN-23

CN-24

CN-25



Table 13-38: Summary of Hard Stop Leach Retention Time Tests


13.2.13 Carbon-In-Leach (CIL) Tests

Carbon-in-leach (CIL) tests were conducted at leach retention times of 24, 30 and 36 hours to further

evaluate retention time. The results of these tests are summarized in Table 13-39 and confirmed that

the gold in solution will load onto activated carbon and that the MDZ ore is not preg-robbing. One gram

of activated carbon was added to each bottle and gold loadings of about 950 g/t Au onto the carbon

were reported.

Table 13-39: Summary of CIL Tests Versus Retention Time with Optimized Leach Conditions


13.2.14 Variability Tests

Testwork to evaluate gravity concentration and cyanidation of the gravity tailing on each of the

variability composites was conducted under optimized conditions which included:


• Pulp density: 45% solids (w/w)





• Retention time: 24 hours

Tests were conducted in duplicate and the results of gold extraction are shown in Table 13-40 and the

results of silver extraction are shown in Table 13-41. Gold recovery into the gravity concentrate ranged

from 60.3 to 82% for the MDZ variability composites with the highest gravity gold recovery being

reported for the deeper MDZ Deep variability composite. Gold extraction from the MDZ gravity tailings

ranged from 91.4 to 92.8% and overall recovery (gravity + cyanidation) ranged from 97.4 to 98.5%

gold. Gold recovery from the Marmato mine variability composite into the gravity concentrate was

reported at 48.7% and gold extraction from the Marmato gravity tailing was 73.7% with an overall gold

CN Feed NaCN % Pre- Air Gravity O'All

Test Size g/L Solids Air or % % Grav

No. P80, µm (w/w) h O2 NaCN CaO Au Au A B Avg. +CN

27 105 0.5 - 0.2 50 4 O2 0.06 0.67 18 h 93.3 58.8 97.2 0.08 0.07 0.08 1.12 3.02 3.74

28 109 0.5 - 0.2 50 4 O2 0.08 0.72 24 h 94.3 97.7 0.07 0.07 0.07 1.23

29 104 0.5 - 0.2 50 4 O2 0.09 0.77 30 h 93.5 97.3 0.08 0.08 0.08 1.23

30 109 0.5 - 0.2 50 4 O2 0.09 0.78 36 h 94.1 97.6 0.07 0.08 0.08 1.26

%

Extraction

AuReagent Cons.Au Residue, g/t

Au Head, g/t

kg/t of CN Feed Calc. Direct

CN Feed NaCN % Pre- Air Gravity O'All

Test Size g/L Solids Air or % % Grav

No. P80, µm (w/w) h O2 NaCN CaO Au Au A B Avg. +CN

31 103 0.5 - 0.2 50 4 O2 0.08 0.70 24 h 92.0 58.8 96.7 0.08 0.10 0.09 1.13 3.02 3.74

32 110 0.5 - 0.2 50 4 O2 0.10 0.67 30 h 92.8 97.0 0.09 0.09 0.09 1.24

33 108 0.5 - 0.2 50 4 O2 0.10 0.76 36 h 92.3 96.8 0.08 0.09 0.09 1.17

Reagent Cons. AuAu Residue, g/t

Au Head, g/t

kg/t of CN Feed Extraction Calc. Direct

%



recovery of 86.5%. Overall silver recovery from the MDZ variability composites ranged from 47.4 to

56.4% and averaged 54.5%. Overall silver recovery from the Marmato mine composite was 70.1%.

Table 13-40: Variability Composites – Gold Recovery Under Optimized Conditions


Sample CN Feed Au Gravity O'All

Test Size Extr'n % % % Calc. Grav Direct

No. P80, µm NaCN CaO 24 h Au Au A B Avg. +CN

Low Grade 34 104 0.09 0.78 91.4 61.5 96.7 0.06 0.06 0.06 0.70 1.80 1.97

Comp (G-3) 35 103 0.10 0.75 92.0 96.9 0.06 0.05 0.06 0.69

91.7 96.8 0.06 0.70

Med Grade 36 110 0.08 0.71 91.6 66.7 97.2 0.06 0.08 0.07 0.83 2.58 3.62

Comp (G-4) 37 114 0.09 0.70 92.7 97.6 0.07 0.06 0.07 0.89

92.2 97.4 0.07 0.86

High Grade 38 112 0.13 0.63 92.3 60.3 96.9 0.13 0.11 0.12 1.56 3.99 3.53

Comp (G-5) 39 114 0.14 0.62 91.3 96.5 0.17 0.11 0.14 1.61

91.8 96.7 0.13 1.59

Deep Zone 48 104 0.09 0.64 91.4 82.0 98.5 0.08 0.06 0.07 0.81 4.52 5.33

Comp (G-7) 49 107 0.09 0.61 91.4 98.5 0.06 0.08 0.07 0.82

91.4 98.5 0.07 0.82

Transition 50 105 0.09 1.00 92.5 63.2 97.2 0.06 0.09 0.08 1.02 2.77 2.67

Comp (G-8) 51 105 0.09 0.99 93.1 97.5 0.07 0.07 0.07 1.02

92.8 97.4 0.08 1.02

New Marmato 54 100 0.09 1.40 74.4 48.7 86.9 0.40 0.42 0.41 1.60 3.15 3.13

Comp (G-12) 55 98 0.12 1.46 72.9 86.1 0.42 0.47 0.45 1.64

73.7 86.5 0.43 1.62

Au Head, g/tReagent Cons.Au Residue, g/t

kg/t of CN Feed



Table 13-41: Variability Composites – Silver Recovery Under Optimized Conditions


13.2.15 CIP Modelling Testwork

Carbon-in-pulp (CIP) modeling was conducted by SGS in order to establish the design parameters

and predict operational performance for the CIP circuit. SGS’s approach to CIP modelling involves

conducting batch gold leaching and carbon adsorption tests with representative samples of ore or

concentrate in contact with commercially available activated carbon or plant carbon. The rate of

leaching is determined in a traditional bottle roll experiment, by taking timed samples of slurry from the

bottle (typically over a 72-hour period) and analyzing the solution phase for gold. The rate of absorption

of the leached gold onto activated carbon is then determined by adding carbon to the same leach

slurry and taking further timed samples of slurry over a further 72-hour period in the same rolling bottle

and analyzing the solution phase for gold. Gold on the carbon is determined by mass balancing the

solution phase, while gold in the leach residue is determined by analysis at the end of the test, to

produce an overall gold balance for the test. As a check, the final test carbon is also assayed for gold.

The leaching and carbon adsorption kinetic data were then fitted to carbon adsorption modeling

equations which generates profiles of gold in solution, on the carbon and in the leach residue across

a series of leaching and adsorption tanks in which carbon is advanced counter-current to the flow of

slurry. The CIP models allow a number of operating parameters to be varied systematically. This allows

the optimum design criteria for the plant to be established. CIP circuit parameters that were modeled

include:

CN Ag Gravity O'All Ag

Test Extr'n % % % Res. Calc. Grav Direct

No. 24 h Ag Ag g/t +CN

34 38.6 15.0 47.8 1.6 2.6 3.1 2.8

35 37.6 47.0 1.7 2.7

38.1 47.4 1.7 2.7

36 47.9 21.2 58.9 1.2 2.3 2.9 2.4

37 48.5 59.4 1.2 2.3

48.2 59.2 1.2 2.3

38 44.5 21.7 56.5 1.6 2.9 3.8 3.6

39 43.3 55.6 1.7 3.0

43.9 56.1 1.7 2.9

48 33.4 34.3 56.2 <0.5 0.8 1.1 0.8

49 34.0 56.6 <0.5 0.8

33.7 56.4 <0.5 0.8

50 41.8 17.0 51.7 2.0 3.4 4.5 3.7

51 45.9 55.1 2.2 4.1

43.9 53.4 2.1 3.8

54 37.8 8.7 72.3 5.6 9.0 10.1 10.3

55 35.5 67.9 6.1 9.5

36.7 70.1 5.9 9.3

Ag Head, g/t



• The percentage of the leachable gold that is in solution prior to the first carbon adsorption tank

(100% generally assumed for CIP designs)

• The number of carbon adsorption tanks

• The volume of the adsorption tanks and pulp residence time

• The amount and concentration of carbon in each adsorption tank

• The carbon advance rate through the CIP plant, which yields the target gold loading on the

carbon going to elution

• The target gold concentration in the solution exiting the last carbon adsorption tank

• The amount of gold remaining on the eluted carbon that is recycled to the last adsorption tank

Base-case circuit modeling included the following design parameters:

• Process plant feed rate: 181 tph

• Slurry feed rate: 288 m3/hr

• Leach retention time: 24 hours

• Number of Leach tanks: 3

• Adsorption tank size: 288 m3

• CIP stages: 6

• Au on stripped carbon: 50 g/t

• Carbon advance rate: 3 t/d

The results of the CIP modeling are presented in Table 13-42 and Table 13-43. Scenario 1 represents

the base-case and other scenarios show the impact of sequential changes in CIP circuit operating

parameters. The data shown in Table 13-42 presents all the key process operating parameters and

the red highlighted values indicate the parameter that has been changed in each scenario. The target

gold barren solution concentration is 0.01 mg/L or less. Key points from the CIP modeling include the

following:

• Scenario 1:

o In Scenario 1 (Base-case) the carbon inventory was 7.2 t per tank, which yielded a carbon

concentration of 25 g/L and the carbon advance rate was set at 3 t/day, which resulted in

a gold loading on the carbon of 1,673 g/t. Soluble gold losses of 0.004 mg/L were predicted

and gold extraction from solution onto the carbon was 99.5% with target barren solution

losses of about 0.01 mg/L achieved in the fifth CIP stage.

• Scenarios 2 to 5:

o Scenarios 2 to 5 illustrated the impact of varying the amount of carbon in the circuit over

the range from 15 to 40 g/L. The barren solution losses increased slightly when decreasing

the amount of carbon in the circuit, and as expected, the gold lock-up (kg of gold)

increased with higher carbon inventories.

• Scenario 6:

o Scenario 6 investigated the impact of decreasing the slurry retention time in each CIP tank

to 30 minutes. This increased the concentration of carbon in each stage to 40 g/L (same

as Scenario 5) but the amount of carbon in each stage was the same as Scenario 2. The

results from Scenario 6 and Scenario 2 were identical, which shows the performance in

CIP is controlled by the amount of carbon in each tank, not the carbon concentration.



• Scenarios 7 and 8:

o Scenarios 7 and 8 were identical to Scenario 6, except that the amount of carbon

transferred to elution each day was decreased to about 2.3 to 2.6 tonnes. As a result, the

loading of gold on the carbon in the first CIP tank increased to 2,156 g/t and 1,923 g/t,

respectively This indicates the capacity of the elution and regeneration circuits could be

reduced from the current design of 3 t/day to 2 t/day with minimal impact on gold recovery.

• Scenarios 9 and 10:

o Scenarios 9 and 10 were run using a 4 t elution circuit since the design engineer (Ausenco)

had noted that a 4 t circuit may be considered in the design. The slurry residence times in

the CIP stages were set at 1 hour and 0.5 hours, respectively in these scenarios. Barren

solution losses were below 0.006 mg/L in both scenarios.

The results from the CIP modelling study were very positive and excellent results can be expected

when processing the Marmato MDZ ore in a standard CIP circuit design. The optimized circuit design

based on the results in this study were as follows:

• Leach retention time of 24 hours

• CIP retention time of six hours (one hour per stage)

• Carbon inventory 6 t/tank (36 t total)

• Elution/regeneration plant capacity 3 t/day. A smaller carbon throughput of 2 t/day could be

considered if there are no plans to increase plant capacity in the future

• Eluted carbon concentration target, 50 g/t

Silver and Copper Carbon Loading

The estimated silver and copper loadings were calculated by SGS using equilibrium isotherm data.

Silver loading is estimated at about 1,400 g/t and the loaded carbon would contain about 0.2% copper.

Based on an expected concentration of silver in solution in the feed to CIP of 1.0 to 1.5 mg/L and

barren solution losses of about 3 mg/L, the extraction efficiency of silver onto carbon is estimated at

about 70 to 80%. Although copper loading on the carbon could be significant and similar to gold and

silver loading on a g/t basis, the amount of leached copper that is extracted in CIP is trivial (less than

1%).



Table 13-42: Modeled Design Parameters for a Multi-stage CIP Adsorption Circuit

* Ramp up time (days) = Gold lock-up (kg)/Gold Produced (kg/day) Source: SGS metallurgical report 2020

Different Scenarios 1 2 3 4 5 6 7 8 9 10

Inputs

Slurry feed rate (m3/h) 288 288 288 288 288 288 288 288 288 288

Solids (t/h) 181 181 181 181 181 181 181 181 181 181

Solution (m3/h) 221 221 221 221 221 221 221 221 221 221

Consider Leach after Carbon addition N N N N N N N N N N

Gold on stripped carbon, g/t 50 50 50 50 50 50 50 50 50 50

Adsorption tank(s) size, m3 288 288 288 288 288 144 144 144 288 144

Carbon frequency advance (% in 24 hours) 42% 52% 69% 35% 26% 52% 40% 45% 35% 69%

Leaching

Au leached before Carbon addition 93.9% 93.9% 93.9% 93.9% 93.9% 93.9% 93.9% 93.9% 93.9% 93.9%

Leach time before Carbon addition (h) 24 24 24 24 24 24 24 24 24 24

Leach only total tankage (m3) 6912 6912 6912 6912 6912 6912 6912 6912 6912 6912

Number of Leaching tanks 3 3 3 3 3 3 3 3 3 3

Volume of Leaching tanks (m3) 2304 2304 2304 2304 2304 2304 2304 2304 2304 2304

CIP/CIL

Leach Kinetic Constant (ks) 0.867 0.867 0.867 0.867 0.867 0.867 0.867 0.867 0.867 0.867

Model output kinetic constant (k) 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003

Model output equilibrium constant (K) 20793 20793 20793 20793 20793 20793 20793 20793 20793 20793

Product of equilibrium and kinetic constants (kK) 69 69 69 69 69 69 69 69 69 69

Number of stages 6 6 6 6 6 6 6 6 6 6

Total CIP/CIL volume (m3) 1728 1728 1728 1728 1728 864 864 864 1728 864

Slurry residence time in each adsorption tank (h) 1.0 1.0 1.0 1.0 1.0 0.5 0.5 0.5 1.0 0.5

Gold grade in residue (g/t) 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074

Gold in final barren solution (mg/L) 0.004 0.006 0.012 0.004 0.003 0.006 0.007 0.007 0.003 0.006

Gold in loaded carbon (g/t) 1673 1669 1660 1675 1676 1669 2156 1923 1269 1265

Carbon residence time/stage (h) 58 46 35 69 92 46 60 53 69 35

Carbon Concentration (g/L pulp) 25 20 15 30 40 40 40 40 40 40

Equivalent transferred carbon unit flowrate (kg/h) 125 125 125 125 125 125 96 108 167 167

Daily carbon transfer / batch elution capacity (kg/day) 3000 3000 3000 3000 3000 3000 2304 2592 4000 4000

Carbon Inventory per stage (kg) 7200 5760 4320 8640 11520 5760 5760 5760 11520 5760

Carbon inventory all stages (tons) 43 35 26 52 69 35 35 35 69 35

Gold Lock-Up on Carbon (kg) 20.4 17.3 14.0 23.5 29.5 17.3 22.5 20.0 22.6 13.2

CIP/CIL Gold recovery per day (g/day) 4869 4858 4829 4874 4877 4858 4853 4855 4878 4861

Overall Gold Leaching Efficiency 93.9% 93.9% 93.9% 93.9% 93.9% 93.9% 93.9% 93.9% 93.9% 93.9%

Overall Gold Adsorption Efficiency 99.5% 99.3% 98.7% 99.6% 99.7% 99.3% 99.2% 99.2% 99.7% 99.4%

Overall Gold Recovery 93.4% 93.2% 92.6% 93.5% 93.6% 93.2% 93.1% 93.1% 93.6% 93.3%

Upgrading ratio 1816 1812 1801 1817 1819 1812 2340 2087 1378 1373

Circuit filling time - slurry (days) 1.3 1.3 1.3 1.3 1.3 1.1 1.1 1.1 1.3 1.1

Ramp-up time (days) * 4.2 3.6 2.9 4.8 6.1 3.6 4.6 4.1 4.6 2.7

* Ramp-up time (days) = Gold lock-up (kg) / Gold Produced (kg/day)



Table 13-43: Modeled Gold Concentrations in Solids, Solution and Carbon in a Multi-Stage CIP Circuit


13.2.16 Cyanide Destruction Testwork

The cyanidation leach residue produced under optimized leach conditions from the MDZ Master

composite was used for cyanide destruction testwork using the industry-standard SO2/Air

detoxification process. The objective was to reduce weak acid dissociable cyanide (CNWAD) in the

residue from about 200 mg/L CNWAD to <1 mg/L CNWAD.

The chemical reaction for the oxidation of (CNWAD) using sodium metabisulphite (Na2S2O5) as the

source of SO2) and air (source of oxygen) is as follows:

2 CN + Na2S2O5 + 2 O2 + 2 OH→2 CNO + Na2SO4 + SO42 + H2O

This reaction is catalyzed by the presence of copper. The feed usually contains some copper (as the

copper cyano complex), and if required, additional copper is added as copper sulfate. Hydrated lime

is added to the reactor to provide the OH- ion for the above reaction. The cyanate ion (CNO-) is unstable,

and slowly hydrolyzes to ammonium and carbonate ions:

CNO- + 2 H2O→CO32- + NH4

+

The carbonate ion precipitates as calcium carbonate and a small amount of the ammonium ion is found

to form ammonia (NH3) which eventually escapes from the solution as NH3 gas.

The cyanide destruction tests were conducted on the leach residue at a slurry density of 45% solids

(w/w). The pH target for the tests was approximately 8.5. All tests were conducted at room temperature

with SO2 additions (as sodium metabisulphite) of 7 to 7.8 g SO2/g CNWAD and retention times that

ranged from about 60 to 90 minutes. The results of continuous detoxification tests are shown in Table

Interstage data 1 2 3 4 5 6 7 8 9 10

Scenario

Gold in ore/stage residues (g/t)

Feed head grade 1.200 1.200 1.200 1.200 1.200 1.200 1.200 1.200 1.200 1.200

Leach tank discharge 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074

Adsorption stage 1 discharge 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074






Gold on carbon (g/t)

Adsorption stage 1 discharge 1673 1669 1660 1675 1676 1669 2156 1923 1269 1265






Stripped carbon feed to last stage 50 50 50 50 50 50 50 50 50 50

Gold in solution (mg/L)

Leach tank discharge 0.921 0.921 0.921 0.921 0.921 0.921 0.921 0.921 0.921 0.921









13-44. The CNpicric assay in three of these four tests was less than 10 mg/L CNWAD. CNWAD

concentrations of less than 1 mg/L were only achieved after additional time (or aging) at the end of the

tests.

Upon completion of the program it was determined the following operating conditions will achieve a

discharge CNWAD concentration of less than10 mg/L.

• 45% solids (w/w)

• Approximately 7 g equivalent SO2 per gram CNWAD

• Approximately 20 mg/L copper addition

• pH 8.5 – lime added as needed (~1.5 kg/t)

• Approximately 80-minute retention time

In order to achieve a CNWAD concentration of <1 mg/L, the design will need to include holding (or aging)

the detoxified leach residue to achieve the discharge target.

Table 13-44: Summary of Cyanide Destruction Tests Conducted on Master Composite Leach Residues

… No Sample Submitted for Assays (1) Cu added using CuSO4 5H2O, SO2 Added Using Sodium Metabisulphite * 2 Stage Reactor Setup (2 x 60 Min Retention Time) ** 7 Day Aged Samples Rerun for Picric Acid in Lab Source: SGS Metallurgical Report, 2020

13.2.17 Tailing Thickening

Tailing thickening testwork was conducted by both SGS and Outotec. The process design engineer

(Ausenco, Section-17) used Outotec’s thickening testwork for process design purposes and, as such,

only Outotec’s test results are presented. Details of SGS’s thickener test results can be found in their

2020 report referenced earlier.

Outotec conducted high rate thickener testwork on detoxed leach tailings generated from MDZ master

and transition composites that were produced using optimized process parameters. The results of

Outotec’s thickener testwork are presented in their report No 32491 dated 2/21/2020. The testwork

was conducted using Outotec’s bench scale 99 mm diameter thickener test unit with Magnafloc 10 as

the flocculant which is a high molecular weight slightly anionic flocculant. All rheological measurements

were carried out using a Thermo Haake VT550 rheometer and an “OK600” 4 blade vane. A constant

Test Test Reten. Product (Solution Phase) Reagent Addition

Dur. Time pH CNT Cu Fe g/g CNWAD g/L Feed Pulp kg/t Solids

Ana. Picric Aged SO2 Lime Cu(1)SO2 Lime Cu(1)

SO2 Lime Cu(1)

Lab Acid Picric** Equiv. Equiv. Equiv.

min min mg/L mg/L mg/L mg/L mg/L mg/L

CND 3 180 60 8.5 … … 0.38 … … … 5.07 2.71 0.150 0.79 0.42 0.023 1.26 0.67 0.037

Continuous

3-1 60 60 8.5 … … 52.6 … … … 5.14 2.85 0.050 0.87 0.49 0.009 1.27 0.71 0.012

3-2 120 60 8.5 2.61 <0.1 12.5 <0.1 27 2.5 7.81 4.58 0.150 1.27 0.76 0.025 1.93 1.13 0.037

3-3 180 79 8.5 2.39 <0.1 7.86 <0.1 16 1.3 7.09 7.02 0.150 1.13 1.15 0.023 1.76 1.74 0.037

3-4 181 84 8.5 2.83 <0.1 8.34 <0.1 18 1.5 7.00 5.53 0.240 1.09 0.88 0.038 1.73 1.37 0.059

3-5* 240 58 8.5 0.27 <0.1 1.26 … 6.8 0.4 7.26 3.91 0.150 1.13 0.62 0.024 1.80 0.97 0.037

… No sample submitted for assays(1)Cu added using CuSO4 5H2O, SO2 added using sodium metabisulphite

* 2-Stage Reactor setup (2 x 60 min retention time)

** 7 day aged samples rerun for picric acid in lab

…

Batch

Feed (CN-47) … … … … … …… … 15.8 1.13 … …225 200

CNWAD by



shear rate of 0.1 sec-1 was used. For each dynamic test underflow sample, a simple un-sheared vane

yield stress was measured.

The tailings samples used for the tailing thickener testwork were characterized as follows:

Master Comp Transition Comp

Density (g/mL): 1.36 1.38

Solids SG (t/m3): 2.66 2.66

pH: 8.70 8.90

Particle size (P80 µm): 107.7 103.3

The results of high rate thickener tests conducted on the MDZ Master composite are summarized in

Table 13-45 and the results of thickener tests conducted on the MDZ Transition composite are

summarized in Table 13-46. An underflow density of 63.5% solids was achieved for the Master

composite tailing sample using Magnafloc10 at a dosage of 50 g/t, which resulted in a flux of 0.80 t/

(m2.h). An underflow density of 63% solids was achieved for the Transition composite tailing sample

using Magnfloc 10 at a dosage of 60 g/t, which resulted in a flux of 0.40 t/ (m2.h). Thickener overflows

were clear with suspended solids reported at less than 100 mg/L and were suitable for recycle back to

the process, although the target underflow density of 64% solids was not achieved in these tests.

Outotec concluded that based on their experience with their testwork and full-scale operation of

thickeners that an estimated 2 to 3% increase in thickener underflow density could be expected when

comparing the testwork to a full-size thickener.

Table 13-45: Summary of High Rate Thickening Test on MDZ Master Composite Leached Tailing

Run No.

Feed Flocculant Underflow Overflow

Flux (t/[m2·h])

Liquor RR (m/h)

Type Dose (g/t)

Meas. Solids (% [w/w])

YS (Pa)

Solids (mg/L)

1 0.80 2.70

Magnafloc 10

40 63.4 32 127

2 0.80 2.70 30 63.0 30 204

3 0.80 2.70 50 63.5 30 <100

4(1) 0.80 2.70 50 69.8 102 <100

5 1.00 3.37 50 62.7 42 <100

6 1.20 4.05 50 61.7 45 <100

Note: (1) Test 4 was run as a high compression test Source: Outotec Report 324931, 02/21/2020

Table 13-46: Summary of High Rate Thickening Test on MDZ Transition Composite Leached Tailing

Run No. Feed Flocculant Underflow Overflow

Flux (t/[m2·h]) Liquor RR (m/h) Type Dose (g/t) Meas. Solids (% [w/w]) YS (Pa) Solids (mg/L)

1 0.80 2.70

Magnafloc10

60 60.2 53 172

2 0.80 2.70 50 60.3 52 338

3 0.80 2.70 70 60.6 61 155

4 0.60 2.03 60 61.5 59 154

5(1) 0.60 2.03 60 69.3 188 154

6 0.40 1.35 60 63.0 21 155

Note: (1) Test 5 was run as a high compression test Source: Outotec Report 324931, 02/21/2020



13.2.18 Tailings Filtration

Outotec conducted filtration testwork on detoxed leach tailings generated from MDZ Master and

Transition composites that were produced using optimized process parameters. Filtration testing was

performed using Outotec’s Labox 100 bench-scale unit and Scanmec leaf dip test apparatus to

examine the filtering characteristics and process suitability. The filter cakes were required to have a

cake moisture of less than 15% and be suitable for dry stacking at the tailings storage facility. The

results of Outotec’s filtration testwork are presented in their report No 32491 dated 2/25/2020.

The thickened tailings samples used for the filtration testwork were characterized as follows:

Master Comp Transition Comp

Density (g/mL): 1.65 1.63

Slurry solids (% w/w): 65.0 62.0

pH: 8.9 8.7

Particle size (P80 µm): 107.7 103.3

Pressure filtration test results on the Master composite tailing sample are shown in Table 13-47 and

the results for the Transition composite tailing sample are shown in Table 13-48. The Master composite

tailing sample achieved 12% moisture contents over the range of cycle times tested (8.5 to 12

minutes). At the 8.5 minute cycle time, a filtration rate of 269.6 kg/m2.hour was reported. The transition

composite tailing sample achieved 14.6 to 15.8% moisture contents over the range of cycles times

tested (9 to 12 minutes). At the 10 minute cycle time, a cake moisture content of 14.6% and a filtration

rate of 214.9 kg/m2.hour were reported. Pressure filtration on both the master and transition composite

tailing samples achieved the required moisture content for disposal in a dry stack DSTF.

Table 13-47: Pressure Filtration Test Results on the Master Composite Tailing Sample

Parameters Units Run #1 Run #2 Run #3 Run #4

Feed Density % w/w 65.0 65.0 65.0 65.0

Filter Cloth Type AITE S400 AITE S400 AITE S400 AITE S400

Chamber Depth mm 50 50 50 50

Cycle time min 12.0 9.5 9.0 8.5

Pumping Pressure bar 6.0 6.0 6.0 6.0

Pressing Pressure bar 12.0 12.0 12.0 12.0

Air Drying bar 10 10 10 10

Cake Thickness mm 48.0 47.6 47.1 47.4

Cake Moisture % w/w 12.2 12.1 12.2 12.1

Filtration Rate D.S. kg/m2h 189.2 239.7 252.2 269.6

Source: Outotec report 324931, 02/25/2020



Table 13-48: Pressure Filtration Test Results on the Transition Composite Tailing Sample

Parameters Units Run #1 Run #2 Run #3 Run #4 Run #5

Feed Density % w/w 62.0 62.0 62.0 62.0 62.0

Filter Cloth Type AITE S400 AITE S400 AITE S400 AITE S400 AITE S400

Chamber Depth mm 50 50 50 50 50

Cycle time min 12.0 11.0 10.0 9.0 10.5

Pumping Pressure Bar 6.0 6.0 6.0 6.0 6.0

Pressing Pressure bar 12.0 12.0 12.0 12.0 12.0

Air Drying bar 10 10 10 10 10

Cake Thickness mm 47.0 47.1 43.9 39.2 47.5

Cake Moisture % w/w 15.8 15.2 14.6 15.4 14.9

Filtration Rate D.S. kg/m2h 191.2 208.6 214.9 209.4 219.4


Vacuum filtration tests were conducted on the Master and Transition tailing composite samples using

the Scanmec Leaf Disc vacuum filtration apparatus. Vacuum filtration test results on the Master

composite tailing sample are shown in Table 13-49 and the results for the Transition composite tailing

sample are shown in Table 13-50. Tests on the Master composite tailing sample were conducted both

with and without filter aid and produced cake thicknesses ranging from 7 to 16 mm. Cake moisture

contents ranged from 17.5 to 22.4%. No vacuum filtration tests achieved the required 15% moisture

content. Tests on the Transition composite tailing sample were conducted both with and without filter

aid and produced cake thicknesses ranging from 3 to 8 mm. Cake moisture contents ranged from 20.7

to 21.7%. No vacuum filtration tests on the Transition composite tailing sample achieved the required

15% moisture content. Based on the results of these tests, vacuum filtration is not an option for filtering

thickened MDZ tailings for disposal in a DSTF.

Table 13-49: Vacuum Filtration Test Results on the Master Composite Tailing Sample

Solids (wt%)

Filter Aid (g/t)

Cycle Times (sec/cycle)

Cake Thickness (mm)

Filtration Capacity (kg/m2h)

Moisture (wt. %)

Cake Cracking

65 None 86 7 560 17.5 No

65 None 40 5 833 18.2 No

65 25 86 10 670 18.6 No

65 50 86 12 1,160 19.4 No

65 75 86 15 1,219 21.2 Yes

65 100 86 16 1,447 22.4 Yes


Table 13-50: Vacuum Filtration Test Results on the Transition Composite Tailing Sample

Solids (wt%)

Filter Aid (g/t)

Cycle Times (sec/cycle)

Cake Thickness (mm)

Filtration Capacity

(kg/m2h)

Moisture (wt. %)

Cake Cracking

62 None 86 3 200 20.7 No

62 None 40 2 431 20.0 No

62 25 86 4 234 20.2 No

62 50 86 5 334 20.2 No

62 75 86 7 351 21.0 No

62 100 86 8 556 21.7 No




13.3 Recovery Estimate

Table 13-51 provides an estimate of achievable gold and silver recoveries from the MDZ based on a

flowsheet that includes gravity concentration and cyanidation of the gravity tailing with optimized PFS

process parameters that include:


• Ret. Time: 24 hours

• Pulp density: 45 to 50% solids (w/w)



• Pulp pH: 10.5 to11 (maintained with lime)


SRK recommends discounting laboratory-reported gold recoveries by 2% and silver recoveries by 5%

to account for inherent plant inefficiencies. Based on the results of the PFS metallurgical program, the

average discounted gold recovery is estimated at 95% and the average discounted silver recovery is

estimated at 51%. This is very similar to the results from the PEA metallurgical program in which the

average discounted gold recovery was estimated at 95% and average discounted silver recovery was

47%. There is little difference in reported gold recoveries for the master and variability composites and

gold recovery appears to be independent of ore grade over the range tested.

Table 13-51: Estimated Gold and Silver Recoveries from the MDZ (PFS and PEA Metallurgical Programs)

Composites

Calc. Head (g/t)

Gravity Recovery (%)

Gravity + Cyan Recovery (%)

Adjusted Overall Recovery (%)

Au Ag Au Ag Au Ag Au Ag

PFS Composites

MDZ Master Comp 3.02 3.80 58.8 16.5 97.7 64.5 96 60

MDZ Variability Comp

Low Grade 1.80 3.10 61.5 15.0 96.8 47.4 95 42

Medium Grade 2.58 2.90 66.7 21.2 97.4 59.2 95 54

High Grade 3.99 3.80 60.3 21.7 96.7 56.1 95 51

Deep Zone 4.52 1.10 82.0 34.3 98.5 56.4 97 51

Transition Zone 2.77 4.50 63.2 17.0 97.4 54.8 95 50

Average Master + Variability Comp.

3.11 3.20 65.4 21.0 97.4 56.4 95 51

PEA Composites

MDZ Master 2.36 4.20 50.6 14.6 96.7 50.6 95 46

MDZ West Zone 1.30 0.90 66.1 24.7 97.3 58.7 95 54

MDZ Center Zone 2.61 2.39 69.0 21.0 97.9 51.6 96 47

MDZ East Zone 1.80 6.70 51.7 15.9 96.7 45.8 95 41

Average Master + Variability Comp.

2.02 3.55 59.4 19.1 97.2 51.6 95 47

Au Adjustment Factor 2

Ag Adjustment Factor 5

Source: SGS, 2019 and 2020



13.4 Significant Factors

The following significant metallurgical and mineral processing factors have been identified:

• The PFS metallurgical program was conducted on an MDZ master composite and on

variability composites representing low, medium and high grade MDZ ore, transition zone and

the MDZ deep zone.

• Native gold was by far the predominant gold carrier and the majority (more than 99%) of the

gold particles occurred within mineral structures that would be readily accessible by leaching

solutions. Gold particles were not often in direct contact with sulfides, yet very commonly

pyrrhotite, chalcopyrite, and bismuth minerals were found in close vicinity to the gold

mineralization

• The metallurgical program optimized process parameters required to recover gold and silver

values from MDZ ore using a process flowsheet that includes gravity concentration followed

by cyanidation of the gravity tailing. Optimized process conditions included:

o Grind size P80: 105 µm

o Pulp density: 45% solids (w/w)

o Pre-aeration: 4 hours

o O2 injection: 20-25 mg/L dissolved O2

o Pulp pH: 10.5-11 (maintained with lime)

o Cyanide Conc: 0.5 g/L NaCN (allowed to attenuate to 0.2 g/L)

o Retention time: 24 hours

• Comminution tests were conducted on the MDZ master composite, MDZ deep zone

composite, three MDZ sub-composites (low grade, medium grade and high grade) and on the

Marmato mine composite. The comminution tests included SAG Mill Comminution (SMC),

SAG Mill Power Index (SPI) and Bond ball mill work index (BWI) tests. In addition, Bond Low

Impact Crushing work index (CWI) and abrasion (AI) tests were conducted on selected ½ HQ

drill core pieces.

o The results of the SMC A x b values ranged from 23 to 29, indicating the ore is hard with

respect to impact breakage.

o The BWI values for the MDZ composites range from 17.7 kWh/t to 19.8 kWh/t, which

places them in the hard range of hardness.

• Gravity recoverable gold (E-GRG) testwork and modeling indicate that about 40% of the gold

contained in the MDZ ore can recovered into a gravity concentrate. Gold contained in the

gravity tailing would be recovered in a standard CIP cyanidation leach circuit.

• An intensive cyanide leach test on the gravity concentrate demonstrated that 99.7% of the

contained gold and 87.9% of the contained silver could be extracted from the gravity

concentrate without regrinding.

• Based on the results of the PFS metallurgical program, overall gold recovery (gravity

concentration + gravity tailing cyanidation) is estimated at 95% and overall silver recovery is

estimated at 51%. This is very similar to the results from the PEA metallurgical program in

which gold recovery was estimated at 95% and silver recovery was 47%. There is little

difference in reported gold recoveries for the master and variability composites and gold

recovery appears to be independent of ore grade over the range tested.



• Cyanide destruction tests demonstrated that weak acid dissociable cyanide (CNWAD) could be

reduced to less than 10 mg/L with the SO2/air process. However, CNWAD levels would further

attenuate to less than 1 mg/L with time.

• Pressure filtration will be required to dewater thickened tailings in order to achieve less than

15% moisture content required for disposal in a DSTF.



14 Mineral Resource Estimate The Mineral Resource Statement presented herein represents the latest mineral resource evaluation

prepared for the Marmato Project reported in accordance with the standard adopted for the reporting

of Mineral Resources of the CIM guidelines, and with NI 43-101 disclosure standards.

SRK has been supplied with electronic databases covering the sampling at the Project, all of which

have been validated by the Company. The databases comprise a combination of historical and recent

diamond core and underground channel samples. In total, there are 1,357 diamond drillholes for a

combined length of 278,945 m and 26,307 individual underground channel samples, inclusive of

current mine sampling contained in the databases. The database contains all the sampling within the

Marmato project and is not limited to license #014-89m, which is the focus of this report.

The resource estimation was completed by Ben Parsons, MSc, MAusIMM (CP), Membership Number

222568, an appropriate “independent qualified person” as this term is defined in NI 43-101. The

Effective Date of the resource statement is March 17, 2020, which is the date the database was

supplied.

The database used to estimate the Marmato Project mineral resources was audited by SRK. SRK is

of the opinion that the current drilling information is sufficiently reliable to interpret with confidence the

boundaries for gold and silver mineralization and that the assay data are sufficiently reliable to support

mineral resource estimation. The use of short channel sampling (sampling length, less than 5 m) has

been limited to the current underground mining domains, defined as the veins, disseminated and splay

domains.

Leapfrog® (version 5.0.4) was used to generate the geological and mineralization models used to

define Marmato model. Datamine™ (version 1.6.75.0) was used to construct the geological solids,

prepare assay data for geostatistical analysis, construct the block model, estimate metal grades and

tabulate mineral resources. Snowden Supervisor software (version 8.12.0) was used for the

statistical/geostatistical analysis and variography.

The estimation methodology involved the following procedures:

• Database compilation and verification

• Construction of wireframe models for the boundaries of the veins

• Construction of wireframe models for the boundaries of the main other domains including:

o Fault network

o Mineralized porphyry

o Low-grade porphyry

o Deeps/feeder structures

• Definition of resource domains

• Data conditioning (compositing and capping) for statistical analysis, geostatistical analysis,

and variography

• Block grade interpolation

• Resource classification and validation

• Assessment of “reasonable prospects for economic extraction” and selection of appropriate

reporting CoGs

• Preparation of the Mineral Resource Statement



14.1 Drillhole Database

SRK was supplied with ASCII files (.csv), extracted from the Company’s Structured Query Language

(SQL) database as seven comma-separated value (.CSV) files with collar, survey, geology, assay,

density, recovery and structure information exported. The exported information contains the latest

drilling and sampling information. These were subsequently imported into Datamine™ and Leapfrog®

for validation.

The database provided included all sampling from the combined drilling database and channel

sampling programs. A summary of the database used in the final estimate is detailed in Table 14-1

Table 14-1: Summary of Number of Records for Each Exported .csv

File Type Number of Records

Collar 27,650

Source 27,650

Survey 86,020

Lithology 159,118

AssayRaw 237,969

AssayLF 237,982

VeinCode 75,015

Density 3,370

Alternation 43,097

Mineralization 45,488

Source: SRK, 2020

A total of 1,357 drillholes has been used to inform the 2020 Marmato MRE including historic drilling

and more recent drilling completed between the 2019 PEA and this PFS (Table 14-2). A total of 40

new drillholes from the exploration and mine developed has been included since the 2019 PEA for a

total of 12,555 m of new drilling. A full description of drilling procedures, sample preparation, sample

analysis and QA/QC is presented in Sections 10 and 11 of this report. Information relating to data

management, and the validation of data is presented in Section 12.

Table 14-2: Summary of Geological Database Information for Drilling Reported by Company

Count Minimum Length (m) Maximum

Length (m)

Average Length

(m)

Sum Length (m)

Company

CGD 20 50.85 559.55 296.67 5,933.35

CGD-GCL 75 16.78 527.40 149.13 11,184.69

CMdC 205 1.20 587.25 226.23 46,377.82

CNQ 47 39.20 600.20 316.45 14,872.95

CNQ-MNL 25 14.00 180.00 72.13 1,803.37

CNQ-PDG

6 70.60 175.00 115.99 695.95

CALDAS 57 127.21 810.20 487.52 27,788.56

MAdO 342 12.00 1,012.10 355.70 121,650.34

MNL 580 4.03 400.35 83.86 48,638.18

Drillhole Subtotal 1,357 1.20 1,012.10 205.56 278,945.21

Grand Total 27,664 0.02 1,012.10 11.61 321,183.22

Source: SRK, 2020

In addition to the drilling information, CGM has captured information from the mine and exploration

channel sampling databases. Limited new sampling has been captured between the 2019 PFS and



the current study for a total of 26,307 channels in the database for a combined sample length of

42,328 m. A summary of the channel sampling programs by Company is shown in Table 14-3.

Table 14-3: Summary of Geological Database Information for Channel Reported by Company

Count Minimum Length

(m) Maximum Length

(m) Average

Length (m) Sum Length

(m)

Company

CGD 165 0.04 13.99 1.02 168.35

CMdC 918 0.03 58.23 3.00 2,749.64

CNQ 39 0.60 154.78 22.22 866.53

MAdO 308 0.50 102.19 9.46 2,912.85

MNL 24,877 0.02 122.21 1.43 35,540.64

Channel Subtotal 26,307 0.02 154.78 1.61 42,238.01

Source: SRK, 2020

14.2 Geologic Model

A 3D lithostratigraphic and structural framework model of the Marmato deposit has been developed.

Within the framework of that model, Resource wireframes have been constructed using Seequent’s

Leapfrog Geo™ v4.5 software. To construct the model, SRK has used a phased approach which

included review of the faults, lithology, and mineralization styles.

14.2.1 Fault Network

TCL recognized two principal deformation stages within the Marmato stock (TCL, 2010):

• Syn-mineralization west-northwest to east-southeast compression reactivated some

basement structures as well as generated a range of second-order shear and extensional

structures along north northwest to west trends, as well as north-northeast-trending thrust

faults.

• Continued post-mineralization compression into the late-Pliocene, (approximately 2 mega

annum [Ma]) that resulted in uplift due to renewed thrusting along the main terrane boundaries,

forming thrust bounded intermontane basins such as the Cauca-Patia depression.

TCL outlined four principal trends of auriferous structures within the Marmato area:

• Northwest-trending steep to sub-vertical faults/fractures

• West-northwest-trending steep to moderately inclined structures

• West-trending structures that tend to have moderate to relatively low angle dips

• East-northeast- to northeast-trending structures that show a range of dips

TCL reported that kinematic indicators show that gold mineralization accompanied a phase of W-NW-

E-SW orientated compression. The N-NE trending reverse faults and conjugate fractures reflect this

compression component. Within this tectonic framework the E-W faults should be predominantly

dextral strike-slip and the W-NW faults should be predominantly sinistral strike-slip. CGM interprets

the rotation of some of these structures to be the result of rotation during progressive compressional

deformation event; however, CGM also noted that there are pre-gold mineralization and post-gold

mineralization phases of fault movement on a number of faults and veins

SRK conducted a site visit to Marmato from December 9 to December 13, 2019. Mr. Blair Hrabi, SRK

Principal Consultant (Structural Geology) conducted the site visit with Dr. Julian Ceballos, CGM

Principal Geologist.



SRK has extruded fault wireframes to surface using regional mapping and underground development

as a guide, and subsequently clipped these to the topography to create a fault network which has been

used to review the vein interpretations during the geological modelling (Figure 14-1). Where possible,

SRK has tied the fault interpretations into the latest geological logging and mapping on levels 16

through 19. During the site visit Mr. Hrabi and Dr. Ceballos visited the Marmato Mine including levels

18, 19, 20, and 21, and reviewed drill core or core photographs for the following drillholes:

• MT-1423

• MT-IU-002

• MT-IU-016

• MT-IU-041

• MT-1430

• MT-IU-009

• MT-IU-017

• MT-IU-045

• MT-1498

• MT-IU-011

• MT-IU-018

• MT-IU-050

• MT-1499-A

• MT-IU-014

• MT-IU-019

• MT-1500

• MT-IU-015

• MT-IU-036

The northwest-dipping Obispo reverse fault (approximately 70°/310°) defines the limit of the known

Marmato gold mineralization to the NW. The N dipping Fault Criminal (approximately 60°/355°) is

interpreted as a dextral strike-slip fault and is truncated to the W by the Fault Obispo. The moderately

W-SW dipping Fault Sur (approximately 50°/190°) cuts the UZ of the Zona Baja but approximately

bounds the southern margin of the MDZ.

Faults 2 (approximately 60°/050°) and 4 (approximately 85°/190°) are found between faults Criminal

and Sur. Fault 2 trends obliquely to, and is cut off by, faults Criminal and Sur. In addition, the composite

Vein (V.) Santa Ines was used in the modelling to define the up-dip extent of Fault Sur.

SRK principal geologist Mr. Giovanny Ortiz visited site on February 10 to 21, 2020 and worked with

both CGM’s exploration and mine geologist to refine the fault interpretation. During the site inspection

and based on mapping on levels 20-21, the decision was made to remove the previously modelled

Fault 4 and replace the structure with a combination of the new structure (Fault 1_3). This fault has

been interpreted to potential offset a portion of the footwall of the MDZ and therefore has been included

in the current estimate during the geological definition phase of the model. SRK and CGM geological

staff undertook a structural interpretation for the deposit using logged faults and breccia in drill core,

underground mapping and surface traces from the digital topography. A series of fault wireframes were

provided to SRK, which were generally localized around areas of structural data.



The final fault network consists of five faults used in the geological model are named:

• Fault Obispo

• Fault Criminal

• Fault Sur

• Fault 2

• Fault 1_3

Source: SRK, 2020

Figure 14-1: Fault Network Compared to Mapping on Level 20 (1056 RL)

Additional faults are in the process of interpretation by CGM which may potentially extend to depth at

the Marmato Project. SRK did not consider these to have sufficient geological confidence to include in

the current estimate but notes that the work should be completed to increase the confidence in these

structures. The additional structures include:

• Cascabel Fault

• Pantanos (Primary)

• Pantanos (Secondary)

Criminal Fault



14.2.2 Topographic Wireframes

CGM commissioned a detailed topographic map with 0.5 and 1 m resolution contour intervals derived

from LIDAR imagery, which was supplied to Datamine™ in 2020. The new topographic map provides

a detailed base map for improved accuracy when plotting the results of the exploration programs, as

well as a high-resolution satellite image. The topography was converted to a solid model in Datamine™

which resulted in excessively large files (more than 2 gb), which was later filter to reduce the number

of points by 50% to enable rapid viewing in various technical software package. This model has been

supplied to SRK by the Company.

14.2.3 Lithological Wireframes

As part of the updated Mineral Resource, SRK initially focused on the creation of a lithological model

(i.e., one encompassing the major geological features inclusive of the current veins being mined). The

lithological database provided to SRK contained 64 separate logging codes, which has been refined

to 14 logging codes by SRK. The main geological features and units modelled by SRK were:

• Major Fault Network

• Porphyry (P1 – P5)

• Meta Schist

• Intrusive

• Volcanic

• Breccia

• Veins

• MDZ

During the definition of the lithological model, SRK noted conflicting interpretations between the

definition of P1 and P2 porphyry units between the mine and exploration teams. To avoid complex

geological coding or detailed relogging, SRK has removed the mine drilling from the lithological

models. SRK notes there are limited differences between the P1 and P2 units which are described as

a dacite porphyry (with Mega Hb-Bi) in P1 compared to a dacite porphyry (common) in P2. SRK does

not consider the removal of the mine drilling to have a material impact on the lithological model or the

associated mineral resource estimates.

The lithological model has been defined in Leapfrog by using the intrusion function within a model with

extents ranging from X: 1,162,550 – 1,166,350, Y: 1,096,350 – 1,099,150, and Z: 0 – 2,000, with the

surfaces clipped to the topography as defined in Section 14.2.2. The fault network has been defined

to generate a total of seven major fault blocks which are used to limit the geological model. A limiting

boundary of 275 m has been used to limit the projection of the geological units as significant areas at

depth still remain undrilled and therefore limit the extension of the veins to depth.

In comparison to the PEA lithological model, SRK has made additional definition of the units within the

P4 – P5 dikes which have been modelled using the vein system utility to form a consistent geological

unit. The other key change is the definition of the breccia units which crosscut the MDZ and may need

consideration for geotechnical criteria prior to mining.



Source: SRK, 2020

Figure 14-2: Cross Section Showing SRK Revised Lithological Model

14.2.4 Veins Model

The vein models have been updated using a combination of the drillhole information, channel

sampling, geological mapping and the initial depletion shapes provided by CGM. The process has

been a collaboration between SRK and CGM to ensure accuracy of the geological conditions mapped

underground are considered. To complete the process CGM geologists supplied SRK with an updated

underground channel database based on the procedures discussed in Section 11.2 of this report.

Additionally, CGM created an initial stope model based on the average dip and strike taken from the

underground long-sections produced by the mine.

SRK imported all the available information into Leapfrog® to aid in the generation of the vein model.

The vein model was created via a staged process. SRK has made a number of modifications to the

geological modelling process in the 2019 update. The changes included combining the vein and halo

domains used during the 2017 process. The decision to combine these domains was based on the

mining methods, typically including both, and therefore a single domain will account for some of the

edge dilution, although not all due to minimum mining widths. Also, the incremental gain from the

model was not deemed to add value, and therefore a single pass combining both mineralization

domains was preferred.

Possibleextensions unknown

Veins extended could possibly be offset by Fault Sur

Extension of veins limited by sampling information



The position of the vein within the channel sampling was based on a combination of lithology logging,

gold grade, geological mapping and (base of) stope wireframe data, with interpretation for the vein in

less well-informed areas guided by visually evident step changes in the gold grade and characteristics

of the vein (i.e., position, thickness and grade continuity) shown in adjacent samples.

The new model includes the differentiation between vein material and disseminated material where

the information was available in the channel samples in the area of the mine. The images below show

the differences with the vein wireframes of the PEA. The impacts of this change include:

• Increased the Au grade and reduction in tonnage in the vein material (MINERALIZATION

STYLE CODES: VEM, VEA, VEN)

• In the previous model, part of the disseminated material (MINERALIZATION STYLE CODES:

VNS, VNA, DSM) was included in the vein wireframes, which could result in the smearing of

high grades

The process was initially completed on the areas with strong geological control in the mine and then

expanded into the upper and lower levels of the deposit which are predominantly supported by drilling

information only.

SRK has used all the available information to define the updated geological model, including mapping,

channel sampling and diamond drilling information. The additional channel (VEN) samples have been

sub-divided per Caldas ‘VEIN’ code:

• Vein samples that influence the wireframe. These were deemed by SRK to be visually spatially

positioned correctly with respect to geological mapping, mining development/stope data and

surrounding drillhole assays and logging and have the following ‘Veins’ code: ‘V_XXX’

• Vein samples that do not influence the wireframe, but have a code corresponding to the

relevant structure for use in statistics evaluation. These are visually spatially offset with respect

to other geological data (which if incorporated would result in ‘pull points’ to the vein

wireframes and potentially overstate the tonnage) and have the following ‘Veins’ code:

‘V_XXX_STM’

SRK has validated the vein model using a series of level plans from the current mining operation for

Levels 16 through 21 and reviewed by CGM geologists for approval. It is the QP opinion that by

combining the geological mapping and channel database from the mine, the lithological model has

been improved with strong controls on the mineralization styles. In the PEA model, the vein model was

combined based on grade with veinlet mineralization in the hangingwall and footwall of the veins which

is termed as the “Disseminated” veins. A summary of the level of information used to generate the

veins model is shown in Figure 14-3.



Source: SRK, 2020

Figure 14-3: Level 20 Geological Mapping Versus Sampling Database and Veins Model, Showing the Level of Information Integrated into the Geological Model

14.2.5 Disseminated Model

The disseminated vein models exist surrounding the main structures and are focused around Levels

16 through 21 of the current mine. The mineralization occurs as veinlets or disseminated mineralization

directly in the hangingwall and footwall of the veins. The mineralization in the disseminated domain

(Coded as DISS) is not logged as VEN but more typically as P1, VNS or DSS in the more recent

drilling. SRK has used a combination of the lithological log and the assays to define the limits of the

disseminated material. The use of the disseminated domains will aid in the definition of diluting grades

surrounding the defined veins and, in some cases, represents higher grades than present within the

veins. The interaction of the veins and disseminated vein domains is shown in Figure 14-4.



Source: SRK, 2020

Figure 14-4: Level Plan (1065 RL), Showing Interaction Between Vein and Disseminated Vein Domains

14.2.6 Splays Model

There are a number of intersections left after the definition of the veins, which still have the logging

code for “VEN”. SRK has reviewed these samples along with the geological mapping to identify a

number of small splays of the main structures. SRK identified a total of 103 structures which show

some degree of geological continuity to be able to define wireframes.

Vein samples were provided by CGM with a structure code but are (based on visual review) more

likely to be located in discontinuous splay veins in the HW or FW of the modelled vein. These are left

unassigned in the background veins coding and have been considered for the SPLAY domain. An

example of the splay is shown in Figure 14-4.

SRK considers the splays to have lower geological confidence to the main veins and further sampling

will be required to confirm potential prior to mining. The wireframe has been created using the vein

tools in Leapfrog®, the boundaries to limit the model are defined using polylines by SRK to crop the

veins as appropriate to the main structures. The extension of the wireframes in strike and dip were

limited to a maximum of 25 m beyond the data.

Group 1000 - Veins

Group 2000 – Disseminated Veins

Group 3000 – Splays



14.2.7 Porphyry “Pocket” Model

The modelling of the porphyry mineralization is hosted within the epithermal and transitional zones

with the mesothermal mineralization. The basis for the mineralization model remains consistent with

the methodology used in the previous model.

SRK generated a series of gold indicator grade shells using a 0.5 g/t Au CoG and an iso-value of 0.45

using 2 m composites. Indicator shells have been sub-divided into the major fault blocks, which have

been modelled independently. Due to the issue of the increase in the channel sampling database, with

many short channel samples located outside of the vein models, a restriction has been placed on the

samples used during the modelling process. The following criteria has been applied prior to modelling:

• Mineralization is flagged as being in the “Epithermal Zone” (not used in Group 5000)

• Vein_N = is null (not used in the vein model Group 1000)

• DISS = is null (not used in the disseminated model Group 2000)

• Splay = is null (not used in the splay model Group 3000)

• Length of hole is greater than 5 m (removes potential influence from narrow veins not

modelled)

• Lithology is not flagged as VEN, VNA, VNS or VOI

SRK has worked under the assumption that the porphyry mineralization will have a relationship to the

orientation of the epithermal vein systems which cross-cut the porphyry, any veinlets would likely be

structurally controlled. To apply this condition, SRK has used a structural trend within Leapfrog® when

generating the Indicator Grade Models. In order to generate the structural trends, SRK identified key

veins and orientations to avoid local variations in the overall trend. The structural trends are then based

on the orientations of these veins and structural orientations and are applied to new interpolants to

force anisotropy along these trends.

SRK ran a series of sensitivities and has monitored the levels of internal waste and the proportion of

samples above cut-off outside of the wireframes. Based on the study, SRK selected to use the

following criteria:

• Indicator Grade: 0.5 g/t

• Composite Length: 2 m with minimum length of 0.5 m (shorter lengths added to the previous)

• Structural Trend applied – with strongest anisotropy along the trend surfaces

• Spheroidal Interpolant, using a sill of 1 and nugget of 0.05

• Base range of 50 m

• Drift set to none

• ISO value of 0.45, with a surface resolution of 5 m

• Discarding any volumes less than 1,000 m3

A summary of the process and wireframe is shown in Figure 14-5.



Source: SRK, 2020

Figure 14-5: Development of Porphyry Pockets Wireframe Methodology

Plan Section



14.2.8 MDZ

In the 2019 PEA geological model, the MDZ was modelled using a broad outline and an approximate

CoG of 0.7 g/t Au. This includes a restricted internal higher-grade core, typically associated with areas

of increased veinlet density, as highlighted by Sillitoe (2019), using a 1.7 g/t Au indicator shell.

SRK has reviewed the PEA mineralization model for the MDZ at routine intervals during the infill drilling

campaign and noted the model performed well in predicting the location of the mineralization. SRK

concluded that the approaches taken in the PEA model were therefore reasonable and applied a

similar process in the PFS updated model. The three step process is to initially define the limits of the

Mesothermal mineralization based on geological logging (using the table “minznMN_LF”) exported

from the database. The key parameters used to define the limit of the mineralization styles are the

Mineralization Zone (MZ1MineralZone) and mineral assemblages (MZ1Mineral, MZ2Mineral).

Internal to the Mesothermal Mineralization domain, SRK has generated a medium and higher grade

indicator defined grade shell. To create the indicator model, the following assumptions have been

used:

• Only drillholes have been used to define the domain to remove potential errors or overstating

tonnage related to isolated short channel sampling

• All holes have been composited to 3 m, with samples lengths less than 0.5 m at the end of

holes appended to the previous sample

• CoGs of 0.7 and 1.7 g/t were used to define the outer limit of the mineralization, which

represents 13.1% of the database being assigned an indicator value of 1, with all other values

assigned an indicator of 0

• SRK has made use of a structural trend to define the orientations for the search ranges during

the indicator estimation process. The structural trends have been defined using input from

televiewer logged geological criteria for VEN (three records), VNA (24 records) and VNS (245

records) and by sectional analysis following key trends. (Figure 14-6);

• A spherical model has been used with a sill of 0.25 and a nugget of 0.05, using a base range

of 100 m for the interpolant.



Source: SRK, 2020

Figure 14-6: Development of MDZ model

VNS

VNA



To define the statistical parameters for the final indicator model, SRK completed a detailed study using

ISO (probability) values ranging from 0.40 to 0.60 at selected increments. SRK has monitored the

volumetric changes of the interpreted grade shells, the mean grade, the internal waste percentage

(samples<cut-off inside the wireframe) and the number of samples above cut-off outside the grade

shell.

Table 14-4: Summary of Leapfrog 0.7 g/t Indictor Grade Shell, ISO Value Sensitivity Study (MDZ Material)

Iso-Value 0.40 0.45 0.475 0.50 0.525 0.55 0.60

Total Samples Inside 75.2% 71.6% 69.7% 67.7% 65.7% 63.7% 59.6%

Internal Samples < cut-off 17% 14% 13% 11% 10% 9% 7%

Mean Grade 2.46 2.54 2.59 2.64 2.68 2.73 2.82

Volume (000 m3) 27,233 25,069 23,930 22,757 21,535 20,277 17,796

Volume % 100% 92% 88% 84% 79% 74% 65%

Source: SRK, 2020

SRK selected an ISO (probability) value of 0.475 for the final model in the 0.7 g/t indicator shell, with

a grid resolution of 5 m to define the wireframes. Upon review, SRK was concerned about potential

blow-outs within the 0.7 g/t indicator models at the edge of the model, where limited data exists. SRK

has utilized the use of control lines and minimum distances to restrict the high-grades from artificially

increasing the tonnage. These areas are also reviewed during the classification to limit the chances of

over estimation by applying a limit to the Inferred boundary. The same process has been used within

the high-grade 1.7 g/t domains to avoid overstating volume. The ISO value was increased to 0.50.

14.3 Domains

All geological surfaces were cut to the topography and the final geological model has been reviewed

by CGM and has been deemed acceptable by SRK for use in determining the MRE. Using the

wireframes, SRK has coded the drilling and block model information into five domains which are stored

in the block model under the field “GROUP”, for the main mineralization styles. A series of sub-codes

(VEIN_N, DISS, SPLAY and KZONE) have been used to distinguish between mineralization style and

individual mineralized structure. A list of the domains used is shown in Table 14-5 and in cross section

in Figure 14-7.



Table 14-5: Summary of Domain Coding Used in the 2017 Mineral Resource Estimate

Group No Subdomains Wireframe Domain Description

1000 95 Group1000.dxf Vein High grade sulfide veins

2000 73 Group2000.dxf Halo Disseminated veinlets and porphyry mineralization adjacent to the main vein structures

3000 103 Group3000.dxf Splays Splays of main structure within limited continuity

4000 7 Group4000.dxf- Grade Shell

Mineralized porphyry material (contained within veinlet), characterized by a mixed population of higher grade above an elevation of 850 m, low grade and barren material, marks the default unit for all material, split by fault domain

5000 3 Group5000.dxf Deeps

High grade core or feeder zone to the main mineralization. Located at depth within the porphyry system with limited veinlet mineralization, split into low, medium (0.7 g/t) and high (1.7 g/t).

Source: SRK, 2020

To validate the mineralization domains as defined, SRK completed a statistical review of the domained

data, which indicates independent populations for the five main domains (Figure 14-7 and Table 14-6).

Source: SRK, 2020

Figure 14-7: Box Plot Showing Raw Sample Statistics Based on Defined Geological Domains (Group)



Table 14-6: Summary Raw Sample Statistics Based on Defined Geological Domains (Group)

Column Domain Field

Domain Count Weight Min Max Mean Total Variance St.

Dev CV

AU 245066 315816 0 1767 0.878 277201 35.98 5.998 6.83

AU GROUP 0 134901 237744 0 1425 0.201 47818 7.41 2.721 13.53

AU GROUP 1000 38996 18751 0 605.1 5.892 110478 180.7 13.44 2.28

AU GROUP 2000 27498 13435 0 947.1 2.349 31562 63.51 7.969 3.39

AU GROUP 3000 7762 3712 0 1767 4.852 18012 785.2 28.02 5.78

AU GROUP 4000 12724 14897 0 891 1.788 26635 78.44 8.856 4.95

AU GROUP 5000 23185 27276 0 345.9 1.565 42697 21.41 4.627 2.96

Source: SRK, 2020

14.4 Assay Capping and Compositing

14.4.1 Outliers

High grade capping is typically undertaken where data is no longer considered to be part of the main

population. Useful discussions on the need for and the application of capping of high grades are found

in Leuangthong and Nowak (2015). Capping is an appropriate technique for dealing with high-grade

outlier values, given that appropriate analysis is undertaken to validate the results of the

implementation of capping. The following procedure is recommended for treating outliers during

resource estimation:

• Determine data validity. Is the data free of sampling, handling, measurement and transfer

errors?

• Review geology logs for samples with high-grade assays. Capping may not be necessary for

assays where the logs clearly explain the presence of high-grade

• Capping should not be considered for deleterious substances that have negative impacts on

project economics

• Decide if capping should be considered before or after compositing

• If high-grade assays unduly affect overall grade average, cap them

• Restrict influence of very high-grade assays during the estimation process if required

Upon review of the domained samples, SRK elected to apply the capping pre-compositing for the

current estimate. To define the appropriate capping levels, SRK completed analysis of the grade

distributions using log probability plots and raw and log histograms Figure 14-8 to Figure 14-12 (with

the selected caps highlighted in yellow) to distinguish the grades at which samples have significant

impacts on the local estimation and whose effect is considered extreme.

SRK reviewed and updated the capping/composite strategy at Marmato as part of the PFS Mineral

Resource update. The updated capping has been based on a disintegration analysis of the log-

probability plots for the veins, disseminated and splay domains (Group 1000 to 3000), and using

percentile analysis of the Au and Ag log-probability plots for the porphyry and MDZ.

In the 2020 estimate, the selected capping limits are as follows:

• Veins: A 60 g/t Au cap in the major veins with large numbers of samples, dropping to 20 g/t

Au in veins with lower sampling density. A standard cap for all veins of 450 g/t Ag has been

selected. Additionally to limit the impact of potential high-grades within the channel sampling



from over-influencing the vein estimates, SRK has used a reduced cap in the second and third

search ranges which is discussed in more detail in Section 14.8.3

• Splay: A 30 g/t Au and 150 g/t silver (Ag) cap for all splays has been used

• Porphyry: Capping has been based on the variable grade profiles shown within the different

fault blocks. The capping values for gold vary from 4.5 g/t to 11.5 g/t Au. In comparison the

silver values were shown to demonstrate more variability across the various fault blocks with

the highest grades reported in the NE (Echandia Licence). The capping for silver therefore

varies from 23 g/t to 205 g/t Ag

• MDZ: Capping has been applied by Indicator grade shells used in the model ranging from 5.5

g/t Au and 25 g/t Ag, to 17.5 g/t Au and 25 g/t Ag, and 40 g/t Au cap and 50 g/t Ag, for the low,

medium and higher-grade components

VEIN_N< 9000

Column Cap Capped Percentile Capped% Lost Total

(%) Lost CV

(%) Count Weight Min Max Mean Variance CV

AU 35728 16849 0 605.1 6.293 194.3 2.22

AU 554.53 1 100% 0% 0.02% 0.40% 35728 16849 0 554.5 6.291 192.6 2.21

AU 493.18 2 99.99% 0.01% 0.08% 1.30% 35728 16849 0 493.2 6.288 188.9 2.19

AU 432.28 3 99.99% 0.01% 0.20% 2.30% 35728 16849 0 432.3 6.283 184.9 2.16

AU 300.02 10 99.98% 0.03% 0.50% 5.80% 35728 16849 0 300 6.263 171.0 2.09

AU 60 379 99.16% 1.10% 6.90% 31% 35728 16849 0 60 5.859 80.9 1.54

Source: SRK, 2020 Selected cap shown in orange

Figure 14-8: Disintegration Analysis Au (g/t) – Veins, Group 1000 (Vein_N<9000)



VEIN_N> 9000

Column Cap Capped Percentile Capped% Lost

Total (%) Lost CV

(%) Count Weight Min Max Mean Variance CV

AU 3221 1783 0 132 2.499 48.72 2.79

AU 103.72 3 99.88% 0.10% 0.90% 4.60% 3221 1783 0 103.7 2.476 43.55 2.66

AU 94.83 4 99.85% 0.10% 1.40% 6.90% 3221 1783 0 94.83 2.464 41.04 2.6

AU 81.44 6 99.78% 0.20% 2.50% 11% 3221 1783 0 81.44 2.436 36.32 2.47

AU 69.13 8 99.73% 0.20% 3.80% 16% 3221 1783 0 69.13 2.404 31.72 2.34

AU 54.44 12 99.64% 0.40% 5.80% 23% 3221 1783 0 54.44 2.355 25.92 2.16

AU 46.76 14 99.62% 0.40% 6.90% 26% 3221 1783 0 46.76 2.327 23.17 2.07

AU 28.92 30 99.26% 0.90% 11% 35% 3221 1783 0 28.92 2.227 16.3 1.81

AU 20 70 98.31% 2.20% 15% 42% 3221 1783 0 20 2.123 11.81 1.62


Figure 14-9: Disintegration Analysis Au (g/t) – Veins, Group 1000 (Vein_N>9000)



DISS < 9000

Column Cap Capped Percentile Capped% Lost Total (%)

Lost CV (%)

Count Weight Min Max Mean Variance CV

AU 26778 13090 0 947.1 2.375 64.5 3.38

AU 453.97 1 99.99% 0% 0.30% 9.10% 26778 13090 0 454 2.367 52.93 3.07

AU 268.4 2 99.99% 0.01% 0.70% 13% 26778 13090 0 268.4 2.36 47.64 2.92

AU 20 414 98.67% 1.50% 14% 53% 26778 13090 0 20 2.039 10.69 1.6


Figure 14-10: Disintegration Analysis Au (g/t) – Disseminated Vein, Group 2000 (Vein_N<9000)



DISS > 9000


Lost CV (%)


AU 705 340.1 0 100.2 1.389 25.25 3.62

AU 53.52 1 99.85% 0.10% 4.70% 18% 705 340.1 0 53.52 1.323 15.27 2.95

AU 33.16 3 99.71% 0.40% 8.90% 30% 705 340.1 0 33.16 1.264 10.37 2.55

AU 26.32 4 99.62% 0.60% 11% 33% 705 340.1 0 26.32 1.239 8.91 2.41

AU 10 15 99.13% 2.10% 24% 53% 705 340.1 0 10 1.055 3.19 1.69


Figure 14-11: Disintegration Analysis Au (g/t) – Disseminated Vein, Group 2000 (Vein_N>9000)




Lost CV (%)


AU 7761 3712 0 1767 4.852 785.3 5.78

AU 927.29 2 99.98% 0.03% 2.20% 16% 7761 3712 0 927.3 4.747 527.2 4.84

AU 685.38 3 99.97% 0.04% 3.90% 26% 7761 3712 0 685.4 4.665 395.5 4.26

AU 405.62 4 99.95% 0.10% 6.70% 40% 7761 3712 0 405.6 4.525 243.8 3.45

AU 30 215 99.90% 2.80% 25% 71% 7761 3712 0 30 3.619 35.67 1.65


Figure 14-12: Disintegration Analysis Au (g/t) – Splays, (Group 3000)



Source: Source: SRK, 2020

Figure 14-13: Percentile Analysis Au (g/t) – Porphyry Domain, (Group 4000)



Table 14-7: Summary of Capping Sensitivity – MDZ Domain (Group 5000), Selected Capping Highlighted in Orange

Column _Filter Cap Capped Percentile Capped% Lost Total

(%) Lost CV

(%) Count Min Max Mean Variance CV

AU FBLOCK = 1 1377 0.001 306.4 1.914 73.36 4.48

AU FBLOCK = 1 20.911 15 98.90% 1.10% 15% 59% 1377 0.001 20.91 1.624 8.87 1.83

AU FBLOCK = 1 12.738 36 98% 2.60% 21% 65% 1377 0.001 12.74 1.509 5.44 1.55

AU FBLOCK = 1 10 45 97.40% 3.30% 25% 68% 1377 0.001 10 1.444 4.16 1.41

AU FBLOCK = 1 9 49 97.30% 3.60% 26% 70% 1377 0.001 9 1.417 3.72 1.36

AU FBLOCK = 1 7.895 56 96.60% 4.10% 28% 71% 1377 0.001 7.895 1.383 3.25 1.3

AU FBLOCK = 1 7.045 68 95.80% 4.90% 29% 72% 1377 0.001 7.045 1.35 2.85 1.25

AU FBLOCK = 1 6.085 83 95.10% 6% 32% 74% 1377 0.001 6.085 1.306 2.39 1.18

AU FBLOCK = 1 4.983 107 93.30% 7.80% 35% 76% 1377 0.001 4.983 1.241 1.84 1.09

AU FBLOCK = 1 4.387 123 92.10% 8.90% 37% 77% 1377 0.001 4.387 1.197 1.53 1.03

AU FBLOCK = 1 3.951 131 91.60% 9.50% 39% 78% 1377 0.001 3.951 1.161 1.32 0.99

AU FBLOCK = 1 - AU > 9

49 9.2 306.4 26.99 1992 1.65

AU FBLOCK = 1 - AU <= 9

1328 0.001 8.71 1.201 2.15 1.22

AU FBLOCK = 2 3095 0 61.1 1.459 8.81 2.03

AU FBLOCK = 2 38.421 7 99.90% 0.20% 1.10% 7.90% 3095 0 38.42 1.443 7.3 1.87

AU FBLOCK = 2 25 15 99.70% 0.50% 2.80% 16% 3095 0 25 1.418 5.84 1.7

AU FBLOCK = 2 15.7 40 99.30% 1.30% 5.80% 26% 3095 0 15.7 1.374 4.25 1.5

AU FBLOCK = 2 11.5 68 98.90% 2.20% 8.40% 33% 3095 0 11.5 1.337 3.34 1.37

AU FBLOCK = 2 9.6 78 98.70% 2.50% 9.90% 36% 3095 0 9.6 1.314 2.92 1.3

AU FBLOCK = 2 8.242 109 98.10% 3.50% 11% 39% 3095 0 8.242 1.292 2.59 1.25

AU FBLOCK = 2 6.92 135 97.30% 4.40% 13% 42% 3095 0 6.92 1.263 2.22 1.18

AU FBLOCK = 2 5.92 169 96.50% 5.50% 16% 45% 3095 0 5.92 1.232 1.91 1.12

AU FBLOCK = 2 4.98 221 95.30% 7.10% 18% 48% 3095 0 4.98 1.194 1.58 1.05

AU FBLOCK = 2 4 294 93.40% 9.50% 22% 52% 3095 0 4 1.138 1.22 0.97

AU FBLOCK = 2 - AU > 9.6

78 9.89 61.1 20.86 149.1 0.59

AU FBLOCK = 2 - AU <= 9.6

3017 0 9.6 1.206 2.05 1.19

AU FBLOCK = 3 226 0.04 53 1.564 18.91 2.78

AU FBLOCK = 3 10.79 4 99% 1.80% 18% 49% 226 0.04 10.79 1.287 3.36 1.42

AU FBLOCK = 3 8.84 7 98% 3.10% 20% 52% 226 0.04 8.837 1.257 2.85 1.34

AU FBLOCK = 3 7.65 9 97% 4% 22% 55% 226 0.04 7.649 1.222 2.36 1.26

AU FBLOCK = 3 5.06 14 96% 6.20% 28% 62% 226 0.04 5.055 1.132 1.43 1.06

AU FBLOCK = 3 4.64 16 95% 7.10% 29% 63% 226 0.04 4.635 1.111 1.28 1.02

AU FBLOCK = 3 3.60 21 94% 9.30% 33% 67% 226 0.04 3.601 1.051 0.91 0.91




(%) Lost CV


AU FBLOCK = 3 3.44 23 93% 10.20% 33% 68% 226 0.04 3.44 1.041 0.86 0.89

AU FBLOCK = 3 3.19 25 92% 11.10% 35% 69% 226 0.04 3.192 1.022 0.78 0.86

AU FBLOCK = 3 2.83 28 91% 12.40% 37% 71% 226 0.04 2.826 0.991 0.65 0.81

AU FBLOCK = 3 2.54 32 90% 14.20% 38% 72% 226 0.04 2.544 0.964 0.56 0.77


13 5.37 53 16.08 264.7 1.01

AU FBLOCK = 3 - AU <= 5.06

213 0.04 5.06 0.971 0.83 0.94

AU FBLOCK = 4 138 0.02 13.64 1.304 2.5 1.21

AU FBLOCK = 4 7.642 4 99% 2.90% 0.90% 2.90% 138 0.02 7.642 1.293 2.31 1.18

AU FBLOCK = 4 7.299 5 98% 3.60% 1.50% 4.50% 138 0.02 7.299 1.284 2.21 1.16

AU FBLOCK = 4 6.136 5 97% 3.60% 4.20% 11% 138 0.02 6.136 1.249 1.82 1.08

AU FBLOCK = 4 4.5 7 96% 5.10% 8.80% 20% 138 0.02 4.5 1.189 1.33 0.97

AU FBLOCK = 4 4.341 8 95% 5.80% 9.40% 21% 138 0.02 4.341 1.181 1.27 0.96

AU FBLOCK = 4 3.433 11 94% 8% 13% 27% 138 0.02 3.433 1.13 0.99 0.88

AU FBLOCK = 4 3.286 12 93% 8.70% 14% 28% 138 0.02 3.286 1.12 0.95 0.87

AU FBLOCK = 4 3.094 13 92% 9.40% 15% 30% 138 0.02 3.094 1.104 0.88 0.85

AU FBLOCK = 4 2.954 14 91% 10.10% 16% 31% 138 0.02 2.954 1.092 0.84 0.84

AU FBLOCK = 4 2.92 14 90% 10.10% 17% 31% 138 0.02 2.92 1.089 0.82 0.83


7 4.61 13.64 6.929 3.53 0.27

AU FBLOCK = 4 - AU <= 4.5

131 0.02 4.44 1.024 0.81 0.88

AU FBLOCK = 5 622 0 47.91 1.787 11.03 1.86

AU FBLOCK = 5 11.87 13 99% 2.10% 7.70% 30% 622 0 11.87 1.65 4.62 1.3

AU FBLOCK = 5 9.732 18 98% 2.90% 9.50% 33% 622 0 9.732 1.617 4.03 1.24

AU FBLOCK = 5 9 20 97.30% 3.20% 10% 35% 622 0 9 1.6 3.76 1.21

AU FBLOCK = 5 6.962 32 96% 5.10% 14% 40% 622 0 6.962 1.536 2.95 1.12

AU FBLOCK = 5 6.13 37 95% 5.90% 16% 42% 622 0 6.13 1.498 2.57 1.07

AU FBLOCK = 5 5.365 45 94% 7.20% 19% 45% 622 0 5.365 1.454 2.2 1.02

AU FBLOCK = 5 5.164 49 93% 7.90% 19% 46% 622 0 5.164 1.441 2.09 1

AU FBLOCK = 5 4.767 57 92% 9.20% 21% 48% 622 0 4.767 1.411 1.88 0.97

AU FBLOCK = 5 4.199 64 91% 10.30% 24% 50% 622 0 4.199 1.363 1.58 0.92

AU FBLOCK = 5 3.891 72 90% 11.60% 25% 52% 622 0 3.891 1.333 1.42 0.89


20 9.06 47.91 15.93 129.4 0.71


602 0 8.68 1.394 2.29 1.09

AU FBLOCK = 6 3651 0 344.4 1.712 64.29 4.68

AU FBLOCK = 6 54.754 7 99.90% 0.20% 8.50% 54% 3651 0 54.75 1.567 11.54 2.17




(%) Lost CV


AU FBLOCK = 6 30 23 99.60% 0.60% 11% 61% 3651 0 30 1.519 7.8 1.84

AU FBLOCK = 6 20 40 99.50% 1.10% 14% 65% 3651 0 20 1.477 5.86 1.64

AU FBLOCK = 6 15 56 99.20% 1.50% 16% 68% 3651 0 15 1.443 4.8 1.52

AU FBLOCK = 6 11.5 91 98.70% 2.50% 18% 70% 3651 0 11.5 1.407 3.95 1.41

AU FBLOCK = 6 10 113 98.30% 3.10% 19% 71% 3651 0 10 1.385 3.55 1.36

AU FBLOCK = 6 9 134 97.90% 3.70% 20% 72% 3651 0 9 1.366 3.24 1.32

AU FBLOCK = 6 8 159 97.40% 4.40% 22% 73% 3651 0 8 1.343 2.9 1.27

AU FBLOCK = 6 7 193 96.70% 5.30% 23% 74% 3651 0 7 1.314 2.55 1.21

AU FBLOCK = 6 5 288 92.50% 7.90% 28% 77% 3651 0 5 1.228 1.75 1.08


91 11.67 344.4 36.18 3849 1.72

AU FBLOCK = 6 - AU <= 11.5

3560 0 11.32 1.28 2.71 1.29

AU FBLOCK = 7 2394 0 116.3 1.744 12.44 2.02

AU FBLOCK = 7 18.73 41 99% 1.70% 5.20% 19% 2394 0 18.73 1.653 7.36 1.64

AU FBLOCK = 7 12.54 71 98.30% 3% 10% 29% 2394 0 12.54 1.57 5.06 1.43

AU FBLOCK = 7 10.00 84 97.90% 3.50% 13% 34% 2394 0 10 1.522 4.13 1.34

AU FBLOCK = 7 9.00 106 97.30% 4.40% 14% 36% 2394 0 9 1.497 3.73 1.29

AU FBLOCK = 7 8.00 122 96.80% 5.10% 16% 39% 2394 0 8 1.468 3.32 1.24

AU FBLOCK = 7 7.00 152 95.80% 6.30% 18% 41% 2394 0 7 1.432 2.89 1.19

AU FBLOCK = 7 6.00 179 94.90% 7.50% 21% 45% 2394 0 6 1.385 2.42 1.12

AU FBLOCK = 7 5.00 232 92% 9.70% 24% 48% 2394 0 5 1.325 1.92 1.05

AU FBLOCK = 7 4.17 290 91% 12.10% 28% 52% 2394 0 4.168 1.259 1.49 0.97

AU FBLOCK = 7 3.76 317 90% 13.20% 30% 54% 2394 0 3.758 1.219 1.27 0.93


106 9.09 116.3 18.04 105 0.57


2288 0 8.82 1.286 2.21 1.16




Source: SRK, 2020

Figure 14-14: Percentile Analysis Au (g/t) – MDZ Domain, (Group 5000)



Table 14-8: Summary of Capping Sensitivity – MDZ Domain (Group 5000), selected capping highlighted in orange


(%) Lost CV


AU

LG MDZ (KZONE 5000)

8157 0 65.54 0.547 3.1 3.22

AU 20 13 99.80% 0.20% 4.60% 27% 8157 0 20 0.522 1.51 2.35

AU 12 20 99.70% 0.20% 7.40% 37% 8157 0 12 0.507 1.04 2.02

AU 10 28 99.70% 0.30% 8.50% 40% 8157 0 10 0.501 0.92 1.92

AU 7 50 99.50% 0.60% 11% 46% 8157 0 7 0.487 0.7 1.72

AU 5.5 74 99.10% 0.90% 13% 50% 8157 0 5.5 0.476 0.58 1.60

AU 4 114 98.60% 1.40% 16% 56% 8157 0 4 0.458 0.43 1.43

AU 3 168 98% 2.10% 19% 60% 8157 0 3 0.441 0.33 1.30

AU 2.5 214 97.40% 2.60% 21% 62% 8157 0 2.5 0.43 0.27 1.22

AU 2 284 96.50% 3.50% 24% 65% 8157 0 2 0.415 0.22 1.13

AU 1.5 443 94.60% 5.40% 28% 68% 8157 0 1.5 0.393 0.16 1.02

AU 1 812 90% 10% 35% 73% 8157 0 1 0.356 0.1 0.88

AU INDZONE = 0 - AU > 5.5

74 5.52 65.54 13.36 139.9 0.88

AU INDZONE = 0 - AU <= 5.5

8083 0 5.49 0.43 0.35 1.37

AU

MG MDZ (KZONE 5001)

8413 0 345.9 1.548 26.43 3.32

AU 75.0 3 99.90% 0.04% 2.80% 35% 8413 0 75 1.505 10.46 2.15

AU 40.0 11 99.90% 0.10% 4.20% 42% 8413 0 40 1.483 8.07 1.92

AU 25.0 26 99.70% 0.30% 6.70% 50% 8413 0 25 1.445 5.79 1.66

AU 17.5 48 99.40% 0.60% 8.80% 55% 8413 0 17.5 1.411 4.48 1.50

AU 13.5 76 99.10% 0.90% 11% 58% 8413 0 13.5 1.381 3.65 1.38

AU 10.0 113 98.60% 1.30% 13% 62% 8413 0 10 1.342 2.83 1.25

AU 9.0 132 98.50% 1.60% 14% 63% 8413 0 9 1.327 2.59 1.21

AU 7.0 202 97.60% 2.40% 17% 66% 8413 0 7 1.289 2.09 1.12

AU 6.0 267 96.80% 3.20% 19% 68% 8413 0 6 1.262 1.8 1.06

AU 5.0 355 95.80% 4.20% 21% 70% 8413 0 5 1.225 1.49 1.00

AU 4.0 515 90% 6.10% 24% 72% 8413 0 4 1.174 1.16 0.92

AU INDZONE = 1 - AU > 17.5

48 18.4 345.9 41.46 2560 1.22

AU INDZONE = 1 - AU <= 17.5

8365 0 17.49 1.319 3.01 1.32

AU

HG MDZ (KZONE 5002)

6615 0.001 246.5 3.841 78.15 2.3

AU 100 10 99.80% 0.20% 2.70% 19% 6615 0.001 100 3.737 48.5 1.86

AU 62.5 20 99.70% 0.30% 4.80% 28% 6615 0.001 62.5 3.658 36.67 1.66

AU 45 38 99.40% 0.60% 6.70% 34% 6615 0.001 45 3.586 29.55 1.52

AU 40 48 99.30% 0.70% 7.50% 36% 6615 0.001 40 3.553 27.01 1.46

AU 35 56 99.20% 0.80% 8.50% 39% 6615 0.001 35 3.515 24.4 1.41




(%) Lost CV


AU 30 72 98.90% 1.10% 9.70% 42% 6615 0.001 30 3.468 21.72 1.34

AU 25 89 98.70% 1.30% 11% 45% 6615 0.001 25 3.408 18.82 1.27

AU 20 136 97.90% 2.10% 13% 48% 6615 0.001 20 3.326 15.71 1.19

AU 15 218 96.70% 3.30% 17% 53% 6615 0.001 15 3.193 11.97 1.08

AU 12.5 290 95.60% 4.40% 19% 56% 6615 0.001 12.5 3.097 9.96 1.02

AU 10 438 90% 6.60% 23% 59% 6615 0.001 10 2.962 7.76 0.94

AU INDZONE = 2 - AU > 40

48 40.08 246.5 79.73 2637 0.64

AU INDZONE = 2 - AU <= 40

6567 0.001 38.92 3.287 17.43 1.27

Source: SRK, 2020



Overall, these levels remain consistent with the capping levels used previously, with application of the

capping restrictions supported by improvements in the definition of the grade populations for the veins

from the channel sampling.

In general, SRK aims to limit the impact of the capping to less than 5% change in the mean value,

however in some cases with extreme outliers, the change in the mean exceeds 5%. The highly

positively skewed nature of the gold distributions and the very high values seen in the population result

in the significant changes in the mean values. A comparison of the raw versus capped values is shown

in Table 14-9.

Table 14-9: Comparison Raw vs Composite Statistics

Element KZONE Count Minimum

(g/t) Maximum

(g/t) Mean

(g/t) Variance

Standard Deviation

CoV

Samples

AU 1000 36,387 0 605.09 6.19 192.07 13.86 2.24

AU 1001 3,312 0 132.02 2.46 46.82 6.84 2.78

AU 2000 26,860 0 947.13 2.38 64.35 8.02 3.38

AU 2001 709 0 100.24 1.38 24.85 4.98 3.60

AU 3000 7,669 0 1,766.57 4.87 793.02 28.16 5.78

AU 4000 12,585 0 891.03 1.78 78.68 8.87 4.99

AU 5000 8,130 0 65.54 0.46 1.85 1.36 2.98

AU 5001 8,386 0 345.92 1.39 14.59 3.82 2.75

AU 5002 6,582 0.001 246.50 3.52 55.39 7.44 2.11

AG 1000 36,140 0 1,995.00 27.82 1,250.56 35.36 1.27

AG 1001 3,295 0 1,160.00 20.24 3,956.96 62.90 3.11

AG 2000 26,747 0 537.50 18.89 511.36 22.61 1.20

AG 2001 708 0 182.56 16.29 414.73 20.36 1.25

AG 3000 7,545 0 1,234.75 21.15 1,182.00 34.38 1.63

AG 4000 12,475 0 5,613.00 14.50 6,874.16 82.91 5.72

AG 5000 8,146 0 7,980.00 3.70 11,318.47 106.39 28.79

AG 5001 8,377 0 290.84 3.51 57.23 7.56 2.16

AG 5002 6,581 0 326.66 5.49 124.61 11.16 2.03

Element KZONE Count Minimum

(g/t) Maximum

(g/t) Mean

(g/t) Variance

Standard Deviation

CoV Difference

(%)

Comp

AU 1000 20,251 0.001 40.00 5.81 51.71 7.19 1.24 -6.1

AU 1001 1,240 0.001 15.00 2.00 8.02 2.83 1.42 -18.9

AU 2000 18,016 0.001 15.00 2.07 7.61 2.76 1.33 -13

AU 2001 347 0.001 7.00 0.96 1.81 1.34 1.40 -30.5

AU 3000 3,007 0.001 20.00 3.83 19.90 4.46 1.17 -21.4

AU 4000 7,708 0 11.50 1.39 2.19 1.48 1.07 -21.9

AU 5000 5,511 0 5.50 0.42 0.28 0.53 1.25 -8

AU 5001 5,114 0.001 17.50 1.32 1.69 1.30 0.98 -4.8

AU 5002 3,636 0.0267 40.00 3.34 12.19 3.49 1.05 -5.2

AG 1000 20,251 0.001 300.00 28.09 821.14 28.66 1.02 1

AG 1001 1,240 0.001 110.00 15.03 541.48 23.27 1.55 -25.7

AG 2000 18,016 0.001 300.00 18.90 426.21 20.64 1.09 0.1

AG 2001 347 0.001 80.00 15.42 321.34 17.93 1.16 -5.3

AG 3000 3,007 0.001 110.00 20.38 403.12 20.08 0.99 -3.6

AG 4000 7,654 0 205.00 10.78 387.59 19.69 1.83 -25.7

AG 5000 5,514 0 25.00 2.17 9.79 3.13 1.44 -41.4

AG 5001 5,113 0 30.00 3.27 17.64 4.20 1.29 -6.9

AG 5002 3,635 0 50.00 5.19 42.46 6.52 1.26 -5.5

Source: SRK, 2020



14.4.2 Compositing

Prior to the undertaking of grade interpolation, samples need to be composited to equal lengths for

constant sample volume, honoring sample support theories.

SRK has undertaken a sample composite analysis for gold in order to determine the optimal sample

composite length for grade interpolation. This investigated both changes in composite length and

minimum composite lengths for inclusion. The analysis compared the resultant mean grade against

the length weighted raw sample mean grades, and the percentage of samples excluded when applying

the minimum composite length. The results for the composite length analysis are summarized in Figure

14-15.

In addition to the analysis completed, SRK has reviewed the histograms of the raw sampling lengths

within the various domains (Figure 14-15). During the review SRK noted the following:

• A review of the sample lengths indicated that the mean sample length is approximately 0.5 m

veins, but 45% of the samples are between 0.5 to 1 m, with a further 5% between 1 and 2 m.

• The average length of the raw sampling in the porphyry and deep mineralization is 1 m, with

the majority of the samples ranging between 1 to 2 m.

Given the narrow nature of the veins, it is SRK’s view that increasing the sample lengths to 2 m is

preferred so only a single composite will exist across in the vein in narrow areas. SRK has elected to

also use the 2 m composite lengths for all other domains.



Source: SRK, 2020

Figure 14-15: Summary Histograms and Cumulative Frequency of Raw Sample Lengths per Domain



Summary descriptive statistics are provided in Table 14-10 for comparison of uncapped and capped

composited data for Au, and Ag. Overall, capping has reduced the highest yield outlier data while not

materially affecting the population of data. It is the opinion of the QP that capping has reduced the

effect of high-yield outlier values and should be considered during the estimation for selected domains,

to ensure there is not over influence of the high-grades beyond first search ranges. This restriction is

described in more detail in Section 14.8.3.



Table 14-10: Comparison Statistics

Field Group Wgt Field

Num Trace

Minimum Maximum Mean Variance Standard Deviation

Cov

Samples

AU 1000 Length 36387 0 605.09 6.19 192.07 13.86 2.24

AU 1001 Length 3312 0 132.02 2.46 46.82 6.84 2.78

AU 2000 Length 26860 0 947.13 2.38 64.35 8.02 3.38

AU 2001 Length 709 0 100.24 1.38 24.85 4.98 3.60

AU 3000 Length 7669 0 1,766.57 4.87 793.02 28.16 5.78

AU 4000 Length 12585 0 891.03 1.78 78.68 8.87 4.99

AU 5000 Length 8130 0 65.54 0.46 1.85 1.36 2.98

AU 5001 Length 8386 0 345.92 1.39 14.59 3.82 2.75

AU 5002 Length 6582 0.001 246.50 3.52 55.39 7.44 2.11

AG 1000 Length 36140 0 1,995.00 27.82 1,250.56 35.36 1.27

AG 1001 Length 3295 0 1,160.00 20.24 3,956.96 62.90 3.11

AG 2000 Length 26747 0 537.50 18.89 511.36 22.61 1.20

AG 2001 Length 708 0 182.56 16.29 414.73 20.36 1.25

AG 3000 Length 7545 0 1,234.75 21.15 1,182.00 34.38 1.63

AG 4000 Length 12475 0 5,613.00 14.50 6,874.16 82.91 5.72

AG 5000 Length 8146 0 7,980.00 3.70 11,318.4

7 106.39 28.79

AG 5001 Length 8377 0 290.84 3.51 57.23 7.56 2.16

AG 5002 Length 6581 0 326.66 5.49 124.61 11.16 2.03

Field Group Wgt Field

Num Trace

Minimum Maximum Mean Variance Standard Deviation

Cov

Composites

AU 1000 Length 20251 0.001 40.00 5.81 51.71 7.19 1.24

AU 1001 Length 1240 0.001 15.00 2.00 8.02 2.83 1.42

AU 2000 Length 18016 0.001 15.00 2.07 7.61 2.76 1.33

AU 2001 Length 347 0.001 7.00 0.96 1.81 1.34 1.40

AU 3000 Length 3007 0.001 20.00 3.83 19.90 4.46 1.17

AU 4000 Length 7708 0 11.50 1.39 2.19 1.48 1.07

AU 5000 Length 5511 0 5.50 0.42 0.28 0.53 1.25

AU 5001 Length 5114 0.001 17.50 1.32 1.69 1.30 0.98

AU 5002 Length 3636 0.0267 40.00 3.34 12.19 3.49 1.05

AG 1000 Length 20251 0.001 300.00 28.09 821.14 28.66 1.02

AG 1001 Length 1240 0.001 110.00 15.03 541.48 23.27 1.55

AG 2000 Length 18016 0.001 300.00 18.90 426.21 20.64 1.09

AG 2001 Length 347 0.001 80.00 15.42 321.34 17.93 1.16

AG 3000 Length 3007 0.001 110.00 20.38 403.12 20.08 0.99

AG 4000 Length 7654 0 205.00 10.78 387.59 19.69 1.83

AG 5000 Length 5514 0 25.00 2.17 9.79 3.13 1.44

AG 5001 Length 5113 0 30.00 3.27 17.64 4.20 1.29

AG 5002 Length 3635 0 50.00 5.19 42.46 6.52 1.26

Source: SRK, 2020

14.5 Density

Density measurements are made routinely by CGM geologists during core logging and sample

preparation. Each geologist tries to make one density measurement daily and to complete the

calculation the following procedure has been used:

• A piece of unbroken core is selected

• A 14 to 15 cm long piece of core from the interval of interest is cut

• As the core is cut, the geologist must ensure that the cut is perpendicular to the core axis and

does not result in the loss of any material along the cut line



• The length of the core is measured, and the diameter of the core is determined with a digital

caliper at 3 to 4 cm intervals and the average diameter is calculated

• The core is weighed on a digital balance and the density is calculated as follows:

o Pi * core diameter * core length = core volume

o Core weight/core volume = density

SRK completed a statistical review of the density measurements in the database provided up to

drillhole MT-IU-031. The database included a total of 3,370 samples with results ranging from 2.01 to

4.75 g/cm3. The majority of the samples have been taken within the various phases of porphyry

mineralization which have a range of 2.04 to 3.85 g/cm3. The highest measured density based on the

drillcore is taken from the vein material which returned an average of 3.39 g/cm3, but also contained

the highest variability. A summary of the measured density per major rock type is shown Table 14-11.

Table 14-11: Summary of Density Statistics by Rock Type and Selected Density

Rock Type Count Minimum Maximum Mean Std. Dev. Selected Density

BX 16 2.62 2.92 2.73 0.08 2.7

FLT 4 2.5 2.80 2.68 0.11

INT 46 2.42 2.80 2.71 0.07 2.71

METASED 99 2.235 3.44 2.82 0.17 2.8

P1 1840 2.1 3.42 2.67 0.10 2.67

P2 1007 2.04 3.85 2.68 0.11 2.68

P3 53 2.29 2.81 2.64 0.09 2.64

P4 200 2.18 3.18 2.70 0.09 2.7

P5 25 2.49 2.78 2.60 0.07 2.6

sap 1 2.92 2.92 2.92 0.00

sedt 12 2.63 2.93 2.77 0.09 2.77

ven 44 2.01 4.75 3.69 0.63 2.95

volc 23 2.4 2.91 2.79 0.10 2.79

Grand Total 3370 2.01 4.75 2.69 0.17

Source: SRK, 2020

SRK has elected to use a lower density for the veins than defined from a pure statistical basis. The

methodology behind the reduction, in conjunction with interviews with onsite personnel during the site

visit, are summarized from the PEA below:

In the 2017 Mineral Resource, the density has been grouped into three main units, which are

comprised of the porphyry, schist and vein rock types. SRK has compared these values to the 2017

model and discussed with the current mining operations team. The porphyry domain has been kept at

the same value as the 2017 block model which used an average density of 2.7 g/cm3. SRK still

considers this to be reasonable but comments that it could fluctuate between 2.65 to 2.70 g/cm3 and

should therefore be reviewed within the MDZ based on the results of the current on-going infill drilling

program being completed by the Company for preparation for a future PFS. If a lower density is

applied, it would impact the overall tonnage less than 2%.



The mine reported that it currently uses a lower density for the vein material based on the feed at the

plant closer to 2.7 to 2.9 g/cm3 range and that there could be risk of overstating the tonnage using a

3.4 g/cm3 as shown from the statistical analysis. The basis for the lower density is likely due to

favorable sampling of higher sulfides in the core which do not reflect the mined stope widths. Therefore,

to provide a more realistic assessment of the density, SRK completed a statistical review of the vein

samples using histograms and log-probability plots which still indicates a density of over 3 g/cm3 and

is reasonable based even on extreme capping of values greater than 3.25 g/cm3.

Cap Capped Percentile Capped% Count Min Max Mean

71 2.01 4.75 3.386

4.622 1 99% 1.40% 71 2.01 4.622 3.384

4.549 2 98% 2.80% 71 2.01 4.549 3.383

4.516 3 97% 4.20% 71 2.01 4.516 3.382

4.495 3 96% 4.20% 71 2.01 4.495 3.381

4.485 4 95% 5.60% 71 2.01 4.485 3.38

4.475 5 94% 7.00% 71 2.01 4.475 3.38

4.461 5 93% 7.00% 71 2.01 4.461 3.379

4.453 6 92% 8.50% 71 2.01 4.453 3.378

4.407 7 91% 9.90% 71 2.01 4.407 3.374

4.336 8 90% 11.30% 71 2.01 4.336 3.366

3.25 33 53.50% 46.50% 71 2.01 3.25 3.029

DEN_MAJOR = VEN - DENSITY > 3.25 33 3.4 4.75 4.018

DEN_MAJOR = VEN - DENSITY <= 3.25 38 2.01 3.25 2.837

Source: SRK, 2019

Figure 14-16: Log Probability Plot of Density Measurements Logged as Vein

To reflect density values more consistent with the mining, SRK has updated the statistical analysis

using a sub-set of the density database and the veins wireframes, generated from the geological model

using a halo of 0.5 m on either side of the veins and re-ran the analysis. The results of the analysis

returned a mean density of 2.97 g/cm3 (rounded to 2.95 g/cm3). In discussion with the CGM geological

team it was felt this was a more reasonable representation of the density for the vein domain. In

summary, the final density values used in the 2020 Mineral Resources are presented in Table 14-12.



Table 14-12: Density assigned per rocktype in 2020 Mineral Resources

ROCK ROCKTYPE COUNT DENSITY

1 P1 1840 2.670

2 P2 1007 2.680

3 P3 53 2.640

4 P4 200 2.700

5 P5 25 2.600

6 METASED 99 2.800

7 BRECCIA 16 2.700

8 INT 46 2.710

9 VOLC 23 2.790

VEN 44 2.950

Source: SRK, 2020

14.6 Variogram Analysis and Modeling

The composite drillhole database was imported into Datamine™ (Groups 1000 to 3000) and Snowden

Supervisor (Groups 4000 to 5000) software for the geostatistical analysis. Semi-variograms have been

completed for both gold and silver values.

SRK tested both omni-directional and directional variograms for all domains and both key elements.

When considering the directional variograms, SRK rotated the key search orientations to be down dip

and strike of the main mineralization directions. Due to the variable directions associated with the

veins, this returned poor results for domains (Groups 1000 to 4000) and therefore an omni-directional

variogram was preferred in the final model (Figure 14-17). SRK considers the use of omni-directional

variograms on the veins to be appropriate as the mineralization has been limited across the width of

the veins by a hard boundary, this results in orientations more like a plate than a ball when used during

the estimation process.

In Group 4000, the mineralization is a result of multiple orientations from the vein models causing

fractures/pockets to host the mineralization. No single primary orientation exists which results in a

stable variogram during the analysis. In order to define variograms of sufficient clarity to be modelled,

the data has been calculated using omni variograms and normal scored transformed variograms

(which removes the influence of some of the variability to a degree).



Source: SRK, 2020

Figure 14-17: Omni Directional Variograms Defined for Au and Ag for Domains Group = 1000 – 4000

SRK utilized directional variography within domain 5000, which has been rotated to a dip of 80° to the

south west (azimuth 210), the modelled experimental semi-variograms are shown in Figure 14-18 and

Figure 14-19.



Source: SRK 2020

Figure 14-18: Group 5000 Au Directional Semi-Variograms



Source: SRK, 2020

Figure 14-19: Group 5000 Ag Directional Semi-Variograms

Spatial continuity was calculated only for Au and Ag by domain as the primary economic variable of

interest. The modelled semi-variograms used in the estimation process are summarized in

Table 14-13.



Table 14-13: Summary of Variogram Parameters per Group

Group = 1000

Group = 2000 Disseminated

Group = 3000 Splays

Group = 4000 Porphyry

Group=5000 MDZ

VREFNUM 1001 1002 2001 2002 3001 3002 4001 4002 5001 5002 5003

VDESC Au Ag Au Ag Au Ag Au Ag Au

MDZ Ag

MDZ Au (LG)

MDZ

Rotation (Azi)

120 120 120

Rotation (dip)

-80 -80 -80

NUGGET 0.2 0.15 0.4 0.2 0.35 0.2 0.47 0.47 0.42 0.242 0.3

Range1 (X) 3 4.5 4 5 2 2 4 13 8 17 4

Range1 (Y) 3 4.5 4 5 2 2 4 13 5 10 5

Range1 (Z) 3 4.5 4 5 2 2 4 13 5 5 2

Sill 1 0.3 0.172 0.4 0.2 0.12 0.25 0.32 0.23 0.36 0.396 0.21

Range2 (X) 20 30.4 28 40 13 14.5 14 30 24 123 16

Range2 (Y) 20 30.4 28 40 13 14.5 14 30 12 32 33

Range2 (Z) 20 30.4 28 40 13 14.5 14 30 12 16 22

Sill 2 0.09 0.121 0.2 0.6 0.08 0.06 0.18 0.23 0.15 0.362 0.3

Range3 (X) 350 400 90 79.1 50 98 85 83

Range3 (Y) 350 400 90 79.1 50 98 60 115

Range3 (Z) 350 400 90 79.1 50 98 30 33

Sill 3 0.11 0.117 0.3 0.49 0.2 98 0.07 0.19

Source: SRK 2020 Note: All structures modelled using a spherical model



14.7 Block Model

SRK has produced a parent block model with block dimensions of 5 by 10 by 10 m (X, Y, Z), as a

function of the sample spacing within the veins. The block size represents a change in the 5 by 5

by 5 m block used in the PEA. Test work undertaken using the variograms for the MDZ looking at

the slope of regression and kriging efficiency suggested increasing the block size to 5 by 10 by

10 m block size was appropriate, the decision was taken to maintain the 5 m block size across

strike to reflect potential for selectivity for mining.

SRK acknowledges a larger block size could be more appropriate in areas of wider spaced drilling, but

the decision was taken to use a uniform block size across the deposit in conjunction with the mining

team. Sub-blocking has been allowed to a resolution 0.5 m along strike, 0.5 m across strike and 1 m

in the vertical direction to provide an appropriate geometric representation.

Given the orientation of the orebody striking to the NW, the decision was made to rotate the database

(for block model grade interpolation) from UTM coordinates through 55° into a N-S local grid

orientation, to enable an improved representation of grade continuity along strike. To rotate the

interpretation the “CDTRAN” Datamine™ command has been utilized. The details of the block model

origin, rotation and local dimensions are shown in Table 14-4.

Table 14-14: Block Model Prototype (DatamineTM format)

Dimension Origin (UTM)

Origin (Local)

Block Size

Number of Blocks

Rotation Min Sub-

blocking (m)

X 1,163,465 0 5 430 - 0.5

Y 1,096,426 0 10 175 - 0.5

Z 0 0 10 210 -55 1

Source: SRK, 2020

Using the wireframes created and described in Section 14.2 several codes have been developed to

describe each of the major geological properties of the rock types. Table 14-15 summarizes geological

fields created within the geological model and the codes used.



Table 14-15: Summary of Key Fields in Block Model

Field Name Description

SVOL Search Volume reference (range from 1 to 3)

KVAR Kriging Variance

MDIST Transformed distance to samples

NSAM Number of samples used to estimate the block

AU Final Gold Estimate using for Reporting

AG Final Silver Estimate using for Reporting

AUOK Gold Estimate using OK

AGOK Silver Estimate using OK

AUIDW Gold Estimate using IDW (Power 2)

AGIDW Silver Estimate using IDW (Power 2)

AUNN Gold NN Methodology

AGNN Silver NN Methodology

RESCAT Classification

GROUP Mineralized structures grouped by domain

VEIN_N Vein coding for individual mineralized structure GROUP1000 coding

DISS Vein coding for individual mineralized structure GROUP2000 coding

SPLAY Vein coding for individual mineralized structure GROUP3000 coding

DENSITY Density of the rock

DEPLETE Mined out areas

ROCK Coding for Major Rock type

LICENCE Mining Licence

MINZONE Mineralization Zone (Epithermal vs Mesothermal)

INDZONE Indicator domain used in MDZ domain

TRDIP Search Orientation information (True Dip)

TRDIPDIR Search Orientation information (True Dip Direction)

Source: SRK, 2020

14.8 Estimation Methodology

14.8.1 Theoretical Analysis

A Kriging Neighborhood Analysis (KNA) exercise has been completed for gold, in order to optimize

the parameters used in the estimation and kriging calculations. Initial grade estimation was undertaken

in Snowden Supervisor using the KNA utility. To complete the exercise, a number of scenarios were

tested using various estimation and kriging parameters. Different input parameters have been changed

and the differences in the slope of regression, kriging efficiency and impact on the negative kriging

weights. The following parameters were adjusted during the analysis:

• Block Size

• Minimum and maximum number of samples

• Search ellipse sizes

• Discretization

• Orientation of search ellipse

The results of the KNA analysis for the MDZ are shown in Figure 14-20. SRK selected to use a 5 by

10 by 10 m block based on the mean slope of regression (0.86) showing a minor improvement on the

10 by 10 by 10 m block (0.85), while still reflecting the potential across strike variability.



Source: SRK, 2020

Figure 14-20: Group 5000 (MDZ), KNA Analysis Using Snowden Supervisor

Within the MDZ, SRK selected a theoretical minimum of five composites and maximum of 20

composites using the example shown. The maximum of 20 composites was selected as the number

of negative weights increased significantly at higher values, for relatively low increases in mean of the

slope of regression.



Once the theorical parameters were defined, SRK completed a number of initial estimates in order to

assess if suitable grade estimates were occurring using the selected parameters. Blocks estimates

were completed in Datamine™ and the following data fields were analyzed to assess quality: kriging

variance; number of samples; and proportion of blocks estimated in each search volume. Additional

fields monitored included the resultant grade in comparison with the sample data.

The Datamine™ block models have been estimated using a nested search format with three searches

used in all estimates. The first search represents an optimized search distance (selected from a kriging

sensitivity analysis), ensuring (in general) that block estimates use a minimum of two drillholes, while

the second and third search volumes use expansion factors that produce more smoothed block

estimates, appropriate to the limit of geological continuity. The third expansion volume was sufficient

to ensure that all appropriate blocks (in areas with reasonable geological confidence) were assigned

grade values. These blocks were generally classified with lower confidence, and in areas of uncertainty

trimmed out of the mineral resources due to a lack of strike or dip continuity.

The optimum parameters selected allowed an appropriate proportion of block estimates in the initial

search volumes, whilst achieving a reduction in variance and a relative increase in slope of regression

(in SVOL 1 and 2) without excessive smoothing. A summary of the analysis is shown in Table 14-16.

Based on the outcome of the validation process, SRK has selected to use either OK algorithm, or ID2

estimates to compile the final grade estimates. Typically zones with larger sample populations are

supported by OK, while zones with less data are supported by ID2 (splays).



Table 14-16: Summary of Datamine Estimates by Search Volume and Validation by Estimation Type (OK, ID, NN)

Group SVOL Tonnes (t) %

Volume AUOK

(g/t) AUID (g/t)

AUNN (g/t)

AGOK (g/t)

AGID (g/t)

AGNN (g/t)

KVAR Transform DIST Number Samples

1000

1 1,571,645 13.0 5.95 5.89 5.66 28.09 27.95 26.97 0.17 0.08 6.5

2 6,106,968 50.6 3.48 3.38 3.20 18.66 18.25 17.42 0.20 0.07 6.4

3 4,399,720 36.4 2.58 2.49 2.45 15.72 15.55 15.24 0.24 0.03 5.6

2000

1 804,485 70.6 1.99 1.98 1.96 18.76 18.81 18.75 0.29 0.06 7.0

2 299,154 26.3 1.79 1.77 1.76 15.38 15.29 14.89 0.37 0.06 7.1

3 35,147 3.1 1.46 1.38 1.42 14.62 14.24 14.16 0.55 0.04 4.1

3000

1 338,954 42.4 3.89 3.89 3.78 20.12 20.15 19.70 0.21 0.05 5.8

2 336,578 42.1 3.09 3.03 2.83 15.24 15.16 14.81 0.33 0.05 5.8

3 124,717 15.6 3.39 3.35 3.49 21.21 21.28 22.42 0.50 0.05 2.6

4000

1 2,809,547 24.0 1.44 1.43 1.42 10.05 9.93 9.76 0.38 0.49 6.7

2 5,846,330 50.0 1.49 1.48 1.42 11.93 11.79 11.84 0.41 0.77 8.1

3 3,032,303 25.9 1.46 1.48 1.49 10.00 9.97 9.90 0.46 0.72 6.6

5000

1 12,087,681 16.5 1.28 1.28 1.27 3.25 3.22 3.21 0.35 0.49 10.4

2 43,481,638 59.3 1.03 1.03 1.00 2.05 2.05 2.09 0.43 1.05 11.8

3 17,698,965 24.2 1.15 1.16 1.13 1.77 1.77 1.86 0.58 1.56 7.0

Source: SRK, 2020



14.8.2 Dynamic Anisotropy

SRK concluded that within all domains there is the presence of multiple strike and dip directions which

could impact the search orientation and result in a bias could be introduced if a single search

orientation was selected per zone. To ensure the block model reflects the nature of the vein

mineralization as accurately as possible, SRK therefore utilized the wireframe interpretation to aid in

determining the search orientations used during the kriging equations on a block by block basis. This

has been done using the dynamic anisotropy function in Datamine™.

The dynamic search orientations have been generated from a series of planes (wireframes) within

DatamineTM, which are used to initially define the dip and dip direction of each planes triangles (used

to create the plane) on a 10 m resolution. These points are then estimated into the block model using

the “Estima” Process within DatamineTM, which is designed to recognize these criteria. In areas of

limited drilling or information SRK has defined a default orientation for the estimation process of dip

and dip direction, for example as shown for the MDZ high-grade wireframes in Figure 14-21.

Source: SRK, 2020

Figure 14-21: Example of Default Search Orientations Used Within the MDZ High-Grade Domain

N

100 m



14.8.3 Threshold Capping

The grade distributions within each of the defined domains can be described as positively skewed log-

normal distributions. When undertaking the capping analysis, SRK noted that high-grade values are

important to the deposit and that excessive capping will likely result in a reduction of metal that could

impact project economics. Counter to this point is that the use of higher capping could over-estimate

the metal for any given areas, which may be a concern especially within the vein domains, which are

more reliant on the channel sampling databases. To tackle this issue, SRK has introduced the use of

threshold capping (or sliding capping), which applies variable capping levels based on the distance

from the drilling. In simple terms, higher capping values are used in the first search volumes, and more

conservative capping values in the second and third volumes. A summary of the application of the

threshold capping strategy is shown in Table 14-17.

Table 14-17: Summary of Domains with Top Capping and Sliding Thresholds for Wider Search Volumes

Domain Capping Level Au (g/t) Capping Level Ag (g/t)

Vein Code Search 1 Search 2 & 3 Search 1 Search 2 & 3

VEINS (Group 1000) >9000 20 15 150 110

<9000 60 40 450 300

DISS (Group 2000) >9000 10 7 90 80

<9000 20 15 334 300

SPLAYS (Group 3000) 30 20 150 110

Porphyry

4001 9.0 No Change 205 No Change








MDZ 5001 17.5 No Change 30 No Change

(Group 5000) 5002 40 No Change 50 No Change

Source: SRK, 2020

14.8.4 Final Parameters

The final kriging parameters selected for gold and silver are presented in Table 14-18. A discretization

grid of 3 by 3 by 3 m has been used within each parent block during the estimation within the Veins,

Disseminated and Splays domains (Group 1000 to 3000). A discretization grid of 4 by 5 by 4 m has

been used within each parent block during the estimation within the porphyry and MDZ domains

(Group 4000 to 5000). The discretization grid ensures that single blocks near the edge of each

estimation zone are assigned a grade that is characteristic of the modelled domain and not just those

values at the block midpoint.



Using the dynamic anisotropy to control the orientation of the searches and based on visual review

of the porphyry pockets, SRK has applied anisotropy to the search ranges within Group 4000. The

ability to vary the search ellipse accounts for the complex geology which could not be accounted for

in the variography for this domain. SRK considers this approach to best reflect the underlying

geological conditions.



Table 14-18: Summary of Estimation Search Parameters Used in Estimation

Group Group 1000 (Veins) Group 1000 (Disseminated) Group 3000 (Splays) Group 4000 (Porphyry) Group 5000 (MDZ)

Sub-Group 1000 1000 2000 2000 3000 3000 4000 4000 5000 5001 5002 5000-5002

SDESC Search

Volume (Au) Search

Volume (Ag) Search

Volume (Au) Search

Volume (Ag) Search

Volume (Au) Search

Volume (Ag) Search

Volume (Au) Search

Volume (Ag)

Search Volume Au

(LG)

Search Volume Au

(MG)

Search Volume Au

(HG)

Search Volume (Ag)

Rotation (Z) 0 0 0 0 0 0 15 15 120 120 120 120

Rotation (Dip) 0 0 0 0 0 0 -80 -80 -85 -85 -85 -85

X Range (m) 25 25 25 25 25 25 25 35 40 45 45 55

Y Range (m) 25 25 25 25 25 25 25 35 50 35 35 55

Z Range (m) 25 25 25 25 25 25 15 15 20 10 10 15

Minimum 4 4 4 4 4 4 4 4 5 5 5 5

Maximum 12 12 12 12 12 12 12 12 24 20 20 20

Dynamic Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Search Volume Factor

3 3 3 3 3 3 2 2 2 2.2 2.2 1.5

X Range (m) 75 75 75 75 75 75 50 70 80 99 99 82.5

Y Range (m) 75 75 75 75 75 75 50 70 100 77 77 82.5

Z Range (m) 75 75 75 75 75 75 30 30 40 22 22 22.5

Minimum 4 4 4 4 4 4 4 4 5 5 5 5

Maximum 12 12 12 12 12 12 12 8 24 20 20 20

Dynamic Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes

SVOLFAC3 6 6 6 6 6 6 4 4 3 3.3 3.3 3

X Range (m) 150 150 150 150 150 150 100 140 120 148.5 148.5 165

Y Range (m) 150 150 150 150 150 150 100 140 150 115.5 115.5 165

Z Range (m) 150 150 150 150 150 150 60 60 60 33 33 45

Minimum 1 1 1 1 1 1 1 1 3 3 3 3

Maximum 8 8 8 8 8 8 8 8 8 8 8 8

Dynamic Yes Yes Yes Yes Yes Yes Yes Yes No No No No

OCTMETH Yes No Yes No Yes No No No No No No No

MINOCT 1 1 1 1 1 1 2 2 2 2 2 2

MINPEROC 1 1 1 1 1 1 1 1 1 1 1 1

MAXPEROC 2 4 2 4 2 4 4 4 4 4 4 4

MAXKEY 2 2 2 2 2 2 3 3 4 4 4 4

SANGL1_F TRDIPDIR TRDIPDIR TRDIPDIR TRDIPDIR TRDIPDIR TRDIPDIR TRDIPDIR TRDIPDIR TRDIPDIR TRDIPDIR TRDIPDIR

SANGL2_F TRDIP TRDIP TRDIP TRDIP TRDIP TRDIP TRDIP TRDIP TRDIP TRDIP TRDIP

Estimation Method

OK OK OK OK ID2 ID2 OK OK OK OK OK OK

Source: SRK, 2020



14.9 Model Validation

SRK has undertaken a thorough validation of the resultant interpolated model in order to confirm the

estimation parameters, to check that the model represents the input data on both local and global

scales and to check that the estimate is not biased. SRK has undertaken this using a number of

different validation techniques:

• Visual inspection of block grades in comparison with drill hole data

• Inspection of block grades in plan and section and comparison with drillhole grades

• Statistical validation of declustered means versus block estimates

• Comparison of estimates using different estimation methods (NN, IDW, OK)

• Swath plots of the mean block and sample grades

The geology model, geostatistical analysis, variography, selection of resource estimation parameters,

and construction of the block model work were completed by SRK. The current drilling information is

sufficiently reliable to interpret with confidence the boundaries of the various veins, and the assaying

data is sufficiently reliable to support Mineral Resource estimation

14.9.1 Visual Comparison

Visual validation provides a local validation of the interpolated block model on a local block scale,

using visual assessments and validation plots of sample grades versus estimated block grades. A

thorough visual inspection of cross-sections, long-sections and bench/level plans, comparing the

sample grades with the block grades has been undertaken, which demonstrates good comparison

between local block estimates and nearby samples, without excessive smoothing in the block model.

Exploration NW



Santa Ines



Mezillos

Source: SRK, 2020

Figure 14-22: Visual Validation of Selected Veins at Marmato



Disseminated Exploration NW

Disseminated Santa Ines Source: SRK, 2020

Figure 14-23: Visual Validation of Selected Disseminated Veins at Marmato



Source: SRK, 2020

Figure 14-24: Section and Level Plan Example of Visual Validation of MDZ (Group 1000)



14.9.2 Comparative Statistics

SRK has completed a statistical validation of the block estimates (NN, OK and ID2) versus the mean

of the composite samples per zone. In general, the results indicate a reasonable comparison

(Table 14-19) between the sample mean grades (declustered) and the block estimates.

SRK notes that the comparison between the mean grades and the raw samples often exceeds the

desired levels of error, but this is often a function of the clustering within the dataset from either channel

sampling, or the fan drilling. SRK has therefore focused on using a comparison to the declustered

mean grades, which have been determined within Snowden Supervisor, by testing multiple

declustering block sizes ranging between 0 to 50 m and selecting points where the mean stabilizes.

In the case of the higher grade domains such as the veins or the high-grades MDZ, this typically

represents a reduction in the average grades, while the lower grade domains have typically been under

sampled and therefore there is a slight increase in the average grades within these zones.

For the comparison, SRK has compared the grouped statistics for the veins, splays, and porphyry

material, but has broken out the MDZ into the three main sub-domains. The zones show satisfactory

correlations between the composite and block estimates, with the highest errors noted within the low-

grade MDZ domain, but a comparison between the OK vs NN returned acceptable results. This is

minor domain in terms of material above cut-off and therefore is not considered to be a material issue.

SRK notes the comparison of the NN and OK return good correlations, which adds support to the

estimates.



Table 14-19: Summary of Statistical Validation of Raw, Declustered, OK, ID2 and NN Block Estimates

Group = 1000 (Veins)

Element Statistic Sample

Data Declustered Sample Data

Block Data 1

OK

OK Vs Sample

% Diff

OK Vs Declustered

%Diff

Block Data 2

ID2

ID2 Vs Sample

% Diff

ID2 Vs Declustered

% Diff

Block Data 3

NN

NN Vs Sample

%Diff

NN Vs Declustered

% Diff

OK vs

NN

AU

Points 22018 22018

Mean 7.01 3.72 3.47 -50% -7% 3.38 -52% -9% 3.25 -54% -13% 6.9%

Std Dev 9.58 6.54 3.27 3.62 5.45

Variance 91.76 42.83 10.66 13.11 29.65

CV 1.37 1.76 0.94 1.07 1.68

Maximum 60.00 60.00 48.48 48.29 60.00

75% 8.40 8.40 4.77 4.58 3.83

50% 3.80 3.80 2.49 2.30 1.54

25% 1.60 1.60 1.40 1.21 0.55

AG

Points 22018 22018

Mean 32 22 19 -42% -13% 19 -43% -15% 18 -45% -18% 5.3%

Std Dev 34.0 33.2 17.0 19.7 27.9

Variance 1155 1103 288 390 780

CV 1.05 1.53 0.90 1.07 1.56

Maximum 450.0 450.0 226.7 288.6 450.0

75% 42.6 42.6 25.5 24.9 22.4

50% 23.6 23.6 14.2 13.0 8.8

25% 10.8 10.8 7.8 6.6 3.0

Group = 2000 (Disseminated)



Block Data 1 OK

OK Vs Sample

% Diff

OK Vs Declustered

% Diff

Block Data 2 ID2

ID2 Vs Sample

% Diff

ID2 Vs Declustered

% Diff

Block Data 3 NN

NN Vs Sample

% Diff

NN Vs Declustered

% Diff

OK vs

NN

AU

Points 18402 18402

Mean 2.02 1.84 1.96 -3% 6% 1.95 -3% 6% 1.95 -3% 6% 0.1%

Std Dev 3.05 2.81 1.43 1.59 2.87

Variance 9.31 7.88 2.06 2.53 8.25

CV 1.51 1.52 0.73 0.82 1.47

Maximum 20.00 20.00 12.48 14.82 20.00

75% 2.18 2.18 2.61 2.56 2.15

50% 1.00 1.00 1.62 1.54 1.03

25% 0.40 0.40 0.94 0.88 0.40

AG

Points 18402 18402

Mean 19 17 18 -7% 3% 18 -7% 3% 18 -7% 2% 0.6%

Std Dev 21.3 19.3 13.7 14.6 20.1

Variance 452 374 188 214 406

CV 1.11 1.12 0.77 0.82 1.14

Maximum 334.0 334.0 125.7 149.1 334.0

75% 24.1 24.1 21.7 21.4 21.5

50% 12.6 12.6 14.0 13.7 12.0

25% 6.0 6.0 9.2 8.8 5.6



Group = 3000 (Splays)


Data

Declustered Sample

Data

Block Data 1

OK

OK Vs Sample

% Diff

OK Vs Declustered

% Diff

Block Data 2

ID2

ID2 Vs Sample

% Diff

ID2 Vs Declustered

% Diff

Block Data 3

NN

NN Vs Sample

% Diff

NN Vs Declustered

% Diff

OK vs

NN

AU

Points 2980 2980

Mean 4.33 3.77 3.47 -20% -8% 3.44 -20% -9% 3.33 -23% -12% 4.1%

Std Dev 5.78 5.63 3.04 3.27 4.77

Variance 33.38 31.72 9.27 10.68 22.78

CV 1.34 1.49 0.88 0.95 1.43

Maximum 30.00 30.00 27.05 27.87 30.00

75% 5.20 5.20 5.11 5.06 4.68

50% 2.34 2.34 2.90 2.76 1.79

25% 0.91 0.91 1.54 1.34 0.45

AG

Points 2980 2980 124391 124391 124391

Mean 22 18 18 -18% 2% 18 -18% 2% 18 -19% 1% 0.8%

Std Dev 23.8 22.2 16.0 16.8 21.9

Variance 566 492 256 282 479

CV 1.06 1.23 0.88 0.92 1.21

Maximum 150.0 150.0 146.4 146.8 150.0

75% 28.3 28.3 24.8 24.8 25.6

50% 15.6 15.6 13.7 13.6 10.8

25% 7.9 7.9 7.2 6.7 2.5

Group = 4000 (Porphyry Pockets)



Block Data1

OK

OK Vs Sample

% Diff

OK Vs Declustered

% Diff

Block Data 2 ID2

ID2 Vs Sample

% Diff

ID2 Vs Declustered

% Diff

Block Data 3 NN

NN Vs Sample

% Diff

NN Vs Declustere

d % Diff

OK vs

NN

AU

Points 7738 7738

Mean 1.39 1.40 1.47 6% 5% 1.47 5% 5% 1.43 3% 3% 2.4%

Std Dev 1.50 1.53 0.71 0.82 1.52

Variance 2.26 2.33 0.51 0.67 2.31

CV 1.08 1.09 0.49 0.56 1.06

Maximum 11.50 11.50 7.05 7.80 11.50

75% 1.69 1.67 1.81 1.81 1.73

50% 0.91 0.91 1.31 1.27 0.96

25% 0.54 0.56 0.97 0.91 0.58

AG

Points 7738 7738

Mean 1 1 1 6% 5% 1 5% 5% 1 3% 3% 2.4%

Std Dev 1.5 1.5 0.7 0.8 1.5

Variance 2 2 1 1 2

CV 1.08 1.09 0.49 0.56 1.06

Maximum 11.5 11.5 7.0 7.8 11.5

75% 1.7 1.7 1.8 1.8 1.7

50% 0.9 0.9 1.3 1.3 1.0

25% 0.5 0.6 1.0 0.9 0.6



Group = 5000 (INDZONE = 0)


Data

Declustered Sample

Data

Block Data 1

OK

OK Vs Sample

% Diff

OK Vs Declustered

% Diff

Block Data 2

ID2

ID2 Vs Sample

% Diff

ID2 Vs Declustered

% Diff

Block Data 3

NN

NN Vs Sample

% Diff

NN Vs Declustered

% Diff OK vs NN

AU

Points 5518 5518

Mean 0.43 0.44 0.48 12% 10% 0.48 13% 11% 0.53 23% 21% -9.0%

Std Dev 0.55 0.56 0.25 0.27 0.65

Variance 0.30 0.31 0.06 0.07 0.42

CV 1.28 1.27 0.52 0.56 1.22

Maximum 5.50 5.50 2.60 3.24 5.50

75% 0.52 0.53 0.61 0.61 0.60

50% 0.28 0.29 0.43 0.42 0.31

25% 0.13 0.13 0.30 0.30 0.16

AG

Points 5518 5518

Mean 2 2 2 -32% -14% 2 -32% -14% 2 -30% -12% -2.1%

Std Dev 3.3 2.5 1.0 1.0 1.8

Variance 11 6 1 1 3

CV 1.47 1.41 0.67 0.69 1.15

Maximum 25.0 25.0 21.4 23.5 25.0

75% 2.2 1.9 1.8 1.8 1.8

50% 1.3 1.2 1.3 1.3 1.1

25% 0.9 0.7 0.9 0.9 0.7



Data

Declustered Sample

Data

Block Data 1

OK

OK Vs Sample

% Diff

OK Vs Declustered

% Diff

Block Data 2

ID2

ID2 Vs Sample

% Diff

ID2 Vs Declustered

% Diff

Block Data 3

NN

NN Vs Sample

% Diff

NN Vs Declustered

% Diff OK vs NN

AU

Points 5113 5113

Mean 1.32 1.26 1.27 -4% 1% 1.29 -2% 2% 1.24 -6% -2% 2.4%

Std Dev 1.33 1.23 0.47 0.53 1.24

Variance 1.76 1.52 0.22 0.28 1.55

CV 1.00 0.97 0.37 0.41 1.00

Maximum 17.50 17.50 5.03 6.35 17.50

75% 1.57 1.51 1.49 1.52 1.47

50% 0.97 0.95 1.18 1.18 0.88

25% 0.59 0.57 0.96 0.93 0.57

AG

Points 5113 5113

Mean 3 3 2 -39% -21% 2 -39% -21% 2 -38% -20% -1.1%

Std Dev 4.5 3.3 1.5 1.6 2.6

Variance 20 11 2 3 7

CV 1.31 1.25 0.74 0.77 1.24

Maximum 30.0 30.0 25.1 26.2 30.0

75% 3.4 2.7 2.5 2.5 2.3

50% 1.8 1.6 1.7 1.7 1.4

25% 1.2 1.1 1.2 1.2 1.0






Block Data 1

OK

OK Vs Sample

% Diff

OK Vs Declustered

% Diff

Block Data 2 ID2

ID2 Vs Sample

% Diff

ID2 Vs Declustered

% Diff

Block Data 3 NN

NN Vs Sample

% Diff

NN Vs Declustered

% Diff

OK vs

NN

AU

Points 3635 3635

Mean 3.35 3.27 3.39 1% 4% 3.40 1% 4% 3.14 -6% -4% 7.8%

Std Dev 3.61 3.60 1.98 2.13 3.54

Variance 13.03 12.93 3.94 4.54 12.50

CV 1.08 1.10 0.59 0.63 1.13

Maximum 40.00 40.00 31.22 31.72 40.00

75% 4.02 3.87 3.70 3.75 3.63

50% 2.37 2.31 2.89 2.83 2.21

25% 1.34 1.33 2.31 2.23 1.24

AG

Points 3635 3635

Mean 5 4 4 -31% -10% 4 -31% -11% 4 -32% -12% 1.8%

Std Dev 6.9 5.4 2.7 2.9 4.8

Variance 48 29 7 9 23

CV 1.28 1.29 0.73 0.79 1.29

Maximum 50.0 50.0 27.9 31.4 50.0

75% 6.1 4.5 4.8 4.8 4.0

50% 2.9 2.4 2.9 2.7 2.2

25% 1.6 1.4 1.9 1.8 1.3

Source: SRK, 2020



14.9.3 Swath Plots

As part of the validation process, the input composite samples were compared to the block model

grades within a series of coordinates. The results of this were then displayed on graphs to check for

visual discrepancies between grades. Figure 14-25 through Figure 14-31 show the results for the gold

grades for the Marmato vein domain and MDZ high-grade domains respectively, based on all three

principal directions. The graph shows the block model grades (black line), ID2 (grey), NN (yellow) and

the declustered composite grades (blue line).



Source: SRK, 2020

Figure 14-25: Swath Analysis Group 1000 Au (g/t)



Source: SRK, 2020

Figure 14-26:Swath Analysis Group 2000 Au (g/t)



Source: SRK, 2020




Source: SRK, 2020

Figure 14-29: Swath Analysis Group 5000 – INDZONE=0 (LG) Au (g/t)



Source: SRK, 2020

Figure 14-30: Swath Analysis Group 5000 – INDZONE=1 (MG) Au (g/t)



Source: SRK, 2020

Figure 14-31: Swath Analysis Group 5000 – INDZONE=2 (HG) Au (g/t)



14.10 Resource Classification


CIM Definition Standards for Mineral Resources and Mineral Reserves (May 2014).

Mineral Resource classification is typically a subjective concept. Industry best practices suggest that

classification should consider the confidence in the geological continuity of the mineralized structures,

the quality and quantity of exploration data supporting the estimates, and the geostatistical confidence

in the tonnage and grade estimates. Appropriate classification criteria should aim to integrate both

concepts to delineate regular areas at similar resource classification.

Data quality, drillhole spacing and the interpreted continuity of grades controlled by the veins have

allowed SRK to classify portions of the veins in the Measured, Indicated and Inferred Mineral Resource

categories.

SRK’s classification system is consistent with those used in the 2019 PEA.

14.10.1 Measure Mineral Resources

Measured Resources are limited to vein material within the current levels being mined by Company.

These areas are considered to have strong geological knowledge as they have been traced both down-

dip and along strike via mapping, plus underground channel samplings provide sufficient data

populations to define internal grade variability.

• Vein and Diss material within the current levels being mined

• Blocks estimated within the first search volume of 25 m which required a minimum of 4

composites and maximum of 12

• Splays were not categorized as measured

14.10.2 Indicated Mineral Resources

SRK has delineated Indicated Mineral Resources using two methods split by the material types:

Veins/Disseminated/Splays

Primarily between Level 16 to 21 currently in operation. Indicated Mineral Resources have been given

at the following approximate data spacing, as function of the confidence in the grade estimates and

modelled variogram ranges. SRK has expanded the limits of the Indicated to also cover areas within

the licensed portion of Echandia where:

• Spacing of 50 m by 50 m (XY) existed from the nearest drillhole

• Multiple holes were enabled to be used during the estimation process

• Support from both diamond drilling and channel sampling was present

MDZ

• Based primarily on 2018 and 2019 drilling

• 50 x 50 m (XY) drillhole Spacing (defined by a distance buffer of 25 m from drilling of UG

levels)

• Enabled multiple holes to be used during the estimation process

• Search Volume less than 2 (i.e. volumes 1 and 2)



• Additional caution has been paid when classifying the dip extensions on the series of holes

drilled to the northeast as limited information is known up and down dip from the current drilling

14.10.3 Inferred Mineral Resources

Inferred Mineral Resources have been limited to within areas of reasonable grade estimate quality and

sufficient geological confidence and are extended no further than 150 m from peripheral drilling on the

basis of modelled variogram ranges.

Notes on Downgrade of Porphyry Pockets

During the site inspection SRK noted and discussed with the mine geologists that some mining has

been attempted within the porphyry pockets. SRK considers this to have uncertainty as no detailed

survey of mining volumes is available. Based on the level of uncertainty SRK has down-graded areas

identified as having potential mining to Inferred.

14.10.4 Final Classification

Mathematical criteria as defined has then been digitized on 50 m sections (across Strike), to smooth

the process with the final wireframe based on interpretation of polylines in Leapfrog to smooth changes

in interpretation between sections. A summary of the classification at Marmato is shown in Figure

14-32 and Figure 14-33.

Source: SRK, 2020

Figure 14-32: Final Classification for the Marmato Project (Looking Northwest Bearing 305)



Source: SRK, 2020

Figure 14-33: Final Classification for License #014-89m Marmato Project (Looking Northwest Bearing 305)

14.11 Depletion

To define the Mineral Resource SRK has created a block model to represent the depletion for the

veins. In order to complete this task SRK has used a combination of AutoCAD™ polylines provided by

CGM and generated Vulcan™ (.00t) files by projecting the strings in perpendicular to the strike. These

wireframes have subsequently been used to copy out the assigned vein. The process is manual and

labor intensive and requires each vein to be individually assessed.

This process may result in some errors of over or under depletion at the edges but given the size of

the deposits is not considered to be material. Once the block model has been established the model

has been combined with the final geological model to code all the blocks for depletion.

SRK and CGM have worked towards generating a 3D digital layout of the mine which has the depletion

assigned to each of the main structures, further work will be required to further validate the models

using underground surveys to improve confidence in the models, but the addition of approximately 18

months on additional depletion from the PEA model has increased confidence in the estimated

depletions for the PFS.

14.12 Mineral Resource Statement

Canadian Institute of Mining, Metallurgy and Petroleum’s (CIM) Definition Standards for Mineral

Resources and Mineral Reserves (May, 2014) defines a Mineral Resource as:

“(A) concentration or occurrence of diamonds, natural solid inorganic material, or natural solid

fossilized organic material including base and precious metals, coal, and industrial minerals in or on

the Earth’s crust in such form and quantity and of such a grade or quality that it has reasonable



prospects for economic extraction. The location, quantity, grade, geological characteristics and

continuity of a Mineral Resource are known, estimated or interpreted from specific geological evidence

and knowledge”.

The “reasonable prospects for eventual economic extraction” requirement generally implies that the

quantity and grade estimates meet certain economic thresholds and that the Mineral Resources are

reported at an appropriate CoG, taking into account extraction scenarios and processing recoveries.

In order to meet this requirement, SRK considers that portions of the vein system are amenable for

underground mining.

To determine the potential for economic extraction, SRK used the following key assumptions for the

costing but notes that the deposit has variable mining costs depending on the mining types resulting

in a range of CoGs (Table 14-20). A metallurgical recovery of 95% Au has been assumed for the MDZ

and 90% for the veins and porphyry material based on the current performance of the operating plant.

Table 14-20: Summary of CoG Assumptions at Marmato Based on Assumed Costs (Averaged for All Mining Styles)

Units Vein Mining

Averaged CoG Deeps Option (Longhole)

Averaged CoG

Assumptions

Gold Price US$/ounce (oz) $1,500 $1,500

Gold Price US$/gram (g) $48.23 $48.23

Au Recovery % 90% 95%

Operating Costs

Mining US$/tonne (t) mined $49.45 $35.00

Processing US$/t ore $13.63 $17.00

Royalties US$/t ore $8.96 $6.80

General and Administrative (G&A) and Other

US$/t ore $12.24 $3.00

Other US$/t ore $0.00 $0.00

Subtotal US$/t $84.28 $61.75

CoG - Head Grade grams per tonne (g/t) 1.9 1.3

Source: SRK, 2020

SRK has defined the proportions of Mineral Resource to have potential for economic extraction for the

Mineral Resource based on two separate CoGs, relating to the different mining methods involved. The

initial cut-off is based on the mining of the veins using the current mining processes and assumed

costs, with a second method (longhole) defined for mining the MDZ and potentially areas of wider

porphyry mineralization in the upper levels.

SRK has reported the tonnage and grades associated with current mine and the MDZ project, which

are the assets owned by CGM. As such, the Mineral Resource includes all material within the #014-

89m license and a sub-portion of the #RPP_357 (Echandia) below an elevation of 1,300 m, which can

be accessed from the existing operation through an agreement with Gran Colombia. SRK has also

included the proportion of Mineral Resources currently under application (Application #KIU-11401)

within the Mineral Resources, but these have been excluded from the Mineral Reserves as the timing

on granting this license remained uncertain at the date of this report (however, SRK understands that

the license was recently confirmed as approved by the Government so will therefore be included in

future technical studies).



The proposed mining plan is predicated on splitting the above Mineral Resources into three styles of

mineralization within three distinct areas. These areas are referred to as the UZ (existing mine levels

16 to 21), the Transition Zone (which includes mining of MDZ material to an elevation of 950 m), and

the MDZ project (which includes all material below the 950 m).

The three styles of mineralization are based on the key geological types defined in the Mineral

Resources of veins, porphyry, and MDZ. Therefore, the estimation domains for the Mineral Resource

Statement have been grouped into veins, porphyry and MDZ mineralization. The veins account for the

veins, halos, and splay material and have used a 1.9 g/t Au cut-off; the porphyry material has also

used a cut-off of 1.9 g/t Au, as the potential mining method will require further investigation; the MDZ

material has used a lower cut-off of 1.3 g/t Au to account for the larger bulk mining methods involved.

SRK highlights to the reader that all Mineral Resources within #CHG_081 (yellow) and upper areas of

*RPP_357 (above 1,300 m) as highlighted in Figure 14-34 in light blue have not been reported and

are excluded from the Mineral Resource statement as they were not transferred to CGM and therefore

are excluded from the Mineral Resources. Any Mineral Resources that may occur within these two

areas are currently held by Gran Colombia.

Source: SRK, 2020

Figure 14-34: Cross-Section Showing License Splits at Marmato

Table 14-21 shows the Mineral Resource Statement for the project, with an effective date of March

17, 2020.

Licence #014-89m

Licence #CHG-081 (above 1300)

RPP #357 -(above 1300)

RPP #357 -(below 1300)

Licence #CHG-081 (below 1300)

Area under applicationNo # KIU-11401Note: Included in Mineral Resources but excluded from Mineral Reserves

Upper Mine(above 950)

MDZ Projectbelow 950)

Transition Zone MDZ(above 950)



Table 14-21: Caldas Mineral Resource(1) Statement with Effective Date of March 17, 2020

Caldas Marmato Project - Effective Date March 17, 2020, Basis for MRE and PFS (Caldas including RPP less than 1,300)(1)

Category Quantity (million tonnes [Mt]) Grade (g/t) Metal (kozs)

Au Ag Au Ag

Upper Mine (2)

Measured 2.1 5.65 27.0 387 1,853

Veins (5) 2.1 5.6 27.0 387 1,853

Porphyry (5) 0.0 0.0 0.0 0 0

Indicated 9.2 4.45 18.7 1,320 5,545

Veins 7.2 5.0 21.1 1,156 4,862

Porphyry 2.1 2.5 10.3 165 682


Veins 9.3 5.2 22.4 1,543 6,715

Porphyry 2.1 2.5 10.3 165 682

Inferred 4.5 3.70 15.5 532 2,224

Veins 2.7 4.4 17.9 386 1,574

Porphyry 1.7 2.6 11.7 145 650

Transition Zone (3) (6)

Measured 0.0 0.0 0.0 0 0

Indicated 3.4 2.68 7.2 294 785


Inferred 0.0 1.95 3.7 2 3

MDZ (4) (6)

Measured 0.0 0.0 0.0 0 0

Indicated 24.7 2.63 3.6 2,085 2,870


Inferred 21.9 2.32 2.1 1,639 1,506

Combined

Measured 2.1 5.6 27.0 387 1,853

Indicated 37.3 3.1 7.7 3,699 9,200


Inferred 26.4 2.6 4.4 2,172 3,733 (1) Mineral Resources are reported inclusive of the Mineral Reserve. Mineral resources are not mineral reserves and do not have demonstrated economic viability. All figures are rounded to reflect the relative accuracy of the estimate. All composites have been capped where appropriate. The Mineral Resources were estimated by Benjamin Parsons, MSc, MAusIMM #222568 of SRK, a Qualified Person pursuant to NI 43-101. (2) Upper Mine is defined as the current operating mines from levels 16-21 using existing mining methodology (cut and fill). (3) “Transition Zone” is defined as mining of MDZ above an elevation of 950 access from the current operations using a modified longhole stoping method. (4) MDZ is defined as mining of MDZ below an elevation of 950 using longhole open stope mining methods. (5) Porphyry and vein mineral resources are reported at a cut-off grade (“CoG”) of 1.9 g/t. CoGs are based on a price of US$1,500/oz Au and gold recoveries of 90% for underground resources without considering revenues from other metals. (6) MDZ mineral resources are reported at a CoG of 1.3 g/t. CoGs are based on a price of US$1,500.oz Au and gold recoveries of 95% for underground resources without considering revenues from other metals within a limiting pitshell.

14.13 Comparison to the Previous Estimate

The 2020 Mineral Resource represents a number of changes in the defined Mineral Resource

compared to the 2019 PEA Mineral Resources, due to the following key factors:

• Infill drilling within the MDZ areas has increased the confidence in the estimates significantly

from the Inferred to Indicated category.

• Minor reduction in the vein domains as a result of additional depletion accounted for between

the PEA and PFS models, plus changes in the geological interpretation of veins and

disseminated material.



SRK highlights that the current MDZ Mineralization represents a notable change in the style of

mineralization and considerations for mining methods at the Marmato Project and has maintained the

use of a high-grade core to the mineralization at depth.

The main changes in the Mineral Resource Statement since the previous estimate can be defined on

the combined Mineral Resource as follows:

• Increase in the Indicated MDZ material from 6.4 Mt at 2.6 g/t Au, for 537 koz, to 28.1 Mt at 2.6

g/t Au, for 2,379 koz, which is an increase of 1,842 koz within the MDZ. This is reflected in a

reduction in the Inferred from 41.2 Mt at 2.1 g/t for 2,812 koz to 22.0 Mt at 2.3 g/t for 1,640

koz, which is a reduction of 1,172 koz.

• Increase in the proportion of Measured and Indicated material within the vein domain from

9.2 Mt at an average grade of 4.6 g/t to 9.3 Mt at an average grade of 5.2 g/t Au, which is an

increase of 180 Koz or 13.2%.

• Reduction in the proportion of Inferred material within the veins from 3.3 Mt at 4.4 g/t Au for

466 koz, to 2.7 Mt at 4.4 g/t Au for 386 koz, which represents a difference of 80 koz.

• Minor increase in proportion for Indicated of porphyry (Pockets) material of 25 koz.

• Increase in the Inferred portion of the porphyry material from 0.3 Mt at 3.1 g/t Au for 34 koz,

to 1.7 Mt at 2.6 g/t Au for 145 koz.

14.14 Mineral Resource Sensitivity

The mineral resource given above is sensitive to the selection of the reporting CoG. To illustrate the

sensitivity the block model quantities and grade estimates for the Marmato Project were classified

according to the CIM Definition Standards for Mineral Resources and Reserves (CIM, 2014).

The reader is cautioned that the figures presented in the tables should not be misconstrued with a

Mineral Resource Statement. Figure 14-35 and Figure 14-36 are only presented to show the sensitivity

of the block model estimates to the selection of cut-off grade. The following tables (Table 14-22 to

Table 14-27) have been split by the mineralization style and classification (Measured and Indicated

have been combined for ease of reporting).



Table 14-22: Grade Tonnage Curve Measured and Indicated - Vein Domains (Group 1000 to 3000)

Cut-Off (Au g/t) Tonnes (kt) Au (g/t) Ag (g/t) Au (koz) Ag (koz)

0.00 14,688 3.72 18.18 1,758 8,584 0.50 14,023 3.89 18.75 1,752 8,453 1.00 12,516 4.26 19.91 1,715 8,011 1.20 11,821 4.45 20.44 1,690 7,769 1.30 11,449 4.55 20.73 1,675 7,631 1.50 10,727 4.76 21.32 1,643 7,355 1.70 10,004 4.99 21.93 1,605 7,054 1.80 9,651 5.11 22.25 1,586 6,904 1.90 9,312 5.23 22.55 1,565 6,752 2.00 8,978 5.35 22.91 1,545 6,612 2.20 8,292 5.62 23.63 1,498 6,299 2.50 7,460 5.99 24.48 1,436 5,871 2.70 7,460 5.99 24.48 1,436 5,871 3.00 6,359 6.55 25.74 1,339 5,262 3.50 5,424 7.12 27.04 1,241 4,716 4.00 4,617 7.71 28.48 1,144 4,228 4.50 3,967 8.27 29.91 1,055 3,815 5.00 3,418 8.84 31.24 972 3,433

Source: SRK, 2020

Table 14-23: Grade Tonnage Curve Measured and Indicated - Porphyry Domain (Group 4000)


0.00 9,059 1.47 7.16 428 2,086

0.50 8,799 1.50 7.33 425 2,074

1.00 6,632 1.73 8.14 369 1,735

1.20 5,218 1.90 8.70 319 1,460

1.30 4,685 1.98 8.93 298 1,345

1.50 3,691 2.13 9.33 253 1,107

1.70 2,747 2.32 9.82 205 868

1.80 2,389 2.40 10.14 185 779

1.90 2,052 2.50 10.35 165 682

2.00 1,731 2.60 10.69 145 595

2.20 1,207 2.81 11.28 109 438

2.50 677 3.19 12.55 69 273

2.70 677 3.19 12.55 69 273

3.00 312 3.75 14.68 38 147

3.50 161 4.24 14.74 22 76

4.00 103 4.55 15.11 15 50

4.50 46 5.02 17.51 7 26

5.00 16 5.54 13.26 3 7

Source: SRK, 2020



Table 14-24: Grade Tonnage Curve Measured and Indicated - MDZ Domain (Group 5000)


0.00 86,787 1.29 2.79 3,606 7,785

0.50 60,574 1.71 3.17 3,329 6,182

1.00 41,281 2.15 3.60 2,860 4,779

1.20 31,695 2.47 3.90 2,522 3,974

1.30 28,050 2.63 4.04 2,375 3,647

1.50 23,049 2.90 4.29 2,152 3,178

1.70 20,181 3.09 4.44 2,005 2,882

1.80 19,016 3.17 4.51 1,940 2,758

1.90 18,067 3.24 4.57 1,883 2,655

2.00 17,127 3.31 4.63 1,824 2,550

2.20 15,419 3.45 4.75 1,709 2,353

2.50 12,888 3.66 4.95 1,518 2,051

2.70 12,888 3.66 4.95 1,518 2,051

3.00 8,583 4.12 5.36 1,138 1,478

3.50 5,416 4.64 5.71 809 994

4.00 3,321 5.22 6.05 557 646

4.50 2,048 5.84 6.47 384 426

5.00 1,316 6.45 6.74 273 285

Source: SRK, 2020

Table 14-25: Grade Tonnage Curve Inferred - Vein Domains (Group 1000 - 3000)


0.00 6,664 2.37 12.51 508 2,681

0.50 5,702 2.72 13.89 498 2,546

1.00 4,652 3.17 15.12 473 2,262

1.20 4,136 3.42 15.74 455 2,093

1.30 3,928 3.54 16.01 447 2,022

1.50 3,485 3.81 16.62 427 1,863

1.70 3,115 4.07 17.15 408 1,718

1.80 2,926 4.22 17.51 397 1,647

1.90 2,741 4.38 17.87 386 1,574

2.00 2,604 4.51 17.99 378 1,506

2.20 2,341 4.78 18.62 360 1,402

2.50 2,020 5.17 19.43 336 1,262

2.70 2,020 5.17 19.43 336 1,262

3.00 1,612 5.78 20.44 300 1,060

3.50 1,319 6.35 21.42 269 909

4.00 1,067 6.97 22.41 239 769

4.50 889 7.53 23.33 215 667

5.00 722 8.16 22.52 189 523

Source: SRK, 2020



Table 14-26: Grade Tonnage Curve Inferred - Porphyry Domain (Group 4000)


0.00 7,626 1.50 8.11 368 1,988

0.50 7,541 1.51 8.17 367 1,981

1.00 5,556 1.77 9.21 316 1,644

1.20 4,449 1.93 9.86 276 1,410

1.30 3,883 2.03 10.26 254 1,281

1.50 2,944 2.24 10.79 212 1,022

1.70 2,293 2.42 11.43 178 842

1.80 1,984 2.52 11.61 161 741

1.90 1,725 2.62 11.71 145 650

2.00 1,516 2.71 11.82 132 576

2.20 1,155 2.91 11.87 108 441

2.50 707 3.27 12.59 74 286

2.70 707 3.27 12.59 74 286

3.00 335 3.91 11.93 42 128

3.50 188 4.47 11.75 27 71

4.00 119 4.93 10.85 19 41

4.50 85 5.24 7.56 14 21

5.00 54 5.60 6.06 10 11

Source: SRK, 2020

Table 14-27: Grade Tonnage Curve Inferred - MDZ Domain (Group 5000)


0.00 108,152 0.94 1.69 3,268 5,876

0.50 67,414 1.30 1.83 2,824 3,965

1.00 32,941 1.93 2.02 2,044 2,139

1.20 25,465 2.17 2.09 1,781 1,709

1.30 21,964 2.32 2.14 1,640 1,509

1.50 16,952 2.60 2.22 1,415 1,208

1.70 13,286 2.87 2.31 1,227 988

1.80 12,052 2.99 2.35 1,158 911

1.90 10,620 3.14 2.42 1,072 827

2.00 9,784 3.24 2.44 1,020 766

2.20 8,069 3.49 2.45 904 634

2.50 6,171 3.83 2.42 761 480

2.70 6,171 3.83 2.42 761 480

3.00 3,630 4.59 2.44 536 285

3.50 2,035 5.67 2.38 371 156

4.00 1,468 6.43 2.23 303 105

4.50 1,071 7.23 2.23 249 77

5.00 812 8.02 2.20 209 57

Source: SRK, 2020



Source: SRK, 2019

Figure 14-35: Grade Tonnage Curves Showing Sensitivity to Changes in Cut-Off for Measured and Indicated Mineralized Material

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

10.00

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00

Au

(g/

t)

TON

NES

(kt

)

Au Cut-Off (g/t)

Grade Tonnage Curve (Veins - M&I)

0.00

1.00

2.00

3.00

4.00

5.00

6.00

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00

Gra

de

Au

(g/

t)

TON

NES

(kt

)

AU Cut-Off (g/t)

Grade Tonnage Curve (Porphyry M&I)

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,000

100,000

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00

Gra

de

Au

(g/

t)

TON

NES

(kt

)

Au Cut-Off (g/t)

Grade Tonnage Curve (MDZ M&I)



Source: SRK, 2019

Figure 14-36: Grade Tonnage Curves Showing Sensitivity to Changes in Cut-Off for Inferred Mineralized Material

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00

Au

(g/

t)

TON

NES

(kt

)

AU Cut-Off (g/t)

Grade Tonnage Curve (Veins - Inferred)

0.00

1.00

2.00

3.00

4.00

5.00

6.00

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00

Gra

de

Au

(g/

t)

TON

NES

(kt

)

Tho

usa

nd

s

AU Cut-Off (g/t)

Grade Tonnage Curve (Porphyry - Inferred)

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

0

20,000

40,000

60,000

80,000

100,000

120,000

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00

Gra

de

Au

(g/

t)

TON

NES

(kt

)

AU Cut-Off (g/t)

Grade Tonnage Curve (MDZ - Inferred)



14.15 Relevant Factors

SRK is not aware of any environmental, permitting, legal, title, taxation marketing or other factors that

could affect resources, however SRK considers that there may be some degree of sensitivity for the

potential to extract mineralization based on the various mining domains.



15 Mineral Reserve Estimate The mine is currently developed to the 1,000 m elevation. A transition is occurring from narrow vein

mineralization to large porphyry mineralized areas (gold associated with pyrrhotite veinlets).

Mineralization is generally vertical with veins widths ranging from more than 1 m to several m. Porphyry


width. For this PFS, there are three different mining methods, separated into three distinct zones.

The first zone is the mineralized vein material between 950 m elevation to 1,300 m elevation, referred

to as the Veins. This is the current mine and will be mined using the current conventional cut and fill

stope method.

The second zone is the wider porphyry material between 950 m elevation and 1,050 m elevation,

referred to as the Transition Area. A modified longhole stoping method will be used in this area. The

stope size is 15 m wide by 15 m high with varying length of up to 26 m. These stopes will be mined in

a primary-secondary sequence with paste backfill for the primary stopes and unconsolidated waste

rockfill for the secondary stopes. Where waste rock is unavailable, hydraulic fill will be used to fill the

secondary stopes.

The third zone is the porphyry material below 950 m elevation, referred to as MDZ. There is a 10 m

sill pillar left in situ between the MDZ and the UZ. The MDZ material is mined using a longhole stoping

method with stope sizes that are 10 m wide by 30 m high, with varying lengths of up to 30 m. The MDZ

area is currently not developed.

The first two zones (veins and transition) are considered the UZ, and the material is processed in the

existing processing facility. The third zone is considered the MDZ and the material is envisioned to be

sent to a new processing facility. Separate mine plans are presented for the UZ area and MDZ area.

Figure 15-1 shows the general locations of the mining areas and process facilities.



Source: SRK, 2020

Figure 15-1: Marmato General Layout

15.1 Conversion Assumptions, Parameters and Methods

Measured and Indicated Mineral Resources were converted to Proven and Probable Mineral Reserves

by applying the appropriate modifying factors, as described herein, to potential mining block shapes

created during the mine design process. Inferred material is treated as waste with zero grade. All

Mineral Reserve tonnages are expressed as "dry” tonnes (i.e., no moisture) and are based on the

density values stored in the block model.

15.1.1 Upper Zone - Dilution

The Veins dilution ranges from 20% to 55% with an average of 26%. Currently the mine has a higher

dilution (55%), however with better grade control practices, the dilution is expected to decrease to

20%. Figure 15-2 shows the change in dilution through the mine life. The Transition dilution is based

on typical dilution associated with the modified longhole mining method (Table 15-1).



Source: SRK, 2020

Figure 15-2: Veins Dilution

Table 15-1: Dilution Assumption

Mining Area Mining Method Dilution (%)

Veins Cut and Fill 20% - 55% (avg. 26%)

Deeps Modified Long Hole 7%

Source: SRK, 2020

15.1.2 Upper Zone - Recovery

Mining extraction ratios/recovery factors are applied to the mine design by area as shown in

Table 15-2. Items considered for the recovery include:

• Veins

o Material loss due to mucking equipment inefficiencies

o Drill and blast inefficiencies

o Geotechnical factors

• Transition Zone

o Mucking blind corners in the stopes

o Material lost into backfill

o Geotechnical factors

Table 15-2: Mining Extraction/Recovery Assumptions

Mining Area Mining Method Dilution (%)

Veins Cut and Fill 90%

Transition Modified Long Hole 90%

Source: SRK, 2020



15.1.3 Upper Zone - Additional Allowance Factors

An additional 5% allowance was applied to development due to overbreak. A 10% extra development

allocation is applied to the ramps in the Transition for items that were not included in the design, such

as muckbays, sumps, safety bays, etc.

15.1.4 Upper Zone – Cutoff Grade Calculation

CoG for Veins material is calculated based on the cost structure provided by Marmato as summarized

in Table 15-3. The Transition cost structure is developed based on the Veins cost as shown in Table

15-4.

Table 15-3: Cut-off Grade Parameters for Veins Material

Parameter Amount Unit

Mining Cost (1) 49.45 USD/t Process Cost 12.24 USD/t G&A 13.63 USD/t Royalties 8.96 USD/t

Total Cost (2) 84.28 USD/t

Gold Price 1,400 USD/oz Silver Price 17 USD/oz Gold Recovery 85 % Silver Recovery 65 %

CoG 2.23 g/t Au

Source: SRK (1) Includes Backfill (2) Values used here may differ from technical economic model, however SRK is of the opinion that the differences are not material.

Table 15-4: Cut-off Grade Parameters for Transition Material


Mining Cost (1) 46.00 USD/t Process Cost 12.24 USD/t G&A 13.63 USD/t Royalties 8.96 USD/t

Total Cost (2) 80.83 USD/t

Gold Price 1,400 USD/oz Silver Price 17 USD/oz Gold Recovery 87 % Silver Recovery 40 %

CoG 1.91 g/t Au

Source: SRK (1) Includes Backfill (2) Values used here may differ from technical economic model, however SRK is of the opinion that the differences are not material.

A grade-tonne curve for the UZ veins area (950 m elevation and above) is shown in Figure 15-3. This

includes only Measured and Indicated material.



Source: SRK, 2020

Figure 15-3: UZ Grade/Tonne Curve Based on Au Cut-Off

15.1.5 MDZ Mine - Dilution

The mining dilution estimate is based on ELOS (Clark, 1997). ELOS is an empirical design method

that is used to estimate the amount of overbreak/slough that will occur in an underground opening

based on rock quality and the hydraulic radius of the opening. ELOS was applied to in situ rock

exposed and to the paste backfill walls wherever mining will occur adjacent to a secondary stope. In

additional to the ELOS allowances, an additional allowance was used to account for backfill dilution

from the floor when mucking a stope.

Dilution assumptions are shown in Table 15-5.

Table 15-5: Dilution Assumptions

Type Value (m)

Sidewall ELOS (rock) 0.35

Sidewall ELOS (backfill) 0.15

Bottom (mucking backfill) 0.10

Back ELOS (rock) 0.35

Endwall ELOS (rock) 0.35

Endwall ELOS (backfill) 0.15

Source: SRK, 2020

The ELOS and floor dilution factors result in a total dilution of 8%. Backfill dilution is added using zero

grade. The rock portion of the dilution (approximately 4.5%) is expected to contain grade. The grade

applied to rock dilution is based on querying block model grades just outside the stope designs in a

representative area. This exercise showed that the dilution was approximately 50% of the stope grade,

and therefore for the reserves, the grade applied to the rock dilution is 50% of the stope grade.



15.1.6 MDZ – Recovery

A stope recovery factor of 92.5% was used. The following items were used to calculate this factor:

• Material loss into backfill (floor) or 0.15 m

• Material loss to mucking along sides and in blind corners

• Additional loss factor due to rockfalls, misdirected loads, and other geotechnical reasons

A development recovery factor of 100% was used for all horizontal development.

15.1.7 MDZ - Additional Allowance Factors

Additional ramp allowance factors were used to account for additional excavations not included in the

PFS design. These items should be designed at the detailed planning stage. Items are summarized in

Table 15-6.

Table 15-6: Additional Ramp Allowance Factors

Type Units Conveyor

Ramp Truck Ramp

Ventilation Drifts

Footwall Accesses

>730 L

Footwall Accesses

<730 L

Average Length(1)

m 500 500 500 375 275

Muckbays m3 916 916 916 892 446

Drillbays m3 - - - - -

Electrical Bays m3 81 81 81 81 81

Pump Stations m3 324 324 - - -

Passing Bays m3 685 685 - - -

Total Additional Allowance

m3 2,006 2,006 997 973 527

Expressed as a % of Representative Length of Development

% 13.4 13.7 8.0 10.4 7.7

(1) Representative length of ramp that the listed allowances are applied to. Source: SRK

15.1.8 MDZ – Cutoff Grade Calculation

Current estimated project costs and the calculated Au CoG are shown in Table 15-7. For reporting

reserves within the design, a minimum cut-off of 1.61 g/t Au was used. A silver recovery of 50% is

expected however it was not used for the CoG calculation.



Table 15-7: MDZ Underground Cut-off Grade Calculation


Mining cost (1) 42.00 US$/t

Process cost 14.00 US$/t

Tailings 3.00 US$/t

Production Taxes 6.75 US$/t

G&A, Other 3.00 US$/t

Total Cost (2) $68.75 US$/t

Gold price 1,400.00 US$/oz

Au Mill Recovery 95% CoG 1.61 g/t

Source: SRK (1) Includes backfill (2) Values used here may differ from the technical economic model, however SRK is of the opinion that the differences are

not material.

A grade-tonne curve for the MDZ area (950 m elevation and below) is shown in Figure 15-4. This

includes only Measured and Indicated material.

Figure 15-4: MDZ Grade/Tonne Curve Based on Au Cut-Off

15.2 Reserve Estimate

Mineral Reserves were classified using the 2014 CIM Definition standards. Indicated Mineral

Resources were converted to Probable Mineral Reserves by applying the appropriate modifying

factors, as described herein, to potential mining shapes created during the mine design process. In

the same manner, Measured Mineral Resources were converted to Proven Mineral Reserves.



A 3D design has been created representing the planned reserve mining areas. The underground mine

design process resulted in 19.7 Mt at an average grade of 3.19 g/t Au and 6.87 g/t Ag. Table 15-8

presents the Mineral Reserve statement as of March 17, 2020.

Table 15-8: Caldas Mineral Reserve Estimate as of March 17, 2020 – SRK Consulting (U.S.), Inc.

Underground Mineral Reserves Cut-Off (1): 1.61 to 2.23 g/t

Area Category Tonnes

(kt) Au

(g/t) Ag

(g/t) Contained Au

(koz) Contained Ag

(koz)

Veins(2)

Proven 762 5.01 21.80 123 534

Probable 3,049 4.20 16.85 412 1,652

Veins Total 3,812 4.37 17.84 535 2,186

Transition (3)

Proven 40 7.63 28.16 10 36

Probable 1,293 3.43 7.92 143 329

Transition Total

1,333 3.56 8.52 152 365

MDZ(4)

Proven - - - - -

Probable 14,556 2.85 3.84 1,333 1,799

MDZ Total 14,556 2.85 3.84 1,333 1,799

Caldas Total

Proven 802 5.14 22.12 133 570

Probable 18,898 3.11 6.22 1,888 3,780

Total 19,700 3.19 6.87 2,021 4,350

Source: SRK, 2020 Notes: All figures are rounded to reflect the relative accuracy of the estimates. Totals may not sum due to rounding. Mineral Reserves have been stated on the basis of a mine design, mine plan, and economic model. Mineral Resources are reported inclusive of the Mineral Reserve. (1): Veins reserves are reported using a CoG of 2.23 g/t Au. The veins CoG calculation assumes a US$1,400/oz Au price, 85% Au metallurgical recovery, US$49.45/t mining cost, US$13.63/t G&A cost, US$12.24/t processing cost, and US$8.96/t royalties. Transition reserves are reported using a CoG of 1.91 g/t Au. The Transition CoG calculation assumes a US$1,400/oz Au price, 95% Au metallurgical recovery, US$46/t mining cost, US$13.63/t G&A cost, US$12.24/t processing cost, and US$8.96/t royalties. MDZ reserves are reported using a CoG of 1.61 g/t Au. The MDZ CoG calculation assumes a US$1,400/oz Au price, 95% metallurgical recovery, US$42/t mining cost, US$14/t processing cost, US$6.75/t production taxes, US$3/t G&A cost, and US$3/t tailings cost. Note that costs/prices used here may be somewhat different than those in the final economic model. This is due to the need to make assumptions early on for mine planning prior to finalizing other items and using long-term forecasts for the life-of-mine plan. (2): The Veins area is currently mined using cut-and-fill methods. Mining dilution ranges from 20% - 55%, averaging 26%, is included in the reserves using a zero grade for dilution. A mining recovery of 90% is applied to stopes. The Veins Mineral Reserves were estimated by Fernando Rodrigues, BS Mining, MBA, MMSAQP #01405, MAusIMM #304726 of SRK, a Qualified Person. (3): The Transition area will be mined using a modified longhole stoping method. A mining dilution of 7% is included in the reserves using a zero grade for dilution. A mining recovery of 90% is applied to stopes. The Transition Mineral Reserves were estimated by Fernando Rodrigues, BS Mining, MBA, MMSAQP #01405, MAusIMM #304726 of SRK, a Qualified Person. (4): The MDZ portion of the Project will be mined by longhole open stoping mining methods. Mining dilution (internal and external) is included in the reserve. Stope dilution is 8%, and a portion of the stope dilution is applied using grade values based on average surrounding block information. A mining recovery of 92.5% is applied to stopes. The MDZ Mineral Reserves were estimated by Joanna Poeck, BEng Mining, SME-RM, MMSAQP #01387QP, a Qualified Person.

15.3 Relevant Factors

An exclusion zone (gap) was considered in the PFS where CGM ownership was not secured at the

beginning of the PFS work. As the PFS was nearing completion CGM informed SRK that the gap was

no longer an issue; however a re-design to include the gap area was not completed. For the reserves

stated here, UZ development mining does go through the gap area. If the development material was

mineralized it is included in the Mineral Reserves. Vein mining for the UZ does not occur in the gap,

though there is mineralized material there that should be evaluated for reserves. The MDZ design

specifically avoided the gap completely. If gap ownership is no longer an issue, the UZ reserves in this

area should be updated and the development layout of the MDZ should be reviewed to ensure it is

optimal. Figure 15-5 shows the license gap.



Source: SRK, 2020

Figure 15-5: License Gap



16 Mining Methods The Project has been in operations in various forms since the mid-1500s. Mineros Nacionales (MN)

was awarded the contract for the concessions in 1989. The Project was originally developed as a 300

t/d underground project in 1997 and has expanded through the years to the existing 1,200 t/d capacity

operation. Table 16-1 shows the production from 2015 to May 2020.

Table 16-1: 2015 to 2020* Production

Year Unit 2015 2016 2017 2018 2019 2020*

Ore Tonnes Processed t 303,279 341,309 365,119 338,902 370,245 119,069

Au Grade g/t 2.79 2.56 2.48 2.67 2.49 2.47

Au Recovered oz 23,954 23,449 25,163 24,909 25,750 8,318

*January through May of 2020 Source: CGM, 2020

16.1 Current Mining Methods

The mine is currently developed and mined to the 1,000 m elevation. A transition occurs from narrow

vein mineralization to large porphyry mineralized areas (gold associated with pyrrhotite veinlets).

Mineralization is generally vertical with veins widths ranging from more than 1 m to several m. Porphyry


width. For this PFS, there are three different mining methods, separated into three distinct zones as

follows:

• The first zone is the mineralized vein material between 950 m elevation to 1,300 m elevation,

referred to as the Veins. This is the current mine and will be mined using the current

conventional cut and fill stope method.

• The second zone is the wider porphyry material between 950 m elevation and 1,050 m

elevation, referred to as the Transition Zone. A modified longhole stoping method will be used

in this area. The stope size is 15 m wide by 15 m high with varying length of up to 26 m. These

stopes are mined in a primary-secondary sequence with paste backfill for the primary stopes

and unconsolidated waste rockfill for the secondary stopes. Where waste rock is unavailable,

hydraulic sand fill will be used to fill the secondary stopes.

• The third zone is the porphyry material below 950 m elevation, referred to as MDZ. There is a

10 m sill pillar left in situ between the MDZ and the UZ (Veins plus Transition area). The MDZ

material can be mined using a longhole stoping method with stope sizes that are 10 m wide

by 30 m high, with varying lengths of up to 30 m. The MDZ area is currently not developed.

The first two zones (Veins and Transition) are considered the UZ, and the material is processed in the

existing processing facility. The third zone is considered the MDZ and the material is envisioned to be

sent to a new processing facility. Separate mine plans are presented for the UZ area and MDZ area.

The UZ of Marmato consists mainly of mineralized veins with varying thickness and geotechnical

conditions. There have been several different mining methods used depending on the lithology and

ground conditions, such as shrinkage stoping, conventional cut-and-fill and caving.

Shrinkage Stoping

In the past, shrinkage stoping (Figure 16-1) is used for areas with high grade vertical veins and

competent backs. The stopes are generally 35 m long and 50 m high. Loading pockets are developed



on the extraction level with vertical raises on the extremity of the stopes for access and ventilation.

This is an overhand mining method using blasted ore left in the stope as a working floor. The blasted

muck also acts as support for the hangingwall and footwall. Marmato has moved away from this mining

method due to safety concerns and dilution control issues.

Source: Atlas Copco, 1980

Figure 16-1: Typical Shrinkage Stoping Diagram

Figure 16-2 shows the typical mining cycle of a CAF panel. The CAF panels are typically 35 m long by

50 m high, with varying thicknesses depending on the vein. The panels are accessed from 2.2 m by

2.2 m haulage levels on the top and bottom. Raises are developed along the vein to break the panel

into discreet mining stopes as well as to provide access and ventilation. Sub-levels are then driven

horizontally along strike. Once the sublevel is opened, vertical holes are drilled up at a length of

approximately 1.7 m to 2.3 m over the width of the vein. After blasting, the mineralized material is

mucked using either slushers, skid steer loaders or microscoops and loaded into trains and hauled

out. Once mucking is complete, concrete walls are built on either end of the stope and the stope is

filled with hydraulic sand fill. When the fill is sufficiently drained, the next slice of mineralized material

can be mined.




Figure 16-2: Conventional Cut and Fill Method Diagram

In areas where the ground condition is poor and does not support vertical blasting, a variant of the

CAF method, called breasting, is used. The backfill goes up to the back and mining advances

horizontally along strike. This method is slower but allows for better dilution control.

Caving

There are certain veins in the upper levels of the mine where a caving method is used. These veins

are typically in very poor ground conditions where the ore begins to cave when opened. The ore is

extracted from the drawpoint until it becomes waste, then mining moves to the level above and the

cycle repeats. While this method can provide ore at a low cost, there is no control for how much of the

vein is extracted, thus making it difficult to plan and schedule. Dilution control is also difficult as it

requires constant grade monitoring from geology.

16.1.1 Mine Layout

The current Zona Baja mining extends approximately 300 m vertically and approximately 900 m along

the vein structure. The mine has been developed with level accesses proceeding horizontally from the

main portal at the surface to horizontal cross cuts that provide access to the veins. There are currently

six production levels and one level in development, the highest production level is Level 16 and the

lowest production level is Level 21 (Figure 16-3). Each level is spaced 50 m apart vertically, with the



exception of Level 20, which is 60 m from Level 19. Table 16-2 lists the elevation of each level and a

brief description of the main activities on the level.

Source: SRK, 2020

Figure 16-3: Marmato Zona Baja Cross Section Looking NE with Active Levels

Table 16-2: Level Elevations and Description

o Level o Elevation o Description

o 16 o 1,260 o Production and ventilation exhaust

o 17 o 1,210 o Production and ventilation exhaust

o 18 o 1,160 o Main entrance/exit, main haulage

o 19 o 1,110 o Production and incline to Level 18

o 20 o 1,050 o Production

o 21 o 1,000 o Production and Transition Zone

o 22 o 950 o In development

Source: CGM, 2020

Level 18 is the main haulage level and the primary access for the mine and is shown in Figure 16-4.

A track drift provides the main haulage for all material. The trains exit via the south portal, unload at

the mill area and enter the mine via the north portal. Personnel and material enter via the north portal

only. All levels can be reached via ladder-ways, and a service and personnel cage hoist is nearing

completion that will provide cage access from levels 18 to 21. Level 16 and Level 17 also have adit

accesses from surface, mainly for ventilation. A rail decline from Level 18 to Level 19 provides the

ability to move material and supplies between levels. Levels below 18 are accessed by apique (vertical

shaft) hoists and skip system that allows transport of material. There are other apiques that transport

supplies to the lower levels.



Source: SRK, 2020

Figure 16-4: Marmato Level 18 with Main Haulage (Mined Out Panels in Cyan)

16.1.2 Reconciliation

Reconciliation information was not provided to SRK and is not carried out on a regular basis. SRK

recommends that production information be reconciled to the mine plan on a regular basis to ensure

the mine plan is predicting appropriate tonnes/grades. Within a known mining area, the tonnes/grades

mined should be compared to the tonnes/grades in the block model. If there are continuous

discrepancies between the mined material and the predicted mine plan, modifications to the mine plan

process should be made to more accurately predict future mining.

16.1.3 Dilution

CGM calculates and tracks the planned dilution, which is calculated by the following formula:

𝑃𝑙𝑎𝑛𝑛𝑒𝑑 𝐷𝑖𝑙𝑢𝑡𝑖𝑜𝑛 = 𝐶𝑢𝑡𝑡𝑖𝑛𝑔 𝑊𝑖𝑑𝑡ℎ − 𝑉𝑒𝑖𝑛 𝑊𝑖𝑑𝑡ℎ

𝐶𝑢𝑡𝑡𝑖𝑛𝑔 𝑊𝑖𝑑𝑡ℎ× 100

Figure 16-5 shows the dilution as provided by CGM. The dilution has increased since 2009 due to a

shift towards higher production with a high of 31.6% dilution in 2017. CGM has introduced better drilling

practices and training to control dilution. Since 2017, the planned dilution has decreased to 20.2% in

2019.



Source: CGM, 2020

Figure 16-5: UZ Planned Dilution

16.2 Geotechnical

The geotechnical data base and stability approaches used at Marmato fulfill the PFS requirements.

The standards used by Marmato for data collection are broadly consistent with industry standards.

The PFS geotechnical field data collection program was developed for obtaining information for both

RMR (Bieniwasky,1989) and Barton Q’-systems (Barton,1974). Both rock mass characterization

methods were used to determine the MDZ rock mass quality (RMQ), which was used to support the

PFS underground mine design and define the type of ground support required for each underground

excavation. All mine design parameters are based on analytic empirical methods, which are

acceptable for a PFS project level design only. More detailed stability modeling should be implemented

at FS and prior to construction. SRK recommends conducting a 3D numerical modelling to determine

the effect of the mine sequence on the overall stope stability and underground infrastructure. Also,

special attention on the major fault interpretation needs to be considered as part of the FS

geotechnical drilling program.

The following sections summarize the key sections of the PFS Geotechnical Technical Report

(SRK, 2020).



16.2.1 Geotechnical Data Base

From June 26, 2018 to March 4, 2020 Marmato’s exploration team, under SRK guidance, conducted

a geotechnical diamond drill hole program composed of nine (NQ/HQ) drill holes totaling 4,505.9 m.

Each drill hole was designed to examine RMQ and structural features in and around the mineralized

zone at different depths and orientations. Drillholes were drilled at varying orientations into the

hangingwall, footwall and mineralized rock. The field investigation included the drilling of core,

structural feature measurements, geotechnical core logging and core sample collection for laboratory

strength testing. In addition to the geotechnical drilling program, a total of 12 exploration drill holes

(4,307 m) were validated and used to support the geotechnical domains. Figure 16-6, shows the MDZ

PFS stope designs and the geotechnical drill hole coverage.

Source: SRK, 2020

Figure 16-6: Location of Geotechnical Drill Holes (As-Builts and MDZ Design Shown)

As part of the geotechnical PFS program (SRK,2020) , Marmato carried out a laboratory testing

program which included 117 uniaxial compressive strength tests (UCS), 90 multiaxial compressive

strength tests (TCS), 46 direct shear tests applied to various discontinuities (DSS), 75 indirect tensile

strength test (BTS, Brazilian tests), 151 elastic constant measurement (Young’s modulus and

Poisson’s ratio) and 200 dry density tests.

To determine the PFS rock mass fabric, Marmato considered 564 valid data sets obtained from drill

holes MT-IU-002, MT-IU-009, MT-IU-015, MT-IU-017 and traverse mapping conducted at Level 21.

The collected data was plotted in a 2D stereogram using Dips, v.8.001 (RocscienceTM,2020).

Televiewer data obtained for drill hole MT-IU-053, was used for data validation only, and to confirm

the structural orientation obtained from traditional alpha and beta angles measurements.



16.2.2 Engineering-Geology

Based on the geotechnical investigations, SRK observed that the Transition and MDZ is mostly

dominated by a unique lithology (P1 Porphyry). However, laboratory tests and RQD values showed

that P1 Porphyry can be subdivided into two geotechnical subdomains called P1A and P1B.

Figure 16-7 shows the location of the geotechnical subdomain and the Marmato Transition and MDZ

stopes locations.

Source: SRK, 2020

Figure 16-7: Geotechnical Subdomains

The geotechnical subdomain P1A, has been characterized as Good Rock Class II (Bieniawski, 1989).

Geotechnical logging showed RQD equal to 83% ± 4%, and a dominant spacing between 300 mm and

1,000 mm. Structural core logging indicates that most structures have joint apertures less than 1 mm

where the dominant wall strength can be considered as hard joint contacts. In terms of intact rock, field

estimated strength indicated that most of the core logging would be equivalent to R4 strength index

(50 MPa to 100 MPa), which was confirmed by laboratory tests (UCS of 80 ± 23 MPa). It was hard to

find a good correlation between intact rock strength and alteration. However, it was observed that most

of the failure mechanisms corresponded to splitting and multiple splitting associated to the micro

fracturing observed in the matrix of the intact rock. In terms of RMQ, SRK estimated the Bieniawski,

1989 RMR (RMR89) of 63 ± 7, which is equivalent to the geological strength index (GSI) equal to 58 ±

10. Based on Barton’s 1974 Q’ System, statistical assessment identified an average Q’ equal to 10.

The geotechnical subdomain P1B has been characterized as Very Good Rock Class I (Bieniawski,

1989). Geotechnical logging showed RQD values between 90% and 100% and spacing between 300

mm and 1,000 mm. Structural core logging indicates that most structures have joint apertures less

than 1 mm where the dominant wall strength can be considered as hard rough joint contacts. In terms

of intact rock, field estimated strength indicated that most of the core logging would be equivalent to

R5 strength index (100 MPa -200 MPa), which was confirmed by laboratory tests (UCS of 158 ± 19



MPa). SRK estimated the Bieniawski, 1989 RMR89 equal to 78 ± 17, which is equivalent to the

geological strength index (GSI) equal to 73 ± 20. Based on Barton’s 1974 Q’ System, statistical

assessment identified Q’ equal to 33.3.

In terms of rock mass fabric, SRK observed that there is not a significant structural set difference

between HW-FW and ore. SRK noted that there are no clear lithological, structural or RMQ boundaries.

Therefore, the structural domain has been considered as a unique structural domain for the MDZ,

which includes HW, FW and ore. This assumption is valid for a PFS project level. Future drilling

programs for FS-level design should confirm and/or adjust this assumption. Structural assessment

revealed the existence of three structural sets, as shown in Table 16-3. Figure 16-8 shows a plan view

of the mine layout and the PFS structural domains.

Table 16-3: Summary of Structural Sets

Domain Set Dip (°) Dip Direction (°) Variability Limit (°)

MDZ

Set 1A 76 034 15

Set 1B 80 202 20

Set 2A 70 350 12

Set 2B 75 170 10

Set 3 20 180 10

Source: SRK, 2020

(a) Structural sets Red lines: Major Faults

1: principal regional stress Source: SRK, 2020

Figure 16-8: Structural Domains

(

a

)



SRK observed that the structural sets orientations are well correlated with the major faults’ orientations.

For example, structural set 1B has a similar orientation to Pantano primary fault and set 2A can be

associated to Obispo faults. Set 1A is influenced by Cascabel, Criminal, Santa Ines and F2 faults.

16.2.3 Stope Stability Assessment

The design PFS stope is 30 m high, 30 m long and 10 m wide, and is located in the P1B geotechnical

subdomain. These dimensions provide acceptable stope stability for the PFS level design. Empirical

stability charts suggest that side walls are in the “unsupported transition zone”, which could involve

increased dilution from sidewalls when stopes are fully opened.

To determine the “effective unsupported span” SRK used the Bieniawski (1993) stand-up time

empirical method. Stand-up time is a function of rock mass properties and excavation technique. Two

geotechnical subdomains were assessed, P1A and P1B. SRK’s conclusion is that the stope span (10

m) is acceptable for stability in both subdomains P1A and P1B.

In term of stope ground support, the empirical stope design charts, indicate that the proposed PFS

stope dimensions will not require systematic ground support to maintain stability. Only spot bolting will

be required due to the formation of specific wedges. To determine technical specifications for ground

support, SRK completed a kinematic assessment using Unwedge 5.0 (Rocscience, 2019). Additional

refined assessments should be completed at the FS level design. SRK recommends that Marmato

perform numerical simulations for a better understanding of the potential for wedge formation during

mucking.

16.2.4 Dilution

Dilution was estimated using the Clark and Pakalnis (1992) method. This method predicts the quantity

of unstable wall rock for a given RMQ and stope size. The parameters plotted on the dilution chart are

the stability number N’ versus hydraulic radius based on case histories of dilution. The dilution is

estimated as an ELOS. The estimated ELOS indicates that MDZ is unlikely to have significant dilution

beyond blasting overbreak due to the good rock mass quality. This empirical approach is valid for PFS

design. Future FS level design should include a numerical simulation for adjusting the ELOS based

on the mine sequence.

16.2.5 Paste Fill Strength Estimation

To estimate the pastefill strength at a PFS level, SRK accepted the analytic solutions developed by

Mitchell et al, 1982. The model was developed to estimate the factor of safety (FoS) of a stope upon

exposure of unsupported backfill. Based on the Mitchell approach, SRK estimates a backfill UCS

strength of 1 MPa for single face fill exposure during mining of the adjacent secondary stope in high

stress conditions. To reach the required pastefill strength for primary and secondary stopes, laboratory

tests conducted by Paterson and Cooke indicted that 7% cement will be sufficient for obtaining high

strength pastefill in 7 days, and 4% cement will be sufficient for low strength pastefill.

At least one day of pastefill curing should be allowed to set the plug prior to completing stope filling. In

secondary stopes where the pastefill will never be exposed, sufficient binder is required to prevent

liquefaction of the pastefill during mining operations.

Based on the Paterson and Cooke test program, the majority of the tailings consist of the silicate

minerals Quartz (18.9%), and the Feldspars Albite (38.2%), Orthoclase (12.5%) and Anorthite (12.0%).



These minerals are inert and do not participate in the hydration dynamics of the cementitious reactions.

As such, they are considered good fillers for backfill material. The remaining tailings consist of the

phyllosilicate minerals Mica (Annite, 8.9%) and Chlorite (9.5%). Annite has weak bonds between the

internal sheet structures of the mineral allowing for failure planes and crack propagation pathways,

potentially lowering paste strength. The magnitude of this effect depends on the weathering of the

material as well as the size fraction of the particles. It has been observed that contents above 5% can

affect the strength of the backfill. The effect of Chlorite is also dependent on the formation of the

mineral, but the internal sheet structures are generally held together much more firmly than that of

Annite, and therefore is not usually an issue in backfill applications.

The water analysis shows that the decant water mainly contains trace amounts of alkali sulfates with

some alkali chlorides. The chloride content (49 mg/L) is within acceptable limits for concrete use and

will not delay the cement setting in the backfill.

There is no large presence of any other metals or problematic compounds reported in the water

analysis (Paterson and Cooke, 2020). As such, the process water is considered acceptable for backfill

purposes.

SRK requested Paterson and Cooke conduct the strength testwork with the objective of evaluating the

feasibility of using the tailings as a pastefill to reach the targeted values of 1.0 MPa for single face

exposure, and 0.5 MPa strengths for low stress conditions at 14-day cure dates. Table 16-4,

summarizes the UCS tests results.

Table 16-4: Uniaxial Compressive Strength Test Results

Mix Binder Content As-Cast Mass Concentration

W:B Ratio UCS (KPa)

7 days 14 days 28 days 56 days

1 11.5% 75.5%m 2.8 943 1,055 1,446 -

2 6.0% 75.7%m 5.6 - 510 621 TBD

3 4.5% 74.1%m 7.8 - 321 347 TBD

4 3.0% 74.1%m 11.7 132 192 230 -

Source: Paterson and Cooke, 2020

16.2.6 PFS Ground Support Requirements

To estimate the PFS ground support requirements SRK used the Norwegian Method of Tunneling

support techniques, modified by Grimstad and Barton (1993). Based on this method, SRK estimated

the PFS ground support for various excavations as shown in Table 16-5.



Table 16-5: Ground Support Requirements

Duration Excavation Type

Dimensions Ground Support type

Bolting spacing

(m)

Depth (m)

Width (m)

Height (m)

600 700 800 900 1000 1100

Long Term Excavations

Decline 5.5 5.6 Spot Bolting (20%, Total Excavation)

Systematic 25 mm Diameter Bolts That Are 2 m Long On 1.2 m Square Spacing and 15 by 15 Steel 5 mm Welded Wire Mesh

1.2

Ramps 5.5 5.5

Main Access 5.0 5.0

Shop Run Around Loops 5.5 5.5 Spot Bolting and Meshing as Needed (10% Of Excavations)

Main Conveyor Ramp 5.5 5.5

Medium Term (One Year)

Foot Wall Access/Haulage Drives

5.0 5.0

Spot Bolting (5% Of Excavations)

Spot Bolting (5% Of Excavations)

1.2 Cross Cuts (Option 1) 4.5 4.5

Cross Cuts (Option 2) 6.5 6.0 Spot Bolting and Meshing (10% Of Excavations)

Systematic Bolting (Above) Cross Cuts (Option 3) 6.5 7.0

Short Term (Less Than One Year)

Stope Access/In Stope Drifts

4.5 4.5 Spot Bolting (10% Of Excavations)

Stope Access/In Stope Drifts

5. 5.0 Spot Bolting (10% Of Excavations)

Production Shafts 5 Systematic Bolting (Above) 1.2

Raise Bore (No Access) 5

Spot Bolting (5% Of Excavations) 4.5

Blast Rise (No Access) 3 3

Emergency Access

5 Diameter

Spot Bolting (5% Of Excavation) Plus Systematic Fiber Reinforced Shotcrete 5 cm Thick

2 4.5

Diameter

Source: SRK, 2020

16.2.7 Sill Pillar design

To estimate the sill pillar dimension SRK used an analytic solution, which included the pillar dimensions

assuming the under-hand stope will have a backfill gap and provide no confinement to the sill pillar.

Based on analytic solution for a 30 m stope span, SRK estimates that a 9.5 m thick sill pillar in

necessary. The pillar is located approximately at 800 m depth and has a FoS of 1.5. The sill pillar could

be optimized and or recovered at the FS project level, which could result in a potential opportunity for

additional ore recovery by using numerical simulations, reducing the rock mass uncertainties and

accepting FoS of 1.3. For a FS level design, numerical simulation should consider the mine induced

stresses, which the analytic solutions have not considered.

16.2.8 Critical Infrastructure Stability Assessment

The stability study for critical infrastructure (e.g., crusher station, underground workshop, transfer

station, long term access and conveyor tunnel) were assessed at PFS level only and should not be

implemented without a more detailed investigation. A simple 2D numerical simulation indicates that

the average distance between stopes and the crusher station (approximately 40 m) does not affect the

stability of the crusher station. At PFS level, the proposed crusher dimension is acceptable. However,

more detailed studies should be implemented in the FS. SRK recommends the following activities:

• Conduct specific geotechnical drilling to characterize the rock mass and structural conditions

• Adjust the major fault model to confirm that no secondary or major fault will affect the crusher

station



• Conduct a 3D numerical simulation to study the effect of the mining induced stresses on

crusher station stability

• Define the required ground support before initiating the civil engineering

• Study the stability and required ground support of the bins

In terms of the underground workshop infrastructure, the stability assessment was conducted using a

tributary area method. The method assumed that the workshop station is located about 750 m deep

and assumed FoS of 1.5. The stability assessment indicates the following:

• Pillar width = 9 m

• Pillar height = 7 m

• Acceptable hydraulic radius (HR) depending of the pillar length

o HR (30 m pillar length) = 2.7



Assuming the maximum bay width is 7.5 m, SRK anticipates needing systematic bolting of 2 m long

bolts, 25 mm in diameter, and spaced 1.2 m. Also, 150 by 150 mm steel welded wire mesh (5 mm

diameter) with 5 cm fiber reinforced shotcrete.

The conveyor tunnel route selection was considered a key part of the PFS design. To select a suitable

tunnel route, high level geological, geotechnical, hydrological, hydrogeological and structural factors

were taken into consideration. Special attention was given to the effect of the modeled major faults on

the tunnel stability. To assess the potential effect of the major faults and expected rock mass quality,

SRK considered the following criteria:

• Reduce the exposure of the tunnel to major faults

• Tunnel trajectory should cross perpendicular to major faults

• Avoid faults shear zones

• Avoid crossing highly clayed materials

Based on these criteria, a feasible tunnel trajectory was identified as shown in Figure 16-9.



Source: SRK, 2020

Figure 16-9: Conveyor Tunnel Trajectory

SRK identified six geotechnical drill hole locations, which should be drilled prior to tunnel construction

see (Table 16-6). Figure 16-10 shows a plan view with the tunnel trajectory, major faults and the FS

proposed geotechnical drilling program.

Table 16-6: FS Geotechnical Drilling Plan (Tunnel Investigation)

Hole ID Easting (m) Northing (m) Elev. (masl) Azimuth (°) Dip (°) Length (m)

MT_IU_CR 1,163,552 1,097,722 1,001 310 -68 280

MT_IU_T01 1,164,557 1,097,589 1,098 250 -85 180

MT_IU_T02 1,164,240 1,097,623 1,094 280 -78 230

MT_IU_T03 1,164,245 1,097,622 1,095 100 -75 220

MT_IU_T04 1,163,985 1,097,670 1,163 280 -85 350

MT_IU_PT01 1,164,854 1,097,822 1,008 230 -10 100

Source: SRK, 2020



Source: SRK

Figure 16-10: FS Drill Hole Location (Tunnel Investigation)

16.2.9 Limitations and Gaps

Although, the limitations described in this section are acceptable for a PFS level design, SRK

recommends addressing the geotechnical model’s limitations and updating the design approach for

FS level design. The main limitations and gaps are listed below:

• The structural domains were defined based on four oriented drill holes and limited structural

data obtained from the underground mine. SRK recommends that Marmato include acoustic

or optical televiewer in future exploration drill hole programs and acquire more structural data

from the underground excavations.

• The PFS major structural model should be updated based on specific drill holes and

underground mapping. Additional large-scale faults not included in the PFS geological model

which have been identified should also be considered. For a FS, shear zones and breccia

zones should be investigated.

• The empirical design charts do not include the effect of the induced stresses, due to the mine

extraction sequence. For FS, SRK recommends reviewing the PFS mine designs and with

further consideration given to mining sequences.

• The PFS ground support requirements are based on an empirical approach. SRK

recommends using a more detailed approach to determine the required ground support at an

FS level.

• The hydrogeological model has not been integrated into the geotechnical model; SRK

considers it important to include the potential pore pressure in future stability models.

• 3D numerical simulations should be conducted as part of FS study.



• The crusher station and underground workshop stability has been defined based on empirical

approaches, which is acceptable at a PFS level. However, the PFS design should not be

implemented without a detailed engineering design.

• Long term access to critical infrastructure such as the crusher station and workshops, should

be investigated in more detail. Specific geotechnical drill holes and numerical simulations

should be considered for FS.

• Specific monitoring information from the current mine operations, such as excavation

displacements and excavation damage, should be collected to be used for future numerical

model calibrations.

16.2.10 Feasibility Study Recommendations

SRK’s opinion is that the current geotechnical data is adequate for a PFS-level design. However, to

advance the design to an FS level, additional characterization data will be required to reduce

uncertainty in the data variation. SRK provides the following recommended characterization activities

to advance the design to a final design level:

• Specific geotechnical drill holes to characterize the rock mass parameters around the critical

underground infrastructures should be drilled.

• Geotechnical core logging and televiewer data in specific exploration drill holes should be

collected and analyzed. The selection of exploration drill holes should be strategically placed

in footwall infrastructure areas and planned stope mining areas to provide sufficient data to

statistically verify the range of expected ground conditions. This includes:

o Collecting RMR/Q data

o Collecting structural orientation data

o Updating the structural model and geotechnical models

o Updating the mine design parameters

• Complete specific geotechnical drill holes to characterize the rock mass parameters around

the conveyor tunnel

• Update the major faults model

• Conduct pre-mining in situ stress measurements

• Collect tiltmeter measurements to confirm that there is minimal subsidence above the

transition zone

• Develop a Ground Control Management Plan with a Triggered Action Response Plan (TARP)

in case of excessive deformation or drift collapse or seepage inflows

• Perform mine scale stress analyses of the planned stoping sequence to evaluate:

o The appropriateness of infrastructure setback distances

o Anomalous stress conditions resulting from the stoping sequence

o Variations in stress and groundwater conditions

o Spatial variations in rock mass strength conditions

• A mine scale hydrogeological pore pressure model should be developed that considers

locations and hydraulic conductivity of specific fault structures as they intersect drifts and

stopes

• Long term access to critical infrastructure should be evaluated, such as the crusher station

and workshops. Specific geotechnical drill holes and numerical simulations should be

considered for the FS.



• Specific monitoring information from the current mine operations should be reviewed for future

numerical simulations and calibrations, such as excavation displacements and excavation

damage. It is important to re-evaluate the ground support based on depth and mine

sequencing.

16.3 Hydrogeology and Mine Dewatering

16.3.1 Hydrogeological Conditions

The mine area is located in the hydrogeological regional area of Magdalena Cauca. The region is

comprised of igneous and metamorphic rocks with limited groundwater storage capacity and hydraulic

conductivity (IDEAM, 2013). The porphyry units represent the main hydrogeological units in the mine

area, with a low hydraulic conductivity and limited groundwater storage capacity. Groundwater flow is

compartmentalized within structural blocks with limited hydraulic communication across fault

boundaries due to fault gouge, weathering, or an offset of geological units.

Previous field campaigns were performed by KP in 2011 and 2012 (Knight Piésold, 2012). The current

field campaign is being performed by SRK, the program began in early 2020 and primarily consisted

of packer isolated interval testing, monitoring well and VWP installations in underground coreholes or

locations distal to the mine area.

SRK analyzed all available hydrogeological data, including:

• KP 172 packer tests, three piezometers installed underground, 11 piezometers installed from

ground surface

• SRK 70 packer tests completed in four coreholes (MT-IU-053, MT-IU-063, MT-IU-065 and MT-

IU-066) and VWP’s installed in MT-IU-063 and MT-IU-066

• SRK 2020 field campaign currently in progress focusing on underground targets from surface

locations

• Historic mine water discharge records in 2017 and 2019

• Historic water level measurements in the hydrogeologic study area

• Water level measurements in vibrating wire piezometers MT-IU-063 and MT-IU-066 installed

in 2020

Hydrogeological Units and Faults

Saprolitic coverage and intrusive fracture rock are the two major hydrogeological units defined in the

Marmato mine area. The saprolite is formed by clay material that has weathered on the top of intrusive

rock units. It can reach over 30 m in some locations and is usually dry in the mine area. The intrusive

fractured rock corresponds to dacite and andesite porphyry stocks and a sheeted pyrite veinlet system

associated with intermediate argillic and propylitic alteration.

The Amaga Formation is present E and SE of the mine area in disconnected pockets and more

extensively W of the mine within the Supia River Valley. The depths of the Supia formation are

unknown. Alluvial sediments are present along creeks and rivers.

The Criminal Fault, which is located to the N of Cerro los Burros and runs toward the Cauca River on

Quebrada Los Pantanes, appears to form compartmentalization of geological blocks with limited

communication across the fault (Knight Piésold, 2012). This fault represents a contact point between

the Dacite Porphyry P1 in Cerro Lo Burros and Dacite Porphyry P2 and Graphitic Schist exposed to



the N. The Criminal Fault is 15 to 20 m wide, has a clayey alteration and has a low water filtration flow.

However, to the north of this contact, horizontal boreholes produce water flow from 7 to 8 L/s in Level

17. Also, water flow has been reported in the same location on Level 21.

From the 2020 drilling program, it is apparent that high-permeability zones (hydraulic conductivity

greater than 0.1 m/d), which may be associated with Fault 2 and Fault 1-3, were encountered in the

vicinity of the planned mine at depths of 600 to 800 m below ground surface (bgs).

Hydraulic Parameters

Measured hydraulic conductivity of the bedrock groundwater system in the vicinity of the mine were

obtained from 2012 and 2020 investigations and are presented in Figure 16-11 and Table 16-7.

Table 16-7: Measured Bedrock Hydraulic Conductivity Values at Depth

Depth (m bgs) Number of Tests

Hydraulic Conductivity (K) (m/d)

Top Bottom Geometric Mean Minimum Maximum

0 200 73 4.05E-02 8.06E-04 8.49E+00

200 400 28 2.35E-02 6.38E-04 1.81E-01

400 850 127 1.30E-02 1.78E-04 1.18E+00

850 1,500 22 1.16E-03 1.12E-04 9.89E-03

Source: SRK, 2020

Source: SRK, 2020

Figure 16-11: Distribution of Measured Hydraulic Conductivity Values vs. Depth



These results indicate:

• Large variability (2 to 3 orders of magnitude)

• General trend of decreasing hydraulic conductivity with depth

• Very low permeability at approximately 850 m bgs where lower part and bottom of the deep

underground mine is planned.

The zone of enhanced hydraulic conductivity values at depths of 600 to 800 m below the ground

surface corresponds to fractured zones associated with Fault 2 and Fault 1-3 in the mine area. Based

on these findings, bedrock units were grouped into four hydrogeological units varying with depth. Base

Case distribution of hydraulic conductivities is based on geometric mean at discrete depth intervals

and was used to predict expected mine inflow.

Measured Water Levels and Direction of Groundwater Flow

Measured water levels show elevations from 661 to 2,022 m Magna Sirgas/Colombia West coordinate

system (MSCW), following the topography at 100 m depth in most of the locations outside the mine

area. Estimated water table and direction of groundwater flow for pre-mining conditions are shown in

Figure 16-12.

Water levels vs. depth and vertical hydraulic gradients were measured in two deep coreholes MT-IU-

063 and MT-IU-066 drilled in 2020 in the vicinity of the planned MDZ mine. Two strings of grouted-in

transducers were installed in these two coreholes, indicating:

• Water level elevations measured from 1,006 to 1,012 m MSCW in MT-IU-063

• Water level elevations measured from 1,065 to 1,072 m MSCW in MT-IU-066

• Mixture of upward and downward hydraulic gradients in both coreholes

Measured water levels and groundwater flow data that were used for model calibration are discussed

in Section 1.3.



Source: SRK, 2020

Figure 16-12: Estimated Water Table and Direction of Groundwater Flow



A depressurization zone was detected in the underground piezometers where the water levels have a

horizontal trend. The shape or extent of the depressurization zone is currently unknown. On a more

regional scale, the groundwater flows W to E, following the topographical gradient to the Cauca River,

located at 692 m MSCW in the proximity of the mine. The Cauca River represents the main discharge

for the hydrogeological system. A conceptual hydrogeological cross-section is shown in Figure 16-13.

Source: SRK, 2020

Figure 16-13: Conceptual Hydrogeological Cross-Section

Current Mine Dewatering

The mine has a series of pumps and tanks from Level 22 to Level 19, where the water is pumped to

the processing plant to be used as makeup water. Each level collects the water produced in its

developments in addition to infiltration coming from levels above. The water is briefly stored in a tank

and pumped to the next level above. Water from Level 16 and Level 17 is collected by gravity and

discharges to Level 18 and through to the process plant. Figure 16-14 shows a simplified scheme of

the current dewatering system.



Source: CGM, 2019

Figure 16-14: Scheme of Current Dewatering System

Preliminary flow measurements were conducted at mine adits and are presented in the environmental

baseline report prepared by CGM (date unknown). The measurement values vary from a total of 6.3

to 15.9 L/s in summer and winter. However, there is no information about which mine levels were in

operation during the flow measurements.

A measurement record of total mine water discharge is available from January 2017. The measured

monthly average of total dewatering in Marmato mine is 37 L/s, varying from 26.8 to 46.4 L/s. Strong

seasonal trends were not observed; however, a decrease of approximately 16 L/s can be detected in

the last 12 months. Figure 16-15 shows the total water discharge from the Marmato mine.



Source: CGM, 2019 Data from July to December 2017 is not available.

Figure 16-15: Measured Mine Water Discharge



The dewatering flow is a combination of groundwater inflows and water content in the hydraulically

placed backfill material (60 to 65% of water). According to Marmato operational personnel, the

contribution of the backfill material is 7 to 14 L/s, depending on the number of hydraulic backfill

equipment in operation. Therefore, the average fresh groundwater inflow into the mine could vary from

23 to 30 L/s. A significant amount of groundwater flow comes from the north section of Level 17

(crossing the Criminal Fault) where horizontal boreholes contribute 7 to 8 L/s.

The existing dewatering system fits the current needs for the mine operations at Marmato mine.

16.3.2 Descriptions of Numerical Groundwater Model

SRK developed a preliminary 3D numerical groundwater flow model using the MODFLOW-USG code,

based on available climatic, geological and hydrogeological data. Historic and proposed underground

mine developments were incorporated into the model. Plan-view and modeled cross-section are

shown in Figure 16-16 and Figure 16-17.

Source: SRK, 2020

Figure 16-16: Model Grid Discretization – Plan View



Source: SRK, 2020

Figure 16-17: Modeled Cross-Section

The simulated hydrogeological units and their hydraulic parameters are shown in Table 16-8.

Table 16-8: Simulated Hydraulic Parameters

Hydrogeologic Unit

Horizontal Hydraulic

Conductivity

Vertical Hydraulic Conductivity

Specific Storage

Specific Yield

m/day 1/m (-)

Alluvium 5 1 1.00E-06 0.10

Amaga Formation 0.1 0.05 1.00E-06 0.05

Amaga Formation (SW of Domain) 0.5 0.01 1.00E-06 0.05

Saprolite 0.5 0.5 1.00E-06 0.05

Bedrock depth<200 m bgs 0.04 0.04 1.00E-06 0.01

Bedrock 200 to 400 m bgs 0.023 0.023 1.00E-06 0.01

Bedrock 400 to 850 m bgs 0.013 0.013 1.00E-06 0.01

Bedrock depth>850 m bgs 0.0012 0.0012 1.00E-06 0.01

Source: SRK, 2020

The World Climate dataset was used to define precipitation within the model domain. Mean annual

precipitation (MAP) varies from about 2,700 mm/yr in the mountains to 1,930 mm/yr in the vicinity of

the Cauca River, with values of 2,400-2,500 mm/yr in the mine area. The relationship between

recharge coefficient vs. ground surface elevation was established during the model calibration process

to match the measured water levels and groundwater flows. Within the mine area, the recharge

coefficient was calculated to be within the range of 0.35 to 0.45 (or from 35% to 45% of MAP).

The rivers, creeks, and groundwater outflows as simulated by the numerical model are shown in Figure

16-18.



Source: SRK, 2020

Figure 16-18: Simulated Rivers, Creeks and Groundwater Outflows

Both historic and future underground developments (tunnels, drifts, declines, stopes, vents and ramps)

were simulated using drain cells, which extract groundwater from the model depending on the water

level elevation above the development and the assigned conductance. The elements of simulated

planned underground mine are shown in Figure 16-19.



Source: SRK, 2020

Figure 16-19: Simulated Developments and Stopes for Planned Mine

The stopes were assumed as “non-restrictive” to groundwater inflow developments due to their size

(similar to the model cell) and this assumption was used for Base Case predictions; i.e., drain cell

conductance for the stope was kept unchanged throughout the LoM.

In SRK’s opinion, planned backfill of the stopes most likely will not restrict groundwater inflow, because

some empty space between the stope roof and backfill would exist due to backfill subsidence. The

Base Case used for predictive simulation does not consider the restrictive effect of the backfilling.

However, the potential of the backfill to restrict groundwater inflow was evaluated, in order to generate

a minimum mine inflow estimate, and was simulated as an additional sensitivity run. In this run, the

unrestrictive stope (drain) conductance of 10,000 m2/d was replaced by a restrictive conductance value

of 0.1 m2/day, six months after mining of the stope starts (six months was the stress period length set

in the model).

The groundwater model was calibrated to:

• Pre-mining water levels (2011 to 2012)

• Historic mine discharge rates (2017 to 2019)

• Water levels installed in piezometers installed recently (2020)

In SRK’s opinion, the developed preliminary groundwater model is conservatively calibrated to the

limited available flow and water level data and is suitable for predictive simulations of mine dewatering

requirements at the PFS level.



16.3.3 Results of Predictions by Groundwater Model

The numerical model was used to predict:

• Passive inflow to the existing and planned underground mine

• Propagation of drawdown as the result of planned dewatering under mining conditions

• Changes in groundwater discharge to river and creeks under mining conditions

The model predictions were made under four scenarios shown in Table 16-9.

Table 16-9: Predictive Scenarios Evaluated by Groundwater Model

Scenario Bedrock Hydraulic Conductivity Inflow to Backfill

Base Case - Expected Inflow Geomean Unrestricted

Sensitivity Runs

Maximum Inflow Average Unrestricted

Minimum Inflow Geomean Restricted

Permeable Faults Geomean (separate for bedrock and faults) Unrestricted

Source: SRK, 2020

The first scenario, in SRK’s opinion, represents the Base Case or expected scenario, while the second,

third, and fourth scenarios were completed as sensitivity analyses to identify possible range of

groundwater inflow scenarios to existing and planned underground mines.

The model predicts:


159 L/s) is expected from the upper levels above 730 m where elevated hydraulic conductivity

values of the bedrock groundwater system were measured.

• Mine inflow to the MDZ planned mine below 730 m is predicted to be lower (15 L/s with an

upper limit of 34 L/s) due to reduced measured hydraulic conductivity with depth.

• Total maximum planned mine discharge is predicted to be up to 88 L/s with a possible range

from 61 to 167 L/s.


is predicted to be up to 111 L/s with a possible range from 89 to 168 L/s.

• The major sources of mine inflow are the depletion of groundwater storage and capturing of

groundwater discharge to surface water bodies (i.e. streams). The model does not predict the




Predicted expected groundwater inflows to the underground mine (Base Case) are shown in

Figure 16-20. Sensitivity runs (maximum inflow, minimum inflow, and inflow under permeable faults

scenarios) are summarized in Figure 16-21.



Source: SRK, 2020

Figure 16-20: Predicted Mine Inflow During Years 2021 to 2032 (Base Case)

Source: SRK, 2020

Figure 16-21: Comparison of Total Predicted Dewatering Requirements for Base Case and Sensitivity Scenarios



In SRK’s opinion, completed predictions of mine inflow are conservative, given that the model:

• Is based on the extrapolation of the measured hydraulic conductivity values in mine areas for

the entire model domain, including topographic high areas outside of the mine area, where

measured water levels are high and hydraulic conductivity values are most likely lower than in

the mine area

• Use of high recharge from precipitation to calibrate the model to measured water levels,

combined with geomean K values in discrete depth intervals that are derived from measured

K values in the mine area

• Use of calibrated conductance values that reproduce measured inflow to the existing, relatively

shallow mine for the simulation of groundwater inflow to the deep underground developments

of the planned mine

• Simulate no restriction of groundwater inflow to the backfilled stopes for Base Case, maximum

inflow, and permeable fault scenarios

Predicted water table at the end of mining on the west to east cross-section through the mine area for

Base Case is shown in Figure 16-22. Predicted drawdown (water table changes) at end of the mining

in plan-view for Base Case is shown in Figure 16-23.

Source: SRK, 2020

Figure 16-22: Predicted Water Table and Direction of Groundwater Flow at End of Mining Shown on West to East Cross-Section through Mine Area



Source: SRK, 2020

Figure 16-23: Predicted Drawdown at End of Mining (Base Case, End of 2032)

The model predicts:

• The lowering of the water table in the mine area of up to 140 m and drawdown propagation of

up to 2 km away from the mine, assuming a 10-m drawdown extent.

• Creating of a “bulb” of depressurization around the planned underground mine.



• Table 16-10 summarizes maximum mine inflows, reduction of groundwater discharge to the

rivers and creeks from the current conditions and potential inflow from the Cauca River.

Table 16-10: Predicted Maximum Mine Inflows and Reduction of Groundwater Discharge to the Rivers and Creeks Under Different Scenarios

Scenario

Predicted Maximum Inflow (L/s)(1)

Maximum Reduction of Groundwater Discharge to

Rivers and Creeks from Current Conditions (L/s)

Maximum Inflow from Cauca

River (L/s) Total Existing

Mine Planned

Mine

Expected Inflow - Base Case

111 40.5 88 74 0

Maximum Inflow Scenario

168 12.2 167 147 0

Minimum Inflow Scenario

90 40.5 61 49 0

Permeable Faults Scenario

150 35.6 138 117 0

Note: (1) Maximum Inflow is predicted at different time Source: SRK, 2020

16.3.4 Hydrogeological Uncertainties


indicates that several uncertainties remain in the understanding of the hydrogeological conditions in

the proximity of the mine. These uncertainties include:

• Hydrogeological role of faults: Extend outside of the mine area and connect to the Cauca River

• Hydraulic properties of bedrock outside of the mine area (especially in areas of topographic

high, where shallow depth to water table has been measured)

• Nature of elevated hydraulic conductivity in the mine area at depth from about 600 to 800 m

bgs (elevation approx. between 700 and 900 m MSCW) in the vicinity of Fault 2 and Fault 1-3;

planned conveyor decline and egress ramp plan to intersect Fault 2 at multiple

locations/elevations

• Recharge estimates from direct precipitation and potential recharge enhancement in the mine

area as result of artisan mine developments

• Limited availability of hydrogeological data related to:

o Groundwater inflow to the current mine (changes in time, spatial and vertical distribution,

and water usage for mining)

o Water table elevation and water level changes due to passive mine dewatering and

seasonal changes in precipitation

• Hydrogeological role of backfill material and possibility to reduce groundwater inflow to mine

developments and stopes

• Groundwater chemistry with depth

Considering the hydrogeological uncertainties mentioned above, SRK recommends planning for a

mine pumping capacity of 168 L/s corresponding to the maximum inflow scenario.



16.4 Upper Zone Mining

16.4.1 Stope Optimization

Stope optimization was completed using Maptek Vulcan’s implementation of Alford Mining Systems’

Stope Optimization program and the PFS block model. Optimization parameters are determined based

on the mineralization geometry, current equipment constraints and geotechnical constraints. These

parameters are listed below.

Veins Stope Optimization

Veins area optimization parameters are as follows:

• CoG of 2.23 g/t Au

• 1 m minimum mining width

• 5 m block heights

• 10 m block length along strike

• Angled stopes based on vein geology

• Elevation between 950 m and 1,300 m

Optimization results are assessed and combined to form 35 m long by 50 m high stopes. Figure 16-24

shows the results of the stope optimization.

Source: SRK, 2020

Figure 16-24: Stope Optimization results for the Veins (Section looking Northeast)

Transition Zone Stope Optimization

Transition Zone optimization parameters are as follows:

• Cut-off of 1.91 g/t Au



• 25 m mining length

• 15 m block heights

• 15 m block width

• Angled stopes based on vein geology

• Elevation 950 m to 1,050 m

Figure 16-25 shows the optimization results for the Transition Zone.

Source: SRK, 2020

Figure 16-25: Stope Optimization Results for the Transition (Section Looking Northeast)

Figure 16-26 shows the overall mine design based on the stope optimization results from the Veins

and the Transition (collectively, the UZ) colored by grade.

Source: SRK, 2020

Figure 16-26: Stope Optimization Results for the UZ Colored by Au Grade (Section Looking Northeast)



16.4.2 Mine Design

Veins Stope Design

The stopes generated from the stope optimizer were amalgamated into 35 m long by 50 m high blocks.

The minimum mining width is 1 m and the stope widths vary between 1 m and 3 m. This is consistent

with the stope size CGM is currently using to mine the veins. There are certain gaps in the blocks

generated by the optimizer, and these are not included as part of the mining plan. These gaps could

potentially be mined as marginal material.

Veins stopes are from Level 22 (950 m elev.) to Level 16 (1,300 m elev.).

Veins Development Design

2.2 m by 2.2 m drifts are designed to follow the veins where possible. A diluted grade was calculated

for this development material and tonnages/grades for development material are tracked separately

from stope material in the production schedule.

Sublevels are typically 1.4 mW x 2.2 mH and it is calculated into the production rate of the stopes.

Vertical raise development at the ends of the stopes is 1.4 mW x 1.4 mL and it is also calculated into

the production rate for the stopes.

Transition Stope Design

The Transition will use a modified long hole stoping method in a bottom up orientation. Stopes are 15

mW x 15 mH x 25 mL. Stopes are mined in a primary-secondary sequence with pastefill as the primary

backfill and waste rock as secondary backfill, as shown in Figure 16-27. Waste rock will primarily come

from development. Where waste rockfill is insufficient or unavailable, hydraulic sandfill will be used as

backfill. Top and bottom transverse access are driven into the stopes and the stopes are drilled down

using a jumbo fitted with a longhole drill (Figure 16-28). The stopes are mucked with an LHD and

loaded into trucks or an orepass. Rail carts haul the ore to the apiques on the main haulage levels. A

7 m sill pillar is left between Level 21 and 22. While pastefill is available, additional work is required in

the next level of study to explore the feasibility of recovering the pillar.

Source: SRK, 2020

Figure 16-27: Transition Mining Method (Magenta is Primary and Cyan is Secondary)



Source: SRK, 2020

Figure 16-28: Plan View of Transition Development

16.4.3 Production Schedule

The production schedule is based on the productivity rates shown in Table 16-11.

Table 16-11: Productivity Rates

Activity Type Dimension Rate

Veins Stopes 18 t/d

Veins Development 2.2 m x 2.2 m 1.19 m/d

Veins Level 2.2 m x 2.2 m 1.19 m/d

Transition Stopes 400 t/d

Transition Development (Access, Ramp and Crosscuts)

3.5 m x 3.5 m 4 m/d

Transition Backfill 220 m3/d

Apique 2 m x 4 m 0.6 m/d

Transition Vent Raise 3.5 m x 3.5 m 1 m/d

Source: SRK, 2020

Veins stope production rate includes the mining of the stope and backfilling. The production rate for

the Transition stopes includes the drilling and blasting of the stope. The backfill rate for the Transition

is 220 m3/d.

The apique development rate of 0.6 m/d assumes that the apiques are in use in the levels above.

The production schedule targets a total production of 1,500 t/d or 525,000 t/y (based on 350 days per

year) to the mill. A gradual ramp up is planned for 1,100 t/d (385,000 t/y) in 2020, 1,250 t/d (437,500

t/y) in 2021, 1,400 t/d (490,000 t/y) in 2022 and 1,500 t/d in 2023. The Transition accounts for 400 t/d

while the remaining UZ production comes from the veins. LoM for the Veins is 12 years for a total of

3.81 Mt at 4.37 g/t Au. LoM for the Transition zone is 11 years for a total of 1.33 Mt at 3.56 g/t Au.



There is a 2 Mt/y permit limit to material moved for the mine. The production schedule prioritizes the

production of the MDZ, therefore the production in the UZ from 2024 onwards is reduced to respect

this limit.

Combined UZ production is 5.14 Mt at 4.16 g/t Au. The production schedule was completed using

iGantt scheduling software from Minemax. Table 16-12 shows the upper mine total production

schedule and Table 16-13 show the total development schedule. Figure 16-29 shows the production

schedule colored by time period.



Table 16-12: Marmato Upper Mine Total Production Schedule

PFS Schedule Unit

2020 2021 2022 2023 2024

Summary 1/1/2020 4/1/2020 7/1/2020 10/1/2020 1/1/2021 4/1/2021 7/1/2021 10/1/2021 1/1/2022 4/1/2022 7/1/2022 10/1/2022 1/1/2023 4/1/2023 7/1/2023 10/1/2023 1/1/2024 4/1/2024 7/1/2024 10/1/2024

Total Ore Tonnes t 30,179 63,460 96,282 96,251 109,369 109,373 109,375 109,370 122,529 122,503 122,512 122,506 131,252 131,247 131,253 131,251 121,935 123,194 123,204 123,202

Ore Au (g/t) g/t 3.73 3.70 3.61 3.65 3.90 3.98 4.05 4.02 3.97 3.96 3.76 4.13 3.97 3.88 4.09 3.96 3.84 3.78 4.07 4.14

Ore Ag (g/t) g/t 19.82 17.27 16.14 16.31 16.57 14.90 14.00 12.57 12.51 12.75 13.36 13.65 14.35 13.90 14.58 16.62 16.22 15.65 16.42 16.70

Au_Oz Oz 3,617 7,558 11,184 11,305 13,717 13,988 14,244 14,138 15,652 15,597 14,797 16,272 16,732 16,390 17,251 16,705 15,049 14,987 16,126 16,399

Ag Oz Oz 19,230 35,229 49,972 50,458 58,262 52,384 49,215 44,596 50,548 51,111 52,622 53,745 60,536 58,639 61,532 70,115 63,574 61,992 65,054 66,159

Waste Tonnes (t) t - 3,535 18,417 29,648 32,544 32,401 31,720 30,591 29,850 30,947 24,290 26,448 16,101 10,838 10,808 8,819 1,247 - - -

Total Development Length (m) m - 413 1,503 2,037 2,258 2,153 1,994 1,959 1,936 2,063 2,035 1,897 1,534 1,002 1,000 801 108 - - -

Total Material Moved t 30,179 66,994 114,699 125,899 141,914 141,774 141,095 139,961 152,378 153,450 146,802 148,954 147,353 142,086 142,062 140,070 123,182 123,194 123,204 123,202

Total Material to Surface t 30,179 66,994 107,770 125,899 134,663 141,774 141,095 133,503 135,130 135,625 137,607 148,954 147,353 134,862 136,780 128,507 123,182 123,194 123,204 123,202

Production Ore Breakout

All Stope Ore Tonnes t 30,179 60,136 85,539 80,601 93,568 98,383 100,025 101,482 112,666 109,901 106,108 110,930 117,717 127,775 127,784 128,639 121,640 123,194 123,204 123,202

All Stope Ore Au (g/t) g/t 3.73 3.75 3.64 3.63 3.80 3.98 4.10 4.02 3.92 3.96 3.89 4.21 4.05 3.91 4.12 3.96 3.84 3.78 4.07 4.14

All Stope Ore Ag (g/t) g/t 19.82 17.56 16.86 17.15 16.92 15.18 14.22 12.79 12.90 13.11 14.06 14.19 14.87 14.01 14.76 16.52 16.20 15.65 16.42 16.70

Veins Stope Tonnes t 30,179 60,136 76,853 78,218 68,141 68,316 71,740 69,987 85,072 84,790 83,110 86,938 91,999 92,878 92,652 93,472 86,594 88,269 88,079 88,071

Veins Stope Ore Au (g/t) g/t 3.73 3.75 3.68 3.64 3.78 4.12 4.18 4.16 4.17 4.16 4.00 4.33 4.16 4.05 4.22 4.10 4.07 4.10 4.34 4.37

Veins Stope Ore Ag (g/t) g/t 19.82 17.56 17.24 17.12 17.25 17.62 15.69 14.79 14.60 14.97 15.25 15.54 16.03 16.08 17.04 18.08 18.62 17.71 18.78 18.59

Trans Stope Tonnes t - - 8,685 2,383 25,427 30,067 28,284 31,495 27,595 25,112 22,997 23,992 25,718 34,896 35,132 35,167 35,045 34,925 35,124 35,131

Trans Stope Ore Au (g/t) g/t - - 3.27 3.33 3.83 3.66 3.91 3.72 3.12 3.27 3.48 3.79 3.63 3.53 3.83 3.61 3.27 2.97 3.41 3.57

Trans Stope Ore Ag (g/t) g/t - - 13.53 18.29 16.02 9.62 10.46 8.36 7.69 6.85 9.73 9.30 10.69 8.52 8.75 12.37 10.21 10.45 10.52 11.98

PFS Schedule Unit

2025 2026 2027 2028 2029 2030 2031 2032 Totals

Summary 1/1/2025 1/1/2026 1/1/2027 1/1/2028 1/1/2029 1/1/2030 1/1/2031 1/1/2032

Total Ore Tonnes t 388,538 452,030 409,596 351,864 386,605 388,893 445,183 91,713 5,144,667

Ore Au (g/t) g/t 4.40 4.30 4.34 4.38 4.30 4.18 4.43 4.22 4.16

Ore Ag (g/t) g/t 16.98 15.75 15.23 14.42 14.33 14.43 17.43 21.69 15.41

Au_Oz Oz 54,909 62,550 57,148 49,501 53,388 52,318 63,448 12,446 687,417

Ag Oz Oz 212,110 228,950 200,573 163,133 178,125 180,418 249,410 63,953 2,551,644

Waste Tonnes (t) t - - - - - - - - 338,205

Total Development Length (m) m - - - - - - - - 24,692

Total Material Moved t 388,538 452,030 409,596 351,864 386,605 388,893 445,183 91,713 5,482,872

Total Material to Surface t 388,538 452,030 409,596 351,864 386,605 388,893 445,183 5,302,186

Production Ore Breakout

All Stope Ore Tonnes t 388,538 452,030 409,596 351,864 386,605 388,893 445,183 91,713 4,997,095

All Stope Ore Au (g/t) g/t 4.40 4.30 4.34 4.38 4.30 4.18 4.43 4.22 4.17

All Stope Ore Ag (g/t) g/t 16.98 15.75 15.23 14.42 14.33 14.43 17.43 21.69 15.55

Veins Stope Tonnes t 287,393 335,352 292,553 259,173 246,341 248,792 402,365 91,713 3,749,179

Veins Stope Ore Au (g/t) g/t 4.51 4.54 4.51 4.65 4.86 4.68 4.57 4.22 4.37

Veins Stope Ore Ag (g/t) g/t 19.46 18.93 18.32 17.25 18.24 18.67 18.49 21.69 17.88

Trans Stope Tonnes t 101,145 116,678 117,043 92,691 140,264 140,100 42,819 - 1,247,916

Trans Stope Ore Au (g/t) g/t 4.06 3.62 3.92 3.60 3.31 3.30 3.14 - 3.56

Trans Stope Ore Ag (g/t) g/t 9.94 6.62 7.51 6.52 7.47 6.91 7.47 - 8.55

Source: SRK, 2020 Note: Numbers may not sum due to rounding.



Table 16-13: Marmato Upper Mine Total Development Schedule

PFS Schedule Unit

2020 2021 2022 2023 2024

Summary 1/1/2020 4/1/2020 7/1/2020 10/1/2020 1/1/2021 4/1/2021 7/1/2021 10/1/2021 1/1/2022 4/1/2022 7/1/2022 10/1/2022 1/1/2023 4/1/2023 7/1/2023 10/1/2023 1/1/2024 4/1/2024 7/1/2024 10/1/2024

Development Ore Breakout

All Development Ore Tonnes (t) t - 3,323 10,744 15,650 15,801 10,989 9,350 7,888 9,862 12,602 16,404 11,576 13,534 3,472 3,469 2,612 296 - - -

All Dev Ore Au (g/t) g/t - 2.97 3.38 3.77 4.52 3.94 3.53 3.97 4.63 3.96 2.92 3.33 3.25 3.03 3.08 3.79 4.84 - - -

All Dev Ore Ag (g/t) g/t - 11.94 10.42 11.93 14.50 12.41 11.65 9.65 7.96 9.55 8.85 8.42 9.82 9.54 8.01 21.25 23.81 - - -

Veins Development Ore Tonnes (t) t - 2,341 9,374 8,375 6,028 6,028 2,532 3,790 2,467 2,784 4,186 452 4,250 3,472 3,469 2,612 296 - - -

Veins Dev Ore Au (g/t) g/t - 2.83 3.35 4.11 6.89 4.66 3.50 3.34 3.89 3.92 2.66 2.62 3.10 3.03 3.08 3.79 4.84 - - -

Veins Dev Ore Ag (g/t) g/t - 8.84 10.45 13.39 22.86 15.50 16.47 13.14 14.24 19.53 10.84 24.08 17.11 9.54 8.01 21.25 23.81 - - -

Trans Development Ore Tonnes (t) t - 983 1,369 7,275 9,773 4,962 6,818 4,097 7,396 9,818 12,218 11,123 9,284 - - - - - - -

Trans Dev Ore Au (g/t) g/t - 3.31 3.57 3.39 3.05 3.07 3.54 4.55 4.87 3.97 3.00 3.36 3.32 - - - - - - -

Trans Dev Ore Ag (g/t) g/t - 19.34 10.20 10.25 9.34 8.65 9.86 6.43 5.87 6.72 8.17 7.78 6.47 - - - - - - -

Lateral Development Breakout

Veins Development -DEV-2.2x2.2 (m) m - 329 971 974 1,161 1,353 1,290 1,346 1,270 1,286 1,333 1,241 1,093 1,002 1,000 801 108 - - -

Trans Ramp Development -RMP-3.5x3.5 (m) m - - 59 59 183 318 299 49 - - - - - - - - - - - -

Veins Level -LVL-2.2x2.2 (m) m - - 31 135 109 - - - - - - - - - - - - - - -

Trans Level Access -ACC-3.5x3.5 (m) m - 1 135 258 187 - - 286 311 316 - - - - - - - - - -

Trans Ore Xcut -XCT2-4x3.5 (m) m - - 36 149 153 137 216 103 196 260 341 324 246 - - - - - - -

Trans Waste Xcut -XCT1-3x3 (m) m - 40 62 170 412 345 189 155 140 176 362 332 195 - - - - - - -

Ventilation Drift -VNT-3.5x3 (m) m - - 20 11 - - - 7 7 8 - - - - - - - - - -

Vertical Development Breakout

Apique Development (m) m - 18 78 64 - - - - - - - - - - - - - - - -

Ventilation Raise -RAR-3x3 (m) m - - 19 7 - - - 12 12 17 - - - - - - - - - -

Backfill Breakout

Veins Backfill (m3) m3 7,521 14,978 19,139 19,532 17,350 18,011 19,143 18,985 23,168 23,529 24,225 26,607 28,482 29,137 29,370 29,959 27,706 28,312 28,187 28,135

Trans Pastefill m3 - - - - 4,109 12,374 12,058 8,927 3,722 1,336 4,308 11,147 10,377 8,676 12,330 5,801 16,545 9,435 - 5,218

Trans Rockfill m3 - - 3,590 - 3,757 - - 3,346 8,937 9,236 4,764 - - 3,743 2,737 5,991 - 4,191 12,121 9,445

Others

Vent Alimak Project m 25 92 210 53



PFS Schedule Unit

2025 2026 2027 2028 2029 2030 2031 2032 Totals

Summary: 1/1/2025 1/1/2026 1/1/2027 1/1/2028 1/1/2029 1/1/2030 1/1/2031 1/1/2032

Development Ore Breakout:

All Development Ore Tonnes (t) t - - - - - - - 147,572

All Dev Ore Au (g/t) g/t - - - - - - - 3.68

All Dev Ore Ag (g/t) g/t - - - - - - - 10.75

Veins Development Ore Tonnes (t) t - - - - - - - 62,456

Veins Dev Ore Au (g/t) g/t - - - - - - - 3.88

Veins Dev Ore Ag (g/t) g/t - - - - - - - 14.34

Trans Development Ore Tonnes (t) t - - - - - - - 85,116

Trans Dev Ore Au (g/t) g/t - - - - - - - 3.53

Trans Dev Ore Ag (g/t) g/t - - - - - - 8.12

Lateral Development Breakout

Veins Development -DEV-2.2x2.2 (m) m - - - - - - - 16,559

Trans Ramp Development -RMP-3.5x3.5 (m) m - - - - - - - 966

Veins Level -LVL-2.2x2.2 (m) m - - - - - - - 274

Trans Level Access -ACC-3.5x3.5 (m) m - - - - - - - 1,494

Trans Ore Xcut -XCT2-4x3.5 (m) m - - - - - - - 2,160

Trans Waste Xcut -XCT1-3x3 (m) m - - - - - - - 2,579

Ventilation Drift -VNT-3.5x3 (m) m - - - - - - - 53

Vertical Development Breakout:

Apique Development (m) m - - - - - - 159

Ventilation Raise -RAR-3x3 (m) m - - - - - - 67

Backfill Breakout:

Veins Backfill (m3) m3 91,691 106,867 93,894 83,156 79,052 79,898 129,203 29,531 1,154,769

Trans Pastefill m3 33,698 33,201 28,972 25,839 31,331 34,946 4,592 - 318,941

Trans Rockfill m3 11,771 12,024 22,650 11,254 24,186 25,959 14,517 - 194,218

Others:

Vent Alimak Project m

Source: SRK, 2020 Note: Numbers may not sum due to rounding.



Source: SRK, 2020

Figure 16-29: Production Schedule Colored by Time Period

16.4.4 Mining Operations

Stoping

CAF stopes are mined using stopers and jacklegs. Once the level access is driven, raises (tambores)

are driven on either side of the stope. A sublevel is driven laterally and used as a drilling platform,

where 1.7 m to 2.3 m slices are drilled up into the back. After blasting and bolting, the stope is mucked

out either using slushers for higher grade stopes or skid steer loaders/microscoops for lower grade

blocks. Once the stope is mucked out, concrete barricades are built on either side of the stope and

filled with unconsolidated hydraulic fill.

• Drilling and blasting

• Mucking or slushing to a raise and removing the mineralized material from the raise and

hauling by train along the production level

• Sand backfill

• Repeat the cycle on top of the sandfill

In stopes where the back is fractured and not amenable to vertical drilling and blasting, the slice is

mined horizontally along strike.

Transition stopes are 15 m by 15 m and will be mined using a modified longhole stoping method in

two sections. The first section is from Level 21 up to Level 20 and the second section is from Level 22

to Level 21 with a pillar in between. The Transition will be mined in a primary-secondary sequence.

The stope is drilled from a top access using a jumbo fitted with an adapter for long hole drilling. A slot

will first be drilled and blasted, before the rest of the stope is slashed into the slot. The material will be

mucked from the bottom using a remote scoop. The primary stopes will be backfilled using pastefill

and the secondary stopes are filled with waste rock.



Development

Development in the Veins is completed using jacklegs. Level accesses are 2.2 m by 2.2 m, sublevels

vary depending on the width of the vein. The apiques will be extended down to level 22.

A 3.5 m by 3.5 m ramp is used to access the Transition stopes. The level access and transverse drifts

are the same size as the ramp. Development for the Transition zone is done using a jumbo.

Haulage

All is hauled using rail haulage on the level. The main haulage level is Level 18. Material from Levels

16 and 17 is brought to Level 18 via an orepass. Material on Level 19 is hauled up using the incline

and material below Level 19 is brought up via the shaft hoist.

10 tonne trucks will be used in the transition zone to haul ore to the orepass, which is then loaded to

rail carts and brought to the apiques.

Backfilling

Backfilling is completed using unconsolidated hydraulic backfill from the plant. Currently,

approximately 55% of the mill tailings are returned to the mine as backfill. There are four tailings

pipelines going underground to different levels with each pipeline having a capacity of 290 m3/d. The

plant’s backfill capacity is 715 m3/d. One limitation of the backfill system is the lack of a surge tank, so

there is limited catch up possibility should a delay occur. Waste rock generated underground that is

not hauled out is used as backfill. The hydraulic backfill system is shown in Figure 16-30.

A paste plant is planned to be installed near the existing plant to provide cemented pastefill to the

Transition Zone. Engineering work is being done by Lara Consulting and is expected to be complete

by October 2020 and construction will begin thereafter.



Source: CGM, 2019

Figure 16-30: Marmato Hydraulic Backfill System

16.4.5 Ventilation

As the mining operations are below 1,500 masl, Colombian regulations state that the minimum airflow

for diesel equipment is 4 m3/min per horsepower (hp) which relates to 0.09 m3/s per kW of engine

power to ensure gaseous and aerosol contaminants from diesel equipment are sufficiently diluted

which is a typical minimum design value for many ventilation systems (although the typical dilution

rates are usually presented as between 0.06 m3/s per kW and 0.08 m3/s per kW). The value of 0.09

m3/s per kW has been used to determine the airflow in the ramps/haulage routes, and on the mining

levels where diesel equipment is used.

Colombian regulations also state that the minimum airflow per worker is at least 0.05 m3/s. This airflow

requirement is typically used in areas without diesel equipment, as the requirements for ventilating

diesel equipment will far exceed this value.

The ventilation system for the upper area draws approximately 139,000 cubic feet per minute (kCFM)

fresh air in from Level 18 and Level 17. Approximately 96 kCFM is exhausted from the portals on Level

16. There is a 43 kCFM difference in the intake and exhaust which CGM attributes to leakage in the

system and the artisanal mining above. For levels below 18, including Level 21, the fresh air is pulled

down the apiques using 30 hp fans onto the level. Secondary ventilation is provided by 15 hp fans and

vent tubing to the face. The flow in the level is mainly controlled by vent doors. In each level, the air

flows west to an exhaust raise, where the air goes up to Level 16. A portion of Level 21 will be an



exhaust collection area (Figure 16-31) that will draw exhaust toward the new exhaust raise and then

to the surface. Figure 16-32 shows the new ventilation system. A new exhaust fan system would be

installed within either the surface exhaust portal drift or on Level 21. The location of the exhaust fan

system would depend on the location of the electrical feed, and the competency of the rock in the area

of the installation.

Source: SRK, 2020

Figure 16-31: Level 21 Exhaust Collection Area

Source: SRK, 2020

Figure 16-32: New Exhaust System

16.4.6 Mine Services

Pumping

The mine has an operating system of ditches, sumps, and small pumps that control water on the

individual mine level and pump water to the main pump system.

The main pumping system used in the mine is a staged system of 10,000 to 15,000 liter sumps/tanks

and pumps that move water from lowest levels of the mine at Level 21 up to the mine portal where the

water is used in the mine processing plant. On Level 21, at the bottom of the currently developed mine,



there is a storage tank with three Krebs pumps that pump through two-4 inch and one-6 inch pipelines

up to Level 20. On Level 20, another tank and pump system with three pumps moves the water to

Level 19 to a concrete lined sump with two Goulds 5500 slurry pumps that pump through 4 inch

pipelines to the portal Level 18 to the process plant water tank. There is redundancy built into the

system with extra pumps on Level 19 and additional locations to place pumps on Level 20. The pump

system handles on average 37 L/s with a range from 26.8 L/s to 46.4 L/ s.

Electrical Supply

The existing project electrical system includes an 8.1 MVA main project substation with six

transformers that provide power to the mine and mill. The mine system power is provided at 33 kV

through transformers that transform power to feed the mine surface and underground facilities. The

three mine related transformers and loads they feed are summarized as follows:

• Transformer 1 (2,000 KVA) steps the 33kV power down to 13.2kV and feeds the three mine

substations that in turn feed the compressors, pumps and offices/shops at 440 VAC

• Transformer 2 (2,000 KVA) feeds the mine at 13.2kV through three separate mine

transformers that in turn feed the various mine levels, hoists, pumps, and mine equipment.

The equipment operates on 440 VAC

• Transformer 4 (1,250 KVA) and 5 (630 KVA) feed two compressors each at 440 VAC

The largest loads at the mine are the compressors, pumps, and hoists which account for approximately

65% percent of the mine load.

Health and Safety

The mine has a mine phone system and emergency egress is provided through stairs in the shaft

declines and a series of ladders to the surface portal level. The mine has a health and safety response

plan and miner safety training sessions for instruction on proper work procedures and safe work

activities.

Manpower

Currently there are 1,158 personnel working at the site; this includes underground staff, process plant

staff and other support staff. The mining staff is approximately 67% of the total staffing. CGM projects

an increase in manpower primarily in the mine underground operations over the next five years (Table

16-14).

Table 16-14: Manpower by Department

Department 2019 2020 2021 2022 2023

Underground Operations 777 868 890 910 910

Process Plant 34 34 34 34 34

Support 217 217 217 217 217

Administration 59 59 59 59 59

Others 71 71 71 71 75

Total 1158 1249 1271 1291 1295

Source: CGM, 2019



Equipment

The mine utilizes a large number of jackleg drills and small electric and air operated equipment. There

are also small diesel microscoops and skid steer loaders. The mine is adding additional small drills

and microscoops in the future.

Development and production of the Transition Zone will be performed by contractors. Jumbos, LHDs

and trucks will be used in the Transition.

Table 16-15 shows the current equipment list as provided by CGM.

Table 16-15: Marmato Equipment List

Equipment Amount

Skid Steer Loaders 27 Jacklegs/Stopers 313 Microscoops 6 Locomotives 27 Mine Pumps 18 Hydraulic Fill Pump 18 Winches 5 Slushers 91 Fans 83 Railcars 215 Compressors 9 Pumps 33 Transformers 12 Electric Grid 6 Jumbo T1D 2 LHD ST2G 2 10t Truck 1

Total Equipment 868

Note: Some equipment is owned by contractors Source: CGM, 2020

16.4.7 Recommendations

SRK notes the following recommendations and opportunities for the UZ mine.

• The UZ mine is currently achieving mined grades that are lower than the grades predicted by

the model. This is due to the mining of veinlets and disseminated material instead of only the

vein. There are on average 50 to 60 panels in production at one time across six levels, but

only one geologist for every two levels. In SRK’s opinion, there are not enough geologists to

mark the face for development or production. Additionally, the turnaround time for assays has

been three to four days, which does not give the geologist time to make decisions on the

heading. The lab has recently moved to operating three shifts per day in an effort to reduce

the turnaround time and CGM has stated they are planning some upgrades to the lab. SRK

recommends that CGM prioritize grade control and mining discipline to improve performance

with regard to mined grades.

• Modified longhole stoping is a new mining method for the site. SRK recommends using a 3D

cavity scanning system to survey the completed Transition stopes. This will allow the site to

evaluate how closely the mined stope shapes align with the planned stope shapes, which is

information that necessary to complete meaningful stope reconciliations. 3D scans can also

be used in the veins for better grade control.



• An opportunity to improve the current planning process is to use 3D modelling software and

block models. This will allow planners to identify additional veins in 3D space and will facilitate

reconciliation. SRK notes that CGM is in the process of transitioning to 3D modeling software.

Using a 3D Gantt type scheduling software to generate a mine plan will more accurately show

periods of potential higher and lower grade.

16.5 MDZ Mining

The MDZ area is currently in the exploration phase and has not been developed. Mineralization is

located approximately 600 to 1,200 m below the surface (480 masl to 1,100 masl). Based on

geomechanical information and mineralization geometry, an underground longhole stoping method

(LHS) is suitable for the deposit. Cut-and-fill vein mining will continue above the MDZ area, but it is not

a method that will be used in the MDZ area.

The MDZ deposit will be mined in blocks where mining within a block occurs from bottom to top with

the use of paste backfill. Sill pillars are left in situ between blocks. The backfill will have sufficient

strength to allow for mining adjacent to filled stopes without the need for dip pillars. The stopes will be

10 m wide and stope length will vary based on mineralization grade. A spacing of 30 m between levels

has been used. In the top mining block, a higher grade core is extracted first, mined from bottom to

top. Subsequently, additional stopes are mined from the bottom of the block up, mining adjacent to

(but not underneath) backfilled stopes.

The mine will be accessed by a decline drift with mineralization transported from stopes via truck to

an underground crusher and then to surface by conveyor. Internal intake and exhaust raises will be

developed using raisebore machines and air will flow into dedicated intake and exhaust ventilation

drifts to surface. A new 4,000 t/d process facility using gravity concentration and cyanidation of the

gravity tailings will be constructed to process material from the MDZ. In addition, a new DSTF will be

constructed to receive approximately 55% of the total LoM tailings from the plant. The other 45% of

tailings will go back underground into the mine as cemented paste backfill.

16.5.1 Stope Optimization

Based on geomechanical information and mineralization geometry an underground LHS method is

suitable for the deposit. Paste backfill will be used to allow for a high recovery of economic material.

Stopes are sized to be large enough to take advantage of bulk mining methods, yet small enough to

minimize dilution. A variety of stope sizes were evaluated using stope optimizer software to make this

determination.

Figure 16-33 shows the resource block model blocks above an Au CoG grade of 1.61 g/t which have

been classified as Measured and Indicated. There are pockets of higher grade material, particularly in

the upper portion of the deposit. Generally, the deposit is approximately 500 m along strike and 125

m in width. This model formed the basis of the stope design.



Source: SRK, 2020

Figure 16-33: Resource Model – AU blocks (g/t) above Mining CoG (Looking North)

Stope optimization within Vulcan software was used to determine potentially economically minable

material. Stope walls were vertical and wall dilution was not applied at the optimization stage, however

the CoG used for design was elevated to account for the expected dilution.



increase/decrease in CoG has a material impact on the material available for design. Figure 16-34 and

Table 16-16 shows stope optimization results for various cut-off grades using a stope size of 10 m

wide by 30 m high.



Source: SRK

Figure 16-34: Undiluted Stope Optimization Results for Varying Cut-off Grades

Table 16-16: Undiluted Stope Optimization Results for Varying Cut-off Grades

Cutoff (g/t)

Material Category

Au (g/t)

Ag (g/t)

Tonnes (kt)

Contained Au (Oz)

Contained Ag (Oz)

1.00 MI 2.04 3.16 37,173 2,441,304 3,771,621

1.50 MI 2.68 3.73 21,048 1,813,107 2,525,177

1.75 MI 2.89 3.93 17,572 1,632,289 2,218,412

2.00 MI 3.09 4.12 14,682 1,457,334 1,945,971

2.25 MI 3.28 4.29 12,217 1,288,234 1,686,879

2.50 MI 3.48 4.47 9,977 1,115,480 1,435,249

3.00 MI 3.94 4.85 6,051 765,529 944,312

3.50 MI 4.53 5.24 3,190 464,441 537,581

4.00 MI 5.15 5.67 1,732 286,498 315,559

4.50 MI 5.89 6.07 903 171,151 176,388

5.00 MI 6.67 6.46 509 109,229 105,757

Source: SRK

Stope optimization results as discussed above are undiluted.

Stope optimization results using a 1.75 g/t Au cutoff were targeted for design work. The reserves CoG

is 1.61 g/t. As stope optimization results did not consider dilution, 8% dilution was factored into the

optimization cutoff which results in a stope optimization cutoff of 1.75 g/t Au (i.e. A 1.7 g/t stope, diluted

by 8%, will give the reserves CoG of 1.61 g/t). Higher grade stopes using 3.5 g/t stope optimization

results were designed as a first pass, with the lower grade stopes added as separate stopes. This

allowed for scheduling of higher grade stopes first.



16.5.2 Mine Design

Stopes are 10 m wide and 30 m high with varying length. Each stope has a 4.5 m by 4.5 m access

located at the bottom of the stope as shown in Figure 16-35. Top accesses are available on most

levels to give access to stopes on the next level and to allow for backfilling. For upper most stopes in

a block or where there is no mining above, it is assumed a hole can be drilled from adjacent

development into the stope for backfilling purposes. The stopes are drilled top down and rings are

blasted from the end of a stope toward the access. The blasted material is remotely mucked from the

stope access. A typical level is made up of approximately 40 stopes along strike.

Source: SRK

Figure 16-35: Stope Cross Section

A primary/secondary stoping sequence will be used, where on any given level, primary stopes must

be separated by a secondary stope. Extraction of the secondary stope can only occur after the two

immediately adjacent primary stopes have been mined, backfilled, and have had time to cure.

Backfilling will be an integral part of the LHS mining cycle, and a seven day cure time is planned.

The stope accesses are connected to a level access which is offset approximately 20 m away from

the end of the stopes. Each stope access typically connects to the level access except in cases where

stopes are small and long development is required to reach the stope. In those instances, a connection

from an adjacent stope is included in the design. This minimizes the amount of development; however,

it also limits the sequencing order.

The level accesses connect to the main ramp which is offset at least 75 m from stoping into the footwall.

The offset may be optimized during the FS, after numerical modelling. On the northeast side of each

level, the level access connects to an intake air ventilation raise and on the southeast side connects

to an exhaust air raise. Figure 16-36 shows a typical level section.



Source: SRK, 2020

Figure 16-36: Typical Level Section

Access and infrastructure development underground was designed to support the mining method and

sized based on mining equipment and production rate requirements. The crusher area was designed

by Ausenco with SRK orienting the various specified sized openings in the design. Figure 16-37 shows

the location of the bin/crusher and their layouts. Trucks will dump into the crusher at an elevation of

790 m. Material will go through the bin into the crusher and then be conveyed out of the mine.



Source: SRK, 2020

Figure 16-37: MDZ Mine Design (Rotated View Looking Southwest)

The decline from the plant site will be the main access to the MDZ for men/materials. This decline is

5.5 m wide by 5.6 m high, excavated at a grade of 17%. A schematic tunnel layout showing conveyor,

services, etc. in the decline is shown in Figure 16-38. There are two ventilation drifts (5 m by 5 m) near

the current UZ process facility. One of these ventilation drifts connects to the main ramp system and

will serve as secondary egress.



Source: Ausenco, 2020

Figure 16-38: MDZ Main Decline Cross Section

Tonnages/grades for the mine design were calculated based on the resource block model. Dilution

and recovery were added to the designed tonnage to account for unplanned stope dilution and

unrecoverable material within the stope as discussed in Section 15.

The MDZ design resulted in 14.6 Mt at an average grade of 2.85 g/t Au and 3.84 g/t Ag and is shown

in Figure 16-39. Figure 16-40 shows the stopes colored by Au grade. Table 16-17 summarizes the

mine design by activity type.



Source: SRK, 2020

Figure 16-39: MDZ Mine Design (Rotated View Looking Southwest)



Source: SRK, 2020

Figure 16-40: MDZ Mine Design, Colored by Au Grade



Table 16-17: MDZ Mine Design Summary – by Activity Type

Total Tonnes Moved (t) 16,362,907

Ore Breakout

Development Ore Tonnes (t) 1,044,054

Stope Ore Tonnes (t) 13,511,892

Development Breakout

Main Conveyor Ramp -RMC-5.5x5.6 (m) 1,680

Main Truck Ramp-RMT-5.5x5.5 (m) 3,650

Drift-FWA-5x5 (m) 7,116

Ventilation Drifts-VMR-5x5 (m) 2,202

Ventilation Connections-VCX-4.5x4.5 (m) 1,032

Stope Drifts Ore-DFA-4.5x4.5 (m) 19,285

Stope Drifts Waste-DFA-4.5x4.5 (m) 11,496

Vertical & Other Development Breakout

Raisebore 5m dia-RS1 (m) 437

Raisebore 5m dia-RS1 (count) 2

Raisebore 4.5m dia-RS2 (m) 403

Raisebore 4.5m dia-RS2 (count) 2

Blasted Raise-BRS-3x3 (m) 87

Blasted Raise-BRS-3x3 (count) 3

Bulk Excavation (m3) 16,570

Source: SRK, 2020

There are several known faults in the area, they range from several cm in width to several m in width.

The Sur and Ines faults impact the MDZ development design in several locations as shown in Figure

16-41. The development design is oriented to cross the faults as perpendicularly as possible. As faults

become more understood, appropriate measures should be taken to minimize the risk to the design in

terms of development rate and costs.



Source: SRK, 2020

Figure 16-41: MDZ Design with Sur and Ines Faults (Rotated View - Looking Northwest)

16.5.3 Production Schedule

The production schedule is based on the mine design and reserves discussed in previous sections.

Productivities were developed from first principles. Input from mining contractors, blasting suppliers

and equipment vendors was considered for key parameters such as drilling penetration rates, blast

hole size and spacing, explosives loading time, bolt and mesh installation time, etc. The rates

developed from first principles were adjusted based on benchmarking and the experience and

judgment of SRK.

The productivity rates used for mine scheduling are shown in Table 16-18, followed by a description

of the general and activity-specific parameters upon which the productivity rates are based.



Table 16-18: Productivity Rates

Activity Type Heading Type Dimensions Rate (1)

Drifting

Main Conveyor Ramp Single 5.5 m x 5.6 m 4.3 m/d

Main Truck Ramp Single 5.5 m x 5.5 m 4.3 m/d

Footwall Accesses Multiple 5.0 m x 5.0 m 6.9 m/d

Ventilation Drifts Single 5.0 m x 5.0 m 5.2 m/d

Ventilation Connections Multiple 4.5 m x 4.5 m 6.6 m/d

Top/bottom stope accesses Multiple 4.5 m x4.5 m 6.6 m/d

Stoping Stoping (2) - 2,1421 t/d

Vertical Development

Ventilation Raise Upper Block 5.0 m diameter 2.8 m/d

Ventilation Raise Lower Block 4.5 m diameter 2.8 m/d

Blasted Raise with escape way 3.0 m x 3.0 m 8.4 m/d

Other Bulk Excavation Various 100 m3/d

Backfill Paste Backfill (3) - 123 m3/hr

(1) All rates are per face. Multiple areas/faces are mined together to generate the production schedule (2) Includes drilling, blasting, and mucking for the slot and the stope. (3) Includes pour time and 3 days of lag (for pouring plug, waiting 3 days for cure, and pouring remainder) Source: SRK, 2020

General schedule parameters applicable to all underground mining activities are presented in

Table 16-19.

Table 16-19: Schedule Parameters for Underground Mining

Schedule Parameters Units Value

Annual mining days(1) days/year 365 Mining days per week days/week 7 Shifts per day shifts/day 3 Scheduled shift length hrs/shift 8

Scheduled Deductions

Shift Change hrs/shift 0.25 Travel Time hrs/shift 0.42 Equipment Inspection hrs/shift 0.25 Lunch Break hrs/shift 0.50 Equipment Parking/Reporting hrs/shift 0.50 Total scheduled deductions hrs/shift 1.92 Operating time (scheduled shift length less scheduled deductions) hrs/shift 6.08 Effective time (operating time reduced to a 50-minute hour, i.e., multiplied by 83.3%) hrs/shift 5.07

Source: SRK, 2020 (1) Actual operational mining days are 360. For simplicity the schedule has been completed assuming 365 with pro-rated productivity rates.

Key assumptions regarding ore and waste material characteristics are detailed in Table 16-20.

Table 16-20: Material Characteristics for Ore and Waste

Characteristic Units Value

In situ density t/m3 2.70

Swell % 40

Loose density t/m3 1.93

Source: SRK, 2020

For the purposes of developing productivity estimates, the ground support requirements detailed in

Table 16-5 were used.



Drifts

The main ramps systems will be developed with a twin-boom jumbo drilling 41 mm diameter blast

holes and 102 mm relief holes. All jumbo holes will be drilled 4.24 m in length, which allows for an

effective advance rate of 4.02 m per round. The drill pattern provides for 51 charged blast holes and

two uncharged relief holes. Drilling times were calculated based on average penetration rates of 1.4

m/min for 41 mm charged holes and 0.4 m/min for 102 mm reamed relief holes. A 10% redrill factor

was assumed.

Use of a bulk emulsion explosive was assumed at a powder factor of 1.08 kg/t. The blasting cycle time

considered mobilization, charging and tying in of holes, clean-up, and demobilization.

Loading will be performed with a 7.3 m3 (17 t) LHD that will transport blasted rock to remuck bays that

are spaced 250 m apart. Load, maneuver and dump times were considered and a 85% bucket fill

factor was assumed. The time associated with loading haul trucks at the remuck bays was accounted

for as an activity that is separate from the main ramp development.

During the pre-production period and for a large portion of the mine life, development waste rock that

is placed in a remuck bay will be loaded into trucks and hauled to the surface for use in construction

of the DSTF. After the DSTF is constructed, development waste will be placed in empty secondary

stopes.

Ground support will be installed as specified in Table 16-5. Time allowances have been included for

mobilization and setup, scaling, bolting/meshing/shotcreting as required, and demobilization.

Utility installation includes piping lines, ventilation tube, electrical cable, messenger cable, and leaky

feeder. Piping, ventilation and electrical utilities will be installed at the end of every other round.

Table 16-21 shows the development rates for the three main ramp types, which are all considered

long term development openings.

Table 16-21: Main Ramp Average Development Rate – Long Term Development Openings

Task Units Conveyor Ramp

(5.5 x 5.6 m) Truck Ramp (5.5 x 5.5 m)

Ventilation Ramp (5x5 m)

Drilling hrs/round 2.67 2.67 2.77

Blasting hrs/round 3.03 3.03 3.33

Mucking hrs/round 2.90 2.85 2.62

Ground Support hrs/round 6.34 5.70 3.49

Utilities/Services hrs/round 1.71 2.11 1.19

Blasting Clear Time hrs/round 0.5 0.5 0.5

Total Cycle Time hrs/round 17.14 16.87 13.90

Total Advance Rate m/day 4.28 4.35 5.28

Source: SRK, 2020

The footwall access, medium term development openings, will be developed much in the same way

as the long-term openings. Table 16-22 shows the development rates for these.



Table 16-22: Footwall Access Development Rate – Medium Term Openings*

Task Units Footwall Access (5x5 m) Ventilation Connection Drifts (4.5x4.5 m)

Drilling hrs/round 3.17 3.23

Blasting hrs/round 3.67 3.86

Mucking hrs/round 2.50 2.57

Ground Support hrs/round 3.24 3.47

Utilities/Services hrs/round 1.26 0.78

Blasting Clear Time hrs/round 0.5 0.5

Total Cycle Time hrs/round 14.33 14.41

Total Advance Rate m/day 5.12 5.10

Source: SRK, 2020 *Advance shown is for a single heading environment. Multiple heading environment using the same assumptions gives a rate of 6.9 m/d for footwall access and 6.6 m/d for ventilation connections.

The drift access openings will be developed much in the same way as the long-term openings.

Table 16-23 shows the development rates for these openings.

Table 16-23: Drift Access Development Rate – Short Term Openings*

Task Units Drift Accesses (4.5x4.5 m)

Drilling hrs/round 3.23

Blasting hrs/round 3.86

Mucking hrs/round 2.57

Ground Support hrs/round 3.51

Utilities/Services hrs/round 1.26

Blasting Clear Time hrs/round 0.5

Total Cycle Time hrs/round 14.94

Total Advance Rate m/day 4.92

Source: SRK, 2020 *Advance shown is for a single heading environment. Multiple heading environment using the same assumptions gives a rate of 6.6 m/d.

Stopes

After top and bottom stope development drifts are established, a slot will be developed at the far end

of the stope. The slot consists of a conventionally blasted drop raise and 28 fan-drilled holes that will

be slashed into the void that is created by the drop raise. Including the fan-drilled holes, the overall

dimensions of the slot will be 15 m wide by 6 m long by 25 m high.

All blasthole drilling for the slot will be at a diameter of 114 mm (4.5 inches) using an in-the-hole (ITH)

drill. A total of 50 holes will be required for the slot (22 holes for the drop raise and 28 holes for

slashing). The estimated penetration rate for the ITH drill is 0.75 meters per minute (m/min) and the

total drilling requirement is 1,066 m (including 10% re-drill).

Stopes will be 30 m in height by 10 m in width and will have varying lengths. An ITH production drill

will be used to fan drill the stope from the upper access drift. Blast holes will be 114 mm (4.5 inches)

in diameter and the estimated drill penetration rate is 0.75 m/min. The total drilling requirement is 206

m per ring (including 10% redrill) and the ore blasted per ring is 2,226 t.

Stope blasting will average 1.8 rings per day, the number of which will be dictated by the length of the

stope. Each three-ring blast will have a total of 39 charged holes (13 holes per ring). A bulk emulsion

product will be used, and the powder factor will be 0.34 kg/t. The estimated blasting cycle time includes

travel/set up, charging and tying in of holes, clean up, and demobilization.



Stope ore will be mucked with a 7.3 m3 (17 t) LHD that will transport blasted ore to re-muck bays that

will be located, on average, 135 m from the stope. Load, maneuver and dump times were considered

and an 85% bucket fill factor was assumed. Ore that is placed in a re-muck bay will be loaded into

trucks and hauled to the grizzlies feeding the underground material handling system; however, the

time associated with loading haul trucks is accounted for as a separate activity.

As shown in Table 16-24, the stope production rate is 2,142 t/d.

Table 16-24: Stope Production Rate

Task Units Slot Stope Total

Drilling hrs 44.6 11.2 55.8

Blasting hrs 9.6 13.2 22.8

Mucking hrs 22.8 90.0 112.8

Total Cycle Time hrs 77.0 114.4 191.4

Days days 3.2 4.8 8.0

Total Production Rate t/d 1,177 2,694 2,142

Source: SRK, 2020

Raisebored Raises

Two 5 m diameter intake/exhaust ventilation raises will be raisebored early in the mine life. One raise

is 237 m and the second is 200 m. The raisebore average advance rate is 2.8 m/d. The rate includes

drilling the pilot hole and reaming the vent raise. Loading will be performed with a 7.3 m3 (17 t) LHD

that will transport cuttings to a re-muck bay that will be located 75 m from the bottom of the raise. Load,

maneuver and dump times were considered, and an 85% bucket fill factor was assumed. Raisebore

cuttings that are placed in a re-muck bay will be loaded into trucks and hauled to the surface during

preproduction. Two additional 4.5 m raises will be needed later in the mine life. The same productivities

are assumed with the same material handling scheme.

Drop Raise with Escapeway

The ventilation connections will be 3 m wide by 3 m long by 30 m high. The advance rate is 8.5 m/day.

All blast hole drilling for the ventilation connections will be at a diameter of 114 mm (4.5 inches) using

an ITH drill. A total of 22 holes will be required for the drop raise (16 charged blast holes and six

uncharged relief holes). The estimated penetration rate for the ITH drill is 0.75 m per minute and the

total drilling requirement is 605 m (including 10% re-drill).

The drop raise will be removed in a series of three blasts using a bulk emulsion product. The first two

blasts will remove the bottom 13 m of the drop raise. The third and final blast will remove the remaining

7.5 m at the top of the drop raise. The blasting cycle time includes travel/set up time, charging and

tying in of holes, clean up, and demobilization.

Development and Production Schedule

The production and development schedule were completed using iGantt software. The production

schedule is based on the rate assumptions shown Table 16-25.

A delay of seven days was used prior to driving on paste fill or mining adjacent to a paste filled stope

to account for curing time. A 45 day delay to install manways in ventilation raises was also assumed.



The mining operation schedule is based on 365 days/year, seven days/week, with three eight hour

shifts each day. A production rate of 4,000 t/d (1.46 Mt/yr) was targeted with ramp-up to full production

as quickly as possible. The schedule timeframe is quarterly for four years and annually for the

remainder of the mine life.

Decline activities begin in October 2021 with mine development through Q4 2023. Stoping begins in

Q4 of 2024, with a one year ramp up period until the mine and plant are operating at full capacity.

Table 16-25 summarizes the MDZ production schedule.

Table 16-25: MDZ Production Schedule

Period Ore (t/d)

Ore Tonnes

(kt)

Ore Au

(g/t)

Ore Ag

(g/t) Au Oz Ag Oz

Waste Tonnes

(kt)

Development Length (m)

2021

Q1

Q2 - - - - - -

Q3 - - - - - -

Q4 - - - - - - 64 874

2022

Q1 - - - - - - 62 855

Q2 - - - - - - 63 865

Q3 - - - - - - 64 888

Q4 - - - - - - 95 1,129

2023

Q1 106 1,332

Q2 107 1,413

Q3 396 36 2.91 4.15 3,412 4,864 58 1,512

Q4 1,876 173 3.14 4.76 17,405 26,421 30 1,822

2024

Q1 2,716 247 3.18 4.57 25,233 36,300 62 1,815

Q2 3,597 327 3.20 4.68 33,718 49,242 36 1,797

Q3 3,997 368 3.01 4.01 35,625 47,357 56 1,833

Q4 4,005 368 3.24 4.52 38,426 53,587 43 1,820

2025 4,004 1,462 3.27 4.47 153,817 210,133 150 5,762

2026 4,002 1,461 3.41 4.85 160,364 227,629 87 2,857

2027 4,003 1,461 2.94 4.54 138,157 213,416 129 3,313

2028 4,003 1,465 2.77 4.04 130,637 190,383 183 3,375

2029 4,003 1,461 2.47 2.87 115,797 134,661 152 4,522

2030 4,001 1,460 2.33 3.11 109,494 146,203 151 4,827

2031 4,000 1,460 2.50 3.01 117,132 141,386 95 3,708

2032 4,000 1,464 2.79 3.22 131,138 151,494 16 1,068

2033 3,678 1,342 2.84 3.85 122,682 166,158 - -

Total - 14,556 2.85 3.84 1,333,037 1,799,234 1,807 47,388

Source: SRK, 2020

Figure 16-42 shows the mine production schedule colored by year.



Source: SRK, 2020

Figure 16-42: Mine Production Schedule Colored by Year



16.5.4 Mining Operations

Stoping

Stopes will be mined using the longhole open stoping method. Individual stope blocks are designed to

be 10 m wide, up to 30 m long, and will have a transverse orientation. Levels are spaced 30 m apart

and each stope block will have a top and bottom access (4.5 m by 4.5 m flat back drifts).

Stopes will be drilled downward from the top access using 114 mm diameter holes (stope slots and

stope production rings will be drilled with an ITH drill). A bottom up, primary/secondary extraction

sequence will be followed. Primary stopes will be backfilled with high strength paste backfill and

secondary stopes will be backfilled with RoM waste from the underground operation and low strength

paste backfill as needed when waste rock is not available.

Stope extraction will occur in two steps. During the first step, a slot will be mined at the far end the

stope using a drop raise and 28 fan-drilled slash holes. The slot is required to create sufficient void

space for the remainder of the stope to be blasted. During the second step, production rings will be

blasted three rows at a time (13 blastholes per ring) until the stope is completely extracted. The number

of three-row blasts in a given stope will depend on the length of the stope. All blasting will be performed

with bulk emulsion.

Ore will be remotely mucked from the bottom stope access using a 7.3 m3 (17 t) LHD. The LHD will

transport the ore to a re-muck bay to maximize the efficiency of the stope mucking operations. A

second LHD and a fleet of 45 t haul trucks will be used to transport ore from the re-muck bays to the

grizzly feeding the underground material handling system. Multiple re-muck bays will be used on each

level to avoid interference between the stope loader and the haul trucks.

UG Material Handling System

The underground material handling system is designed to size the rock, provide surge and storage

capacity, and be an efficient, automated system for moving the rock to surface via conveyor.

During operations, ore will be brought to the dump point and fed via a rock-breaker protected grizzly

into an infeed ore pass that loads into the crusher station. The ore pass from the grizzly is sized to

hold approximately 2,000 t (approximately ½ half day mill feed). A feeder will load a vibratory grizzly

which will separate fine and coarse sections and feed the coarse (oversize) material into a jaw crusher.

Undersize will be fed to the main transfer conveyor and then onto the main conveyor to surface.

Development

Lateral development includes main conveyor ramp, interlevel truck ramps, ventilation drifts, level

accesses, stope accesses, and short connecting drifts for ventilation. The conveyor ramp system will

be 5.5 m wide by 5.6 m high with an arched back at a maximum 17% gradient. The interlevel truck

ramps will be 5.5 m wide by 5.5 m high with an arched back at a maximum 14% gradient. Level

accesses will be 5 m wide by 5.5 m high with a flat back and will be mined higher at the re-muck bays

to allow the haul trucks to be loaded by the LHD. Stope access drifts will be flat back 4.5 m wide by

4.5 m high.

Interlevel ramps and levels accesses will be located in the footwall and have been designed to avoid

crossing fault zones to the maximum extent possible. Stope accesses are oriented perpendicular to

the strike of the orebody.



The lateral development is sized for the operation of the mining equipment fleet that has been selected

for the operation. The development profiles include allowances for ventilation ducting and services.

Conventional drop raising will be used for the two lowest levels to establish ventilation connection and

secondary egress.

Haulage

The mine plan assumes that 7.3 m3 (17 t) LHDs will load 45 t haul trucks from re-muck bays that will

be strategically located throughout the development workings. Ore and waste haulage distances and

cycle times were calculated using the haulage profile module in Vulcan and are based on estimated

underground truck speeds as shown in Table 16-26. The outputs from the Vulcan haulage profile

module are a one-way haulage distance and an average truck cycle time (round trip).

Table 16-26: Truck Hauling Speeds

Road Grade (%) Speed (1) (km/hr)

Loaded

0-2.5 11.0

2.5-5.0 10.5

5.0-7.5 10.3

7.5-10.0 10.2

10.0-12.5 10.1

12.5-15.0 7.4

15.0-20.0 7.4

Empty

0-2.5 11.0

2.5-5.0 10.5

5.0-7.5 10.5

7.5-10.0 10.5

10.0-12.5 10.5

12.5-15.0 9.0

15.0-20.0 7.5

Source: SRK, 2020 (1) Uphill and downhill assumed speeds the same.

The ore haulage distances were evaluated from the mine production schedule. Based on this

evaluation, ore haulage pathways were created to approximate the location of ore development and

stope mining in each time period. Vulcan haulage profile was then used to generate a one-way ore

haulage distance and an average cycle time (round trip) using the speed parameters shown in

Table 16-26.

All waste material mined through the end of 2028 is sent to surface for surface construction purposes.

In 2029, and through the end of the mine life, the availability of mined-out secondary stopes was

evaluated to determine haulage distances for waste material. These waste haulage pathways were

created to approximate the location of development waste mining and waste rock dumping for each

time period. Vulcan haulage profile was then used to generate a one-way waste haulage distance and

an average cycle time (round trip) using the speed parameters shown in Table 16-26.

The average one-way ore haulage distances are approximately 400 m early in the mine life through

2027 and increase to approximately 1,570 m later in the mine life. The LoM average is 732 m. Waste

haulage distances are approximately 2,000 m when hauling material to surface and 800 m thereafter.

At the peak, four haul trucks are required to transport the ore and waste. Figure 16-43 and Figure

16-44 show the haulage distance and cycle time by period. SRK notes the cycle times reflected in this

summary are indicative as there is a fixed component including the loading time, dumping time,



positioning time and additional delays that are included in the productivity and equipment quantity

determinations and not included in the information summarized in these figures.

Source: SRK, 2020

Figure 16-43: Haulage Distance – One Way Length



Source: SRK, 2020

Figure 16-44: Haulage Cycle Time - Roundtrip

A small bypass stockpile near the portal is available for short term storage if needed, but there is

limited stockpiling space on surface and as soon as ore material is produced it is expected to be

processed.

Backfilling

The mine production sequence includes the use of cemented paste backfill to fill the voids left by the

stopes to maintain the mine structural integrity. The mine utilizes a high strength backfill paste that has

a 7% cement content in the primary stopes. A lower strength paste with 4% cement is used to backfill

the secondary stopes. Section 18.13 discusses the surface plant and system to move the pastefill

underground to the stopes. A backfill operations crew installs barricades in the lower access drift to

the stopes, extends the pipe delivery system in the upper access drift into the stopes, and monitors

the backfill as the stope fills. Once the stope is filled the backfill is allowed to cure (seven days) to

design strength of over 1 MPa prior to blasting on the adjoining stope.

The LoM backfill breakdown by volume and type is shown in Table 16-27.

Table 16-27: Backfill Volume Summary – By Type

Backfill Type Volume (m3)

Total Backfill 5,429,425

Low Strength Backfill (Waste rock or 4% Cement Pastefill) 1,722,163

High Strength Backfill (7% Cement Pastefill) 3,707,262

Source: SRK, 2020



Ground Support

The current knowledge of the geotechnical characteristics indicates that ground support will be

required in the ramps, in areas of faulting and in primary access drifts as well as the crusher and shop

areas. The stope access drifts will require minimum ground support except at brows of the stopes. The

ground support plan includes use of grouted rebar and split set style bolts as a standard. The bolting

will be supplemented with wire mesh, shotcrete and additional support where required. The plan

includes allowances for areas of full shotcrete, that will be used in longer term active mine areas. A

bolter will be utilized as normal practice and shotcrete equipment are included in the estimate.

Additionally, a cable bolter has been included in the equipment fleet to allow for cable bolting if

necessary, in intersections and at the stope brows.

Grade Control

As part of the routine mining sequence, CGM will conduct infill drilling on routine sampling grids, and

will execute a grade control program to monitor the mining production. The infill drilling will be

conducted from established drilling stations on each level, with underground diamond drilling using

NQ core diameter drilled across the width of the known mineralization. Drilling will be logged for basic

geological and geotechnical parameters and sampled using the current established protocols by CGM.

SRK envisions up to three holes in a fan pattern can be drilled from each station to gain knowledge for

levels above and below as required.

The aim of the grade control program is to deliver the most economic material to the mill via accurate

definition of “ore” and waste contacts. The basis of a successful program in an underground

environment will be completed via detailed geological mapping and grade sampling ahead of the

mining. Grade control strategy is related to mining method and orebody type. For underground

operations sampling methods include chip, channel and panel samples, grab/muck pile samples, and

drill-based samples.

The aim of the program will be to identify variations in dip, strike and width, impact on local scale from

faulting effects, and grade continuity/type. Variations in geometry at the edge of the mineralization will

require geological understanding to ensure optimum grade, minimal dilution and maximum mining

recovery.

The current proposed mining methods involve development cross-cuts at regular intervals across the

width of the mineralization at the top and bottom of a stope prior to mining. Samples should be taken

across the full width of the exposed mineralization via cut channel sampling (using the CGM

exploration protocols) with sufficient volume to ensure accurate assay. The aim of the sampling should

be to achieve a sample weight the equivalent of at minimum half NQ core for the sampling interval.

The samples should be logged geologically marking the width of the mineralization and any hanging

wall or footwall mineralization. The samples should be processed at an onsite facility to enable quick

turnaround, with sufficient QA/QC samples inserted to monitor the quality of the laboratory and routine

external laboratory checks to test for bias. Further to the channel sampling programs the mine

geologist will perform daily mapping, as well as define the ore/waste contact for the mining teams. The

mapping should be incorporated into a digital format to further improve the geological model and

enable the development of short term estimation.



Additionally, blasted material will be available for grab sampling to test grades, which should then be

input to a production database and be used to confirm head-grades and for reconciliation purposes.

Note that grab sampling can result in selection bias if not conducted following a routine defined

protocol.

SRK recommends that CGM create a series of protocols to cover all grade control tasks from mapping

to sampling and integration with the database. The local sampling should be used in the generation of

a short term grade estimation model for use in short term planning. The use of short-term models will

also aid CGM in the ability to complete routine reconciliation studies to monitor the performance of the

grade estimation and to identify any potential issues.

On-going quality assurance/quality control monitoring and review will allow protocols and staff to be

updated as required.

16.5.5 Ventilation

The ventilation configuration of the MDZ is considered best practice design as it is configured to

minimize series ventilation. The ventilation design for the project includes dedicated ventilation splits

for fixed facilities so that all air used in the shops and crusher dump areas is transferred directly to

exhaust and away from working levels. The design also provides fresh air to each active mining level.

Five stages of mine development were modeled. Each stage accounting for worst-case operating

conditions. The following sections provide an overview of the MDZ ventilation design.

Ventilation Modeling Criteria

Several factors were considered when determining the airflow requirements for the mine such as gas

dilution, diesel particulates, maintaining minimum air velocities and meeting government regulations.

These factors need to be applied to targeted areas to determine the total mine airflow requirement.

Fixed facilities underground (including shop and crusher area) will also demand a dedicated airflow

split. SRK applied general mine ventilation best practices to the ventilation design and the Colombian

Underground Mining Safety Code (UGMSC) for specific ventilation requirements for the project

location.

Gases can be broken down into three categories; strata gasses, exhaust fumes and blasting fumes.

Harmful strata gases such as methane, carbon dioxide and hydrogen sulfide are not projected to be

encountered at this mine and, therefore, the dilution of strata gasses is not included in this study.

The gaseous components of the equipment exhaust will need to meet the UGMSC. These values are

usually met if the standard airflow criteria are achieved in the individual mining areas for diesel

equipment dilution. However, if the airflow cannot be achieved at the mining areas then the gas

concentration will increase. This acts as an operational limitation and criteria.

Air velocity limitations vary according to airway type. In areas such as return airways and apiques

where personnel are not expected to work, higher velocities are acceptable. Airway velocities typically

used by SRK for various airway types are shown in Table 16-28. Air velocity limits and recommended

values for travelways are established to accommodate work and travel by people and equipment,

optimizing dust entrainment and temperature regulation.



Table 16-28: Recommended Maximum Air Velocities for Various Airway Types

Airway Type Air Velocity (m/s) Maximum

Travelways 6

Primary Ventilation Intake and Exhaust Entries 10

Primary Ventilation Raises 20

Ventilation Shaft With Conveyance or Escape 10

Conveyor 3

Source: SRK, 2020

Colombian mining regulations specify a minimum factor of 0.09 m³/s per kW of diesel engine power to

ensure gaseous and aerosol contaminants from diesel equipment are sufficiently diluted. This is the

recommended minimum airflow to ensure sufficient dilution of contaminants with new equipment.

There is no Colombian specific criteria or legislative limit for Diesel Particulate Matter (DPM).

The total number of equipment, motor power and a factor 0.09 m3/s/kW for dilution of diesel was used

to estimate the total required airflow in the mine. An airflow utilization factor was also incorporated.

Leakage through bulkheads and doors must also be accounted for. Airflow through facilities must also

be ventilated, this includes the underground shop and crusher area. The total minimum airflow for the

mine is estimated to be near 340 m3/s from the calculations in Table 16-29. A target airflow was also

developed for each scenario by zone and the maximum number of equipment to be utilized. This is

used as a guideline for the airflow distribution regime.

Table 16-29: Equipment List and Airflow Requirement

Equipment Type Qty Power

(kW)

Diesel Engine

Utilization (%)

Utilized Diesel Power

(kw)

Airflow Requirement

(m3/s)

Sandvik DD422i - Jumbo, 2 boom 3 119 20% 71 6.4

Sandvik DS411 - Mechanical Bolter 4 110 20% 88 7.9

Sandvik DU421 - Production Drill 4 130 20% 104 9.4

Orica MaxiCharger 5344 – Production 2 120 20% 48 4.3

Normet Spraymec 1050 - Shotcrete Sprayer 4 110 20% 88 7.9

Getman A64 HD R60 - Transmixer Truck 2 129 20% 52 4.6

Orica Handiloader 1120 – Development 2 120 20% 48 4.3

Sandvik LH517 - LHD, 7.3 m3, 17 t 4 256 75% 768 69.1

Sandvik LH307 - LHD, 3.7 m3, 7 t 1 160 70% 112 10.1

Sandvik TH545i - Haulage Truck, 45 t 4 450 70% 1260 113.4

CAT UG20M – Grader 1 105 20% 21 1.9

Getman A64 - Scissor Lift 2 129 10% 26 2.3

Getman A64 - Boom Truck 1 129 10% 143 12.9

Getman A64 - Flat Deck Truck 1 129 10% 13 1.2

4x4 Pickup - Light Vehicles 4 75 10% 30 2.7

4x4 Pickup - Light Vehicles 2 75 10% 15 1.4

Getman A64 - Fuel/Lube Truck 2 129 10% 26 2.3

Getman A64 - Personnel Carrier, 16 per. 1 129 10% 13 1.2

Kubota RTV 1120D - Personnel Carrier 3 19 10% 6 0.5

CAT 1255D - Forklift/Telehandler 1 106 10% 11 1.0

Getman A64 - Explosives Truck 2 129 10% 26 2.3

CAT 272D - Skid Steer 2 73 10% 15 1.3

Total Airflow Required for Diesel Dilution (0.09 m3/s/kW)

2982 268.4

Fixed Facilities 40.0

Leakage (10%) 30.8

Total Airflow Requirement 339

Source: SRK, 2020



Ventilation Model Development

VentSIM Visual ventilation simulation software was used to generate the ventilation model. The

ventilation design is configured to intake air through the main decline, internal ramps, and intake raises

and exhaust air through the exhaust raise system. This setup allows the mine to use on-shift blasting

since blast fumes will be mainly isolated to each working level and directly exhausted instead of

contaminating the ramp.

There are two mining blocks to be ventilated: an upper block to be mined first and a lower block that

is mined later in the mine life.

• Upper Block- The main decline is to be developed initially to a raise bore to establish flow-

through ventilation in the mine. The conveyor decline is planned to connect at the midway

point in the orebody. From this point the upper block is developed with mining starting at the

bottom of the block (near the conveyor decline) and ending at 940 elevation.

• Lower Block—The lower mining block starts with developing the ramp down to the bottom

(480 elevation) and then mining up from the bottom to the main decline.

The mine is to be ventilated using a push-pull system with a primary fan at the surface exhaust portal

and a smaller fan at the intake portal. The conveyor decline is also to be an air intake into the mine.

Since the main decline is assumed to include a conveyor, velocities are to be limited by controlling the

flow through the intake fan such that airflow velocity in the conveyor is limited to 3 m/s to mitigate dust

creation. Plans are to have a single 5 m intake raise, a single 5 m return raise for the UZ, and 4.5 m

intake and return raises for the MDZ.

For the primary ventilation circuit, air enters the mine from the intake raise and is regulated across a

level to the return raise system where it is exhausted from the mine. In the secondary circuit, intake

air flows down the main decline into the mine. Part of this air is used to ventilate the shop and

underground crusher; the rest of this air is fed to the internal ramp and onto the working levels.

The crusher area requires dedicated ventilation because of the high generation of dust. 20 cubic

meters per second (m³/s) is used for dilution of heat and dust from the crusher area and provides

perceptible movement of air in the large galleries. Air is to flow over the crusher dump and down to the

crusher level. The crusher area is to have a regulated connection to the main exhaust raise.

An underground shop is also planned. A direct connection to exhaust is to be made at the back end

of the shop. A regulator will be used to limit the airflow to around 20 m3/s. This direct connection to

exhaust will ventilate the shop and help limit harmful fumes generated in the shop from entering the

mine.

Compressibility of air and leakage through stoppings and doors must also be accounted for in the total

mine flow. A general layout of the airflow distribution is shown in Figure 16-45.



Source: SRK, 2020

Figure 16-45: Marmato Project General Ventilation Scheme

Auxiliary Ventilation

For auxiliary ventilation, a standard duct size of 1.2 m was used to provide enough clearance between

the ducting and equipment. Short headings on levels can be ventilated with 50 kW fans, while longer

development headings will require 75 kW fans. During development of the conveyor decline and lower

portion of the ramp, it will be challenging to provide the target airflow due to the length of duct (1.7 km)

and leakage. For these long drives it is recommended to use two 1.2 m ducts in parallel using 75 kW

fans in series. A summary of the ducting and fan requirements is provided in Table 16-30.

Table 16-30: Auxiliary Ventilation Fan Summary

Duct Dia. (m) Fan Pressure (kPa) Inlet Airflow (m³/s) Air Power (kW) Motor Power (kW)(¹)

1.2 1.4 26 35 46

1.2 2.4 23 54 72

Source: SRK (1) Assumes 75% fan efficiency



Main Fans

Primary ventilation circuits are established with two main fan installations. The primary exhaust fan is

the main driver of air for the mine while an intake fan is used to balance the amount of airflow in the

conveyor decline and ramp system. Table 16-31 shows the operating point during the maximum power

outputs for each fan installation. The fans are assumed to be located underground near the surface

portals.

Table 16-31: Fan Operating Points*

Description Pressure

(kPa) Quantity

(m³/s) Motor Power

(kW)¹ Inlet Density

(kg/m³)

Main Intake Fan Installation 0.50 240 160 1.04

Main Exhaust Fan Installation 2.02 342 920 1.03

Source: SRK, 2020 (1) Assume 70% fan efficiency *based on the maximum motor power output for each fan

Fan Power Demand

Power demand from the main and auxiliary fans were estimated with a peak demand of 1,525 kW (the

highest amount of power expected to be used at one time) occurring in year 2031 as shown in Figure

16-46. The installed power is the total power of all the fan installations combined. The total installed

power is approximately 2,000 kW for two main fans and twelve auxiliary fans.

Source: SRK, 2020

Figure 16-46: Estimated Fan Power Demand



16.5.6 Mine Infrastructure & Services

The mine infrastructure includes a power distribution system, mine dewatering systems, underground

service water supply system, an underground paste backfill pumping system, an underground ore

handling system, diesel fueling system, an underground shop, warehouse and storage, offices,

explosives storage, and communications system. The major systems are described in the following

sections. Figure 16-25 shows the location of the major infrastructure systems.

Power Distribution

The main power for the MDZ mining area will be supplied by two 13.2 kV power lines down the decline

to the crusher area. The power then will be distributed through 13.2 kV power lines through the

remainder of the mine to stepdown portable substations at the shop, paste plant pump station, mine

pump stations and operating mining and development areas. The main ventilation fans will receive

power through the ventilation declines near the bottom of the UZ and will be fed from the existing

Marmato UZ substation. The mining connected loads, including the surface backfill plant, total 9.2 MW.

The average running load is approximately 5.3 MW. The major loads include mine ventilation and mine

pumping systems. The mine pumping system runs intermittently and is the main reason the connected

load and average running load are substantially different.

Backup generation of approximately 3.6 MW will be provided by the backup generator at the MDZ

plant. The fans will use the backup generation capacity at the existing Marmato UZ plant.

Mine Dewatering

The MDZ mine pumping system is developed in stages. Declines will be constructed at the MDZ and

UZ sites. Skid mounted pump systems (60 l/s capacity) will be installed in the active development

declines and will be staged to control water in the declines. Once the declines are completed the skid

mounted pump systems will be used on the development ramps as the mine is expanded.

At the 730 m level of the mine, a permanent pump station will be established with a sump and agitator

system. The pump system will include two pump trains of four 225 kW pumps in series with a capacity

of 177 l/s at a total dynamic head (TDH) of 270 m. One train will be operational with one on standby.

The pump discharges to a 30.5 cm steel pipe that carries mine water from the pump station to the

surface to the plant storage water tank. Based on the current knowledge of the hydrogeology, the

pump system is planned to operate at 60 l/s on average.

Later in the mine life, at the current planned bottom of the mine, a second permanent pump station will

be constructed at the 480 m level. The second permanent pump station will be a duplicate of the

system installed at the 730 m level and will operate in the same manner. The system will pump through

30.5 cm schedule 40 steel pipe to the pump station on the 730 m level where the 730 m level station

will pump to the surface. The 480 m level station will operate on average at 13 l/s.

The system is designed to handle maximum flows and operate at the much lower operational flow rate.

Underground Water Supply

Mine service water is supplied from the supply tank located at the surface backfill plant via 10.2 cm

HDPE pipeline down the MDZ decline to the mine operational areas.



Underground Backfill System

The surface portion of the cemented paste backfill system is described in Section 18.12. Cemented

paste is moved via 20.3 cm schedule 120 pipe from the surface plant to an underground booster pump

station that pumps the paste from the 730 m level to the stopes above this level. The booster pump

station includes a paste hopper and piston pump, pump hydraulic power unit, flush and clean up water

storage tank, and a high pressure flush pump and clean up pump. The booster pump station moves

the paste through the underground paste reticulation system to the stopes. For the stopes below the

730 m level, boreholes have been included to house the piping that will transport the paste to the lower

level stopes.

Ore Handling System

Ore is moved by 45 t truck from the stopes to a truck dump with grizzly and rock breaker that feeds

into an ore pass that feeds a jaw crusher. The ore is crushed and fed onto a conveyor that transports

the ore to the surface where it then crushed further in a secondary crusher feeding into the mill.

Additional detail is in Section 17.3.6.

UG Fuel Storage and Distribution

Fuel from the surface will be transported to the underground storage system via fuel trucks. One fuel

station is included in the design, located near the shop area. The station will be a bladder type, with

up to 150% containment, complete with the following safety functions; 4 hr rated UL approved roll up

door, thermal activated fuel shut off valve to dispensing system, anti-syphon valve, and a dry chemical

automatic fire suppression system with detection and actuation. The station should be alarmed, by

means of a PLC with level alarms, and a level switch.

Additionally, fusible link fire doors are also included in the underground layout, these twin fire doors,

upon actuation will isolate the fueling area from the main shops.

Underground Shop

The maintenance area consists of three large bays to accommodate vehicular traffic. A service trench

runs the length of each bay to allow access to the undercarriage of the vehicles. One wash bay is also

included in the workshop layout. A drainage trench with covering grating will run the length of the bay

to carry water to a nearby oil capture sump. Grading of the area will help reduce the possibility of oil

contamination. Each maintenance bay will also come equipped with an overhead crane to help

facilitate the maintenance work on vehicles.

Warehouse and tool cribs are also included within walking distance of the maintenance bays.

Three rollup doors will separate the two maintenance bays from the rest of the mine. An office will be

located at the end of a drift located in the maintenance area.

Explosives Storage

Underground powder and primer magazines are included in the mine design. The mine explosives will

be stored off site by the military and deliveries will be on as needed basis with the underground

magazines providing the capacity required for production needs.



Safety Infrastructure

The mine design includes portable and permanent refuge stations. The portable refuge stations will

be staged near the operating faces during development. An emergency hoist is included in the design

for the bottom section of the mine to allow emergency egress. A stench warning system through the

ventilation system will notify workers of emergency conditions.

Communications System

The mine will be equipped with a leaky feeder system that will allow internet, phone, and radio

communications underground. The mine will have standard underground call phones with intercom. A

control system will allow remote operation of the rock breaker and CCTV system to monitor dump

points, crusher, and key material handling locations.

16.5.7 Mine Labor

Labor levels are estimated based on the production schedule and equipment needs. The productivities

used reflect a mix of local and skilled labor with an experienced management team.

The personnel will work one of four rotations that are summarized in Table 16-32.

Table 16-32: MDZ Shift Schedule and Rotation

Roster Type Shift Area

6x2 Rotating 8 hour Underground

12x4 Rotating 8 hour Underground

14x7 Rotating 12 hour Surface

5x2 Days 9 hour Manager/Technical

Source: CGM, 2020

The estimate is based on owner mining using an operating schedule consisting of eight hours per shift

for underground workers, three shifts per day and seven days per week. The management and

technical team are planned to work five nine hour days per week. Surface personnel will work a 12

hour shift. The eight hour and 12 hour shifts are supported by a four crew rotation.

The overall mine staffing at full production is summarized in Table 16-33. The workforce will increase

over time through the addition of staff to operate additional equipment. The total number of personnel

ranges from eight in 2021 to the maximum of 429 in 2024.

Table 16-33: MDZ Mining Labor Summary

Total Mine Roster (On and Off Site) Maximum Staffing

Mine Technical Staff Roster 47

Mine Operations Labor Roster 256

Mine Maintenance Staff Roster 38

Mine Maintenance Labor Roster 88

Total Labor Roster 429

Source: SRK, 2020

Table 16-34 shows the required workforce at the maximum staffing level.



Table 16-34: MDZ Mining Labor

Department/Section Category Shift Hours Max Staff

Mine Technical Staff 47

Mine Superintendent Salary 9 1

Chief Mining Engineer Salary 9 1

Long Term Planning Engineer Salary 9 1

Stope Designer Engineer Salary 9 1

Production Reporting Supervisor Salary 9 1

Short Term Planning Salary 9 2

Grade Control Engineer Salary 9 2

Surveyors Salary 9 4

Technician Salary 9 2

Senior Geotechnical Engineer Salary 9 1

Geotechnical Engineer Salary 9 3

Geotechnical Technician Salary 9 2

Ventilation Engineer Salary 9 1

Ventilation Technician Salary 9 2

Backfill Engineer Salary 9 1

Backfill Technician Salary 9 2

Chief Mine Geologist Salary 9 1

Grade Control Geologist Salary 9 3

Infill Drilling Supervisor Salary 9 3

Backfill Coordinator Salary 9 1

Backfill Plant Supervisor Hourly 12 4

Senior Modelling Geologist Salary 9 1

Senior Field Logging Geologist Salary 9 1

Project Lead Salary 9 1

Mechanical Engineer Salary 9 1

Civil Engineer Salary 9 2

Clerk Salary 9 2

Mine Operations Labor 256

Shiftboss Development Hourly 8 4

Jumbo Operator Hourly 8 13

Bolter Operator Hourly 8 9

Cablebolter Operator Hourly 8 9

Shotcrete Operator Hourly 8 9

Crusher Operator Hourly 8 5

Production Drill Operator Hourly 8 18

Blaster Hourly 8 9

UG Explosive storage personnel Hourly 8 9

Blaster Helper Hourly 8 9

Crusher Helper Hourly 8 4

LHD Operator Hourly 8 18

Truck Driver Hourly 8 18

Transmixer Driver Hourly 8 5

Service Crew Hourly 8 9

Shotcrete Helper Hourly 8 9

Utility/Laborer/Nipper Hourly 8 13

Service Crew Helper Hourly 8 13

Conveyor System Operator Hourly 8 4

Production Driller Helper Hourly 8 18

Shiftboss Production Hourly 8 4

Shiftboss Stoping Hourly 8 4

Shiftboss Blasting Hourly 8 4

Infill Drilling Operator Hourly 8 9

Infill Drilling Operator Helper Hourly 8 9

Pastefill Piping Crew Hourly 8 18



Department/Section Category Shift Hours Max Staff

Pastefill Barricade Crew Hourly 8 6

Pastefill Supervisor Hourly 8 1

Pastefill Pour Watcher Hourly 8 6

Pastefill Operations Crew Hourly 8 12

Mine Maintenance Staff 38

Maintenance Superintendent Salary 9 1

Maintenance General Foreman Salary 8 1

Maintenance Planning Supervisor Hourly 8 2

Maintenance Planning Engineer Hourly 8 2

Maintenance Planning Technician Hourly 8 5

UG Maintenance General Foreman Salary 8 3

UG Shop Shiftboss Hourly 8 5

Surface Truck Shop Shiftboss Hourly 12 5

Pastefill Plant Operator Hourly 12 7

Pastefill Plant Helper Hourly 12 7

Mine Maintenance Hourly Labor 88

UG Prod Mechanic Hourly 8 9

UG Prod Electrician Hourly 8 9

UG Prod Mechanic Helper Hourly 8 9

UG Prod Electrician Helper Hourly 8 9

UG Shop Mechanic Hourly 8 9

UG Shop Electrician Hourly 8 5

UG Shop Welder Hourly 8 5

UG Shop Mechanic Helper Hourly 8 9

UG Shop Electrician Helper Hourly 8 5

UG Shop Welder Helper Hourly 8 0

Surface Shop Mechanic Hourly 8 5

Surface Shop Electrician Hourly 8 5

Surface Shop Welder Hourly 8 5

Surface Shop Mechanic Helper Hourly 8 5

Surface Shop Electrician Helper Hourly 8 0

Surface Shop Welder Helper Hourly 8 0

Backfill Mechanic Hourly 12 3

Total Labor 429

*This value represents peak production staffing (max equipment) Source: SRK, 2020

16.5.8 Equipment

The underground equipment used, shown in Table 16-35, is typical for the sublevel stoping mining

method with the number of pieces of equipment calculated from the production rates and typical

availabilities for underground mines.

The estimate uses an equipment availability of 85%, a job efficiency factor of 83% (50 minute hour),

and an activity efficiency factor that varies by activity (85% to 100%). Each shift of eight hours is

reduced by 1.92 hours to represent shift change, breaks, lunch, fuel/grease/inspect time and travel to

and from working areas. This provides an equivalent working day of 15.21 hours or 5.07 hours per

shift. The resulting reductions result in 5,552 effective hours per year of mining time. It should be noted

that the layout of this mine and mining on multiple levels requires the addition of equipment to reduce

equipment move time. This reduces the overall utilization of the equipment fleet.

The equipment totals by pre-production and production year are summarized in Table 16-35.



Table 16-35: Mine Equipment by Period

Equipment Fleet Requirements Type Diesel

Engine (kW)

Electric Power

(kW)

Max Number

Date

Jan-21

Apr-21

Jul-21

Oct-21

Jan-22

Apr-22

Jul-22

Oct-22

Jan-23

Apr-23

Jul-23

Oct-23

Jan-24

Apr-24

Jul-24

Oct-24

Jan-25

Jan-26

Jan-27

Jan-28

Jan-29

Jan-30

Jan-31

Jan-32

Jan-33

Underground Mobile Equipment

Sandvik DD422iE - Jumbo, 2 boom DD422iE 119 180 3 0 0 0 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 2 1 1

Sandvik DS411 - Mechanical Bolter DS411 110 70 3 0 0 0 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 2 1 1

Sandvik DS422i - Cablebolter DS422i 120 75 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Sandvik DU411 ITH - Production Drill DU411 ITH 130 80 3 0 0 0 0 0 0 0 0 0 0 0 2 2 3 3 3 3 3 3 3 3 3 3 3 0

Getman Orica - Explosives Truck w/Orica MaxiCharger 5344 Orica 129 0 2 0 0 0 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Getman Orica - Getman Explosives Truck with Orica Handiloader Orica 129 0 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0

Getman SST Shotcrete Unit - Shotcrete Sprayer SST Shotcrete Unit

110 0 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0

Getman A64 HD R60 - Transmixer Truck A64 HD R60 129 0 2 0 0 0 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 0 0

Sandvik LH517i - LHD, 7.3 m3, 17 t LH517i 256 0 4 0 0 0 2 2 2 2 2 2 2 2 3 3 4 4 4 4 4 4 4 4 4 4 2 2

Sandvik TH545i - Haulage Truck, 45 t TH545i 515 0 4 0 0 0 2 2 2 2 2 2 2 2 2 3 3 4 4 4 4 4 4 4 4 4 2 2

Auxiliary Equipment

CAT UG20M - Grader UG20M 105 0 1 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Sandvik LH307 - LHD, 3.7 m3, 7 t LH307 160 0 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Getman A64 - Scissor Lift A64 129 0 2 0 0 0 0 0 0 0 0 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Getman A64 - Boom Truck A64 129 0 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Getman A64 - Flat Deck Truck A64 129 0 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

4x4 Pickup - Light Vehicles 4x4 Pickup 75 0 12 0 0 0 3 4 4 4 5 5 5 6 6 6 6 6 12 12 12 12 12 12 12 12 6 5

Getman A64 - Fuel Truck A64 129 0 2 0 0 0 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Getman A64 - Personnel Carrier, 16 per. A64 129 0 3 0 0 0 0 1 1 1 1 1 1 1 2 3 3 3 3 3 3 3 3 3 3 3 3 2

Kubota RTV 1120D - Personnel Carrier RTV 1120D 19 0 12 0 0 0 4 4 4 5 6 7 8 8 8 8 8 12 12 12 12 12 12 12 12 12 5 4

CAT 1255D - Forklift/Telehandler 1255D 106 0 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Getman A64 - Service Mechanic Truck A64 129 0 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

CAT 272D - Skid Steer 272D 73 0 2 0 0 0 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Sandvik DE142 - Underground Core Drill DE142 0 0 2 0 0 0 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

CAT 982M - Front End Loader - 6.7m3/8.75 yd3 982M 325 0 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Miller - Skid Mounted Pump system (each) - 7500gal tank, 2ea 112 kW (1op/1stby) pumps

0 0 143 6 0 0 0 0 2 4 6 6 6 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

Miller - 730 Level Sill Pump System - 8ea 400kW (4 op/4stby) pumps; VFD; 4 submersible 22kW pumps with 11kW starters (4 op)

0 0 293 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Miller - 480 Level Sill Pump System - 8ea 400kW (4 op/4stby) pumps; VFD; 4 submersible 22kW pumps with 11kW starters (4 op)

0 0 65 1 0 0 0 1 1 1 1

Getman A64 - Lube Truck A64 129 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Getman A64 - SLHanger Scissor Lift A64 129 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Getman A64 - SLWing Scissor Lift A64 129 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Getman A64 - Pallet Handler A64 129 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Atlas Copco XATS 400 - 350 CFM Mobile Compressor XATS 400 115 0 2 0 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Miscellaneous Equipment

Getman A64 - Flat Pallet for Pallet Handler 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Getman A64 - Personal Carrier for Pallet Handler 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Getman A64 - Water Sprayer for Pallet Handler 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Fans Main - 1500kVA XFMR & SWGR (VFD $ included with Fans)

1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Surface Backfill - 2500kVA XFMR &SWGR 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Conveyor Decline Main PWR Feed - 13.2kV 2x3C#500MCM Cable, 15kV JBs - 2x1.7km

1 0 0.25 0.5 0.75 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Conveyor Decline Main PWR Feed - 13.2kV UG SWGR Skid 4 0 1 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 4 4 4 4 4

Crusher - 1000kVA XFMR & 5kV SWGR 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

UG Shop - 500kVA XFMR & MCC + Distribution 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Mobile Substation 1000kVA - 13.2kV:440V 4 0 1 1 1 1 2 2 3 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

Mobile Substation 750 KVA - 13.2kV:440V 8 0 2 2 2 2 2 3 4 6 6 6 8 8 8 8 8 8 8 8 8 8 8 8



Equipment Fleet Requirements Type Diesel

Engine (kW)

Electric Power

(kW)

Max Number

Date

Jan-21

Apr-21

Jul-21

Oct-21

Jan-22

Apr-22

Jul-22

Oct-22

Jan-23

Apr-23

Jul-23

Oct-23

Jan-24

Apr-24

Jul-24

Oct-24

Jan-25

Jan-26

Jan-27

Jan-28

Jan-29

Jan-30

Jan-31

Jan-32

Jan-33

Pump Starter Box - 30hp 0 0

Pump Starter Box - 150hp 0 0

Pump Starter Box - 350Hp 0 0

MineArc - Refuge Chamber - 12 person 0 0

MineArc - Refuge Chamber - 8 person 3 0 1 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

MineArc - Refuge Chamber - 40 person built in 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Allowance - Underground Shop Tool and Equipment 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Allowance - Powder and Primer Magazines 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Allowance - Surface Shop Tool and Equipment 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Allowance - Engineering Tools and Software 1 0 0.25 0.5 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Allowance - Shipping Container Storage-2 ea 2 0 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Allowance - Communications Infrastructure and Mine Automation 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Allowance - Miscellaneous Lighting 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Fueling - UG Fueling System - supply lines, pump and storage 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Fueling - Surface Fuel Bay 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

MDZ - Conveyor Decline Portal 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

MDZ - Veins side - Vent Decline Portal 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

MDZ - Veins side - Vent Decline Portal 2 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Allowance - Underground Emergency Hoist 1 0 1 1 1 1

Surface Backfill - Tailings, water delivery, return water 1 0 0.5 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Surface Backfill - Surface paste plant 1 0 0.5 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Underground Backfill - UG Booster pump station 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Underground Backfill - UG Reticulation System (paste delivery) 1 0 0.5 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Surface Shotcrete - Modular Shotcrete Plant 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Surface Backfill - Earthworks Estimate for Surface Backfill Plant Pad

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Underground - Water Piping for 730 Level Sill Pump System - 2260 m

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Underground - Water Piping for 480 Level Sill Pump System - 270 m

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Allowance - Fault Mitigation in Declines (30m @US$2500/m) 1.262 0 0 0 0 0 0 0 0.73

8 1.26

2 1 0.29 0.71 1 1 0 0 0 0 0 0 0 0 0

Allowance - Mine Rescue Unit 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Ventilation Equipment

Howden 12300-AMF-8000 - Main Ventilation Exhaust Fan (1000 KW)

0 1000 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Howden 11200-AMF-6100 - Main Ventilation Intake Fan (200 KW) 0 200 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Fan 75 kW - Development Fan 0 75 8 0 4 4 4 6 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8

Fan 50 kW - Auxiliary Fan 0 50 4 0 2 2 2 2 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

Fan 25 kW - Auxiliary Fan 0 25 4 0 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

Ventilation - Single Equipment Door 0 0 9 0 4 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9

Source: SRK, 2020



16.6 Combined UZ and MDZ Production Schedule

Figure 16-47 and Figure 16-48 summarize the combined UZ and MDZ schedules. This combined

schedule is used in the economic model results shown in section 22.

Source: SRK, 2020

Figure 16-47: Combined UZ and MDZ Mining Profile – Tonnes and Grade



Source: SRK, 2020

Figure 16-48: Combined UZ and MDZ Mining Profile - Contained Metal



17 Recovery Methods CGM operates a 1,200 t/d process plant to recover gold and silver values from material produced from

current Marmato mining operations in the UZ and plans to expand this facility to a 1,500 t/d capacity

in the next couple of years. In addition, CGM is evaluating the development of the MDZ, which is below

the current mining operations and the construction of a new 4,000 t/d plant to process material solely

from the MDZ. Recovery methods currently in use for processing Marmato material and recovery

methods are being evaluated for processing the MDZ material and are presented and discussed in

this section.

17.1 Marmato Process Plant (Current Operations)

The Marmato process plant flowsheet incorporates unit operations that are standard to the industry

and include:

• Three-stage crushing

• Closed circuit ball mill grinding


• Flotation

• Flotation and gravity concentrate regrind

• Cyanidation of the flotation and gravity concentrates

• Counter-current-decantation

• Merrill-Crowe zinc precipitation

• Smelting of precipitates to produce final doré product

The Marmato process plant flowsheet is shown in Figure 17-1 and a list of major equipment is shown

in Table 17-1. Each of the process unit operations is briefly described in this section.



Figure 17-1: Marmato Process Flowsheet

Source: CGM, 2020



Table 17-1: Equipment List for Marmato Process Plant

Flowsheet No. Equipment Description Quantity Hp

1 Mine rail cars 1.5 t

2 Pre-hopper 100 t 1

3 Winch 2 60

4 Feed hopper 5 m x 7 m 1

5 Feed hopper with hydraulic gate 1

6 Vibrating grizzly 5 ft x 13 ft (5/16") 1 20

7 Primary jaw crusher 25" x 40" 1 125

8 Conveyor belt 30" 1 25

9 Vibrating screen (double-deck) 20 ft x 8 ft (7/8" x 3/8") 1 40

10 Conveyor belt 24" x 14 m 1 12

11 Secondary cone crusher Omincone -1352 1 250


13 Tertiary cone crusher HP300 1 300

14 Spiral Classifier 30" x 17 ft 1 7.5

15 Fine ore bin 7 m x 5.8 m 1


17 Conveyor belt (with belt scale) 24" x 9 m 1 7

18 Primary ball mill (Allis Chalmers) 9.5 ft x 14 ft 1 600

19 Tapezoidal jig 17 ft2 4 7

20 Cyclone feed pump 6" x6" 2 75

21 Hydrocyclone Gmax 20" 2

22 Flash flotation cell SK-80 1 20

23 Secondary ball mill 7.5 ft x 10 ft 1 300

24 Gravity concentrate pumps 3" x 3" 6 20

25 Flotation conditioner 12 ft x 12 ft 1 30

26 Rougher flotation cell KCF/KYF 30 2 75

27 Scavenger flotation cell (circular) 10 ft x 10 ft 3 30

28 Cleaner flotation cell 2 m x 2 m 2 7.5

29 Thickener 24 ft x 10 ft 5 3

30 Regrind cyclone feed pump Wilfley 5K 2 60

31 Regrind hydrocyclone 6" 4

32 Regrind ball mill Hardinge 7 ft x 5 ft 2 200

33 Pretreatment agitated tank 12 ft x 12 ft 2 12


35 Static Thickener 12 ft x 35 ft 1

36 Peristaltic pump Bredel SPX-65 1 1

37 Leach tank 20 ft x 20 ft 2 30


39 Leach tank 12 ft x 12 ft 6 12

40 PLS tank 12 ft x 12 ft 2

41 Clarifier 1

42 Deaeration tower 1

43 Vacuum pump Hydral 1 12

44 Zinc dust dosing cone feed pump Halberg Nowa 1 25

45 Zinc dust dosing cone 1

46 Filtro Press pump feed Halberg Nowa 1 25

47 Filtro Press 2

48 Precipitate receiving tray 1

49 Flux mixing tray 1

50 Precipitate smelting furnace 1

51 Ingot molds 2

52 Dore’

53 Barren solution tank 12 ft x 12 ft 1

Source: CGM, 2019



17.1.1 Crushing Circuit

RoM ore is hauled by rail from the mine and dumped into a hopper where a slusher is used to move

the material to a 5 m by 7 m feed hopper that feeds a vibrating grizzly to remove the -3/8 inch material,

prior to feeding the primary jaw crusher. The discharge from the jaw crusher is conveyed to a double-

deck vibrating screen fitted with a 7/8 inch upper deck and a 3/8 inch lower deck. The screen oversize

from the top deck is conveyed to a Nordberg 1352 Omnicone which is operated in closed circuit with

the vibrating double-deck screen. Ore retained on the second deck is conveyed to a Nordberg HP300

cone crusher, which is also operated in closed circuit with the vibrating screen. The -3/8 inch screen

undersize discharges to the fines bin. The -3/8 inch undersize from the vibrating grizzly is further

classified in a spiral classifier. The classifier oversize is fed directly into the primary ball mill and the

classifier undersize is thickened and then pumped to the primary hydrocyclones. The crushing circuit

has an operating capacity of 1,600 t/d.

17.1.2 Grinding and Gravity Concentration Circuit

Crushed ore (-3/8 inch) is fed from the fines bin and then transported on a conveyor fitted with a belt,

scale to the 9.5 ft by 14 ft primary ball mill (600 hp). The primary ball mill discharges to a Knelson

gravity concentrator (model QS-40) to recover coarse gravity recoverable gold. The tailings from the

gravity concentrator is pumped to the cyclones where a size separation at P50 75 µm is made. The

cyclone underflow discharges to the 7.5 ft by 10 ft secondary ball mill (300 hp), which is operated in

closed circuit with the cyclones and the overflow advances to the flotation circuit. The gravity

concentrate is combined with the flotation concentrates prior to advancing to the regrind and

cyanidation circuits.

17.1.3 Flotation and Concentrate Regrind Circuit

The cyclone overflow from the grinding circuit is advanced to the flotation circuit where it is first

conditioned with the required flotation reagents and then subjected to one stage of rougher flotation

followed by one stage of scavenger flotation, which provides a total flotation retention time of 40

minutes to recover the contained gold and silver values. The scavenger flotation concentrate is

upgraded in one stage of cleaner flotation and combined with the rougher flotation concentrate. The

rougher + scavenger cleaner flotation concentrates are combined with the gravity concentrate, and

then thickened to about 55% solids and reground to about 80% passing (P80) 44 µm. A portion of the

flotation tailings is pumped to an agitated storage tank and then pumped back underground with a

positive displacement pump for use as hydraulic backfill in the mine.

17.1.4 Cyanidation and Counter-Current-Decantation (CCD) Circuit

The reground gravity and flotation concentrates are re-thickened and then advanced to a conventional

two-stage cyanidation circuit, which provides a total of 30 hours of leach retention time. The first stage

of leaching consists of two 20 ft by 20 ft agitated leach tanks operated in series at a cyanide

concentration of 700 mg/L NaCN and at a pH of 10.5 adjusted with lime. The second-stage leach

circuit consists of six 12 ft by 12 ft agitated leach tanks in which the cyanide concentration is allowed

to attenuate through the circuit from 700 to 400 mg/L NaCN. The leached slurry is then passed through

a counter-current-decantation circuit (CCD) which serves to wash the pregnant leach solution (PLS)

from the leached solids. The leached solids are discharged from the last CCD thickener and then



pumped to the DSTF. The PLS is processed in the Merrill-Crowe circuit to recover the solubilized gold

and silver values.

17.1.5 Merrill-Crowe Circuit and Smelter

The PLS is pumped to the Merrill-Crowe circuit where it is first clarified to remove any remaining

suspended solids and then de-aerated to less than 1 mg/L dissolved oxygen in a vacuum tower. Zinc

dust is then added to the de-aerated PLS in a controlled manner which results in the precipitation of

the gold and silver values, which are then recovered in a filter press. The resulting gold and silver

precipitate are removed from the filter press on a scheduled basis and then smelted using a flux with

the following composition:

• Borax: 40%

• Sodium nitrate: 30%

• Soda ash: 20%

• Silica: 10%

Approximately 600 kg of flux is blended with 600 kg of precipitate and smelted in a diesel-fired furnace

to produce a final doré product.

17.1.6 Process Plant Consumables

Process plant consumables are shown in Table 17-2 and includes grinding media, wear materials and

process reagents. Consumable costs during 2019 (Jan to July) averaged US$3.05/t processed.

Table 17-2: Marmato Process Plant Consumables

Item kg/t US$/Kg US$/t

Grinding Balls (1.5 inch) 0.165 1.13 0.19

Grinding Balls (2 inch) 0.312 1.15 0.36

Grinding Balls (3 inch) 0.392 1.15 0.45

Wear Liners 0.32

Sodium Cyanide 0.370 2.48 0.92

Zinc Dust 0.020 5.03 0.10

Lime 0.625 0.18 0.11

Copper Sulfate 0.015 2.35 0.04

Xanthate (Z-11) 0.011 3.58 0.04

Aerofroth (A65) 0.028 5.56 0.16

Collector MX 5160 0.005 11.12 0.06

Aero AR1404 0.003 11.22 0.03

Lead Acetate 0.002 6.49 0.01

Flocculant (EGA 1203) 0.015 5.09 0.08

Silica 0.095 1.17 0.11

Borax 0.015 0.74 0.01

Soda Ash 0.008 0.60 0.00

Potassium Carbonate 0.001 1.65 0.00

Soloun K 0.015 1.37 0.02

ACPM 0.017 2.47 0.04

Total Consumables 3.05 Source: CGM, 2019



17.1.7 Operating Performance

The current Marmato process plant performance is summarized in Table 17-3 for the period from 2013

to 2020 (Jan to May). During this period mineralized tonnes processed has increased from 274,191 to

370,245 t/y while grades have declined slightly from 2.90 g/t Au in 2013 to 2.49 g/t Au in 2019 and

silver grades have ranged from 12.36 to 9.13 g/t Ag. Overall gold recovery has ranged from 83.7 to

88.9% and has averaged about 87.1% during the period 2019 to 2020 (Jan to May). Silver recovery

has ranged from 33 to 41.1 and has averaged 33.2% during the period of 2019 to 2020 (Jan to May).

Annual gold production has increased from 22,566 ounces in 2013 to 25,750 ounces in 2019.

Table 17-3: Summary of Marmato Process Plant Operating Performance and Recovery Estimate

Parameter 2013 2014 2015 2016 2017 2018 2019 2020 (Jan-May)

Ore Tonnes 274,191 295,023 303,279 341,309 365,119 338,902 370,245 119,069

Ore Grade

Au (g/t) 2.90 2.85 2.79 2.56 2.48 2.67 2.49 2.47

Ag (g/t) 12.36 9.13 9.33 9.24 9.61 10.53 9.98 9.49

Metal Recovery

Au (%) 88.6 89.0 88.0 83.7 86.8 85.5 87.1 88.9

Ag (%) 36.6 41.1 37.9 35.8 34.9 33.2 33.0 33.3

Metal Produced

Au (Ounces) 22,566 24,113 23,954 23,449 25,163 24,909 25,750 8,318

Ag (Ounces) 39,916 34,753 34,490 36,318 39,524 37,522 39,558 11,972

Source: CGM, 2020

17.1.8 Operating Costs

The Marmato process plant operating costs reported for 2019 and 2020 (Jan to May) are summarized

in Table 17-4. During 2019, operating costs were reported at US$12.25/t of ore processed and during

2020 (Jan to May) plant operating costs were reported at US$13.09/t. For the period 2019 to 2020

(Jan to May) plant operating costs have averaged US$12.43. The exchange rate has averaged 3,345

COL per US$1.00 during this period.

Table 17-4: Marmato Process Plant Operating Costs: 2019 - 2020 (Jan-Apr)

Production Total (2019) Total 2020 (Jan-Apr) Total 2019 -2020 (Jan-Apr)

Au Oz 25,750 7,104 32,854

Ore Tonnes 370,495 100,608 471,103

Process Cost (COP)

Crushing 3,147,524,778 1,057,755,849 4,205,280,627

Grinding 5,316,634,662 1,655,877,068 6,972,511,730

Flotation 2,394,649,787 792,698,640 3,187,348,427

Cyanidation & Merrill-Crowe 3,517,471,235 1,076,654,318 4,594,125,553

Refining 485,287,000 143,020,457 628,307,457

Total Process Cost (COP) 14,861,567,462 4,726,006,332 19,587,573,794

Total Process Cost (US$) 4,538,616 1,316,963 5,855,579

US$/Oz 176 185 178

US$/tonne 12.25 13.09 12.43

Exchange Rate (COP/US$) 3,275 3,621 3,345

Source: CGM, 2020



17.2 Expansion Plans

CGM plans to expand the Marmato process plant capacity to 1,500 t/d over the next two years

(complete by Q4 2021), which represents approximately a 35% increase in capacity. The incremental

improvements to the existing process plant would include:

• Install a refurbished 15.5 ft by 22 ft ball mill (3000 hp) to replace the current primary ball mill

(600 hp) and secondary ball mill (300 hp)

• Install new 15 inch hydrocyclones in the upgraded grinding circuit

• Increase flotation circuit capacity by 50% with the installation of an 80 m3 flotation cell

• Install a new Knelson gravity concentrator (QS-40), which would replace the existing jigs

• Recommission an existing Hardinge regrind ball mill

• Increase leach circuit volume to 800 m3 (35% increase) by replacing two leach tanks with two

5.9 m diameter by 6 m high leach tanks

• Install two new thickeners (30 ft diameter)

• Upgrade the Merrill-Crowe circuit

• Install new water tank

• Upgrade fresh water supply system

CGM’s capital cost estimate to complete the Marmato plant expansion is summarized in Table 17-5,

which shows a total capital expenditure of US$11.6 million over the next four years. SRK has applied

a 25% contingency, which brings the total capex estimate to US$14.5 million over this period.



Table 17-5: Summary of Marmato Process Plant Expansion Capex

Equipment 2020 2021 2022 2023 2024 Total

New Apron Feeder 440,000 440,000

Replace Secondary Cone Crusher

550,000 550,000

Replace Tertiary Cone Crusher

770,000 770,000

New Samplers 55,000 55,000

Install Refurbished Ball Mill (15.5 ft by 22 ft)

2,640,000 2,200,000 4,840,000

New Primary Mill Pumps and Piping

180,000 180,000

Upgrade Concentrate Pumps and Piping

50,000 50,000

New Hydrocyclones (15 Inch) 77,000 77,000

New Linear Trash Screen (Delkor 20 M2)

88,000 88,000

New Flotation Cell (80 M3) 250,000 250,000

Knelson Concentrator (QS40) 660,000 660,000

Recommission Hardinge Regrind Mill

120,000 120,000

Install Two New Leach Tanks (5.9 m Diameter by 6 m High)

275,000 275,000 550,000

Install New Thickeners (30 ft Diameter)

660,000 660,000 660,000 1,980,000

New Merrill-Crowe Filter Presses

132,000 132,000

PLS Clarifier 66,000 66,000

New Merrill-Crowe Deaeration Tower

44,000 44,000

New Water Tank (150 M3) 110,000 110,000

New Water Supply System 220,000 444,000 664,000

Subtotal 5,035,000 3,511,000 1,210,000 440,000 1,430,000 11,626,000

Contingency (@25%) 1,258,750 877,750 302,500 110,000 357,500 2,906,500

Total 6,293,750 4,388,750 1,512,500 550,000 1,787,500 14,532,500

Source: CGM, 2020

17.3 MDZ Process Plant

17.3.1 Processing Methods

The process plant for the MDZ Project is based on a flowsheet with unit operations that are well proven.

The proposed flow sheet uses standard processes for:


• Gravity/Leach/Adsorption


• Cyanide Detoxification

• Tailings Thickening/Filtration

Metallurgical test programs involved SGS Lakefield, Ausenco’s industry experience, and input from

equipment suppliers and were contemplated in the design of the overall proposed flowsheet diagram

shown in Figure 17-2.

ELUTIONELECTROWINNING

CELL

GOLD SLUDGEFILTER PRESS

ACIDWASH

COLUMN

ELUTIONCOLUMN

CARBONDEWATERING

SCREEN

CARBONREGENERATION KILNPROPANE

DRYING OVEN

BARRINGFURNACE

SLUDGETROLLEY

CASCADEPOURING TABLE

DORECARBONQUENCH

TANK

ELUTIONPROPANE HEATER

RECOVERYHEAT

EXCHANGER

HCLWATER

STRIPELUATE TANK

KILN FEEDHOPPER

ICUELECTROWINNING

CELL

CIP TANKS (X6)

BL

PRIMARYCRUSHER

TRASHSCREEN

LEACH TANKS (X4)

TO PLANT

LIMECYANIDE

DETOX TANKS

CARBONSAFETYSCREEN

CARBON SIZINGSCREEN

LOADED CARBONSCREEN

WATER

CuSO4SMBSLIME

CONTINUOUS FLOW INTERMITTENT FLOW

FUTURE

LEGEND

EQUIPMENT

PRE-AERATION TANK

GRAVITYSCREEN

GRAVITYCONCENTRATOR INTENSIVE

LEACHINGFEED TANK

ICU PREGNANTSOLUTION TANK

INTENSIVE LEACHINGREACTOR

ROM ORE

BALL MILL

CYCLONE FEEDHOPPER

U/F

O/FCYCLONECLUSTER

OXYGEN OXYGEN OXYGEN

FLOCCULANT

O/F

THICKENER

PROCESSWATER TANK

OXYGEN OXYGEN

FILTER 2

CHAIN GATE

BACKFILL(BY OTHERS)

FILTER 1

SAG MILL

CuSO4SMBSLIME

SAG MILLFEED CONVEYOR

CRUSHED ORESTOCKPILE

SECONDARYCRUSHERSURGE BIN

SECONDARYCRUSHER

STOCKPILEFEED CONVEYOR

PRIMARY CRUSHERDISCHARGE CONVEYOR

UNDERGROUND

BELTFEEDERS

SCATS BUNKER

SECONDARYCRUSHER

PAN FEEDER

TAILINGS FILTERFEED TANK

FILTRATE TANK

SECONDARY CRUSHERSURGE BIN

CONVEYOR No. 2

PRE-SOAKTANK

TO TAILINGS

NaOHNaCN

WATER

SECONDARYCRUSHERSCREEN

WATER

OXYGEN

METALDETECTOR

AusencoCOPYRIGHT c

A1Ausenco

Toronto, Ontario,

Canada

T +1 416 363 6518

W www.ausenco.com

Jun

23,20

20 -

6:48p

m C:

\Use

rs\luc

y.liu\

Desk

top\10

4317

Mar

mato\

1043

17-0

000-

F-00

1.dwg

- luc

y.liu

THE DRAWING AND ITS CONTENTS ARE CONFIDENTIAL FOR THEPRIVATE INFORMATION OF MARATHON GOLD INC. FOR USEONLY FOR THE PROJECT IT WAS PREPARED AND NOT TO BERELIED UPON OR USED IN WHOLE OR IN PART FOR OTHERPURPOSES OR BY OR FOR THE BENEFIT OF OTHERS WITHOUTPRIOR ADAPTATION AND SPECIFIC WRITTEN VERIFICATION BYAUSENCO.

PRELIMINARY

VIBRATINGGRIZZLY FEEDER

TRAMP METALMAGNET

SELF CLEANINGMAGNET

SECONDARYCRUSHER

DIVERSION CHUTE

TRASHBIN

SCATS BUNKER

HIGH ANGLEPEBBLE CONVEYOR

PEBBLE CRUSHERDIVERSIONCHUTE No.1

PEBBLE CRUSHERDIVERSIONCHUTE No.2(FUTURE)

PEBBLECRUSHER(FUTURE)

TRASHMAGNET

BL

OXYGEN

LIMELIME

BL

TAILINGS FILTER CAKE TAILINGS FILTER CAKE

POLISHING PONDSITE RUNOFF POND

WATERTREATMENT

PLANT

BL

TO ENVIRONMENT

WASHWATER TANK

SECONDARY CRUSHERSURGE BIN

CONVEYOR No. 1

PLANT SITERUN-OFF

Figure 17-2



17.3.2 Plant Design and Equipment Characteristics

The MDZ process plant has a design for an ore treatment of 1,460,000 dry t/y or 4,000 dry t/d based

on a plant availability of 8,059 h/y, or 92%. The process design criteria summary is provided in

Table 17-6.

Table 17-6: Process Design Criteria Summary

o Description o Units o Value

o Ore Throughput o Dry t/a o 1,460,000

o Operating Schedule

o Crusher Availability o % o 67

o Plant Availability o % o 92

o Crusher Operating Time o h/a o 5,869

o Plant Operating Time o h/a o 8,059

o Throughput, Daily - Average o Dry t/d o 4,000

o Plant Capacity, Hourly o Dry t/h o 181

o Gold Grade - Design Mill Head o g/t o 3.48

o Overall Gold Recovery o % o 95.4

o Silver Grade – Design Mill Head o g/t o 5.19

o Overall Silver Recovery o % o 57.8

o Crushing (Two Stage)

o Primary Crusher o Type o Jaw Crusher

o Secondary Crusher o Type o Cone Crusher

o Impact Crushing Work Index (75th Percentile)

o kWh/t o 23.4

o Crushed Ore Stockpile Residence Time (live)

o H o 24

o Crushing Product Size, F80 o Mm o 32

o Grinding

o Circuit Type o o 2C-SAB with Hydrocyclones

o Axb (75th Percentile) o o 24.5

o Bond Ball Mill Work Index (75th Percentile) o kWh/t o 19.0

o Product Particle Size, P80 – Equipment Design

o µm o 105

o Pebble Recirculating Load o % of Fresh Feed o 37

o Gravity Concentration

o Overall Gravity Gold Recovery o % o 35

o Intensive Leaching Batches per Week o # o 7

o Leach-Adsorption

o Leach o o

o Total Leach + Adsorption Time o H o 24

o Number of Pre-aeration Tank o # o 1

o Number of Leach Tanks o # o 4

o Leach Operating Density o % w/w o 43.2

o Cyanide Addition to Leach o kg/t o 0.61

o Pre-aeration + Leach pH Target o o 10.5-11

o Pre-aeration + Leach DO Target o mg/L o 20-25

o CIP

o Number of Carbon-in-Pulp Tanks o # o 6

o Carbon Concentration o g/L o 25

o Overall Leach Gold Extraction o % o 95.1

o Gold Losses

o Soluble o mg/L o <0.015

o Carbon o % o 0.08

o Desorption and Carbon Regeneration

o Desorption o o

o Elution Method o o AARL



o Description o Units o Value

o Carbon Batch Size o T o 4

o Elution Cycles per Week o # o 7

o Electrowinning (EW) o o

o Gravity EW Cathode Type o o Stainless Steel

o Gravity EW Number of Cells o # o 1

o Gravity EW Number of Cathodes/Cell o # o 12

o Elution EW Cathode Type o o Stainless Steel

o Elution EW Number of Cells o # o 1

o Elution EW Number of Cathodes/Cell o # o 24

o Cyanide Destruction

o Method o o SO2/ Oxygen

o Number of Tanks, Parallel o # o 2

o Residence Time, Each o Min o 60

o Detoxification Feed CNWAD o mg/L o 200

o Detoxification Discharge Target CNWAD o mg/L o < 1.0

o SO2 Addition o SO2:g CNWAD o 7.81

o Oxygen Addition o g O2/ g CNWAD o 3

o Lime Addition o g CaO/g SO2 o 0.58

o Tailings Thickening

o Tailings Thickener Underflow Density o % w/w o 60

o Flocculant Addition o g/t o 50

o Tailings Filtering o o

o Tailings Filtration Rate o kg/m2h o 181

o Tailings Filter Cake Moisture Target o % w/w o <15

o Tailings Filter Type o o Horizontal Plate and Frame Pressure

Filter

o Number of Units o # o 2

o Plate Size o Mm o 2640 x 3050

o Effective Filtration Area o m2 o 811

o Number of Chambers per Unit o # o 87

o Chamber Depth o Mm o 50

o Filter Plate Material of Construction o o Polypropylene


17.3.3 Process Plant Description

The MDZ process plant is located NE of the town of Marmato, Colombia. Access to the plant will be

via the plant roads off National Route 25. The primary crusher is located underground, and the

secondary crusher positioned at the surface near the entrance to the tunnel portal. The crushed ore

stockpile is located east of the main plant. The main plant is outdoors and contains the grinding, gravity

recovery, leach/CIP tanks, reagent, elution/carbon regeneration, cyanide detoxification, and tailings

thickening. The electrowinning and refining area are in a separate building to the south section of the

main process plant layout. The thickened tailings are pumped to a filter plant, located next to the main

plant, to be filtered and transferred to the tailings storage facility. Figure 17-3 shows the

mill/leach/reagents general arrangement in plan view.



Figure 17-3: Mill/Leach/Reagents General Arrangement



17.3.4 Primary/Secondary Crushing and Stockpile

RoM ore is transported to the primary crusher located underground and dumped into a RoM bin. The

RoM bin has a 600 mm by 600 mm static grizzly installed to keep large oversize from plugging the

primary crusher feed cavity, and rock breaker to treat any oversize. A vibrating grizzly feeder with 100

mm bar spacing ahead of the of the primary jaw crusher is used to screen out the finer material and

feed the jaw crusher the grizzly oversize material. The jaw crusher will produce a product with an 80%

passing size of 128 mm. The crushing circuit is designed for an annual operating time of 5869 h/a or

67% availability.

The primary crusher product along with the grizzly feeder undersize is conveyed along a short sacrificial belt and transferred to a longer conveyor to the surface and discharge into a surge bin for the secondary crusher. A pan feeder will feed the secondary crusher screen, with the oversize fed to a secondary cone crusher, which produces a product with an 80% passing size of 36 mm. The screen undersize and the secondary cone crusher discharge (fine ore product) are combined and conveyed to the crushed ore stockpile. The stockpile allows for 24 hours of continuous milling operation at the nominal feed rate. Crushed ore is 100% passing 65 mm and 80% passing 32 mm. The crushing and grinding circuits are configured in a 2C-SAB circuit, which is two stage crushing followed by SAG (semi-autogenous grinding mill) and ball mill.

Figure 17-4 shows the primary and secondary crushing arrangement and the stockpile.

Figure 17-4: Primary/Secondary Crushing and Stockpile

Crushed ore will be withdrawn from the stockpile by two variable speed belt feeders. The belt feeders

are sized such that during maintenance, one of the feeders can provide the full mill-feed capacity of

181 tonnes per hour (t/h).



17.3.5 Grinding

Crushed ore from the stockpile is transferred to the SAG mill via the mill feed conveyor. The conveyor

is equipped with a belt scale to provide feed rate data for feed control to the SAG mill.

A SAG mill media bin located adjacent to the mill will add media to the mill on a pre-set scheduled

basis. A similar media bin is located adjacent to the ball mill to add media to the mill at a determined

charging schedule.

Primary grinding is provided by a 6 m diameter by 3.6 m EGL SAG mill, with an installed pinion power

of 2 MW. The SAG mill trommel undersize will report to the cyclone feed hopper, with the oversize

recirculated back to the SAG mill feed conveyor by means of a high-angle pebble conveyor. There is

a contingency for a future pebble crusher installation at the head end of the high-angle pebble

conveyor. Secondary grinding is provided by a 4.9 m diameter by 7.3 m ball mill, with an installed

pinion power of 3.1 MW and operated in closed circuit with hydrocyclones. The classification circuit

will operate at a nominal circulating load of 300% which is a typical for ore of similar characteristics

and target grind size of 80% passing 105 µm. To avoid damage to the cyclone feed pumps and cyclone

clusters, ball mill discharge is screened through a trommel screen to scalp off oversized particles and

broken grinding media. The trommel screen undersize slurry from the ball mill discharges to the

cyclone feed hopper. The slurry is pumped by the cyclone feed pump to the classification cyclone

cluster.

Slurry from the cyclone cluster underflow launders is split, with 40% of the flow feeding the gravity

concentration and intensive cyanide leach circuit and 60% of the flow returning directly back to the ball

mill. The tails of the gravity concentration and intensive cyanide leach circuit return to the ball mill. The

cyclone overflow slurry from the cyclone cluster gravity flows to a trash screen, with the screen

undersize reporting to the pre-aeration tank prior to leaching. Trash screen oversize discharges to a

collection area for removal. A sump pump will be provided in the grinding area to facilitate clean-up.

The pump will discharge into the cyclone feed hopper. Figure 17-5 shows the SAG, ball mill and the

high-angle pebble conveyor, along with the future pebble crusher location.

Figure 17-5: Grinding and Gravity Concentrate and Intensive Leach Area



17.3.6 Gravity Concentration and Intensive Cyanide Leach Circuit

A portion of the cyclone underflow reports to the gravity circuit scalping screen with an aperture of 2

mm. The scalping screen oversize returns to the ball mill feed. The scalping screen undersize reports

to the centrifugal concentrator, which is operated in a semi-batch process. The gravity concentrate is

collected in the concentrate storage cone and is flushed into the intensive cyanidation leach reactor

for leaching. The tails of the centrifugal concentrator reports to the ball mill feed.

The intensive cyanidation leach unit (ICU) extracts the contained gold in the gravity concentrate by

intensive cyanidation. The leach solution from the ICU, containing a mixture of NaCN, NaOH, and

Leach Aid, is prepared in a heated ICU reactor vessel feed tank. The leached residue from the reaction

vessel is washed and returned to the cyclone feed hopper. The pregnant solution from the ICU is sent

to the ICU pregnant solution tank in the gold room for electrowinning and smelting. Figure 17-5 shows

the gravity & intensive leach area adjacent to the grinding area.

17.3.7 Leach and Adsorption Circuit

The leach and adsorption circuit consists of a pre-aeration tank, four leach tanks, and six CIP tanks.

The circuit is gravity fed from the trash screen undersize to the pre-aeration tank. The pre-aeration

and leach tanks are the same dimensions, providing a total residence time of 18 hours at 43% solids

slurry density in the leach tanks. Pre-aeration passivates reactive sulphides such as pyrrhotite and

pyrite, which increases available oxygen and improves cyanide consumption in the leach stage. The

leach tanks allow the gold-bearing solids to encounter cyanide and oxygen, dissolving the gold from

the ore into solution in the form of stable gold-cyanide complex. The tanks are oxygen sparged to

maintain elevated dissolved oxygen levels up to 20 mg/L. Hydrated lime is added to the pre-aeration

tank and first leach tank to reach the target pH range of 10.5 to 11. The discharge of the leach tanks

gravity flows to the first CIP tank. Figure 17-6 shows the aeration and leach tanks in series.

Figure 17-6: Pre-Aeration and Leach Tanks

The six CIP tanks provide a total residence time of 6 hours at 43.2% w/w density. Carbon from the

carbon regeneration circuit is returned to the last CIP tank. The carbon is then circulated counter

current to the slurry flow with carbon transfer pumps. Carbon is maintained at a concentration of 25 g/l



of slurry. Mechanically swept carbon retention screens in each CIP tank keeps carbon in the tank,

while allowing slurry to continue to flow by gravity downstream. The carbon in the first CIP tank is

pumped to the loaded carbon screen in the carbon acid wash, elution, and regeneration circuit. The

slurry from the last CIP tank gravity feeds to the cyanide detoxification tanks. Figure 17-7 shows the

CIP tanks, along with the detoxification tanks, acid wash & elution columns, regeneration kiln and gold

room areas.

Figure 17-7: CIP and Detoxification Tanks, Acid Wash and Elution Columns, Regeneration Kiln, and Gold Room Areas

17.3.8 Carbon Elution and Regeneration Circuit

Slurry from the first CIP tank is pumped to the loaded carbon screen. The oversize from the loaded

carbon screen flows by gravity and is directed to the acid wash column with 4 t carbon capacity. Screen

undersize is returned to the No. 1 CIP tank. Prior carbon elution, acid soluble foulants that have loaded

onto the carbon surface are dissolved in a dilute acid washing stage. Hydrochloric acid is diluted with

fresh water in an in-line mixer to provide the required acid wash solution concentration and injected

into the acid wash column. Acid solution is circulated through the acid wash column and then rinsed

with fresh water, prior to transfer to the elution column. The spent acid solution and rinse water are

drained to the acid wash area sump and transferred to the cyanide detoxification tanks.

The Anglo American Research Laboratory (AARL) process is used for the gold stripping (elution) from

loaded carbon. The elution system comprises an elution column, strip eluate tank, strip eluate pump,

and an elution heater package. This equipment operates in a closed loop with the electro-winning cell

located inside the gold room.

The elution process begins with filling the elution column with a set volume of water, along with cyanide

and sodium hydroxide to obtain a strong NaCN and NaOH solution. The strong solution is heated to

120°C, and allowed to-soak in the column for 30 minutes.



The pre-soak solution is transferred to the pregnant eluate tank. Elution water is heated to 120°C and

pumped through the heat exchanger and elution heater. The heated water is then cycled through the

elution column to the pregnant eluate tank.

After completion of the elution process, stripped carbon is transferred from the elution column to the

carbon dewatering screen. The screened carbon is fed into the kiln feed hopper and a screw feeder

meters the carbon into the carbon regeneration kiln. The carbon regeneration kiln is propane-fired,

and is a horizontal, rotary unit designed to regenerate 100% of the stripped carbon at a temperature

between 700-750°C.

Regenerated carbon discharges by gravity from the kiln to a quench tank to cool down and is then

transferred to the carbon sizing screen. The barren carbon is screened and the oversize carbon reports

to the last tank in the adsorption circuit. Fine carbon is discarded to the cyanide detoxification tanks.

17.3.9 Electrowinning and Gold Room

Gold recovery is performed by electrowinning pregnant solutions from the ICU and the elution circuits

and smelted into doré bars. The process takes place within the gold room, a secure area that is

equipped with access control, intruder detection, and closed-circuit television equipment.

The pregnant solution from ICU and elution are each processed in dedicated electrowinning cells, with

their own pregnant solution tanks and pumps recirculating the solution through the electrowinning cells

fitted with stainless steel mesh cathodes. Gold is deposited onto the cathodes and the resulting barren

solution is pumped to the first leach tank.

Gold flake washed from the cathodes and cell floor sludge are drained from the electrowinning cell to

a sludge hopper. The sludge is then pumped to a plate and frame filter. The filter cake (gold/silver

sludge) is loaded from the sludge filter into trays on the electrowinning sludge trolley. The sludge is

then oven dried, combined with fluxes, and smelted in a barring furnace to produce gold doré bars.

17.3.10 Cyanide Detoxification

The CIP tails gravity flows to the cyanide detoxification distribution box, which is combined with

washdown water from different areas in the plant and feeds two detoxification tanks in parallel. The

tanks provide a total residence time of one hour each to reduce the weak acid dissociable cyanide

(CNWAD) concentration from 200 mg/L to less than 1 mg/L to comply with the environmental

requirements prior to deposition in the tailing storage facility.

The cyanide detoxification method used is the SO2/O2 process. The process requires the use of

oxygen, lime, copper sulphate, and sodium metabisulphite (SMBS) as the SO2 source. Oxygen is

sparged into each tank bottom and the tanks have intensive agitation to ensure completion of the

reactions. Once the slurry is treated, the tailings gravitate to the carbon safety screen. The carbon

safety screen oversize is collected into a carbon bag for re-use or disposal, while the screen undersize

transfers to the tailings thickener distribution box.

17.3.11 Tailings Thickening and Filtration

Prior to deposition to the DSTF, the tailings must be thickened and filtered to achieve the 15% w/w

moisture content for dry stacked tailings. The tailings thickening occurs at the MDZ process plant, with

the combination of the tailings slurry, returning filter press filtrate water, and any other washdown water



from the plant areas in the tailings thickener distribution box. Flocculant is added to improve settling

rates and achieve an underflow density of 60% w/w solids. The tailings thickener overflow gravity feeds

to the process water tank for recirculation back into the process. The thickened tailings report to the

filter feed tank for pumping to either the paste backfill plant or the tailings filter plant located next to the

main plant. Figure 17-8 shows the tailings thickener and the thickener underflow storage tank area.

Figure 17-8: Tailings Thickener and Thickener Underflow Storage Tank Area

The tailings filter plant receives thickened tailings in the filter feed tank. Two horizontal plate and frame

pressure filters are used to treat the tailings and reduce the moisture content below the 15% w/w target

for dry stacked tailings. The filters are situated to allow the filter cake to drop below the filter floor to a

filter cake loadout bunker and transferred by a front-end loader to the DSTF equipment. Figure 17-9

shows the tailings pressure filter plant area.



Figure 17-9: Tailings Pressure Filter Plant Area

17.3.12 Reagents

Lime

The hydrated lime is supplied in bulk bags and is lifted using a frame and the mobile crane onto the

bag splitter located above the lime mix/storage tank. The tank will be partially filled with process water

at the beginning of the mixing sequence to achieve a slurry concentration of 20% w/w. The lime slurry

is used as a pH modifier and is pumped by the lime distribution pumps to the various locations in the

plant through a ring main. Unused lime returns to the lime mix/storage tank. Spillage in the lime area

is collected in the lime sump and pumped to the cyanide detoxification distribution box.

Sodium Cyanide

Sodium cyanide is used as a gold lixiviant and will be supplied in briquette form in 1 t bulk bags shipped

in boxes. The boxes will be offloaded by forklift and stored in a limited access cyanide storage facility

that is part of the cyanide mixing area. The sodium cyanide will be dissolved into a 20% w/w solution.

Sodium cyanide is in briquettes and is pre-buffered with sodium hydroxide to ensure high solution pH

and prevent hydrogen cyanide formation in the mixing system.

The mixed solution gravitates to the cyanide holding tank that is sitting under the mix tank. The cyanide

solution is pumped by the sodium cyanide pumps to dosing points in leaching and elution. Spillage in

the sodium cyanide area is collected in the sodium cyanide sump and pumped to the cyanide

detoxification distribution box.

Copper Sulphate

Copper Sulphate is used as a catalyst for the cyanide detoxification process and will be supplied in 25

kg bags. The copper sulphate mixing tank will be partially filled with process water and the copper

sulphate will be manually loaded into the tank by way of a bag splitter. The copper sulphate will be

dissolved into 20% w/w solution concentration and then gravitates to the copper sulphate storage tank

under the mixing tank. The copper sulphate solution is distributed by the copper sulphate distribution

pump to the cyanide detoxification distribution box. Spillage in the copper sulphate area is collected in

the copper sulphate sump and pumped to the cyanide detoxification distribution box.

Sodium Metabisulphite (SMBS)

SMBS is the source of SO2 in the SO2/O2 process for cyanide detoxification and will be supplied in

bulk bags. The SMBS mixing tank will be filled with process water and the SMBS will be lifted by mobile

crane on to the bag splitter above the mixing tank. The SMBS will be dissolved into a 20% w/w solution

concentration and the solution will be transferred to the SMBS storage tank by the SMBS transfer

pump. The SMBS solution is pumped by the SMBS dosing pump to the cyanide detoxification

distribution box. Spillage in the SMBS area is collected in the SMBS sump and pumped to the cyanide

detoxification distribution box.

Sodium Hydroxide

Sodium hydroxide is used as a pH modifier and will be supplied in liquid bulk containers at a

concentration of 50% w/w. A sodium hydroxide container and dosing pump will be located at the

intensive cyanide leach area to dose directly to the ICU. A second container will be located at the

reagent area for dosing to the pregnant elution and ICU solution tanks in the electrowinning & gold



room area and to the sodium cyanide mixing tank. Spillage in the intensive cyanide leach area is

collected in the intensive cyanide leach sump and pumped to the first leach tank. Spillage in the sodium

hydroxide area is collected in the sodium cyanide sump.

Hydrochloric Acid

Hydrochloric acid is used in the acid wash cycle to remove any foulants that may be present on the

carbon surface prior to elution. The hydrochloric acid will be supplied in liquid bulk containers at a

concentration of 33% w/w and will be dosed to the acid wash column. Spillage in the hydrochloric acid

area is collected in the hydrochloric acid sump trap.

Sulphamic Acid

Sulphamic acid is used in the strip elution to remove calcium carbonate scale from heat exchangers.

The sulphamic acid will be supplied in liquid bulk containers at a concentration of 33% w/w. Spillage

in the sulphamic acid area is collected in the sulphamic acid sump trap.

Flocculant

Flocculant is used to help improve the settling of solids in the tailings thickener and will be supplied in

a powder form in bulk bags. A self-contained flocculant metering and mixing system will be installed

for controlled batch mixing at a solution strength of 0.5% w/w using process water. The mixed

flocculant solution will be piped to the flocculant storage tank below and dosed to the tailings thickener

distribution box, where it will be further diluted at a ratio of 1:10 flocculant to process water to aid in

flocculant dispersion. Spillage in the flocculant area is collected in the flocculant sump and pumped to

the tailings thickener distribution box.

Gold Room Fluxes

Borax, Nitre, Silica, and Sodium Carbonate are the fluxes used in the gold room and will be supplied

as dry solids in 25 kg bags.

Anti-Scalant

Anti-Scalant is used to inhibit scale build-up in the process water lines and will be supplied in liquid

form in drums. The drum will be located beside the process water tank and dosed into the process

water tank by metering pump.

Activated Carbon

Activated carbon will be supplied in solid granular form in bulk bags. The activated carbon will be

added to the carbon quench tank, by way of bag breaker, when needed to make up any carbon losses.

Oxygen

Oxygen is injected into the pre-aeration and leach tanks to help enhance the cyanide reaction with the

slurry and improve the dissolution of gold. Oxygen is also used in the cyanide detoxification to react

with SO2 and cyanide to breakdown the cyanide compounds. Oxygen is supplied in liquid form and

delivered to the oxygen storage tanks. The liquid oxygen flows through a vaporizer to transform the

liquid to gas for distribution. Dosing is done through spargers to the bottom of the pre-aeration, leach,

and cyanide detoxification tanks.



17.3.13 Services and Utilities

Process/Instrument Air

High-pressure air is produced by compressors and is stored in an air receiver tank to meet the plant

requirements. A portion of the high-pressure plant air is passed through an air dryer to supply

instrument air to the various field instruments in the plant. The primary crusher is equipped with a local

air compressor and air receiver tank to provide the underground installation with compressed air. The

tailings filter plant is equipped with high- and low-pressure air compressors and receivers to provide

the necessary air for diaphragm pressing and air blowing cycles for the vertical plate pressure filters.

Diesel Fuel

The diesel storage area is supplied and maintained by the supplier. The diesel is used in various

mobile equipment and vehicles related to operations and maintenance in the process plant.

Propane Gas

The propane storage area is supplied and maintained by the supplier. The propane gas is used as the

energy source for the elution column heating and the kiln heater.

17.3.14 Water Supply

Fresh Water Supply System

Fresh water is supplied to the process plant fresh/fire water tank. The fresh water is used in various

reagent areas and locations in the process that require low suspended solids and salt content. The

fresh/fire water tank also provides emergency fire water. A freshwater truck supplies the tailings filter

plant with makeup water for the gland, fire, and potable water needs.

Process Water Supply System

Process water is mainly supplied from the tailings thickener overflow launder, with makeup water being

provided by the site runoff pond and/or the fresh water tank. Process water is used throughout the

plant, mostly in the grinding circuit for pulping the ore to allow for transport, pumping, and classification.

A dedicated process water pump provides the tailings filter plant with process water for cloth wash and

washdown water.

Gland Water Supply System

Gland water is used to keep all slurry pump glands clean from abrasive solids. A dedicated pump is

used to provide most of the slurry pumps, with an additional booster pump for the tailings pipeline

pumps to achieve the higher operating pressure required for the pump application. The tailings filter

plant has a separate gland water pump to provide water to the filter feed slurry pumps

17.3.15 Operating Costs

Summary

The operating cost estimates for the process is presented in Q2 2020 US Dollars (USD or $). The

estimate was developed to have an accuracy of 25%. An additional operating cost of $6.38/oz Au sold

will be applied in the financial model for the refining and transportation charges, based on the current

rates applicable to the client from their existing operation in the area.



The operating cost estimates for the life of mine are included in Table 15.1. The overall life-of-mine

operating cost is $16.8M over 16 years, or $11.51/t of ore milled.

Basis of Estimate

The assumptions made in developing the operating costs include:

• Cost estimates are based on Q2 2020 pricing without allowances for inflation.

• For material sourced in Colombian Pesos (COP), an exchange rate of 3300 COP per USD

was assumed.

• Fuel costs were provided by the client based on the contracts for the client’s existing plant.

The diesel cost used was $0.63/L and propane cost used was $0.90/Kg.

• The power costs were provided by the client based on the contacts for the client’s existing

plant. The unit price of $0.09/kWh is from the hydro provider (285 COP/kWh).

• The labor is assumed to primarily come from the local region around Marmato

Processing Operations

The costs for processing are generated from the labor, power, reagents and consumables,

maintenance, laboratory and assays. An overall average annual cost was estimated to be $18M/y or

$12.31/t milled.

A breakdown of the Opex and unit costs are presented below in Table 17-7.

Table 17-7: Operating Cost Summary

Cost Center US$M US$/t

Reagents & Consumables 10.1 6.95

Plant Maintenance 0.86 0.59

Power 5.34 3.66

Laboratory 0.23 0.16

Labor (O&M) 1.30 0.89

Vehicles 0.10 0.07

Total 18.0 12.31


Consumables

The consumables were based on reagent consumptions, process consumables, and propane utilities

for heating, using nominal consumption rates. Individual reagent costs were obtained through existing

pricing from the client’s operation and alternative sources. The total consumables costs are $10.1M/y

or $6.95/t milled.

Maintenance

The maintenance costs for the process were calculated based on process capital costs for each area

using a factor between 4 and 5%. The total maintenance costs are $0.86M/y or $0.59/t.

Power

The power consumption was estimated from the average power utilization of all major equipment

included in the process. The annual power usage was calculated at 61,562.3 MWh/y at a unit cost of

$0.09/kWh (285 COP/kWh) based on the current utility rates provided to the client from the energy

provider ISAGEN. The total power costs are $5.34M/y or $3.66/t.



Laboratory

The laboratory costs were based on client and in-house data. The costs do not include the mine

samples or labor to operate the laboratory. The total laboratory costs are $0.23M/y or $0.16/t.

Labor (Operation & Maintenance)

The process labor requirements for the operation are 107 employees over multiple shifts. Most of the

workers will be local, with specialized positions potentially drawing from other regions in Colombia. For

the purpose of the labor costs, the compensation used was a salary rate with a 60% burden.

An organizational roster listing the labor requirements for the process operation and maintenance is

shown in Table 17-8. The total labor costs are $1.3M/y or $0.89/t.

Table 17-8: Operations and Maintenance Manpower Schedule

Labor/Contractor Summary Rotation #/Shift # Shifts Quantity

Total Process Labor 36 76 107

Process Upper Management

Mill Manager 10.5-hour shifts 1 1 1

Mill Maintenance Superintendent 10.5-hour shifts 1 1 1

Mill Chief Metallurgist 10.5-hour shifts 1 1 1

Mill Administrative Assistant 10.5-hour shifts 1 1 1

Subtotal 4 4 4

Mill Operations

Gold Room Supervisor 12 hour shifts 1 3 3

Gold Room Operator 12 hour shifts 1 3 3

Shift Supervisors 12 hour shifts 1 3 3

ROM / Crushing Operator 12 hour shifts 1 3 3

ROM / Crushing Assistant Operator 12 hour shifts 2 3 6

Control Room Operator 12 hour shifts 1 3 3

Grinding / Gravity Operator 12 hour shifts 1 3 3

Tailings Operator 12 hour shifts 2 3 6

CIP Operator 12 hour shifts 1 3 3

Reagent Operator 12 hour shifts 1 3 3

Subtotal 12 30 36

Technical Services

Junior Metallurgical Engineer 10.5 hour shifts 1 1 1

Metallurgical Technician 10.5 hour shifts 2 1 2

Chemist 10.5 hour shifts 2 1 2

Sample Preparers 12 hour shifts 2 3 6

Analytical Technicians 12 hour shifts 2 3 6

Subtotal 9 9 17

Mill Maintenance

Maintenance Planner 10.5 hour shifts 2 1 2

Maintenance Foreman 12 hour shifts 1 3 3

Mechanics 12 hour shifts 3 3 9

Welders 12 hour shifts 1 2 2

Electricians 12 hour shifts 2 3 6

Instrument Technician 12 hour shifts 1 3 3

Trades Assistants 12 hour shifts 1 3 3

Subtotal 11 18 28




Vehicles

The vehicle costs are based on a scheduled number of light vehicles and mobile equipment, including

fuel, maintenance, replacement parts, and annual insurance fees. Fuel costs were established based

on utilization and a diesel cost of $0.55/L. A breakdown of the vehicles and equipment is shown in

Table 17-9.

Table 17-9: Light Vehicles and Mobile Equipment Summary

Category Vehicle Type Number Utilization

Light Vehicles

4 X 4 Crew Cab 1 2h/d 5 days per week

4 X 4 Pickup 1 2h/d 7 days per week

4 X 4 Crew Cab 1 2h/d 5 days per week

4 X 4 Pickup 1 2h/d 7 days per week

Mobile Equipment

Skid Steer Loader 1 5h/d 7 days per week

3 t All Terrain Forklift 1 5h/d 7 days per week

35 t All Terrain Crane 1 0.5h/d 5 days per week

It 30 Front End Loader 1 5h/d 5 days per week

Elevated Work Platform 1 3h/d 7 days per week

5t Hiab Flat Bed Truck 1 3h/d 7 days per week




18 Project Infrastructure

18.1 General Site Access

The Project is in the Municipality of Marmato in Caldas. The Marmato Project is located on the eastern

side of the Western Cordillera (Cordillera Occidental) of Colombia on the west side of the Cauca River.

The Marmato Project is located approximately 125 km south of Medellín, the capital of the department

of Antioquia, Colombia. Medellín is the second largest city in Colombia with a population of

approximately 2.5 million. The Project is located in the department of Caldas near El Llano.

Figure 18-1 shows the location of the Project.

Source: CGM, 2017

Figure 18-1: Marmato Project Location

Primary access to the site is via the Pan American Highway, Colombia Highway 25. The road is a

paved two lane improved highway that winds through the mountainous area south of Medellín and

then follows the Cauca River to the turn off to the Project. The route from Medellín is via Itaguí (7 km),

Caldas (12 km), Alto de Minas (13 km), Santa Barbara (27 km), La Pintada (26 km), La Guaracha del

Rayo (32 km), and then a turn onto a secondary road to the community of El Llano that is the

community closest to the Project. From El Llano, the road is paved but partially single lane another 2

km to the site security gate. An improved dirt road continues up the mountain another 4 km to the

community of Marmato, where the artisanal miners are working the Zona Alta portion owned by Gran

Colombia. Approximately 40% of the 1,200 employees currently employed by the Project live in El

Llano, with the remainder traveling to work from various communities in the area.

The Pan American Highway continues south and east 90 km to another large regional city, Manizales,

and then on another 270 km to Bogotá, the capital of Colombia. Air access through international and

regional airports is available in Medellín, Manizales, and Bogotá.



Field personnel for the exploration program have been employed from the towns of Marmato and El

Llano and neighboring municipalities. In the long term, personnel currently working on the large

number of small scale mines and from the surrounding region would be able to supply the basic

workforce for any future mining construction and operation.

18.2 Marmato Existing UZ Operations Infrastructure

The existing operating Marmato UZ operations have fully developed site infrastructure. The UZ

operations have an access road, onsite roads to access facilities, mine portals, processing facility,

administrative and offices, shops, warehousing, electrical maintenance buildings, helicopter pad, camp

facility, compressed air systems, ventilation systems, sand backfill system for underground, tailings

storage facilities, water supply, mine pumping systems, electrical supply and distribution, solid waste

handling facilities, septic systems, security, and communications systems.

18.2.1 Existing Project Access

The general Marmato Project access is described in section 18.1. The secondary road to the existing

operation and the general layout of the facilities is shown in Figure 18-2.



Source: SRK, 2019

Figure 18-2: Marmato General Access and Major Facilities

18.2.2 Existing Project Facilities

The UZ operations have a fully developed infrastructure and facilities that include a security checkpoint

that provides access to the office and administrative office area. The facilities include employee

motorcycle parking, meeting area, multiple shops and warehouses, a camp with cafeteria, exercise

and sports field, equipment storage yards, compressor station, welding shop, a 500 kW backup

generator, processing plant, underground mine, explosives storage a short distance from the mine that

is managed by the military, main power substation and distribution powerlines with motor control

centers at key loads. The site has three portals that access the mine workings. A yard that has rail

through it near the portal allows servicing of the mine cars and locomotives. Figure 18-3 shows the

overall plant and camp area.



Source: CGM, Modified by SRK, 2019

Figure 18-3: Marmato Existing Project Site Map



18.2.3 Energy Supply and Distribution - Existing Marmato Project

Power to the site is provided through the Colombian power company Central Hidroeléctrica de Caldas

(CHEC), a subsidiary of Empresas Públicas de Medellín (EPM) through existing local substations.

Substantial transmission capacity is available in the region around the Project, with energy provided

over the transmission system by the third largest electricity producer in Colombia, ISAGEN.

The main substation feeding the Project has a capacity of 40 MVA and supplies the Project at 33 kV

through the El Dorado substation to the CGM principal substation. The 8.1 MVA main CGM substation

has six transformers that provide power to the mine, mill, and other facilities.

The loads supported by each transformer are provided as follows:

• Transformer 1 (2,000 KVA) steps the 33 kV power down to 13.2 kV and feeds the three mine

substations that in turn feed the compressors, pumps and offices/shops at 440 VAC

• Transformer 2 (2,000 KVA) feeds the mine at 13.2 kV through three separate mine

transformers that in turn feed the various mine levels, hoists, pumps, and mine equipment.

The equipment operates on 440 VAC

• Transformer 3 (600 KVA) feeds the ball mill at the processing plant at 4,160 VAC

• Transformer 4 (1,250 KVA) and 5 (630 KVA) feeds two compressors each at 440 VAC

• Transformer 6 (1,600 KVA) provides 440 VAC for the beneficiation plant

The Project one-line electrical diagram is shown in Figure 18-4.



Source: CGM, 2019

Figure 18-4: Marmato Electrical System Schematic



18.2.4 Site Water Supply

Water supply for the existing Marmato mining activities is currently provided by a combination of

underground dewatering and reclaim from the existing DSTF. The current Project has adequate water

supplies but can be challenged during dry portions of the year. The Project has contingency plans to

draw water from either the Cascabel or Cauca River, with the Cauca River being the preference and

a water supply pumping system is included in future plans for the MDZ that will support the existing

UZ operations too.

18.3 MDZ Introduction

The overall site plan can (Figure 18-5) shows the major project facilities, such as underground crusher,

underground portal, stockpiles, process plant, DSTF, mining services, accommodations, access

roads, and office buildings.

18.4 MDZ Process Plant Site Location

Process plant site is located about 3 km west by road of the ore body, on a naturally occurring plateau

having 30% east‐west and 15% north-south grade slope.

The location of the plant was selected based on the following factors:

• Proximity to the Marmato deposit, resulting in shortest length of the decline

• Utilizing a naturally occurring plateau, relatively flat comparing to surrounding terrain

• Ground stability for process plant foundations and avoiding fluvial deposit

• Minimizing upstream watershed and land slide risk

• Land availability and land agreement

The proposed plant site location was visited on Dec 5, 2019 during a site visit to the existing operation.




Figure 18-5: Overall Site Plan



18.4.1 Site Geotechnical

At the time of completion of this report, there is no site-specific geotechnical investigation report

available for the proposed plant site. A bore hole and test pit plan were prepared based on the

proposed PFS process plant layout and location of major structures and foundations. For field

geotechnical drilling and investigation, a campaign by the geotechnical consultant is planned to be

carried out in the coming months. The PFS level design of the foundations was based on preliminary

recommendations received from IRYS Ingeniería de Rocas y Suelos S.A.S., which relied on local

geotechnical knowledge and nearby historic drillings.

Based on previous geotechnical investigations by KP (Project No. EL202-00153/07) in 2012, west of

the process plant, the ground geology consists of

• Clay/silt, some sand, medium plasticity, brown and speckled red and white, residual soil, some

visible rock over burden at varying thickness

• Rock, intrusive igneous, moderately to highly weathered, moderate to highly fractured, weak

to moderately strong, white and brown, oxidized mostly feldspar and quartz fabric

18.5 MDZ On-Site Roads and River Crossings

18.5.1 Site Access Road

The access to the process plant site will be possible via three new access roads:

• A new 400 m long, 8 m wide, E-W gravel road connecting the process plant lowest bench to

the Marmato public road will be constructed. This road will be constructed at the start of

construction to access the site.

• A new 750 m long, 8 m wide, N-S gravel road connecting the process plant lowest bench to

the camp and administration area located SW of the process plant will be constructed,

adjacent to the town of El Llano.

• A new 400 m long, 8 m wide, E-W gravel road connecting the Marmato public road the 750 m

long camp access road described above. This road will be the main access to the site.

Designed access roads incorporate use of Mechanically Stabilized Earth (MSE) retaining walls,

Gabion basket retaining walls, soil nailing and slope stabilization.

18.5.2 River Crossing

The new site access roads are crossing seasonal streams and creeks at several locations. Corrugated

steel culverts will be placed at water crossing. The culverts are sized to accommodate the peak flow

of water during a storm event.

18.6 MDZ Water Supply

18.6.1 Water Requirements

Overflow from the tailings thickener and site runoff pond decant water meet the main process water

requirements. Fresh water provides any additional make-up water requirements.



18.6.2 Run-Off Water Collection and Treatment System

Process plant surface water will be graded to naturally drain water to connection swells and ditches

through the plant. The collected water from drainage ditches will be discharged to a 6,000 m3 capacity

lined storm water management pond (SWMP) located to the west of the plant. Water quality in the

SWMP will be monitored and tested for compliance with local environmental discharge requirements.

A Water Treatment Plant (WTP) will be placed adjacent to the SWMP to treat the water, if required,

prior to discharge to environment.

18.6.3 River Water Collection and Treatment System for MDZ and UZ

A new water collection and pumping facility designed by CGM will be constructed to remove water

from the Cauca River from a basin with a floating suction and pump into two separate systems that

will feed the UZ plant and the MDZ plant. Each of the separate systems include a raw water tank, a

sand filter, and pumps capable of pumping water from the tank through two separate pipelines, one to

the MDZ and one to the UZ. The pumps will operate at 35 L/s. The dynamic head on the UZ system

will be 207 m and the MDZ system is 294 m.

18.7 MDZ Power Supply

18.7.1 Electrical Power Source

Major electrical power will be required at the MDZ plant as all process facilities and major infrastructure

buildings are situated there. Electrical power to the MDZ plant is planned to be supplied by Central

Hidroeléctrica de Caldas S.A. (CHEC) from the Salamana 115 KV substation located 15 km away.

Site power will be obtained from a 115 KV HV line that will be provided by the local power authority up

to the MDZ plant outdoor substation. A peak demand of 28 MW is required for the facility, of which

approximately 12 MW are required for the process plant and 16 MW are required for the mining loads.

18.7.2 Electrical Distribution

The plant electrical system is based on a 4.16 kV, 60 Hz distribution. The 115 kV feed from the local

power authority will be stepped down to a 4.16 kV by 2 x 10/13.3 MVA ONAN/ONAF transformers at

the plant main substation and will supply the plant main 4.16 kV switchgear housed in the switch room

next to the plant main substation through a 4.16 kV cable bus.

For the mining load substation, the 115 kV feed from the local power authority will be stepped down

to a 13.2 kV by 1 x 25/33.25 MVA ONAN/ONAF transformer and will supply the mining substation

through a 13.2 KV overhead line.

The following substations/electrical rooms will be provided:

• Plant main HV switchyard (outdoor substation)

• Process Plant E-Room

• Grinding Plant E-Room

• Primary Crusher E-Room

• Secondary Crusher E-Room

• Reagents & Process plant E-Room

• Tailings & Filtration E-Room



Electrical rooms will house the 4.16 KV switchgear, MV VFDs, 440 V motor control centers (MCCs),

LV VFDs, plant control system cabinets, lighting transformers, various distribution boards and UPS

power distribution.

Overhead power lines of 4.16 kV will provide power to various remote facilities, ancillary buildings and

camp. Pole mounted and/or pad mounted transformers will step down the voltage at each location,

and supply 440V to the respective remote facilities, ancillary buildings and camp.

18.7.3 Electrical Rooms

Electrical buildings will be prefabricated panel buildings to minimize installation time on site. The

buildings will be installed on a structural framework over 2 m above ground level to allow for bottom

entry of cables into electrical switchboards, panels, MCCs and cabinets. The electrical buildings will

be installed with HVAC units and suitably sealed to prevent ingress of dust. They will be in the process

plant area and as close as possible to the main load points, to reduce cost. In order to reduce the size

of the E-Rooms and avoid heat issues, the transformers for each electrical room will be oil natural air

natural (ONAN) type and installed just outside each electrical room.

18.7.4 Transformers

The main power transformers are 115 kV/13.2 kV and 115 kV/4.16 kV and will be ONAN, with

provisions for future oil natural air forced (ONAF), cooling configuration and will have either OLTC (on-

line tap changer) or external voltage regulators. All plant 4.16 kV/440 V distribution transformers will

be ONAN, with provisions for future ONAF, cooling configuration and will have a de-energized tap

changer.

18.7.5 Standby/Emergency Power Supply

The site is provided with a 1 MW standby diesel generator sized to supply critical process loads and

life safety systems. The standby diesel generator is located close to the process plant main substation

and connected to the 4.16 kV main substation switchgear. In this way, a single generator set can

supply standby power to all facilities using the normal power distribution system.

18.7.6 Ball and SAG Mill Drives

In this study, the SAG and ball mill motors are equipped with liquid rheostat and use a slip energy

recovery starting method to minimize voltage drop impact on the utility supply system during motor

start-up. A PWM-based slip energy recovery system on the SAG mill motor will provide variable speed

and energy recovery.

Two 1 MVAR harmonic filters and capacitor banks have been added on the 4.16 kV main busbars to

achieve a 0.95 PF at the supply end, to minimize harmonic impact on the power distribution system.

18.7.7 Redundancy

Redundancy for the site electrical power distribution system has been minimized to optimize capital

costs. Warehouse spare transformers are being carried to minimize long shutdowns in the unlikely

event of transformer failure, rather than installing standbys.



18.8 MDZ Mine Operations Support Facilities

18.8.1 Mine Administration and Dry Building

The main mine administration building will be a two-story modular construction building, 850 m² each

floor, located at the camp and administration area. 50% of the ground floor will be used as the mess

hall, and the balance of the first floor and second floor will house cubical, shared and private offices

for mine administrative personnel.

The mine dry building will be a single-story modular construction building with a 550 m² footprint,

located at the camp and administration area. This building is used primarily by the local workforce

employed as miners and plant operators. The workforce will arrive to this facility at the beginning of

each shift and will be transported from the dry facility to the work area (process plant and mine portal)

by bus. This is intended to limit traffic of private vehicles and motor vehicles within the process facility.

18.8.2 General Maintenance Building

The general maintenance building is a 20 m wide, 50 m long, and 12 m high, pre-engineered (metal)

building with 20 tonne capacity overhead crane. All the process plant and mine equipment repair and

maintenance will be performed in this facility.

18.8.3 Truck Wash Facility

The underground mining trucks will be washed on a designated truck wash pad area adjacent to the

general maintenance building. The pad consists of concrete slab-on-grade with a water collection

sump.

18.8.4 Truck Fuel Facility and Equipment Ready Line

The area in front of the general maintenance shop will be the designated ready line parking for mobile

equipment.

The diesel fuel storage and dispensing facility will be located in this area, with adequate fire separation

to comply with NFPA and local fire code.

18.8.5 Explosives Storage

The explosive storage shed will be placed at an isolated secure area located between the process

plant and camp/administration area.

18.9 MDZ Process Support Facilities

18.9.1 Mill Administration Office and First Aid Facility

The mill administration office and first aid facility will be a single-story modular construction building,

with a 250 m² footprint, located near the process plant.

18.9.2 Laboratory

The laboratory will be an assortment of prefabricated, single-story, modular buildings on precast

concrete blocks, with a 140 m² footprint. It will house the equipment for typical site assays.



18.9.3 Warehouse and Storage Yard

The warehouse building is a 16 m wide, 28 m long and 12 m high, pre-engineered (metal) building

with a 10 t capacity overhead crane. This building will be used for the storage of process reagents,

consumables and sensitive equipment that require storage indoors.

The area south of the future pebble crusher structure is designated as an outdoor storage yard which

will be used as a laydown area during construction.

18.9.4 Gatehouse and Weigh-Scale

The security and gatehouse building are a 12 m wide, 3 m long and 3 m high, prefabricated building

located near the process plant at the main entry to the process area.

18.10 Common Support Facilities

18.10.1 Man Camp

The camp will be located south of the process plant, and north of the town of El Llano. The camp will

comprise a three-story modular construction building which will house construction personnel during

construction and will transition to housing plant operation personal after commissioning.

The camp is sized for a total of 466 personnel. There will be 58 personnel in single-bed rooms, 18 in

two-bed rooms and 390 in six-bed rooms. The camp facility also includes a kitchen and recreation

area.

18.11 MDZ Support Facilities

18.11.1 Communications

All PLCs within the plant will be linked to a common ethernet communications network, where practical.

The backbone of the communications network will be via fiber optic cabling, with spare cores provided

for future use. Copper CAT 6 cabling will be used for the final connection to individual points.

The main hub for all PLC communication will be installed in the plant services switchroom. Fiber optic

breakout boxes, ethernet switches, media converters and power supplies will be provided and installed

wherever required.

All I/O signals for the PLCs will be via standard digital and analog modules. The PLC equipment

installed within each area will function autonomously, such that a failure of the PLC in one plant area

will not affect the other areas.

18.11.2 Wastewater Treatment

Two modular containerized sewage treatment plants will be collecting and treating the sanitary sewage

for the project. One at an average 50 m3/day processing capacity located in the camp/admin area and

one at an average 10 m3/day located at the process plant.



18.11.3 Solid Waste Disposal

Solid wastes will be separated at the site by type and stored in a designated waste storage area for

disposal in regulatory approved disposal sites. Underground waste rock will be utilized for construction

of the DSTF facilities (see section 18.13). All other waste rock will be used for backfill in secondary

stopes underground.

18.12 MDZ Site Preparation

18.12.1 Site Earthwork

The process plant area will be graded to five cascading benches following the natural topography as

described in section 18.4:

• Bench 1: Filtration plant

• Bench 2: Process plant

• Bench 3: Reclaim tunnel and future pebble crusher

• Bench 4: Secondary Crusher

• Bench 5: Mine Portal

The pads will be mainly cut in west of each pad and filled in east of each pad. The transition between

each pad will be constructed of 1V:1.5H sloped grade, soil nailed stabilized slopes and Mechanically

Stabilized Earth (MSE) retaining wall as required, following the footprint for process plant layout.

The pads are accessed via plant roads at the south edge of the process plant pads. The in-plant roads

have a slope of 10% to 14% max.

18.12.2 Site Foundations

The grading of the process plant benches is designed to have all major foundations placed in cut,

allowing the foundation to rest on undisturbed competent ground or bed rock. Only light foundations

are placed on compacted fill.

All foundations are designed as shallow foundations. No deep foundation (piles) or rock are considered

in the design. Suitability of shallow foundations shall be confirmed based on the findings of the field

geotechnical investigations

18.13 MDZ Cemented Paste Backfill Plant

The mining method selected for the MDZ requires the use of cemented paste backfill to support the

stopes and achieve the required mineralized material recovery. The paste backfill plant is located on

the surface of a bench near the portal of the MDZ. The paste backfill is made from a combination of

tailings, water and cement binder.

Agitated tailings from the filtered tailings feed tanks at the MDZ plant are pumped approximately 400 m

by two 6 by 4 centrifugal transfer slurry pumps in series to the agitated paste filter feed tank at the

paste plant through a 200 mm NB HDPE DR9 overland pipeline.



The thickened tailings are dewatered in vacuum disc filters using a flocculant as a filter aid. The filter

cake is weighed and transferred to the paste backfill continuous twin-shaft mixer. The filter cake, water,

and cement binder (either 4% for low strength or 7% for high strength paste) are combined in the mixer

and then a hydraulic piston pump pumps it underground through a pipeline down the MDZ decline to

the underground booster pump station discussed in Section 16.5.6.

The plant has a process water tank, a freshwater tank and two 170 m3 binder (cement) storage silos

that store approximately 24 hours of cement required for the operation. Cement is delivered in 25 t

bulk cement trucks that are available seven days per week and 24 hours per day from a local supplier.

A shotcrete plant is located at the south end of the pad adjacent to the paste plant.

The plant general arrangement is shown in Figure 18-6. The surface plant 3D layout can be seen in

Figure 18-7.



Source: Paterson & Cooke, 2020

Figure 18-6: Paste Plant General Arrangement



Source: Paterson & Cooke, 2020

Figure 18-7: Paste Plant 3D Layout

18.14 Site Water Management

Water supply for the existing Marmato mining activities is currently provided by a combination of

underground dewatering and reclaim from the existing DSTF. During the wet season, water from the

existing tailings and excess mine dewatering flows are discharged to the Cascabel River under existing

permits. Additionally, water from the flotation tailings is recovered and used as part of the water supply

for the existing UZ mine. The current operation has adequate water supplies but can be challenged

during dry portions of the year.

Additional water management infrastructure is planned to be in-place prior to the start of the MDZ

project, including additional pumping from the UZ in the short term, a pumping station on the Cauca

River to supply both the UZ and the MDZ, and a water treatment plant to be constructed in 2021/2022

to manage tailings discharges.

Site water management for water discharges from the DSTF are discussed in detail in the DSTF design

section (Section 18.15.3), but in general have been developed to collect seepage water and retain

runoff within the DSTF footprint for reuse in the process and divert run-on to the facility from upgradient

areas around the DSTF. Concurrent reclamation of the DSTF surface will allow non-contact runoff

from the facility to be discharged to the downstream drainages.

18.14.1 Water Supply

The water for the MDZ plant and mine will be supplied by a combination of groundwater from the mine,

recycled water from the existing DSTF and planned DSTF, and fresh water from the Cauca River.

Water Balance Modeling

The water balance was developed to produce a makeup demand for water at the existing UZ

processing plant and the proposed MDZ processing plant. Water to meet the demands could be

sourced from excess water at the existing DSTF, contact water produced by the planned DSTF,



expected underground dewatering as described in Section 16.3 or an external freshwater source,

prioritized in that order. The model also estimated if excess water would result from the existing DSTF

or planned DSTF contact water and/or the underground dewatering exceeding the makeup demand.

SRK assumed that discharges from the underground can be discharged if monitored, and discharges

from the DSTFs can be discharged with monitoring and control of suspended solids and cyanide. This

is further addressed in Section 20, Environmental Studies.

The model assumes that tailings from the MDZ Plant will be dewatered using filters and the resulting

water recirculated into the process. Conversely, once the DSTF are operating, tailings from the UZ

plant will be delivered to the existing DSTF sites, drained and dried before being re-handled to the

planned DSTF. The resulting moisture content delivered to the DSTF will be roughly similar for both

streams of tailings.

The water balance indicated that the water produced at the existing DSTF, either during operations of

the existing DSTF or during the re-handling stage, will be consumed at the UZ Plant during operations,

but discharges will be required after the UZ plant operations have ceased. Drain down from the

planned DSTFs will also be consumed at the process plants, but surface water runoff from the planned

DSTF during large storm events will occasionally exceed the makeup demand at the process plants.

With limited storage available at site, excess water from the planned DSTF would need to be

discharged during unusually wet periods.

Source: SRK, 2019

Figure 18-8: Makeup and Demand at the Upper Zone Process Plant

0

500

1000

1500

2000

2500

2020 2022 2024 2026 2028 2030 2032 2034 2036 2038

Flo

ws (

m3

/da

y)

Time

Water Supply and Demand at the Upper Zone Plant

Mean

Upper Zone Plant Water Demand Reclaim from Existing DSTF to UZ PlantReclaim from DSTF to UZ Plant Dewatering to UZ PlantFreshwater to UZ Plant



Source: SRK, 2019

Figure 18-9: Makeup and Demand at the MDZ Process Plant

The projected underground dewatering flows (3,500 to 7,500 m3/day) consistently exceed the total

demand for water at the plant (500 to 3,600 m3/day) by a factor of two or more and excess underground

dewatering is expected to be discharged at rates of 3,000 to 7,500 m3/day. Makeup demand vs

available water from the existing DSTF, planned DSTF and underground dewatering predicted by the

model at the UZ and MDZ plants are shown in Figure 18-8 and Figure 18-9 respectively.

Overall, the water balance model indicates a net surplus of water for the Project due largely to an

overall increase in available water from the MDZ dewatering and a decrease in makeup demand

resulting from the lower moisture contents in the dewatered tailings.

Although the water balance does not indicate there will be a need for additional water supply sources,

during dry periods the underground dewatering will be the only available source of water. A back-up

supply from the nearby Cauca River is recommended as a secondary supply source for time when

underground dewatering flows are unavailable.

18.15 Tailings Management Area

The following sections summarize reasonably available information regarding existing tailings

generation and management and describe the conceptual design and operation of new DSTFs for

filtered (dewatered) tailings. Conceptual design details for DSTF 2 and DSTF 1 are shown on the

drawing set in Appendix B.

If the existing Cascabel 1 and 2 tailings management facilities can be proven or redesigned to meet

internationally accepted standards of practice, they may be able to provide between one and two years

0

200

400

600

800

1000

1200

1400

1600

2020 2022 2024 2026 2028 2030 2032 2034 2036 2038

Flo

ws (

m3

/da

y)

Time

Water Supply and Demand at the MDZ Plant

Realization #200

MDZ Plant Water Demand Reclaim from DSF to MDZ PlantDewatering to MDZ Plant Freshwater to MDZ Plant



of additional capacity, after which a new DSTF(s) would need to be commissioned to accommodate

tailings through the currently-estimated 16-year LoM. Where appropriate, recommendations for

additional investigations or expansion of existing baseline data collection programs are provided.

On January 27th, 2020, Breese Burnley, a Qualified Person in accordance with Companion Policy 43-

101CP to NI 43-101, conducted a personal inspection of the Marmato site under Section 6.2 of the

Instrument. This inspection was intended to familiarize Mr. Burnley with the conditions at the mine site

and any potentially available material information that could affect mine development/expansion in this

location. Information collected on site in 2020 was supplemented by CGM during 2020, as necessary.

In addition, SRK contracted with in-country geotechnical consulting company Dynami Geoconsulting

SAS (Dynami) to perform site reconnaissance, assist with slope stability and hydrological modeling,

participate in the geotechnical site investigation and coordinate with site personnel.

The following sections describe the existing operation and the conceptual design of new DSTFs

required to provide for tailings management through the LoM.

18.15.1 Existing Tailings Management Facilities

The processing plant currently sends tailings slurry from the cyanide leach circuit to unlined settling

ponds located on deposited tailings within the footprint of Cascabel 1. The draindown flows from the

ponds are directed via a gabion wall and drain system to small collection basins downgradient of the

tailings disposal piles. Clarified overflow water is pumped back to the plant for use in the process.

Excess water not needed at the plant is discharged under permit to the adjacent stream, Quebrada

(Qda.) Cascabel. Once sufficiently dewatered to allow for mechanical handling, the tailings are

excavated from the ponds and transported via truck to existing Cascabel 1 and placed and compacted.

Cascabel 2 is a lateral expansion of the existing operation into a separate natural drainage channel

north of Cascabel, as shown on Figure 18-10 below.

At the time of report preparation, CGM was expediting the additional characterization and analysis of

Cascabels 1 and 2 recommended by Dynami (Dynami, 2020c) with the short-term goal of bringing the

design of both Cascabels 1 and 2 up to internationally accepted standards of practice. For the

purposes of PFS tailings management, it was assumed that the design of Cascabels 1 and 2 could

either be shown to meet internationally accepted standards of practice, or could be redesigned to do

so, such that the final design configuration would provide sufficient additional capacity until a new

DSTF could be commissioned. These assumptions must be reviewed following completion of the

recommended additional characterization and analyses and the results used to update or revise the

plan and costs for tailings management for the next level of study. SRK recommends that CGM identify

other options for filtered tailings storage that may provide additional interim storage capacity in case

Cascabel 1 and 2 cannot be shown to be stable to internationally accepted standards of practice.



Source: IRYS, 2018

Figure 18-10: Proposed Configurations of Cascabels 1 and 2

18.15.2 New Tailings Storage Facility Siting Study

SRK completed a siting study of potential options for tailings management facility siting in the vicinity

of the existing Cascabel 1 tailings storage facility and the proposed portal and plant location. Factors

considered in the siting study included such considerations as topography, permitting requirements for

stream crossings, property ownership and acquisition potential, distances between the proposed

process plant location and the prospective DSTFs and municipality boundaries.

Conventional slurry DSTFs typically required very large areas to achieve a reasonable rate of rise and

corresponding deposited tailings dry density. A suitable slurry tailings management site could not be

identified in the steep and incised terrain around the project site. Based on this and the relatively high

precipitation in the area, SRK concluded that a DSTF would be a more favorable tailings management

method for the Marmato site. Recent advances in tailings filtration technology and performance favors

tailings filtration and dry stacking for this project. Maximizing process water recycling and minimizing

makeup water requirements was also considered an additional benefit of the DSTF method.

As part of the siting study, SRK developed conceptual designs for seven potential DSTF locations.

From that analysis, only three locations, sites 1, 2 and 6, were identified within the area of study as



potentially feasible for providing the capacity required through the mine life and achieving global

stability in the steep terrain in the site vicinity. The locations of the feasible DSTF locations are shown

on Figure 18-11 relative to the proposed process plant location. A trade-off study (SRK, 2020a) was

prepared to facilitate a comparison of high-level costs for each of the three sites for only major cost

items.

Based on the results of the SRK trade-off study (ToS) for sites 1, 2 and 6, CGM indicated a preference

to evaluate the feasibility of developing DSTF 2 and then DSTF 1 to achieve the desired tailings

storage capacity through the currently predicted mine life. The combination of DSTF 2 and DSTF 1

provides just enough capacity based on current projections. DSTF 6 provides sufficient capacity on its

own for the currently predicted mine life, although the distance to the plant adds some additional

planning and access complexities.

Source: SRK, 2020

Figure 18-11: Potential DSTF Sites Identified Through Siting Study

18.15.3 New Dry Stack Tailings Storage Facility Design

Under the PFS scenario using both DSTF 1 and DSTF 2, DSTF 2 would be constructed first and

receive dewatered and dried tailings from the UZ and filtered tailings from the MDZ after Cascabel 1

and 2 reach capacity. Current assumptions allow for between one and two years of additional capacity

at the Cascabel sites, as described above.

UZ tailings are assumed to be dewatered as described above at Cascabel 1, hauled to a designated

drying area and disked to a target moisture content of 15% by weight, then hauled to the planned

DSTF (DSTF 2 or DSTF 1) and placed, amended with cement and compacted. MDZ tailings will be

routed through a new filter plant to a target moisture content of 15% by weight. The filtered tailings will

be hauled directly to DSTF 2 or DSTF 1, placed, amended with cement and compacted.

Current calculations consider DSTF 2 is in operation by early-2021 and operational through 2027.

DSTF 1 construction is completed in 2026 and commissioned in 2027 and receives the balance of the

tailings through the end of the projected 13-year mine life in 2033. Construction of DSTF 1 would



commence one year prior to reaching capacity in DSTF 2 to ensure the availability of tailings disposal

capacity is uninterrupted.

Design Criteria

Conceptual designs for both DSTF 2 and DSTF 1 are based on an assumed 13-year LoM with a total

mined tonnage of 26 Mt and a maximum annual mining rate of 2 Mt/y for 13 years. Of the tailings

produced, approximately 9.4 Mt (approximately 48% of total mill throughput) would be mixed with

cement and sent back to underground workings as paste backfill. The remaining 10.3 Mt

(approximately 52% of total mill throughput) of tailings would be dewatered at the Cascabel site and

drying area or using filter presses to a target moisture content of 15%, then trucked to the DSTFs,

mixed with 1% cement by weight and placed and compacted in controlled lifts to a specified minimum

compacted density. The dry density of the placed tailings was assumed to be 1.8 t/m3 for volumetric

calculations. Key DSTF design criteria are summarized in Table 18-1 below.

For the purposes of the PFS, the outer slope for DSTF 2 was designed at a 2.5H:1V

(horizontal:vertical) slope, while the outer slope of DSTF 1 was designed at 2H:1V. Cement addition

is assumed to be required to achieve global stability of the tailings mass at both sites. For the purposes

of PFS cost estimating, cement amendment at 1% by weight was assumed. The actual required

amount of cement addition must be determined through a detailed laboratory testing program. Overall

slope angles should be revised for global stability in the next stage of the Project based on the results

of cement/tailings admixture testing.

Due to the tailings being filtered and placed and compacted with cement amendment, it is not

anticipated that an appreciable amount of draindown from the tailings will contribute flows to the

underdrain. However, the global stability of the DSTFs will depend on the ability to prevent the

development of elevated pore pressures within each stack. The DSTFs are therefore assumed to be

unlined but with an internal drain system designed to prevent the development of elevated pore

pressures within the tailings mass. Both sites are also assumed to require an underdrain system in the

natural drainage bottoms designed to intercept any potential upwelling groundwater or spring/seep

flows.

Conceptual design details for both DSTF 2 and DSTF 1 are shown on the drawing set in Appendix B.

Each DSTF is designed with a rock starter embankment designed with 2H:1V upstream and

downstream slopes and constructed from a combination of imported rock and non-potentially acid

generating (non-PAG) waste rock from underground workings or an approved on-site borrow source.



Table 18-1: DSTF Design Criteria

Criteria Units Description

General

Total Annual Mined Tonnes tpy 2,000,000

Total Mineralization Mined t 26,000,000

Life of Mine yr 13

Paste Backfill - percent of total tailings generated % ~48

Required Tailings Storage Capacity t 10,300,000

m3 5,700,000

100-year 24-hour Storm Event mm 143

Filtered Tailings

Placed Dry Density t/ m3 1.8

Target Moisture Content w/w 15%

Tailings Acid Generation Potential PAG/NAG NAG

Tailings Transport and Stacking System

Tailings Transport System from Filter to DSTF Truck

Tailings Spreading within DSTF - Dozer

Tailings Cement Amendment - Tractor with Discs

Compaction of Tailings - Dozer and Vibrating Smooth Drum Roller

Overall Slope Angle XH:1V DSTF-1 – 2H:1V

DSTF-2 – 2.5H:1V

Rock Starter Embankment

Rock Source Imported Rock, Waste Rock or Local

Borrow

Downstream/Upstream Slopes XH:1V 2H:1V

Source: SRK, 2019

Rock Starter Embankments and Foundation Preparation

For the PFS, it was assumed that phased construction of the DSTF 2 rockfill starter embankment

would require approximately 100,000 m3 of imported rock before the first waste rock arrives from MDZ

development in late 2021. It was also assumed that the waste rock from the underground would be

non-PAG and suitable for rockfill starter embankment construction. The rockfill embankments would

be constructed between abutting ridges that form the primary valleys within which the DSTFs would

be located. These ridges should consist of bedrock and constitute solid foundations for embankment

foundations, and unsuitable soils should be removed from the embankment footprints. A geotechnical

site investigation must be completed as part of the next phase of study to confirm suitable foundation

conditions and adjust the costs described herein.

The main features of the rock starter embankments are as follows:

DSTF 1 Rock Starter Embankment:

• Crest Elevation: 800 m

• Crest Width: 20 m

• Upstream Slope: 2H:1V

• Downstream Slope: 2H:1V

• Maximum Starter embankment Height at Centerline: 35.3 m

• Maximum Starter embankment Height from Toe: 58.9 m



DSTF 2 Rock Starter Embankment:

• Crest Elevation: 940 m

• Crest Width: 20 m

• Upstream Slope: 2H:1V

• Downstream Slope: 2H:1V

• Maximum Starter embankment Height at Centerline: 27.2 m

• Maximum Starter embankment Height from Toe: 46.9 m

The rockfill embankment footprints will require foundation preparation prior to placement of the rock

fill materials, to include clearing and grubbing and removal and stockpiling of salvaged topsoil for later

use in reclamation, removal of unsuitable foundation materials, and excavation of a foundation key for

embankment stability. For PFS costing, it was assumed that the unsuitable material requiring removal

would be removed to an average depth of 2m over the footprint of the rock starter embankment, and

the foundation key would be 20 percent of the rockfill embankment at both DSTFs.

A similar approach was taken for the foundation of the filtered tailings. An average topsoil thickness of

1m over the footprint of the proposed filtered tailings placement area was considered as an initial

capital cost. Excavation of benches into the prepared foundation slope to key in filtered tailings and

construct perimeter drains was considered an operational cost.

DSTF Slope Stability

Due to the COVID-19 virus and associated restrictions on international travel, SRK was unable to

execute the originally planned geotechnical site investigation prior to preparation of the PFS study.

This inability to characterize the foundation conditions beneath the conceptual DSTF footprints means

the designs for those facilities could change significantly during the next phase of study, with possible

resulting recommendations ranging from a redesign of required foundation preparation measures to

recommendations against development at the selected sites. In addition, it could be determined that

more cement is required to achieve global stability than has been provided for in the PFS cost estimate,

resulting in significantly different operating costs than considered herein.

In the absence of detailed testing and characterization data for both the foundation and the tailings

and tailings/cement admixtures, Dynami (Dynami, 2020b) prepared slope stability analyses under

SRK’s direction that started with reasonably achievable strength and material properties developed

based on SRK’s experience with recent dry-stack projects in similar geological and climatic

environments. The goal of the analyses was to determine what minimum material strength parameters

would be required to achieve global stability of the conceptual DSTF configurations, and those

numbers were evaluated relative to reasonably achievable values. The PFS designs currently reflect

the results of these analyses and SRK’s experience at similar sites, however a complete slope stability

study must be completed as part of the next phase of study and prior to construction and must consider

the results of the site geotechnical investigation and tailings characterization. The results of that study

must be used to revise the PFS design to demonstrate stability in accordance with internationally

accepted standards of practice.



DSTF Water Management

Underdrain System

Seeps, springs or upwelling groundwater within the DSTF footprint would be controlled by the

installation of an underdrain system consisting of perforated central header pipes placed in channels

in the base of existing natural drainages with connecting perforated transverse drains in smaller

tributaries to the main drainages. Any seepage that is intercepted by the underdrain would be

considered non-contact water and routed to tanks of the natural channels below each DSTF’s rock

starter embankment for release to the natural drainage downstream. If routed to the tanks, the

captured fluids would be periodically sampled and tested to confirm suitable chemistry for discharge

and either allowed to overflow into the natural channel downstream or pumped back to the process

water tank for incorporation back into the process circuit.

For the purposes of PFS costing, the underdrain was considered to consist of a network of 0.6m

diameter perforated and solid wall pipes surrounded by a 6 m2 cross section of drain gravel, wrapped

in a geotextile with a layer of filter sand between the underdrain and filtered tailings.

The underdrain system would be installed in phases as the DSTF advances up the slopes and

drainages with temporary stormwater controls in place to ensure that contact or upgradient stormwater

is not allowed to enter the underdrain system.

Internal Drainage System

A system of inter-bench drains would be installed within the placed tailings to prevent the development

of elevated pore pressures within the tailings mass. The drains would consist of 100 mm perforated

pipes installed in gravel-filled drains wrapped in geotextile on 10 m centers and routed to the closest

underdrain at the back of each DSTF (Figure 18-12). The drainpipes would be connected via a series

of solid pipes through the underdrain system that would route collected seepage flows into the water

management tanks at the toe of the rockfill embankment. The described internal drainage system

would be constructed every 10 m vertically as the DSTF advances. PFS costs for drain construction

were considered as part of DSTF operating costs.

Source: SRK, 2020

Figure 18-12: Internal Drainage System



Stormwater Management

The upper operational deck surface would be graded so that contact runoff reports to a lined

operational pond at the back of the DSTF (i.e. as far as possible from the outer slope crest) to facilitate

pumping back to the plant site process water tank for reuse. Contact stormwater must be managed on

the DSTF top deck and not allowed to enter the underdrain system. As the filtered tailings surface

progresses up the slope at the back of each DSTF, the lined operational pond would be relocated to

the lowest point on each lift.

Concrete-lined diversion channels must be constructed around the DSTFs before any other

construction is started. Reinforced concrete cutoff walls are recommended where natural channels

intersect each concrete channel. The diversion channels would route non-contact stormwater run-on

around the DSTFs and back into natural drainages downstream. Conceptual channel alignments are

shown in Figure 18-11 above. The possible channel alignments are significantly affected by current

permitting limitations which prohibit rerouting flows from one watershed into an adjacent watershed.

The ability to do so would result in a much more feasible diversion channel west and south of DSTF 2

and should be further explored.

A combination of reinforced concrete energy dissipation structures or a stilling basin and riprap aprons

would be constructed to slow the flow down before being discharged into the natural drainages. Interim

diversion channels were included in PFS operating costs to minimize the amount of run-on reporting

to the topdeck stormwater management system during operations.

The concrete-lined diversion channel around the west and south side of the process plant intersects

the haul road from the plant to DSTF 1 and the municipal road to Marmato. To manage stormwater

flows at these intersections, pre-cast concrete spans with wingwalls from Contech were included in

PFS costing (Figure 18-13). Estimated costs include demolition of the existing public road, reinforced

concrete footings, installation of pre-cast arches and reconstruction of public road to local standards.

Source: Comtech/SRK, 2020

Figure 18-13: Pre-Cast Concrete Span Channel Crossing



Reclamation of the outer face of the DSTFs should be carried out concurrently during active operations

per Section 18.4.9 and, therefore, stormwater contacting the downstream face would be non-contact

stormwater that could be discharged into the natural channel downstream. A non-contact stormwater

collection system was designed to collect and direct non-contact stormwater from the face of each

DSTF with a series of riprap-lined channels on benches routed to perimeter downslope diversion

channels. The channels were classified into three groups as shown:

• Bench channels (or ditches): Drain from the DSTF towards the northern or southern perimeter

downslope channels.

• Perimeter downslope channels: north and south channels running along the intersection of

the tailings outer slope and native ground.

• Temporary channel above active tailings bench to drain runoff from the area below the

upgradient diversion channel.

Dynami conducted preliminary hydrologic and hydraulic analyses (Dynami, 2020a) for each of the

conceptual DSTF The diversion channels dimensions used in PFS costing are summarized in Table

18-2. Watersheds for each DSTF are shown in Figure 18-14.

Table 18-2: Stormwater Diversion Channel Summary

Location Return Period

Ditch Slope Min. (%)

Bottom Width (m)

Depth (m)

Sideslope (m)

DSTF-1

Bench Channel-1 1:100 2% 1 0.35 05H:1V

Bench Channel-2 1:100 2% 0.5 0.3 05H:1V

Perimeter Channel South 1:100 2% 1.5 1 05H:1V

Perimeter Channel North 1:100 2% 1.5 1 05H:1V

Diversion Channels PMP 2% 4 0.6 05H:1V

DSTF-2

Bench Channel-1 1:100 2% 0.6 0.6 05H:1V

Bench Channel-2 1:100 2% 0.6 0.6 05H:1V

Perimeter Channel South 1:100 2% 1 1 05H:1V

Perimeter Channel North 1:100 2% 1 1 05H:1V

Diversion Channel around Plant PMP 2% 4.5 1.5 05H:1V

Diversion Channel above DSTF PMP 2% 4.5 1.3 05H:1V

Source: Dynami, 2020b



Source: Dynami, 2020

Figure 18-14: Watersheds for DSTF 1 (Top) and DSTF 2 (Bottom)



As described above, a significant design limitation requires that water from one watershed cannot be

diverted into an adjacent watershed. The southern diversion channel alignment attempts to

accommodate this limitation; however, the alignment will be difficult to construct and likely expensive

to maintain. SRK recommends moving ahead with an application for permission to divert water from

the Los Indios Creek watershed into the channel immediately south of the plan area to eliminate a

significant portion of the southern diversion channel. The flows from both channels terminate at the

Caucas River and discharge very close to each other, so the cumulative impacts of the diversion are

not anticipated to be significant.

Access and Haul Roads

The DSTFs would be accessed by dedicated haul and access roads (Figure 18-15) that include

additional width to support stormwater management and safety berms. For the purposes of PFS cost

estimation, the platform for both the road and drainage channels was assumed to be 15 m wide with

a maximum grade of 12%.

The DSTF-1 access road is currently routed from the filter plant to the bottom of the rockfill

embankment and to the portal. The road will be adjusted as deposition advances up the valley to

maintain access.

The DSTF-2 access road is currently routed from the filter plant to about mid height up DSTF-1 and

then to the base of the rock embankment. Stormwater culverts with riprap at the entrance and exits

were incorporated into the design. Additional methods of road stability were assumed to be needed

for about 20% of the access roads linear length.

Source: SRK 2020

Figure 18-15: DSTF Haul and Access Roads



Tailings Stacking

Trucks would be used to transport the dried tailings from the UZ or filtered MDZ tailings from the filter

plant to either of the DSTFs via dedicated haul roads, as shown in (Figure 18-15). The tailings would

be end dumped, spread in approximately 30 cm-thick loose lifts with a dozer, mixed with cement using

a disc harrow or similar equipment, and compacted using a vibrating smooth drum roller or sheepsfoot

compactor, as appropriate. It is currently anticipated that the cement amendment would be required to

achieve global slope stability at the currently proposed design slopes. For the purposes of PFS costing,

it was assumed that a 1% amendment of cement by weight will be required to achieve global stability.

Global slope stability should be evaluated in detail following completion of the geotechnical site

investigation and tailings materials characterization programs.

Temporary Stockpile and Tailings Storage

Temporary stockpiles for topsoil, removed unsuitable soils, and imported or waste rock for rock starter

embankment construction were assumed to be located southwest of the plant site. For PFS costing,

these areas were assumed to require foundation preparation and underdrain construction to manage

existing drainages and ensure stockpile stability. Costs were developed based on assumed lengths of

0.6 m diameter HDPE culverts and gravel-filled underdrains wrapped in geotextile. Because

geotechnical foundation information was not available at the time of PFS report preparation, as

described above, the foundation areas for all stockpiles should be evaluated in the next phase of study

to ensure stockpile stability.

A temporary filtered tailings storage area was assumed to be incorporated into the final plant layout to

accommodate up to three days of filtered tailings storage during periods of excess precipitation when

placement and compaction of tailings to the minimum specified relative density is not feasible. For PFS

costing, SRK included costs for a sprung structure (high-tension fabric building).

Finally, a temporary holding pond for slurry tailings storage was assumed to be required but a suitable

site was not identified. Given the current plans for two filter presses, and assuming at least one should

be operational at any given time, the potential need for a slurry holding pond should be evaluated at

the next phase of study. For PFS costing, a placeholder cost was included in the initial capital costs.

Closure and Reclamation

Reclamation of the outer slopes of the DSTFs would be undertaken concurrent with DSTF

construction. As successive lifts of dewatered tailings are placed and compacted, the outer slope face

would be covered with rock cladding, topsoil, and vegetated. The final top surface of the DSTF would

be graded back to the natural slope and stormwater diversion channels at not less than 2%, covered

with topsoil and revegetated.

18.15.4 Tailings Risks and Opportunities

Currently identified risks and opportunities with respect to the costs developed for the PFS include the

following:

• The inability to characterize the foundation conditions beneath the conceptual DSTF footprints

means the designs for those facilities could change significantly during the next phase of

study, with possible resulting recommendations ranging from a complete redesign of required

foundation preparation measures to recommendations against development at the selected

sites.



• Geochemical characterization of both waste rock and ore/tailings was ongoing at the time of

preparation of the PFS. Preliminary indications are that at least a portion of the waste rock

and tailings may be acid generating. If acid generation potential turns out to be significant,

there may be additional costs incurred to import suitable rock for rockfill starter embankment

construction and for operational and long-term water management.

• SRK understands that CGM is expediting the characterization and analysis of Cascabels 1

and 2 recommended by Dynami in 2020. If the facility cannot be demonstrated to be stable

with respect to internationally accepted standards of practice, CGM will need an alternative

site for interim storage of UZ tailings until a new DSTF is ready to accept tailings.

• Only a very small amount of tailings was available for testing during the preparation of the

PFS. More extensive testing should be performed at the next level of study to confirm tailings

geotechnical characteristics and cement addition requirements. The results of that testing

should be used to evaluate the final configuration of each DSTF in accordance with

internationally accepted standards of practice. The results may indicate that more or less

cement is required to achieve the stability requirements than were assumed for PFS costing.

• Stormwater maintenance requirements at both DSTF 1 and DSTF 2 may constitute higher

costs through operations and closure than is currently allowed for in the PFS costs given the

regulatory limitations on stormwater routing between adjacent watersheds and the size of the

contributing watershed at DSTF 2. The actual development and operation of DSTF 6 could

ultimately be comparable or less expensive in cost through the life of the mine to develop and

operate given the more amenable site topography and lower stormwater management

requirements.

• Several geological faults crossing the tailings facilities are identified in available geological

studies (IRYS, 2019). The information indicates that the majority of the footprint area for the

three DTSFs corresponds to Miocene age rocks and residual soils, except for a colluvial

deposit on DTSF-2. SRK recommends a trenching study within DTSF-2’s footprint to assess

the activity of the faults crossing this quaternary unit.

18.16 Off-Site Infrastructure and Logistics Requirements

The Project has no significant off-site infrastructure needs and this section is provided for reference

only.

18.16.1 Port

Marmato is 200 km east of the Pacific Ocean and 300 km south of the Caribbean Sea and Atlantic

Ocean. The nearest port is Buenaventura on the Pacific Ocean (320 km by the Pan American Highway

to the south west). The port will not be used for the Project other than as support for delivery of out of

country equipment.

18.16.2 Rail

There is an abandoned railway cutting along the east side of the Cauca River opposite Marmato, which

previously formed part of a railway network between the Pacific and Atlantic Oceans which ran

between Buenaventura and Puerto Berrio on the navigable Magdalena River. The middle section

between Medellín and La Felisa (Caldas, 10 km south of Marmato) was completed in 1942 and closed

in 1972. Ferrocarriles del Suroeste S. A. (Southwestern Railways) applied for a concession to rebuild



this 185 km line between La Felisa and Envigado, near Medellín, in November 2007, at a cost of

US$140 million. This would become integrated with the national railway network. Ferrocarriles del

Oeste S. A. (Western Railways) were awarded the contract to operate the 499 km Buenaventura to

La Felisa railway in November 2007. The concession is in two stages. In July 2009 the 388 km railway

between Buenaventura and Cartago and La Tebaida which has been rehabilitated was opened. In the

second stage the new concessionaire will take over operation of the 119 km section between Cartago

and La Felisa once this has been rebuilt by Tren de Occidente (Western Train). Currently the

concession contractor (Western Railway) is in liquidation, so another company would have to develop

the rail if needed for the Project. Currently there are no plans to use the rail.



19 Market Studies and Contracts

19.1 Commodity Price Projections

Gold and silver markets are mature, global markets with reputable smelters and refiners located

throughout the world. Demand is presently high with prices for gold showing an increase during the

past year. Markets for doré are readily available. Marmato possess a gold room for the production of

doré.

Assumed prices are based on the long-term outlook for gold and silver. This projection is well below

the current spot prices and the long-term views of relevant market analysts in the precious metal

sector. Table 19-1 presents the prices used for the cash flow modelling and resources estimation.

Table 19-1: Marmato Price Assumptions

Description Value Unit

Gold 1,400 US$/oz

Silver 17.00 US$/oz

Source: CGM, 2020

19.2 Contracts and Status

CGM currently has a long-term supply agreement for the sale of its products to an international refinery

who take delivery of doré from the mine at designated transfer points in Colombia. The refinery is

responsible for shipping the products abroad. The refining costs and discounts associated with the

sales of the products are based on this agreement. This study was prepared under the assumption

that the Project will sell doré containing gold and silver.

Treatment charges and NSR terms are summarized in Table 19-2.

Table 19-2: Marmato Net Smelter Return Terms

Description Value Units

Payable Gold 100%

Doré Smelting & Refining Charges 6.38 US$/oz-Au

Source: CGM, 2020



20 Environmental Studies, Permitting and Social or Community Impact The following section discusses reasonably available information on environmental, permitting and

social or community factors related to the Marmato Mine (Upper Zone) and the proposed MDZ

expansion. Where appropriate, recommendations for additional investigation(s), or expansion of

existing baseline data collection programs, are provided.

On December 1, 2016, Mark Willow, a QP in accordance with Companion Policy 43-101 to NI 43-101

– Standards of Disclosure for Mineral Projects, conducted a personal inspection of the Marmato site

under Section 6.2 of that Instrument. This inspection was intended to familiarize Mr. Willow with the

conditions on the properties, and any potentially available material information that could affect mine

development/expansion in this location. Information collected on site was supplemented by CGM

during 2019/2020, as necessary.

20.1 Environmental Studies

20.1.1 Environmental Setting

The Marmato Project is located in the Municipality of Marmato, Department of Caldas, Republic of

Colombia, and is approximately 125 km due south of the city of Medellín, the capital of the Department

of Antioquia. The town of Marmato was founded in 1540 and has a population of approximately 10,000

people. It is one of the most historic gold-mining regions of the hemisphere. The Marmato Project has

excellent infrastructure, being located along the Pan American Highway with access to Medellín to the

north and Manizales (the capital of Caldas) to the south and has access to the national electricity grid

which runs near the property.

The west central Colombian Department of Caldas is situated in the Cordillera Central of the Andes

Mountains and is bounded by the Magdalena River on the east and the Cauca River on the west.

Marmato lies at an elevation of 1,050 masl, just west of Río Cauca, which joins the Magdalena River

near Magangué in Bolívar Department, before eventually flowing out into the Caribbean Sea.

The local topography is characterized by steep, incised valleys. The climate is tropical with an annual

average temperature of 21°C, that typically varies from 14°C to 24°C, and average annual rainfall of

approximately 2,162 mm/y, predominantly falling between April and November, with a negligible

difference of 153 mm of precipitation between the driest and wettest months. The drainage pattern

across Marmato is dendritic; the license area drains east into the Cauca River, which is heavily

influenced by artisanal mining operations. The vegetative cover across the landscape consists of

disturbed grassland (used mainly for mining and livestock rearing activities) interspersed with

fragmented forest patches, mainly along drainage lines within the incised valleys. Forest patches

provide important habitat for wildlife.

The operations are located within the town of Marmato, which has been a center for gold mining for

more than 500 years and the environmental and social setting is strongly influenced by this. Mining,

both formal and informal, is the main economic activity in Marmato and the neighboring towns of El

Llano and La Garrucha. Artisanal mining with informal processing operations using basic technology

has resulted in poor health and safety conditions and widespread water contamination from discharge



of tailings and waste directly into the environment. This has led to a prevalence of mercury-related

health problems in the local populations. Health issues related to population influx are also common.

20.1.2 Management Procedures and Baseline Studies

The existing Marmato Project predates the regulatory requirements to prepare an environmental

impact assessment as part of the permitting process. Instead, the operations were authorized through

the approval of an Environmental Management Plan (Planes de Manejo Ambiental or PMA). The PMA

for Marmato was approved by the regional environmental authority (Corporación Autónoma Regional

del Caldas or Corpocaldas) on October 29, 2001 under Resolution 0496, File No. 616. The site-specific

PMA covers environmental studies and required management procedures and practices associated

with:

• Reclamation in the area of the production plant

• Reclamation and closure of the tailings settling ponds

• Management of unstable zones (including erosion control)

• Water management in the mines

• Management of stormwater runoff

• Management and protection of watersheds

• Control planning and use of explosives

• Reforestation and revegetation programs

• Reclamation and closure planning

• Management of tailings

• Containment structures

• Cyanide destruction (detoxification)

• Management of wastewater (domestic)

• Water usage

• Air resource management

• Physical risk management measures (including toxic substances)

• Social management

• Biological management (i.e., biodiversity)

In addition, a comprehensive baseline data collection program was initiated in 2019 to gather relevant

and appropriate site information with respect to the existing Marmato Project and the proposed MDZ

expansion. The study area for these investigations covered the underground portal, and processing

plant site, as well as all of the proposed tailings disposal areas. The data was compiled and reported

in Capítulo 20: Caracterización Ambiental y Social del Proyecto, Caldas Gold Marmato S.A.S., Título

Minero #014 – 89m (May 2020). Included in this up-to-date report is:

• Comprehensive soil characterization of the study area, including surface uses, chemical

characteristics, and agronomic properties

• Hydrology of the study area, including flows and quality

• Climatology gathered from 10 regional stations which include winds, temperature,

precipitation, relative humidity, cloudiness (solar radiation), and evapo-transpiration potential

• Air quality from stations in El Llano, El Atrio, and La Plaza recording particulate matter (PM10),

sulfur oxides (SOx), nitrogen oxides (NOx), carbon monoxide (CO), volatile organic carbons

(VOCs), and ozone (O3)



• Climatology and meteorology

• Ambient noise levels from seven monitoring locations

• Baseline ecological data, including ecosystems and cover types, flora, and fauna resources

• Archeological and cultural resource assessment of the surrounding areas using personnel

from the Archeology Laboratory at the University of Caldas

• An assessment of the current socioeconomic situation within the study area, including

demographics, sanitation conditions, energy/power infrastructure and usage, and education

The report goes on to describe the environmental management programs employed by the current

operations (including costs) and provides an overview of the closure plan and costs for the mine and

its associated infrastructure (i.e., tailings, plant, etc.). Assessment of potential impacts associated with

the MDZ expansion project can only begin in earnest once the PFS mine plan has been finalized, at

which point, CGM will initiate engagement with Corpocaldas (anticipated in Q3 2020).

20.1.3 Geochemistry

SRK directed a sampling and analytical program to generate environmental geochemistry data for

tailings and waste rock for the existing operations and MDZ expansion project. Samples of future

tailings were collected from the PFS metallurgical program, and two samples of existing tailings were

collected from site (one consisting of conventional tailings, and one of cyanide tailings). 60 samples of

future waste rock were collected from exploration drill core. As of this writing, the waste rock and

existing tailings samples are still undergoing analysis at the laboratory.

Data from SRK’s metallurgical program indicates that tailings will be discharged with a neutral to

alkaline supernatant. However, the tailings solids will be PAG with the potential to eventually exceed

the alkaline supernatant and produce acidic drainage in the longer term. Detoxified cyanide tailings

are anticipated to have elevated concentrations of arsenic, sulfate, and total dissolved solids in

potential leachates. Testing on paste backfill tailings suggest that the material will be acid-neutralizing

in the short term, but in the long term, the material could become acidic.

A waste rock analytical program completed in 2012 in support of the defunct open pit mine design

indicated that a significant fraction of waste rock could be potentially acid generating (KP, 2012).

Effective management of both tailings and waste rock will be a critical issue for success of the project.

20.1.4 Known Environmental Issues

SRK is not aware of any known environmental issues that could materially impact CGM’s ability to

extract the mineral resources or mineral reserves at the Marmato Project. While there will be some

challenges associated with land acquisition and surface water control during operations, the Marmato

Project has not had, nor does it currently have any legal restrictions which affect access, title, mining

rights, or capacity to perform work on the property. Likewise, in regard to environmental compliance,

the operation is covered by the PMA and associated environmental permits, which further reduces

environmental risks.

Informal processing operations related to artisanal mining in this location using basic technology (many

of which are unpermitted), has resulted in poor health and safety conditions and widespread water

contamination from discharge of tailings and waste directly into the environment. This has led to a

prevalence of mercury-related health problems in the local populations. These operations may also

increase the potential for environmental risk in terms of soil stability impacts to other associated



resources. There are periodic review protocols which allow CGM to identify potential damages by third

parties and report them to Corpocaldas. The operational areas are generally protected to prevent

access by unauthorized third parties and their activities to protect against risks and environmental

liabilities.

20.2 Mine Waste Management and Monitoring

20.2.1 Waste Rock Management

Very little waste rock is generated by the underground operations at Marmato, and this is expected to

continue with the MDZ expansion. What little waste rock is generated is used as backfill in the

underground workings or on the surface for construction projects, such as maintenance of roads. As

part of the MDZ expansion, some of this waste rock will be needed for construction of the starter

embankment(s) for the proposed tailings disposal areas.

A preliminary study conducted by Knight Piésold (2012) suggested that approximately two-thirds of

the waste rock would likely have to be treated as PAG for conservative waste rock handling purposes.

Additional testing is currently underway to better refine this prediction, but some of the waste rock may

have the potential for Acid Rock Drainage and Metal Leaching (ARDML) and may not be suitable for

surface disposal/construction. However, since so little waste rock is brought to the surface, only

opportunistic sampling and testing of construction materials is probably necessary. At the moment,

Corpocaldas does not require the testing of mine waste materials; only effluent discharges. This is

likely to change in the future, as source control becomes more of the norm in this jurisdiction.

20.2.2 Tailings Management

The current gold processing plant at Marmato is fed with mill feed which is milled and processed

through a cyanide (CN) leach circuit using underground dewatering water, minimal water recycled from

the tailings facilities and fresh make-up water from the surface. Approximately 55% of the tailings from

the operations is returned to the underground workings as sand fill. The remaining approximately 45%

(including fines) are slated for surface disposal. The MDZ expansion will use a similar processing

methodology.

The CN leach circuit includes a hydrogen peroxide (H2O2) and copper sulfate (CuSO4) destruction unit

on the tail end to reduce CN concentrations to below the Colombian mine effluent discharge limit of

1 mg/L (Article 10 of Resolution 0631, dated March 17, 2015) before sending the residual tailings to

the unlined settling ponds (Cascabel 1 and Cascabel 2). Alternatively, hydrogen peroxide and copper

sulfate may be used in conjunction with ultraviolet (UV) treatment, if deemed prudent and economical.

The underdrain water from these ponds is directed to small collection basins downgradient of the

tailings surface disposal piles. Flocculant is added to this water on an as-needed basis to remove

residual suspended solids; the underflow (solids) from this process is directed back to the tailings

settling ponds, while the clarified overflow water is pumped back to the plant for use in the process.

Excess water, not needed at the plant, is discharged under permit to the adjacent stream, Quebrada

(Qda.) Cascabel.

Once sufficiently dewatered to allow mechanical handling, the tailings are excavated from the ponds

and transported via truck to the final disposal location. Monitoring of the residual tailings to determine

whether or not they are classifiable as ‘hazardous’ is accomplished through Corrosive, Reactive,

Explosive, Toxic, Inflammable, Pathogen (biological) (CRETIP) analyses. Toxicity analyses were



carried out by the Universidad Pontificia Bolivariana on cyanides and metals (chromium, mercury and

lead). The results support the classification of the tailings as non-toxic for the metals based on

comparisons to the maximum concentration thresholds established by Decree 4741 of 2005. The

analyses also showed that Total CN was below the threshold allowed in Decree 1594 of 1984 for water

discharges. SRK did not perform a comprehensive audit of all testing data; however, given the vintage

of these results, SRK suggests that more frequent sampling and analysis be conducted by the

operation going forward, and especially for the expansion project.

Based on the mineralogy of the orebody, process methodology, and limited analytical testing, SRK

anticipates that the Marmato tailings could be acid generating and will require appropriate

management during operations and post closure.

For the MDZ expansion of the Marmato Project, CGM intends to continue with a similar approach to

tailings and tailings water management. However, rather than using less efficient settling ponds, the

tailings will be filter pressed. The dry stack tailings will be transported to the new disposal facility, mixed

with cement, and stacked in a configuration that minimizes surface runoff. To the extent practicable,

contact water collected on the deck of the DSTF will be reused in the process; excess water will be

discharged under permit.

A critical driver of environmental impacts from tailings is whether contact water will be contained. The

current and predicted future quality of contact water needs to be determined. Two components of

potential chemical loading need to be estimated:

• Loading to surface water due to tailings runoff

• Loading to groundwater through seepage from the base of the tailings facilities

Current metallurgical testing includes geochemical characterization to provide data that will assist in

forecasting tailings solids and potential runoff chemistry. Water re-use/recycling is recommended to

the extent feasible. The operation will seek to maximize water recycling and minimize treatment and

discharge.

20.2.3 Site Monitoring

Various mitigation and monitoring programs are discussed in the approved PMA. Additional monitoring

will likely be requested by Corpocaldas as part of the permitting of the MDZ expansion. CGM routinely

verifies their compliance with provisions of the environmental mining guide adopted by Resolution 18-

0861 of 2002 by the Ministry of Mines and Energy, for activities associated with underground mining

and exploration.

Water Quality and Monitoring

As part of the current operations, CGM has seven domestic wastewater discharges and three non-

domestic wastewater discharges for which monitoring is conducted. Table 20-1 lists each discharge

point with its respective analytical parameters to be measured. The results of the monitoring are

provided to the regional environmental authority (Corpocaldas).



Table 20-1: Water Discharges

o ID o Wastewater

Discharge o Discharge

Point o Parameters

o 1

o Non-Domestic

o Tailings dam o Turbidity, pH, temperature, total solids, total suspended solids, DB05, DQ0, grease and oils, flow, conductivity, sedimentary solids (for treatment systems – bulk tails, if total cyanide and lead are taken into account).

o 2 o Thickener

o 3 o Sedimentation

Ponds

o 4

o Domestic

o Camp 1

o pH, temperature, total solids, total suspended solids, DB05, DQ0, grease and oils, flow, fecal coliform and total coliform.

o 5 o Camp 2

o 6 o Camp 3

o 7 o Offices

o 8 o Mines

o 9 o Mine Dry

o 10 o Contractor

Lodging

Source: Mineros Nacionales S.A.S., 2017

CGM anticipates requiring additional wastewater discharge points at the proposed Main Camp,

Offices, Process Plant, Raw Water Pump station, and possibly at the DSTF facilities.

Air Quality and Monitoring

Air quality emissions from stationary sources at Marmato are currently regulated and monitored by the

Air Pollution Unit (Unidad de Contaminación Atmosférica or UCA) according to Table 20-2.

Table 20-2: Stationary Emission Sources

Unit Parameter UCA Degree of Significance

Monitoring Frequency

Metallurgical Laboratory

Particulate matter (PM)

0.01 Very low 3 years

Sulphur dioxide (SO2) 0.00 Very low 3 years

Nitrogen oxides (NOX)


Lead (Pb) 0.014 Very low 3 years

Smelter/Foundry

Particulate matter (PM)


Sulphur dioxide (SO2) 0.37 Low 2 years

Nitrogen oxides (NOX)


Lead (Pb) 0.47 Low 2 years


To date, CGM does not have any have any additional defined air quality monitoring points as part of

the MDZ expansion project but is anticipating the need to install points at the Mine Portal and around

the DSTF facilities. The precise locations will depend upon the results of the environmental impact

analysis.

20.2.4 Environmental Procedures and Permissions

Environmental protection measures and procedures followed by CGM at the Marmato operations

include those shown in Table 20-3.



Table 20-3: Environmental Procedures

ID Resource Environmental Procedure

Location Approval

Current State of Process

Competent Authority New Valid

Valid for

(years)

Renewal or Modification

Filing Date

1 Air Atmospheric emission permit

Smelter

Resolution 270 of April 27, 2009

5 ×

February 21, 2014. Renewal awaiting Ministry Indigenous Peoples determination before Corpocaldas renews permit.

Corpocaldas

Laboratory furnace

Shedding filter bag

2 Water

Surface water concession

La Maruja portal

× 5

Corpocaldas

Aguas Claras

Resolution 0046 of March 09, 2004, amended by Resolution 127 of May 5, 2004

×

February 07, 2014 Renewal awaiting Ministry Indigenous Peoples determination before Corpocaldas renews permit.

Zaparillo

Guineo

Domestic water discharge permit

Camp 1

Resolution 270 of April 27, 2009 amended by resolution 254 of February 28, 2014

× February 21, 2014

Camp 2

Camp 3

Office

Mine

Contractor

Mine dry

Non-domestic water discharge permit (industrial)

Tailings

Thickener

Sediment ponds

Channel Occupation

Charco Hondo

Resolution 0062 of February 15, 2006

Corpocaldas




CGM will need to request an additional surface water concession for the Cauca River as part of the

MDZ expansion project

20.2.5 General Water Management

Operational water for the existing Marmato operation is provided through a combination of

underground mine dewatering water, reclaim water from the existing DSTF, and several surface

freshwater sources. Even with the high precipitation experienced by the site, only nominal effort

appears to be directed toward stormwater management and the prevention of contact with mine

equipment and facilities at the existing operations. Some concrete channels and energy dissipation

structures for the management of run-off are already constructed, and some others are being

considered.

The MDZ expansion includes long-term stormwater diversion structures designed to convey 2/3 of the

Probable Maximum Precipitation (PMP) event, while internal (operational) conveyances are designed

to 100-year event requirements.

Surface water runoff control represents a significant water management challenge to the Project

considering the difficulties in distinguishing between the impacts from the artisanal mining activities

and those of the Project. The geochemical and hydrogeological/hydrological impacts should be

evaluated prior to closure, when dewatering ceases and water levels rebound within the mine

workings. Of critical importance is the possibility of mine water discharging to surface water or

groundwater and potentially impacting users. There are reports that dewatering effluent carries

elevated concentrations of metals. The water quality of dewatering effluent must be well characterized

in the event that treatment is needed before it is used or discharged. A forecast of closure water quality

is needed.

20.2.6 Environmental Management Budget

The operational costs for environmental management for the remainder of 2020, as provided by CGM,

are COL$1,591,779,394 (US$482,357). With respect to the MDZ expansion, CGM notes that the

environmental impact analysis and its requisite environmental management programs have not yet

been completed, from which the revised management budget will need to be developed covering the

larger operation.

20.3 Project Permitting Requirements

20.3.1 General Mining Authority

Since 1940, the Ministry of Mines and Energy (MME), formerly the Mines and Petroleum Ministry, has

been the main mining authority with the legal capacity to regulate mining activities in accordance with

the laws issued by the Colombian Congress. The MME can delegate its mining related powers to other

national and departmental authorities. Mining regulations in Colombia follow the principle that (except

for limited exceptions) all mineral deposits are the property of the state and, therefore, may only be

exploited with the permission of the relevant mining authority, which may include the MME, the

National Agency for Mining or the regional governments designated by law.

In 2001, the Congress issued Law 685 (the Mining Code). This law established that the rights to

explore and exploit mining reserves would only be granted through a single mining concession



agreement (the 2001 Concession Agreement). This new form of contracting did not affect the pre-

existing mining titles (licenses, ‘aportes’, and concessions) which continue to be in force until their

terms lapse. The 2001 Concession Agreement includes the exploration, construction, exploitation, and

mine closure phases and are granted for periods of up to 30 years. This term may be extended upon

request by the title holder for an additional 30-year term. According to the Mining Code, the initial term

was divided into three different phases:

• Exploration – During the first three years of the concession agreement, the title holder will

have to perform the technical exploration of the concession area. This term may be extended

for two additional years upon request;

• Construction – Once the exploration term lapses, the title holder may begin the construction

of the necessary infrastructure to perform exploitation and related activities. This phase has

an initial three-year term which may be extended for one additional year; and

• Exploitation – During the remainder of the initial term minus the two previous phases, the title

holder will be entitled to perform exploitation activities.

20.3.2 Environmental Authority

In 1993, Law 99 created the Environmental Ministry and then in 2011 the Decree 3570 modified its

objectives and structure and changed the name to Environment and Sustainable Development

Ministry. The Ministry is responsible for the management of the environment and renewable natural

resources and regulates the environmental order of the territory. Also, the Ministry defines policies and

regulations related to rehabilitation, conservation, protection, order, management, use, sustainable

use of natural resources. Article 33 of the same Law created the regional environmental authorities

(including Corpocaldas) with the responsibility to manage the environment and renewable natural

resources. Under this same law, Regional Autonomous Corporations were created and others, which

preceded the law, were ratified. These regional authorities function in the same way as the Ministry,

but with jurisdiction over specific territories.

In 2011, Decree 3533 created the National Authority of Environmental Licenses (Autoridad Nacional

de Licencias Ambientales, ANLA). ANLA is responsible to ensure all project, works or activities subject

to licensing, permit or environmental procedures comply with the environmental regulations and

contribute to the sustainable development of the country. ANLA will approve or reject licenses, permits

or environmental procedures according to the law and regulations, and will enforce compliance with

the licenses, permits and environmental procedures.

With regard to the licensing process of mining projects, the competence of either ANLA or Corpocaldas

is determined by the annual volume of material to be exploited. For projects exploiting more than 2

Mt/y the responsibility will be with ANLA. Both ANLA and Corpocaldas can enforce project compliance

with the terms of their licenses or permits. Up to now and in the foreseeable future, based on the

annual production and transport of materials at Marmato, the environmental authority that controls

operations is Corpocaldas.



20.3.3 Environmental Regulations and Impact Assessment

Colombian laws have distinguished between the environmental requirements for exploration activities,

and those that have to be fulfilled for construction and exploitation works. During the exploration phase,

the concession holder is not required to obtain an environmental license, for now. However, the

concession holder requires environmental permits which will be obtained from the regional

environmental authority. The concession holder will have to comply with the mining and environmental

guidelines issued by the MME and the Environmental Ministry.

In order to begin and perform construction and exploitation operations, the concession holder must

obtain an environmental license or the approval of an existing PMA either from ANLA if the Project

exploits more than 2 Mt/y or from the regional environmental authority (Corpocaldas) if the mineral

exploitation is less than 2 Mt/y.

The approval process begins with the request for Terms of Reference (ToR) to prepare an

Environmental Impact Statement (EIS) or update an existing PMA. The approval of the EIS and PMA

by the jurisdictional environmental authority includes all environmental permits, authorizations and

concessions for the use, exploitation or affectation, or all of the above, of natural resources necessary

for the development and operation of the Project, work or activity. Additionally, other permits and

requirements (non-environmental) are required in order to begin construction and operation of the

Project.

Non-Governmental Organizations (NGOs) and the local communities have the opportunity to

participate in the environmental administrative procedures leading up to the issuance of an

environmental license. The environmental process will include participation of, and information to, all

communities in the project area including indigenous communities and Afro-descendant communities.

To date, the Marmato Project has removed less than 2 Mt/y; therefore, the environmental authority

responsible for issuing environmental licenses and permits, as well as for monitoring and controls, is

the Autonomous Regional Corporation of Caldas – Corpocaldas. Corpocaldas has approved the site’s

Environmental Management Plan under Resolution 496 of 2001, and has issued environmental

permits for the use of natural resources.

CGM maintains internal management files to identify environmental impacts associated with

exploration activities, which are not covered under the mining management plan, but which are also

addressed for its control, mitigation, and correction (if needed). Management files include resources

such as water, air, soil, flora and fauna, waste management, hazardous materials, socio-economic

components in terms of employment and economic transformation due to the demand for goods and

services. In accordance with the provisions of the environmental mining guidelines, impacts associated

with the biotic, abiotic, and socio-economic components which the activity may generate, or cause

have been identified, as well as the different measures which have been established to address these

impacts.



20.3.4 Water Quality and Water Concessions

The Colombian regulations that principally govern water quality, including discharge permitting and

requirements, are Decree 2811 of 1974, Decree 1541 of 1978, Decree 1594 of 1984, Decree 3930 of

2010 and Resolution 631 of 2015, that establish the maximum permissible limits for discharges to

surface water.

Resolution 631 of 2015 (new parameters and maximum limits on point discharges) is being used as a

guideline for this project. The regional environmental authority (Corpocaldas) enforces compliance

with these regulations.

In preparation for the construction of a new CGM workers camp, the Company is carrying out the

necessary studies to file for an additional wastewater discharge point. The camp will be built with a

capacity for 150 people with a total of 10 sanitary units. A single point of domestic wastewater

discharge is currently being considered, which will be treated by a domestic wastewater treatment

system prior to being discharged into the Marmato stream. The domestic wastewater treatment system

would consist of a grease trap, an integrated septic system with a capacity of 15,000 liters and two

inspection boxes for the tributary and the effluent.

Water rights for mining activities are granted by means of a water concession which is granted by

Corpocaldas and which is independent of the mining concession or to land ownership. The water rights

related to mining activities are included in the environmental licenses or in the approved PMA and are

normally granted for five years. The terms and conditions under which a water concession is granted

may depend, amongst others, on the amount of water available in the specific region, the possible

environmental impact of the concession, water demand, the ecological flow and the different users

that the water source services. The water concession is accompanied with a discharge permit.

Water concessions held by CGM for the Marmato Project are shown in Table 20-4.

Table 20-4: Surface Water Concessions

Location Approval Term

(Years) Renewal of

Modification Filing Date

Competent Authority

Bocamina La Maruja

Resolution 345 of March 17, 2014

5 ×

January 29, 2019 (In process extension request)

Corpocaldas Aguas Claras

Resolution 0046 of March 9, 2004 (Amended by Resolution 127 of May 5, 2004)

10 ×

February 7, 2014 (In process extension request)

Zaparillo

Guineo

Source: CGM, 2019

Renewal or extension requests of these water concessions are currently under review by Corpocaldas

but are expected to be re-issued. Under the provisions of Article 35 of Decree Law 019 of 2012, the

environmental permits issued to the Marmato Project are automatically extended until Corpocaldas

officially acts on the request, and the originally terms and conditions of the permit remain in effect in

accordance with the provisions of Decrees 948 of 1995, 3930 of 2010 and 1541 of 1978, compiled in

Decree 1076 of 2015.



A new water concession and impoundment on the Cauca River are being considered to capture 50

liters of water per second to cover the MDZ expansion project. The mine’s perched aquifers, which

have served as the main water supply for current operations, are experiencing low water levels

resulting from long periods of drought and anthropogenic pressures.

20.3.5 Air Quality and Emissions

Decree 948 of 1995, Resolution 650 of 2010 and Resolution 2154 of 2010 provide the main regulations

on protection and control of air quality. These regulations set forth the general principles and

regulations for the atmospheric protection, prevention mechanisms, control and attention of pollution

episodes from fixed, mobile or diffused sources. These regulations also provide emission levels or

standards. Among the emission sources regulated are:

• Controlled open burnings

• Discharge of fumes, gases, vapors

• Dust or particles through stacks or chimneys

• Fugitive emissions or dispersion of contaminants by open pit mining exploitation activities

• Solid, liquid and gas waste incineration

• Operation of boilers or incinerators by commercial or industrial establishments, etc.

Also, Resolution 627 of 2006 regulates noise emissions in terms of ambient noise. The parameters

regulated are

• SO2

• NO2

• CO

• TSP

• PM10

• O3

• Noise

CGM enforces compliance with these regulations at Marmato.

20.3.6 Fauna and Flora Protection

The main regulations for the protection of fauna and flora are contained in the Natural Resources Code

and the Agreement about Biological Diversity entered into in Rio de Janeiro on June 5, 1992, within

the framework of the Rio Convention. Also, forest management and use is regulated by Decree 1791

of 1996 and the compensation for biodiversity loss is regulated by Resolution 1517 of 2012. In addition,

there are other important regulations on the matter such as the Cartagena Protocol on Biotechnology

Security of the Agreement about Biological Diversity entered into in Montreal on January 29, 2000,

and the Convention on International Trade of Threatened Wild Fauna and Flora Species (CITES).

Endangered species are protected by environmental and criminal law.

In order to perform biodiversity studies, a permit for scientific investigation must first be obtained from

Corpocaldas.



20.3.7 Protection of Riparian Areas and Drainages

Resolution No. 077 of March 2, 2011, regarding riparian and water channel protection, strictly prohibits

the filling of perennial water courses except under very specific terms: road and pipeline crossings,

bank and slope protection measures, and installation of public service networks (Title III, Article 9).

The backfilling of intermittent or ephemeral channels can be authorized under permit by Corpocaldas,

provided that the design is appropriate for the conditions, and that surface water and groundwater are

properly managed. Application of this prohibition directly influenced the siting of the future tailings

disposal areas, in that it (they) cannot be located in perennial drainages.

20.3.8 Protection of Cultural Heritage or Archaeology

Cultural and natural heritage protection in Colombia is stated in the political constitution and developed

through several international treaties and laws of the state. There are strict legal provisions, such as

Law 397 of 1997 and Decree 763 of 2009, whereby the heritage is safeguarded and protected. For

example, if a citizen finds an archeological specimen, he or she must inform the Ministry of Culture of

the discovery within 24 hours; otherwise he or she could be sanctioned by the competent authority.

20.3.9 Marmato Permitting

The Marmato Project is authorized under a number of resolutions issued by Corpocaldas in the name

of CGM’s predecessor, Mineros Nacionales S.A.S. These are identified in the Environmental Studies

and Management section (above), and include, among others:

• Environmental Management Plan or PMA (Resolution No. 496)

• Various water concessions

• Discharge permits (Resolutions 270 modified by 255)

• Air emissions (Resolution 270)

CGM is currently in the process of modifying the PMA to include a second DSTF disposal area

(Cascabel 2). To this end, CGM has presented the impact assessment and technical documentation

for this modification to Corpocaldas for review. Corpocaldas has evaluated the request and is waiting

for the Ministry of the Interior to certify the presence, or not, of ethnic communities in the area of the

new facility prior to issuing its final decision. Once Corpocaldas authorizes the Cascabel 2 modification,

a new modification request will be submitted for the construction of a third tailings disposal facility (El

Guaico) for agency approval.

The PMA will require a major modification to allow for the proposed MDZ expansion project, which

envisions an increase in production in a second processing plant to be constructed. By regulation, the

total of mined material (including waste and material) cannot exceed 2 Mt/y in order for Corpocaldas

(Regional Environmental Authority) to remain as the permitting authority. If more than 2 Mt/y is mined,

then the PMA will need to be submitted to, and authorized by, the national authority, ANLA

(Environmental License National Authority). At this time, CGM does not propose a combined

excavation rate of greater than 2 Mt/y between the existing and MDZ expansion operations, thus

maintaining Corpocaldas as the permitting authority.

During construction, Channel Occupancy Permits will need to be obtained for the new tailings site, the

process plant site, and the site of the underground portal (bocamina). Likewise, a Forest Exploitation

Permit will be needed for areas of proposed surface disturbance with trees (Diameter at Breast Height

or DBH more than 10 cm).



These new facilities and operation will be subject to the environmental licensing process described

above, which will require the submittal of comprehensive design reports, hydraulic and hydrogeological

investigation reports, geotechnical reports on stability, and an environmental impact assessment will

need to be prepared. This will require that adequate baseline data be collected from which the

significance of potential impacts can be assessed. Much of this information has been collected and

reported in the Capítulo 20: Caracterización Ambiental y Social del Proyecto, Caldas Gold Marmato

S.A.S., Título Minero #014 – 89m (May 2020). Initiation of the environmental impact assessment will

begin upon finalization of this PFS, detailing the proposed expansion facilities and operations.

Operationally, the existing discharge, emissions (if applicable), and water concession permits may

also require modification to suit the new mining conditions. As with the minor modifications discussed

above, the Ministry of the Interior will again need to be engaged with respect to the certification of the

presence of ethnic communities to ascertain if the MDZ project modification and expansion will require

special consultation.

The final environmental impact assessment deliverable includes the application for all the

environmental permits that will be required for the construction and operation phases of the project.

Once the EIA is officially delivered to Corpocaldas, the review process can begin based on the agreed-

upon terms of reference. This review process can take anywhere from six to 24 months to complete,

depending on the complexity of the project and the quality of the information provided. An incomplete

application is immediately rejected. CGM estimates that a minimum of six months will be required for

review of the complete application and issuance of the Resolution by which Corpocaldas approves the

modification requested for the MDZ expansion of the Marmato Project. However, this process has

been delayed as a result of the COVID-19 pandemic, and CGM does not anticipate fully reengaging

Corpocaldas with the submittal of the EIA until Q1 of 2021. Optimistically, permissions to initiate

construction could be received by Q3 2021.

20.3.10 Performance and Reclamation Bonding

The termination of a mining concession can happen for several reasons: resignation, mutual

agreement, and expiration of the term, the concession holder’s death, free revocation and reversion.

In all cases, the concession holder is obliged to comply or guarantee the environmental obligations

payable at the time the termination becomes effective.

The 2001 Mining Code requires the concession holder to obtain an Insurance Policy to guarantee

compliance with mining and environmental obligations which must be approved by the relevant

authority, annually renewed, and remain in effect during the life of the Project and for three years from

the date of termination of the concession contract. The value to be insured will be calculated as follows:

• During the exploration phase of the Project, the insured value under the policy must be 5% of

the value of the planned annual exploration expenditures

• During the construction phase, the insured value under the policy must be 5% of the planned

investment for assembly and construction

• During the exploitation phase, the insured value under the policy must be 10% of the value

resulting from the estimated annual production multiplied by the pithead price established

annually by the government

According to the Law, the concession holder is liable for environmental remediation and other liabilities

based on actions and or omissions occurring after the date of the concession contract, even if the



actions or omissions are by an authorized third-party operator on the concession. The owner is not

responsible for environmental liabilities which occurred before the concession contract, from historical

activities, or from those which result from non-regulated mining activity, as has occurred on and around

the Marmato Project site.

In accordance with the terms and conditions of the PMA, CGM maintains an Environmental Insurance

Policy for the current operation. That policy is renewed annually with Corpocaldas as beneficiary. This

policy is intended to cover the entire Marmato operations and all aspects of environmental compliance.

According to CGM, the current amount covered by the policy is COL$302,835,000 (USD$91,768). This

amount will be reviewed and adjusted during the modification process of the PMA for the MDZ

expansion project.

20.4 Social or Community Related Requirements

The 2001 PMA for Marmato specifically requires the management of the social component of the

Project through two programs (MM17 and PGS1, see section 4.5). CGM is required to maintain records

on all community activities (including number of participants, topics, duration, etc.), which is to be

turned over to Corpocaldas every six months as part of the ongoing monitoring programs.

20.4.1 Social Investment

As part of the social management and monitoring program, CGM has developed a social investment

model which seeks to promote the development of communities in the area of influence, with the

purpose of contributing to the consolidation of society and fostering economic development (Economic

Development), guaranteeing the care and respect for the environment (Environmental Development),

and supporting and participating in actions aimed at improving the quality of life and well-being of its

inhabitants (Social Development and Promotion of Solidarity Actions). Activities in 2017 included,

among others (for example):

• Direct economic compensation in excess of COL$3,850,000,000 (US$1.17 million) (including

state royalties and a payment of COL$2,079,000,000 (US$630,000) to the municipality of

Marmato)

• Support of education programs, such as Mining Training School with the National Learning

Service (Servicio Nacional de Aprendizaje) (SENA), Mine Rescue Training, Nutrition and

Safety Training, etc.

• Support for traditional festivities of local municipalities like Ferias de El Oro and Fiesta of El

Barequero

• Support for the Afro-Colombian meeting

• Inauguration of the C.E.S. San Antonio hospital and other health programs, including

vaccination days, sexual education workshops, drug addiction prevention workshops, etc.

• Leisure activities with educational institutions

• Music school educational programs

• A film exhibition of Marmato

• Employee human resource initiatives, such as the Bono Social and Novenas de navidad

programs

According to CGM, the company has a complaints and petitions handling procedure to record

grievances both at the Company offices and community office in El Llano. The grievance recording



and response procedures follow international good practices. SRK was not made aware of any current

or ongoing complaints in need of resolution.

20.4.2 Community Relations

Between 2014 and 2018, CGM developed and implemented a social engagement program at Marmato

specifically designed to focus on the well-being of the community and care for the environment. These

initiatives are incorporated in the Community Relations Plan (Plan de Relaciones con la Comunidad),

and include:

• Biodiversity and water for the future

• Education for development

• Protection of culture

• Health and well-being

• Productive chains of small mining in our value chain

• Women leaders and entrepreneurs

• Eradication of child labor

• Infrastructure for development

These initiatives are a collaborative effort within CGM and not necessarily the responsibility of

individual departments or organizations.

20.4.3 Employment

The Marmato Project currently operates with 152 administrative employees, 1,090 operating workers

and 54 apprentice workers, most of whom are from the municipalities surrounding the project, including

Supía, Riosucio, La Pintada, and Marmato. Skilled labor, such as engineers, geologists, surveyors,

some supervisors, and mechanics are from the cities of Medellín, Pereira, Manizales, Cartago, and

some from areas of the Department of Boyacá.

With the MDZ expansion, CGM anticipates hiring approximately 900 temporary workers during

construction and around 350 permanent employees as part of the new operations.

20.4.4 Artisanal and Small-Scale Mining Operations

The area has been exploited since pre-Colonial times by the Quimbaya people. The Spanish colonists

assumed control of the Marmato mines in 1527 and the area has been in almost continuous production

ever since. This majority of the mining is informal/artisanal in nature (sometimes referred to as

“traditional”), which is the general characteristic of the mining sector in Colombia. A recent census

revealed that 72% of all mining operations in Colombia are classed as ‘artisanal and small-scale

mining’ (ASM), and 63% are ‘informal’, lacking a legal mining concession or title. Large-scale mining

(LSM) only accounts for 1% of operations. Over 340,000 Colombians depend directly on ASM and

medium-scale mining (MSM) for their income. This informality deprives the state of important financial

resources, while the current poor conditions (environmental, social, health and safety, labor, technical

and trading) prevent the sector from delivering on important social objectives, such as generating

formal employment and improving the quality of life in mining communities (Echavarria 2014).

In 2013, a decree (933) was enacted to address the legal void for almost 4,000 requests for

formalization from Law 1382 of 2010, which was promulgated, in part, with the objective of combating



illegal mining, while recognizing the traditional nature of informal ASM. This decree redefined

traditional mining as a form of informal mining. It set out formalization procedures for ASM in LSM

mining concessions and titles, notably including procedures for concession-owners to cede areas to

ASM, and included tax incentives. For the first time, it also provided options for areas returned to the

state to be reserved for ASM formalization. In addition, Mercury Law No. 1658 of 2013, introduced

incentives for the formalization of ASM such as: granting of soft credits and financing programs to

facilitate access to resources; and created a sub-contract intended to formalize illegal mining activities

with the registered license-holder. Under Article 11 of Law 1658, concession owners can sign

subcontracts with ASM operating in their concessions without the liability associated with normal

operating contracts. These subcontracts will legally allow these ASM to operate in an agreed upon

area with no oversight by the concession owner. Instead these ASM will be under the control of the

Colombian mining and environmental authorities.

20.5 Mine Closure, Remediation, and Reclamation

Article 209 of Law 685 of 2001 requires that the concession holder, upon termination of the agreement,

shall undertake the necessary environmental measures for the proper reclamation and closure of the

mining operation. To ensure that these activities are carried out, the Environmental Insurance Policy

(see above) shall remain in effect for three years from the date of termination of the contract. Little else

regarding the specifics of mine closure is provided in the Law. Decree 2820 Article 40 Paragraph 2 of

2010 specifically indicates that the concession holder must submit a plan for dismantling and

abandonment of the Project.

While a formal closure plan is not legally required at this stage of the operation, currently there is a

closure plan for Marmato, Plan de Cierre y Abandono de Mina La Maruja – Gran Colombia Gold

Marmato S.A.S. (May 2019) which discusses basic reclamation and closure actions including aspects

of temporary, progressive, and final closure. More detailed, site-wide closure actions have not yet been

defined, as these will be developed through five-year updates to help identify potential closure risks

that CGM may need to manage and finalize closer to the end of operations. The below discussion

focuses on final closure and post-closure.

Some surface facilities (e.g., tailings storage facility) will be progressively reclaimed over the duration

of the mine site operations, albeit on a limited basis, as there are relatively few surface facilities suitable

for concurrent reclamation and closure. In addition, progressive reclamation and closure can result in

the development of expertise on the most appropriate reclamation methods. Progressive reclamation

and closure will be undertaken, however, without posing impediments on day-to-day operations of the

site. Final closure of the mine site will be undertaken following completion of all mining operations.

Final closure of the Marmato facilities and MDZ expansion facilities will entail the following activities, if

not undertaken during progressive closure phases:

• Reclamation of tailings storage facilities:

o The DSTF will be covered with growth media and revegetated.

o Concrete structures will be properly decommissioned.

o Metal fences will be removed.

• Underground workings:

o All equipment with resale value will be removed and salvaged.

o All portals, ventilation apiques, etc., will be sealed to exclude public access.



• Plant and other buildings:

o Plant equipment will be decommissioned and removed for transportation to, and storage

in, Medellín.

o Buildings will be demolished.

• Erosion control measures will be taken where there is evidence of erosion.

• Human resources: mine workers’ contracts will not be renewed (no extra costs to be included

in the reclamation and closure cost estimate) or the contracts will be terminated (which would

incur additional costs).

The May 2019 closure plan discusses post-closure activities which include monitoring for physical and

chemical stability. Physical stability monitoring will include monitoring for ground movements which

would indicate subsidence. The plan currently assumes monitoring of physical stability twice annually

for three years and then annually for three years if no movement is detected. Chemical stability will

include monitoring of water quality of mine effluent as well as tailings draindown. Monitoring of water

quality will continue twice a year for the first three years and then once annually for at least three years

afterwards until such time that permissible limits are met, or flows diminish (in the case of draindown

from tailings).

20.5.1 Reclamation and Closure Costs

Reclamation and closure costs for the current operation are provided in the May 2019 reclamation and

closure plan. These costs are based on percentages of costs to build the facilities. The plan does not

provide the basis for the percentages, and SRK did not independently calculate or validate this

estimate; however, the amount is in keeping with the closure of other moderate-sized underground

mining operations in South America. The reclamation and closure cost estimate provided totals

COL$20,128,000,000 (US$6.1 million based on exchange rate of 3,300 to 1). A requirement for long-

term post-closure water treatment, if deemed necessary, could significantly increase this estimate.

Based on limited PFS design information for the MDZ expansion project, an additional cost of US$3.1

million was included in the technical economic model (for a total of US$9.2 million) to account for the

increase in production anticipated for the new operations and the construction of a new plant and

tailings storage facilities. The lower additional costs can be attributed to the requirement for concurrent

reclamation of the tailings disposal facility, which, due to the construction method, requires reclamation

during operations as opposed to post closure. Numerous assumptions were used in order to calculate

a reasonable estimate, though a more robust assessment of the facilities is recommended as part of

any FS of the project:

• The river water pumping station was assumed to remain post closure for use by the

community.

• The structures and facilities associated with the main camp were also assumed to remain post

closure for use by the community. This includes the domestic wastewater package plant. No

costs were included for demolition and removal.

• Most of the newly created roads servicing the mine would remain post closure to facilitate

access to the portal area, tailings, and stormwater diversion structures, all of which are likely

to require inspection, and possibly care and maintenance in the future. All process-related

equipment and structures would be dismantled and removed from site. No salvage value is

given.



• Roads from each DSTF to the borrow area will be reclaimed.

• No new stormwater diversion structures are expected to be constructed post closure. Those

used during operations will remain.

• The portal conveyor system would be dismantled and removed, but the underground crusher

is expected to be left in place. While this equipment may have salvage value, no credit was

included in the closure cost estimate.

• Both ventilation drives and primary portal entry will be sealed.

• Regrading, placement of growth media and seeding costs for the portal and process areas is

included.

The costs for placement of cover material on the tailings storage facilities was calculated for the entire

surface of each facility, assuming that limited concurrent reclamation had occurred during operations.



21 Capital and Operating Costs SRK visited Marmato’s mine site and office in early 2020; during these visits SRK reviewed budget

estimates and site-specific cost data, this review included data regarding both capital and operating

costs. These reviewed budgets and site-specific cost data are the base support for the capital and

operating cost models prepared for this PFS.

The cost models prepared for the Marmato UZ operation are mostly based on the reviewed budgets,

as this will be a continuation of the current operation. The MDZ cost estimates are based on cost

models prepared by SRK and Ausenco and are based on PFS level designs, estimates and site-

specific data provided by Marmato’s staff.

The mine is currently owner operated and the projections prepared for this PFS assume that this will

be maintained. Common prices for consumables, labor, fuel, lubricants and explosives were used by

all engineering disciplines to derive capital and operating costs. Included in the labor costs are shift

differentials, vacation rotations, all taxes and the payroll burdens.

21.1 Capital Cost Estimates

21.1.1 Marmato Upper Zone

The Marmato UZ is a currently operating underground mine; the estimate of capital includes some



projected production schedule.

The sustaining capital cost estimate developed for this mining area includes the costs associated with

the engineering, procurement, construction and commissioning. The cost estimate is based on

budgetary estimates prepared by CGM and reviewed by SRK. The estimate indicates that the Project

requires a sustaining capital of US$59.5 million to support the projected production schedule

throughout the LoM. Table 21-1 summarizes the LoM sustaining capital estimate and Table 21-2 and

Table 21-3 present the same estimate by year.

Table 21-1: Marmato UZ Sustaining Capital (LoM)


Infill Drilling 11,847,000


Mine Sustaining 10,049,860

Plant Expansion 11,626,000

Plant Expansion Contingency 2,906,500

Plant Sustaining 3,600,000

Dewatering 2,275,706

DTSF 4,744,900


Total 59,546,192





Description 2020 2021 2022 2023 2024 2025 2026

Infill Drilling 2,200,000 2,200,000 2,200,000 2,200,000 2,200,000 121,000 121,000

Development 1,187,325 2,998,025 1,986,625 224,250 - - -

Mine Sustaining 2,127,399 1,777,862 1,852,400 3,858,800 - 154,000 279,400

Plant Expansion 5,035,000 3,511,000 1,210,000 440,000 1,430,000 - -

Plant Expansion Contingency 1,258,750 877,750 302,500 110,000 357,500 - -

Plant Sustaining 300,000 300,000 300,000 300,000 300,000 300,000 300,000

Dewatering 135,000 713,569 1,427,137 - - - -

TSF 1,032,700 2,792,200 170,200 170,200 239,200 42,550 42,550

Closure Costs - - - - - - -

Total 13,276,174 15,170,406 9,448,862 7,303,250 4,526,700 617,550 742,950



Description 2027 2028 2029 2030 2031 2032

Infill Drilling 121,000 121,000 121,000 121,000 121,000 -

Development - - - - - -

Mine Sustaining - - - - - -

Plant Expansion - - - - - -

Plant Expansion Contingency - - - - - -

Plant Sustaining 300,000 300,000 300,000 300,000 300,000 -

Dewatering - - - - - -

TSF 42,550 42,550 42,550 42,550 42,550 42,550

Closure Costs - - - - - 6,100,000

Total 463,550 463,550 463,550 463,550 463,550 6,142,550


Most of this sustaining capital estimate is supported by a budget forecast prepared by CGM and

reviewed by SRK; the following items are covered by this budget:

• A yearly infill drilling expenditure of US$2,200,000

• Mine equipment maintenance and replacement schedule

• Other mine sustaining capital includes improvements and maintenance of existing mine

infrastructure and stationary equipment

• Surface sustaining capital includes improvements and maintenance of infrastructure located

at surface, such as the camp, laboratory, filter presses, detox system, etc.

• Plant expansion capital

• Plant sustaining capital is an estimate of a yearly maintenance cost

• Dewatering structures developed by CGM

• DTSF costs prepared by SRK

• Closure costs estimate prepared by CGM

Development costs are derived from the mining schedule prepared by SRK. The prepared mining

schedule includes meters of development in waste, this schedule of meters was combined with unit

costs, based on site specific data, to estimate the cost of this development operation. This cost was

then capitalized into the sustaining capital.

Table 21-4 presents the assumed unit costs for this development, while Table 21-5 presents the yearly

schedule of capital development meters.



Table 21-4: Marmato UZ Capital Development Unit Costs

Description US$/m

Trans Ramp Development -RMP-3.5x3.5 (m) 1,200

Veins Level -LVL-2.2x2.2 (m) 1,050

Trans Level Access -ACC-3.5x3.5 (m) 1,200

Trans Waste Xcut -XCT1-3x3 (m) 1,150

Ventilation Drift -VNT-3.5x3 (m) 1,175

Apique Development (m) 3,071

Ventilation Raise -RAR-3x3 (m) 1,900


Table 21-5: Marmato UZ Capital Development Meters (2020 to 2023)

Description 2020 2021 2022 2023

Trans Ramp Development -RMP-3.5x3.5 (m) 118 849 - -

Veins Level -LVL-2.2x2.2 (m) 166 109 - -

Trans Level Access -ACC-3.5x3.5 (m) 394 473 627 -

Trans Waste Xcut -XCT1-3x3 (m) 272 1,101 1,010 195

Ventilation Drift -VNT-3.5x3 (m) 31 7 15 -

Apique Development (m) 160 - - -

Ventilation Raise -RAR-3x3 (m) 26 12 29 -


21.1.2 MDZ



the years of 2021 and 2023.

The capital cost estimates prepared for the expansion into this mining area also include estimates for

EPCM and the Owner’s cost to manage it. The cost estimate is based on cost models prepared by

SRK and Ausenco with site specific inputs from CGM. The estimate indicates that the expansion will

require an investment of US$269.4 million; this includes an estimated capital of US$237.2 million plus

13.6% contingency of US$32.2 million. Table 21-6 summarizes the expansion capital estimate.



Table 21-6: MDZ Construction Capital (US$)


Development 19,719,753 - 2,279,534 10,343,401 7,096,818

Mining Equipment Purchases 52,430,929 - 16,868,012 15,295,229 20,267,68

8 Mining Services 11,589,225 - 1,288,744 6,206,511 4,093,970 Infrastructure 33,201,830 - 16,600,915 16,600,915 - Process Plant 42,371,769 - 21,185,884 21,185,884 - DSTF 19,660,473 - 17,212,986 1,279,528 1,167,958 Temporary Power Line 272,727 - 272,727 - - Mining EPCM 9,276,559 - 2,883,922 4,999,126 1,393,512 Mining Owner's 15,721,708 - 3,978,018 7,881,638 3,862,053 Infrastructure + Plant EPCM 10,484,229 - 5,242,114 5,242,114 -

Infrastructure + Plant Owner's 13,602,581 1,087,62

5 4,663,472 5,298,567 2,552,917

Infrastructure + Plant Other Indirect

8,860,555 - 4,430,278 4,430,278 -

Sub-Total 237,192,33

7 1,087,62

5 96,906,605 98,763,190

40,434,916

Mining Contingency 15,091,967 - 2,508,648 5,950,365 6,632,954 Plant + Infrastructure Contingency 14,237,757 - 7,118,879 7,118,879 - DSTF Contingency 2,871,944 - 2,581,948 191,929 98,067

Total Contingencies (13.6%) 32,201,668 - 12,209,474 13,261,173 6,731,021

Total 269,394,00

5 1,087,62

5 109,116,07

9 112,024,36

3 47,165,93

7

Source: CGM/Ausenco/SRK, 2020

The MDZ construction capital is supported by a mix of a PFS study prepared by Ausenco, budgetary

estimates and cost models developed for the installation of the new operation. Budget estimates were

prepared by CGM and reviewed by SRK and cost models were prepared by SRK and Ausenco. The

following items are covered by this budget:

• Schedule of mine equipment purchases prepared from SRK

• Cost model estimate from SRK to install surface facilities like portal, ventilation system, power

distribution, ancillary building, etc.

• Cost model estimate from SRK to install underground facilities like shops, ventilation systems,

refuge chambers, pumping systems, paste distribution, fuel distribution, ancillary equipment,

etc.

• Cost model from Ausenco to install other infrastructure including power supply, access road,

camp and pumping station

• Cost model estimate from Ausenco to install mineral processing plant, including EPCM and

Owner’s costs

• A PFS study prepared by SRK build a tailings storage facility

Development costs are derived from the mining schedule prepared by SRK. The prepared mining

schedule includes meters of development during pre-production, this schedule of meters was

combined with unit costs, based on site specific data, to estimate the cost of this development

operation. Table 21-7 presents the assumed development unit costs and Table 21-8 presents the

scheduled meters for the pre-production period.



Table 21-7: MDZ Pre-Production Development Unit Costs

Description US$/m

Lateral Development

Main Conveyor Ramp -RMC-5.5x5.6 (m) 2,998



Ventilation Drifts-VMR-5x5 (m) 2,286


Bulk excavation equivalent meters (m) 7,724


Blasted Raise-BRS-3x3 (m) 1,570

Raisebore 5m dia-RS1 (m) 4,755


Table 21-8: MDZ Pre-Production Development Meters


Lateral Development

Main Conveyor Ramp -RMC-5.5x5.6 (m) 1,680 - 396 1,284 -

Main Truck Ramp-RMT-5.5x5.5 (m) 1,608 - - 440 1,167

Drift-FWA-5x5 (m) 1,307 - - 68 1,239

Ventilation Drifts-VMR-5x5 (m) 2,202 - 478 1,724 -

Ventilation Connections-VCX-4.5x4.5 (m) 171 - - 44 126

Bulk excavation equivalent meters (m) - - - - -


Blasted Raise-BRS-3x3 (m) 6 - - 6 -

Raisebore 5m dia-RS1 (m) 383 - - 154 229


Ausenco prepared an engineering study at a PFS level to design the mineral processing facilities and

other supporting infrastructure at the mine site. The resulting cost estimate is presented in Table 21-9.



Table 21-9: MDZ Processing Plant and Infrastructure Capital

Description Value (US$)

Buildings (Non Process) 2,475,260

Site Development 6,517,067

Plant Roads 1,460,997

Camp 4,538,557

Administration 2,021,881

Off Site Infrastructure Site Development 313,695

Backfill Paste Plant 505,855

U/G Conveying 4,470,587

U/G Power Generation / Distribution 4,029,202

Sewer (incl WWT plant, ponds) 1,859,489

Power 5,009,239

Total Infrastructure Direct $33,201,830

Crushing/Conveying 7,652,569

Milling 12,079,464

Leach/CIP 6,273,515

Detox 1,068,900

ADR/Elution 1,995,010

Goldroom 1,215,914

Reagents 1,832,347

Air/ Water 1,676,444

Tailings Thickening 2,667,196

Tailings Filter 5,910,410

Total Process Plant Direct $42,371,769

EPCM Cost 10,484,229

Contractor Indirect 4,518,958

Freight 2,378,746

Vendor Representatives 255,305

Pre-Commissioning/Commissioning (Craft Support) 20,095

Spares/First Fills 1,687,451

Total Indirect Cost $19,344,784

Total Project Cost Ausenco Scope 94,918,382

Owner's Cost -

Contingency 14,237,757

Total Project Cost Ausenco Scope incl Contingency $109,156,139


Ausenco did not provide SRK with an expenditure curve associated with its capital estimate. SRK

applied this capex evenly thorough the quarters of year 2021 to be conservative.

The MDZ will require sustaining capital to maintain the equipment and all supporting infrastructure

necessary to continue operations until the end of its projected production schedule. The sustaining

capital cost estimate developed for this mining area includes the costs associated with the engineering,

procurement, construction and commissioning. The cost estimate is based on PFS designs and cost

models prepared by SRK with site specific inputs from CGM. The estimate indicates that the Project

requires sustaining capital of US$131.3 million to support the projected production schedule through

the LoM. Table 21-10 summarizes the LoM sustaining capital estimate and Table 21-11 and

Table 21-12 present the same estimate by year.



Table 21-10: MDZ Sustaining Capital (LoM)


Drilling -


Mine Equipment Purchases 17,166,844

Mine Equipment Rebuilds 26,862,004

Mining Services -

Mining Owner's Cost 5,892,624

Mining Contingency 14,671,389

DSTF Sustaining 23,806,666

Rio Sucio Power Line 5,614,521


Total $131,299,895



Description 2023 2024 2025 2026 2027

Drilling - - - - -

Development 2,735,635 4,986,400 3,834,632 2,168,451 3,433,459

Mine Equipment Purchases 6,646,459 3,972,308 - - 1,186,305

Mine Equipment Rebuilds - 1,162,732 2,300,471 4,985,557 4,601,468

Mining Services - - - - -

Mining Owner's Cost 1,689,596 943,372 402,871 227,366 487,479

Mining Contingency 1,322,664 1,704,601 1,307,595 1,476,275 1,799,109

DSTF Sustaining 6,817,007 21,934 64,320 15,054 13,150,673

Rio Sucio Power Line 280,726 561,452 561,452 561,452 561,452

Closure Costs - - - - -

Total $19,492,087 $13,352,799 $8,471,341 $9,434,155 $25,219,945



Description 2028 2029 2030 2031 2032 2033

Drilling - - - - - -

Development 6,412,653 3,918,836 3,897,980 2,467,849 429,949 -

Mine Equipment Purchases 208,000 4,232,979 920,793 - - -

Mine Equipment Rebuilds 681,459 4,291,695 2,278,851 6,399,725 160,047 -

Mining Services - - - - - -

Mining Owner's Cost 454,151 875,725 507,741 258,506 45,817 -

Mining Contingency 1,540,853 2,184,960 1,382,954 1,825,216 127,163 -

DSTF Sustaining 166,510 2,714,184 502,892 166,510 187,582 -

Rio Sucio Power Line 561,452 561,452 561,452 561,452 561,452 280,726

Closure Costs - - - - - 3,000,000

Total $10,025,079 $18,779,831 $10,052,664 $11,679,257 $1,512,011 $3,280,726


In the case of development sustaining capital, these costs are derived from the mining schedule

prepared by SRK. The prepared mining schedule includes meters of development in waste; this

schedule of meters was combined with unit costs, based on site specific data, to estimate the cost of

this development operation. The development in waste costs assumes that this operation will be

performed by a mix of contractors and the owner. Table 21-13 presents the assumed development

unit costs and Table 21-14 and Table 21-15 present the scheduled meters for the sustaining period.



Table 21-13: MDZ Development Sustaining Capital Unit Costs

Description US$/m

Lateral Development




Stope Drifts Waste-DFA-4.5x4.5 (m) 2,640


Blasted Raise-BRS-3x3 (m) 2,292

Raisebore 5m dia-RS1 (m) 7,618

Raisebore 4.5m dia-RS2 (m) 5,184


Table 21-14: MDZ Development Sustaining Capital Meters (2023 to 2027) (US$)

Description 2023 2024 2025 2026 2027

Lateral Development

Main Truck Ramp-RMT-5.5x5.5 (m) - - - - 861

Drift-FWA-5x5 (m) 396 793 492 374 213

Ventilation Connections-VCX-4.5x4.5 (m) 23 69 109 98 -

Stope Drifts Waste-DFA-4.5x4.5 (m) 501 1,149 945 425 411


Blasted Raise-BRS-3x3 (m) 21 - - - -

Raisebore 5m dia-RS1 (m) 54 - - - -

Raisebore 4.5m dia-RS2 (m) - - - - -


Table 21-15: MDZ Development Sustaining Capital Meters (2028 to 2032)

Description 2028 2029 2030 2031 2032

Lateral Development

Main Truck Ramp-RMT-5.5x5.5 (m) 1,136 23 - - -

Drift-FWA-5x5 (m) 433 534 438 264 -

Ventilation Connections-VCX-4.5x4.5 (m) 73 83 84 26 -

Stope Drifts Waste-DFA-4.5x4.5 (m) 265 926 1,034 686 163


Blasted Raise-BRS-3x3 (m) 17 17 - - -

Raisebore 5m dia-RS1 (m) - - - - -

Raisebore 4.5m dia-RS2 (m) 403 - - - -


SRK prepared a schedule of required mining equipment and services and estimated mining sustaining

capital including equipment replacements, rebuilds and related services and administration

(Table 21-16 and Table 21-17).



Table 21-16: MDZ Mining Sustaining Capital (2023 to 2027) (US$)

Description 2023 2024 2025 2026 2027

Lateral Capital Development 2,276,490 4,986,400 3,834,632 2,168,451 3,433,459

Vertical Capital Development 459,145 - - - -

Mining Equipment 5,346,144 2,941,369 - - -

Auxiliary Equipment 1,092,315 418,689 - - 1,186,305

Miscellaneous Equipment 208,000 612,250 - - -

Ventilation Equipment - - - - -

Mining Equipment Rebuilds - 347,627 949,777 1,104,952 3,470,255

Auxiliary Equipment Rebuilds - 815,105 1,244,014 1,983,045 1,131,213

Miscellaneous Equipment Rebuilds - - - - -

Ventilation Equipment Rebuilds - - 106,680 1,897,560 -

Mine Services - - - - -

Owner's Cost 1,689,596 943,372 402,871 227,366 487,479

Sub-Total 11,071,691 11,064,812 6,537,974 7,381,375 9,708,711

Mining Contingency (17.4%) 1,322,664 1,704,601 1,307,595 1,476,275 1,799,109

Total $12,394,354 $12,769,413 $7,845,569 $8,857,650 $11,507,820


Table 21-17: MDZ Mining Sustaining Capital (2028 to 2032) (US$)

Description 2028 2029 2030 2031 2032

Lateral Capital Development 4,284,988 3,878,792 3,897,980 2,467,849 429,949

Vertical Capital Development 2,127,666 40,044 - - -

Mining Equipment - 2,762,379 920,793 - -

Auxiliary Equipment - 1,380,600 - - -

Miscellaneous Equipment 208,000 90,000 - - -

Ventilation Equipment - - - - -

Mining Equipment Rebuilds 552,476 2,540,692 1,340,827 3,867,556 -

Auxiliary Equipment Rebuilds 128,983 1,751,002 938,024 2,385,168 160,047

Miscellaneous Equipment Rebuilds - - - - -

Ventilation Equipment Rebuilds - - - 147,000 -

Mine Services - - - - -

Owner's Cost 454,151 875,725 507,741 258,506 45,817

Sub-Total 7,756,264 13,319,234 7,605,365 9,126,079 635,814

Mining Contingency (17.4%) 1,540,853 2,184,960 1,382,954 1,825,216 127,163

Total $9,297,116 $15,504,194 $8,988,320 $10,951,295 $762,976


SRK prepared PFS level DSTF designs to support the MDZ operation. The yearly capital cost

estimates for the DSTF infrastructure is presented in Table 21-18 and Table 21-19.

Table 21-18: MDZ DSTF Sustaining Capital (2023 to 2027) (US$)

Description 2023 2024 2025 2026 2027

Earthworks, Direct - 14,672 - - 7,734,609

Supervision/Overheads/Profit - 4,402 - - 2,320,383

Equipment - - 55,930 13,090 1,380,376

SubTotal - 19,073 55,930 13,090 11,435,368

Contingency (15%) - 2,861 8,390 1,964 1,715,305

Total - 21,934 64,320 15,054 13,150,673




Table 21-19: MDZ DSTF Sustaining Capital (2028 to 2032)

Description 2028 2029 2030 2031 2032

Earthworks, Direct 111,378 111,378 130,421 111,378 115,404

Supervision/Overheads/Profit 33,413 33,413 39,126 33,413 34,621

Equipment - 2,215,369 267,750 - 13,090

SubTotal 144,791 2,360,160 437,297 144,791 163,115

Contingency (15%) 21,719 354,024 65,595 21,719 24,467

Total 166,510 2,714,184 502,892 166,510 187,582


Additional power requirements to support MDZ will be provided by the Rio Sucio Power infrastructure.

An estimate was prepared by CGM and a 10 year payment schedule is included in the sustaining

capital. Table 21-20 presents the assumptions of this payment schedule.

Table 21-20: MDZ Rio Sucio Power Line Sustaining Capital (2028 to 2032)


Rio Sucio S/S 115 kV 1,314,324 US$

15 km Lines 115kV 1,315,387 US$

MDZ S/S 115 kV 608,617 US$

HV - Total Capex 3,238,328 US$

Year Interest 11.50 %

Period 10 Years

Starting Year 2,023 US$

Down-Payment - US$

Yearly Payment 561,452 US$/year

Source: CGM, 2020

Project closure costs are based on a budget estimate prepared by CGM and reviewed by SRK.

21.2 Operating Cost Estimates

SRK, Ausenco and CGM prepared the estimate of operating costs for the PFS production schedule.

Marmato UZ LoM cost estimate is presented in Table 21-21 and MDZ LoM cost estimate is presented

in Table 21-22

Table 21-21: Marmato UZ Operating Costs Summary


Mining 48.45 249,251

Process 13.86 71,283

G&A 13.82 71,086

Total Operating 76. 12 391,620

Source: CGM/SRK/, 2020

Table 21-22: Marmato MDZ Operating Costs Summary


Mining 35.19 512,288

Process 13.68 199,113

G&A 8.23 119,771





21.3 Basis for Operating Cost Estimates

The prepared estimates that compose the operating costs consist of domestic and international

services, equipment, labor, etc. Where required, the following were included:

• Value added tax

• Freight

• Duty

It was assumed that the mill operates 350 days per year under a daily schedule of two shifts of 12

hours.

The operating cost estimates are based on the quantities associated with the production schedule,

including the following:

• Development meters

• Stope ore tonnage

• Ore tonnage

All operating costs include supervision staff, operations labor, maintenance labor, consumables,

electricity, fuels, lubricants, maintenance parts and any other operating expenditure identified by

contributing engineers.

21.3.1 Marmato UZ

Site-specific 2019 budget estimates were used to estimate the LoM operating costs of the Marmato

UZ. The following costs were used to estimate the operating cost of this mining area:

• Vein mining: US$47.00/t-stope, which includes backfill costs

• Transition mining: US$42.00/t-stope, which includes backfill costs

Additionally, development operating costs are derived from the mining schedule prepared by SRK.

The prepared mining schedule includes meter of development in ore, this schedule of meters was

combined with unit costs, based on site specific data, to estimate the cost of this development

operation. The development in ore costs assume that this operation will be performed by the owner.

Table 21-23 presents the assumed development unit cost, while Table 21-24 presents the scheduled

meters for the operating years.

Table 21-23: Marmato UZ Operating Development Unit Costs

Description US$/m

Veins Development -DEV-2.2x2.2 (m) 1,050

Trans Ore Xcut -XCT2-4x3.5 (m) 1,500


Table 21-24: Marmato UZ Operating Development Meters (2020 to 2024)

Description LoM 2020 2021 2022 2023 2024

Veins Development -DEV-2.2x2.2 (m) 16,558 2,274 5,150 5,130 3,896 108

Trans Ore Xcut -XCT2-4x3.5 (m) 2,161 185 609 1,121 246 -




The Marmato UZ mineral processing costs were modeled with fixed and variable components. These

are based on the recent years of operation and are presented in Table 21-25.

Table 21-25: Marmato UZ Mineral Processing Operating Costs


Fixed Processing Cost 1,801,402 US$/year

Variable Processing Cost 7.29 US$/t


Marmato’s UZ tailings and G&A costs were modeled as fixed costs. Table 21-26 and Table 21-27

present the estimated yearly costs.

Table 21-26: Marmato UZ TSF And G&A Operating Costs (2020 to 2026) (US$)

Description 2020 2021 2022 2023 2024 2025 2026

TSF 778,882 1,550,161 1,756,531 1,523,013 1,234,802 899,824 1,081,687

G&A 5,468,187 5,468,187 5,468,187 5,468,187 5,468,187 5,468,187 5,468,187


Table 21-27: Marmato UZ DSTF And G&A Operating Costs (2027 to 2033) (US$)

Description 2027 2028 2029 2030 2031 2032

TSF 1,010,560 845,920 991,408 974,133 1,087,847 211,547

G&A 5,468,187 5,468,187 5,468,187 5,468,187 5,468,187 5,468,187


21.3.2 MDZ

Cost models prepared by SRK and Ausenco based on site-specific inputs from CGM were used to

estimate the LoM operating costs of the MDZ.

SRK prepared a cost model to estimate the mining operating costs associated with the MDZ mining

schedule. This model estimates the cost associated with each operation that composes the mining

area and the estimated yearly costs by process are presented in Table 21-28 and Table 21-29.

Table 21-28: MDZ Mining Operating Costs (2023 to 2027) (US$)

Description 2023 2024 2025 2026 2027

Operating Development 2,895,921 5,819,528 4,707,591 2,056,179 2,063,319

Production Drilling 293,228 3,109,452 3,650,647 3,921,192 3,922,850

Production Blasting 675,392 7,162,013 8,408,547 9,031,694 9,035,514

Production Mucking 87,085 923,466 1,084,194 1,164,542 1,165,034

Production Backfill 677,230 7,881,996 8,930,047 9,733,981 9,476,936

Hauling 659,658 3,085,207 3,106,364 2,933,110 3,330,938

Mine Services and Maintenance 4,225,975 9,522,563 7,581,680 7,699,826 7,845,331

Rehabilitation - 52,500 140,000 140,000 210,000

Definition Drilling 316,817 600,000 600,000 600,000 600,000

Operating and Maintenance Hourly 3,321,537 6,934,112 6,934,112 6,934,112 6,934,112

Operating and Maintenance Staff 1,527,717 3,055,433 3,055,433 3,055,433 3,055,433

Sub-Total 14,680,559 48,146,271 48,198,615 47,270,068 47,639,468

Contingency 835,612 3,589,613 3,749,438 3,823,601 3,803,514

Total $15,516,171 $51,735,884 $51,948,053 $51,093,669 $51,442,981




Table 21-29: MDZ Mining Operating Costs (2028 to 2033)

Description 2028 2029 2030 2031 2032 2033

Operating Development 830,800 2,811,103 3,262,893 3,044,206 1,156,400 -

Production Drilling 4,061,297 3,845,843 3,797,217 3,818,841 4,024,168 3,799,470

Production Blasting 9,354,399 8,858,144 8,746,142 8,795,950 9,268,879 8,751,332

Production Mucking 1,206,151 1,142,164 1,127,723 1,134,145 1,195,124 1,128,392

Production Backfill 9,827,813 7,805,596 8,022,274 8,044,576 9,683,728 9,118,607

Hauling 3,624,384 3,459,024 4,000,399 3,925,646 3,419,127 1,837,274

Mine Services and Maintenance 7,881,397 7,917,462 7,838,608 7,838,608 6,875,608 6,610,040

Rehabilitation 280,000 280,000 280,000 280,000 280,000 280,000

Definition Drilling 685,648 605,100 450,000 - - -

Operating and Maintenance Hourly 6,934,112 6,787,922 6,681,992 6,681,992 5,709,442 4,221,780

Operating and Maintenance Staff 3,055,433 3,055,433 3,055,433 2,843,575 2,075,173 1,801,518

Sub-Total 47,741,435 46,567,791 47,262,682 46,407,538 43,687,649 37,548,413

Contingency 3,861,244 3,499,229 3,566,475 3,527,063 3,636,942 3,245,212

Total $51,602,678 $50,067,020 $50,829,157 $49,934,601 $47,324,590 $40,793,625


Ausenco estimated the MDZ mineral processing operating cost as part of the PFS. This cost was

considered entirely variable and its detail is presented in Table 21-30.

Table 21-30: MDZ Mineral Processing Cost

Processing US$/t-milled

Labor 0.89

Comminution Consumables 2.69

Reagents/Consumables 4.25

Power 3.66

Maintenance & Lubrication Supplies 0.59

Laboratory 0.16

Vehicles 0.07

Water Treatment 0.09

Total Process Operating Costs 12.40

Source: CGM/Ausenco, 2020

MDZ tailings and G&A costs were modeled as fixed costs. Table 21-26 and Table 21-27 present the

estimated yearly costs.

Table 21-31: MDZ DSTF And G&A Operating Costs (2023 to 2027)

Description 2023 2024 2025 2026 2027

DSTF 509,413 1,718,748 1,820,976 1,690,050 1,721,338

G&A 7,370,93 12,880,949 11,551,961 11,551,974 11,551,974


Table 21-32: MDZ DSTF And G&A Operating Costs (2028 to 2033)

Description 2028 2029 2030 2031 2032 2033

DSTF 1,745,584 2,029,329 2,057,835 1,991,088 1,741,293 1,593,741

G&A 11,552,136 1,546,385 11,540,807 11,345,639 10,855,370 8,023,006




22 Economic Analysis SRK prepared a cash flow model to evaluate Marmato’s ore reserves. This model was prepared in

quarterly periods from the beginning of the year 2020 to the end of 2024, after which it was modeled

in yearly periods. This section presents the main assumptions used in this cash flow model and its

outcomes. All financial data is first quarter 2020 and currency is in U.S. dollars (US$), unless otherwise

stated.

22.1 External Factors

Assumed prices are based on the long-term outlook for gold and silver. This projection is well below

the current spot prices and the long-term views of relevant market analysis in the precious metal sector.

Table 22-1 presents the prices used for the cash flow modelling and resources estimation.

Table 22-1: Marmato Price Assumptions


Gold 1,400 US$/oz

Silver 17.00 US$/oz

Source: CGM, 2020

All cost inputs prepared for the cash flow model used a mixture of costs estimated in U.S. Dollars and

in Colombian Pesos. All cost estimates that were denominated in Colombian Pesos were converted

to US Dollars using a foreign exchange conversion rate of 3,300 Colombian Pesos per U.S. Dollars.

Marmato currently has a long-term supply agreement for the sale of its products to an international

refinery who take delivery of doré from the mine at designated transfer points in Colombia. The refinery

is responsible for shipping the products abroad. The refining costs and discounts associated with the

sales of the products are based on this agreement. This study was prepared under the assumption

that the Project will sell doré containing gold and silver.

Treatment charges and NSR terms are summarized in Table 22-2.

Table 22-2: Marmato NSR Terms

Description Value Units

Doré

Payable Gold 100%

Doré Smelting & Refining Charges 6.38 US$/oz-Au

Source: CGM, 2020

22.2 Production Assumptions

Marmato’s operation is supported by the production of doré bars containing gold and silver. The doré

bars result from the mineral processing of RoM containing gold and silver from an underground mining

operation. Table 22-3 presents this PFS’s life of mine production summary.



Table 22-3: Marmato Production Summary

LoM

Marmato Upper Zone

Stope Ore (t) 4,997,091

Dev Ore (t) 147,572

Total Ore (t) 5,144,663

Au Head Grade (g/t) 4.16

Ag Head Grade (g/t) 15.41

Au Contained (oz) 687,339

Ag Contained (oz) 2,549,213

Waste (t) 338,204

Total Material Mined (t) 5,482,867

Marmato Deeps Zone

Stope Ore (t) 13,511,892

Dev Ore (t) 1,044,054

Total Ore (t) 14,555,946



Au Contained (oz) 1,332,795


Waste (t) 1,806,960

Total (t) 16,362,906

Total Mine Production

Stope Ore Tonnes (t) 18,508,983

Dev Ore Tonnes (t) 1,191,626

Total Ore Tonnes 19,700,609



Au Contained (oz) 2,020,134


Waste (t) 2,145,164

Total Material Mined (t) 21,845,773

Source: SRK, 2020

The Marmato operation is composed of two major mining areas, namely the Marmato UZ and the

MDZ.

The Marmato UZ is a mining operation that exploits the upper portion of the mineral deposit, it is

represented by the current operation, which is composed by an underground mining operation and an

existing mineral processing plant. The production schedule prepared for the Marmato UZ does not

consider a pre-production period, as this part of the operation is currently producing. The production

schedule assumes an expansion of the mining and mineral processing operations starting in 2021.

This expansion is planned to ramp-up in the years of 2021 and 2022 and reach full capacity in the year

2023.



The MDZ is a future mining operation that will exploit the lower portion of the deposit that will support

an expansion of the Marmato operation. This will require the installation of a mining operation and a

mineral processing facility dedicated to this area of the mineral deposit. The mineral processing circuit

will also extract doré bars containing gold and silver from run of mine from the MDZ. The mine schedule

for the MDZ assumes a pre-production period of three and a half years, which consists of a delay of

six months followed by a construction period of another three years.

The Marmato UZ mine production is based on a LoM assumed average mining rate of 1,155 t/d and

maximum mining rate of 1,648 t/d, including movement of both ore and waste and based on a

denominator of 365 days per year. The MDZ mine production is based on a LoM assumed average

mining rate of 4,115 t/d and maximum mining rate of 4,515 t/d, this includes movement of both ore and

waste and is based on a denominator of 365 days per year. The combined total mine movement is

limited to a maximum of 2 Mt/y.

The mine schedule does not include any stockpiling, all of the blending of RoM is done in the mine or

in a designated area before the plant feed. Table 22-4 and Table 22-5 presents the yearly LoM mine

production assumptions by area.



Table 22-4: Marmato Yearly (2020 to 2026) Mine Production Assumptions

2020 2021 2022 2023 2024 2025 2026

Marmato UZ

Stope Ore (t) 256,454 393,457 439,606 501,914 491,238 388,538 452,030

Dev Ore (t) 29,717 44,028 50,444 23,087 296 - -

Total Ore (t) 286,171 437,485 490,050 525,001 491,534 388,538 452,030

Au Head Grade (g/t) 3.66 3.99 3.95 3.97 3.96 4.39 4.30

Ag Head Grade (g/t) 16.83 14.51 13.06 14.86 16.25 16.98 15.75

Au Contained (oz) 33,668 56,077 62,288 67,074 62,579 54,875 62,529

Ag Contained (oz) 154,890 204,032 205,843 250,803 256,794 212,132 228,933

Waste (t) 51,600 127,256 111,535 46,566 1,247 - -

Total Material Mined (t) 337,771 564,741 601,585 571,567 492,781 388,538 452,030

MDZ

Stope Ore (t) - - - 103,599 1,098,587 1,289,794 1,385,379

Dev Ore (t) - - - 105,415 212,080 171,776 75,270

Total Ore (t) - - - 209,014 1,310,667 1,461,570 1,460,649

Au Head Grade (g/t) - - - 3.10 3.15 3.27 3.41

Ag Head Grade (g/t) - - - 4.65 4.43 4.47 4.85

Au Contained (oz) - - - 20,831 132,913 153,659 160,137

Ag Contained (oz) - - - 31,272 186,520 210,048 227,761

Waste (t) - 63,504 284,055 299,957 196,560 149,907 87,318

Total (t) - 63,504 284,055 508,971 1,507,227 1,611,477 1,547,967

Total

Stope Ore Tonnes (t) 256,454 393,457 439,606 605,513 1,589,825 1,678,332 1,837,409

Dev Ore Tonnes (t) 29,717 44,028 50,444 128,502 212,376 171,776 75,270

Total Ore Tonnes 286,171 437,485 490,050 734,015 1,802,201 1,850,108 1,912,679

Au Head Grade (g/t) 3.66 3.99 3.95 3.72 3.37 3.51 3.62

Ag Head Grade (g/t) 16.83 14.51 13.06 11.95 7.65 7.10 7.43

Au Contained (oz) 33,668 56,077 62,288 87,905 195,492 208,534 222,666

Ag Contained (oz) 154,890 204,032 205,843 282,075 443,314 422,180 456,694

Waste (t) 51,600 190,760 395,590 346,523 197,807 149,907 87,318

Total Material Mined (t) 337,771 628,245 885,640 1,080,538 2,000,008 2,000,015 1,999,997

Source: SRK, 2020



Table 22-5: Marmato Yearly (2027 to 2033) Mine Production Assumptions

2027 2028 2029 2030 2031 2032 2033

Marmato UZ

Stope Ore (t) 409,596 351,864 386,605 388,892 445,184 91,713 -

Dev Ore (t) - - - - - - -

Total Ore (t) 409,596 351,864 386,605 388,892 445,184 91,713 -

Au Head Grade (g/t) 4.34 4.37 4.30 4.18 4.43 4.22 -

Ag Head Grade (g/t) 15.23 14.42 14.33 14.43 17.43 21.69 -

Au Contained (oz) 57,171 49,475 53,418 52,299 63,442 12,443 -

Ag Contained (oz) 200,574 163,168 178,148 180,463 249,477 63,956 -

Waste (t) - - - - - - -

Total Material Mined (t) 409,596 351,864 386,605 388,892 445,184 91,713 -

MDZ

Stope Ore (t) 1,385,965 1,434,879 1,358,758 1,341,578 1,349,218 1,421,761 1,342,374

Dev Ore (t) 75,157 30,299 102,341 118,804 110,823 42,089 -

Total Ore (t) 1,461,122 1,465,178 1,461,099 1,460,382 1,460,041 1,463,850 1,342,374

Au Head Grade (g/t) 2.94 2.77 2.47 2.33 2.50 2.79 2.84

Ag Head Grade (g/t) 4.54 4.04 2.87 3.11 3.01 3.22 3.85

Au Contained (oz) 138,110 130,485 116,029 109,399 117,354 131,308 122,570

Ag Contained (oz) 213,272 190,311 134,819 146,022 141,294 151,546 166,160

Waste (t) 129,281 182,953 152,291 150,710 94,775 15,649 -

Total (t) 1,590,403 1,648,131 1,613,390 1,611,092 1,554,816 1,479,499 1,342,374

Total

Stope Ore Tonnes (t) 1,795,561 1,786,743 1,745,363 1,730,470 1,794,402 1,513,474 1,342,374

Dev Ore Tonnes (t) 75,157 30,299 102,341 118,804 110,823 42,089 -

Total Ore Tonnes 1,870,718 1,817,042 1,847,704 1,849,274 1,905,225 1,555,563 1,342,374

Au Head Grade (g/t) 3.25 3.08 2.85 2.72 2.95 2.87 2.84

Ag Head Grade (g/t) 6.88 6.05 5.27 5.49 6.38 4.31 3.85

Au Contained (oz) 195,281 179,960 169,447 161,698 180,795 143,751 122,570

Ag Contained (oz) 413,846 353,478 312,968 326,485 390,770 215,502 166,160

Waste (t) 129,281 182,953 152,291 150,710 94,775 15,649 -

Total Material Mined (t) 1,999,999 1,999,995 1,999,995 1,999,984 2,000,000 1,571,212 1,342,374

Source: SRK, 2020



Marmato UZ is burdened by development waste only during its first four years of production, its gold

and silver grades are mostly stable through the life of the mining area. MDZ presents more

development in waste between the years of 2021 and 2023 to open the mining area for production.

Once ore production ramps-up the waste development is mostly stable throughout the life of the mining

area; its gold and silver grades present a slight decline as the production advances.

The resulting combined mine production profile presents modestly increasing ore production between

the years of 2020 and 2023 and a large expansion in the production capacity once the MDZ ramps-up

in 2024. The silver grade declines through the life of mine, which is related to MDZ having a much

lower silver grade than Marmato UZ. The gold grade declines from around 4 g/t down to around 3 g/t

over the course of the production life.

Figure 22-1 and Figure 22-2 show each mine area’s RoM ore production, while Figure 22-3

consolidates the two mining operations into a single production profile.

Source: SRK, 2020

Figure 22-1: Marmato UZ Mine Production Profile

Source: SRK, 2020

Figure 22-2: MDZ Mine Production Profile



Source: SRK, 2020

Figure 22-3: Marmato Combined UZ and MDZ Mine Production Profile

The Marmato operation’s doré production is supported by two mineral processing facilities, a plant that

beneficiates the run of mine from Marmato UZ and a second plant that will beneficiate the run of mine

from MDZ.

The Marmato UZ is currently producing and is supported by an existing mineral processing plant. The

mineral processing production schedule assumes an expansion of the mineral processing operations

starting in 2021. This expansion is planned to ramp-up in the years of 2021 and 2022 and reach full

capacity in the year of 2023.

The processing of the MDZ run of mine will require the installation of a mineral processing facility

dedicated to this area of the mineral deposit. The mineral processing circuit will also extract doré bars

containing gold and silver from run of mine from MDZ.

The Marmato UZ mineral processing production is based on a LoM assumed average processing rate

of 1,084 t/d and maximum ore mining rate of 1,438 t/d, based on a denominator of 365 days per year.

The MDZ mineral processing production is based on a LoM assumed average processing rate of 3,677

t/d and maximum ore mining rate of 4,014 t/d, based on a denominator of 365 days per year

Table 22-6 presents the projected LoM plant production for the Marmato UZ, MDZ and the combined

mineral processing throughput.



Table 22-6: Marmato Mill Production Assumptions

LoM

Marmato UZ

Ore Feed (t) 5,144,663

Ore Feed Au Head Grade (g/t) 4.16

Ore Feed Ag Head Grade (g/t) 15.41

Ore Feed Contained Au (oz) 687,339

Ore Feed Contained Ag (oz) 2,549,213

Gold Recovery (%) 87%

Silver Recovery (%) 33%

Recovered Gold (oz) 598,939

Recovered Silver (oz) 846,780

MDZ

Ore Feed (t) 14,555,946



Ore Feed Contained Au (oz) 1,332,795




Recovered Gold (oz) 1,266,155

Recovered Silver (oz) 719,609

Total

Ore Feed (t) 19,700,609



Ore Feed Contained Au (oz) 2,020,134




Recovered Gold (oz) 1,865,094

Recovered Silver (oz) 1,566,389

Source: SRK, 2020

Table 22-7 and Table 22-8 present the projected yearly plant production for Marmato UZ, MDZ and

the combined mineral processing throughput.



Table 22-7: Marmato Mill Production Schedule (2020 - 2026)

2020 2021 2022 2023 2024 2025 2026

Marmato UZ

Ore Feed (t) 286,171 437,485 490,050 525,001 491,534 388,538 452,030

Ore Feed Au Head Grade (g/t) 3.66 3.99 3.95 3.97 3.96 4.39 4.30

Ore Feed Ag Head Grade (g/t) 16.83 14.51 13.06 14.86 16.25 16.98 15.75

Ore Feed Contained Au (oz) 33,668 56,077 62,288 67,074 62,579 54,875 62,529

Ore Feed Contained Ag (oz) 154,890 204,032 205,843 250,803 256,794 212,132 228,933

Gold Recovery (%) 87% 87% 87% 87% 87% 87% 87%

Silver Recovery (%) 33% 33% 33% 33% 33% 33% 33%

Recovered Gold (oz) 29,338 48,865 54,277 58,448 54,531 47,817 54,487

Recovered Silver (oz) 51,450 67,774 68,375 83,310 85,300 70,465 76,045

MDZ

Ore Feed (t) - - - 209,014 1,310,667 1,461,570 1,460,649

Ore Feed Au Head Grade (g/t) - - - 3.10 3.15 3.27 3.41

Ore Feed Ag Head Grade (g/t) - - - 4.65 4.43 4.47 4.85

Ore Feed Contained Au (oz) - - - 20,831 132,913 153,659 160,137

Ore Feed Contained Ag (oz) - - - 31,272 186,520 210,048 227,761

Gold Recovery (%) 0% 0% 0% 95% 95% 95% 95%

Silver Recovery (%) 0% 0% 0% 40% 40% 40% 40%

Recovered Gold (oz) - - - 19,789 126,268 145,976 152,130

Recovered Silver (oz) - - - 12,509 74,608 84,019 91,104

Total

Ore Feed (t) 286,171 437,485 490,050 734,015 1,802,201 1,850,108 1,912,679

Ore Feed Au Head Grade (g/t) 3.66 3.99 3.95 3.72 3.37 3.51 3.62

Ore Feed Ag Head Grade (g/t) 16.83 14.51 13.06 11.95 7.65 7.10 7.43



Gold Recovery (%) 87% 87% 87% 89% 92% 93% 93%

Silver Recovery (%) 33% 33% 33% 34% 36% 37% 37%



Source: SRK, 2020



Table 22-8: Marmato Mill Production Schedule (2027 - 2033)

2027 2028 2029 2030 2031 2032 2033

Marmato UZ

Ore Feed (t) 409,596 351,864 386,605 388,892 445,184 91,713 -

Ore Feed Au Head Grade (g/t)

4.34 4.37 4.30 4.18 4.43 4.22 -

Ore Feed Ag Head Grade (g/t)

15.23 14.42 14.33 14.43 17.43 21.69 -

Ore Feed Contained Au (oz) 57,171 49,475 53,418 52,299 63,442 12,443 -

Ore Feed Contained Ag (oz) 200,574 163,168 178,148 180,463 249,477 63,956 -

Gold Recovery (%) 87% 87% 87% 87% 87% 87% 0%

Silver Recovery (%) 33% 33% 33% 33% 33% 33% 0%

Recovered Gold (oz) 49,818 43,112 46,548 45,573 55,282 10,843 -

Recovered Silver (oz) 66,625 54,200 59,176 59,945 82,869 21,244 -

MDZ

Ore Feed (t) 1,461,12

2 1,465,17

8 1,461,09

9 1,460,38

2 1,460,04

1 1,463,85

0 1,342,37

4 Ore Feed Au Head Grade (g/t)

2.94 2.77 2.47 2.33 2.50 2.79 2.84


4.54 4.04 2.87 3.11 3.01 3.22 3.85



Gold Recovery (%) 95% 95% 95% 95% 95% 95% 95%

Silver Recovery (%) 40% 40% 40% 40% 40% 40% 40%



Total

Ore Feed (t) 1,870,71

8 1,817,04

2 1,847,70

4 1,849,27

4 1,905,22

5 1,555,56

3 1,342,37

4 Ore Feed Au Head Grade (g/t)

3.25 3.08 2.85 2.72 2.95 2.87 2.84


6.88 6.05 5.27 5.49 6.38 4.31 3.85



Gold Recovery (%) 93% 93% 93% 92% 92% 94% 95%

Silver Recovery (%) 37% 37% 36% 36% 36% 38% 40%



Source: SRK, 2020

The Marmato UZ presents a steady ramp-up in throughput capacity between 2020 and 2023. However,

after reaching its maximum capacity in 2023 the schedule does not use the full plant capacity in the

subsequent years. The reduction in plant throughput is related to a maximum total mine movement

permit that will only allow CGM to mine a total of 2 Mt/y of combined ore and waste. Gold and silver

production present a slight decrease.

The mineral processing facility supporting MDZ ramps up between 2023 and 2025. Once mineral

processing production ramps-up, the facility is run at a stable capacity until the end of mine life. Gold

and silver production decline slightly due to declining gold and silver grades.

The resulting combined mineral production profile presents a modest increase of plant capacity

between 2020 and 2023 and a large expansion in plant capacity once the MDZ ramps-up in 2024.

MDZ silver and gold production increases significantly in 2024 and then gradually decreases over the

remaining mine life.



Figure 22-4 and Figure 22-5 show the production profiles for the two mineral processing plants.

Figure 22-6 consolidates the two mineral processing operations into a single production profile.

Source: SRK, 2020

Figure 22-4: Marmato UZ Processing Production Profile

Source: SRK, 2020

Figure 22-5: MDZ Processing Production Profile



Source: SRK, 2020

Figure 22-6: Marmato Processing Production Profile

22.3 Taxes, Royalties and Other Interests

The analysis of the Marmato Project includes an effective corporate income tax rate of 30%. A

depreciation schedule was calculated by SRK assuming a 10 year straight-line depreciation.

Royalties are also deductible from taxable income. The Project includes payment of governmental

production royalties on both gold and silver sales as outlined in Section 4.3. The total royalty due,

excluding the related party royalty to Croesus, is calculated as 9.2% of gross metal sales deducted by

the costs of transportation and metal refining.

22.4 Results

SRK evaluated the Marmato Operation’s cash flow with three separate cash flows. One for each major

mining area, Marmato UZ and MDZ and one for the entirety of the Marmato Operation combining both

mining areas. This was done to confirm that the mineral reserves for each area are economic on a

stand-alone basis as well as on a combined basis.

The Marmato UZ economic modeling presents a healthy cash flow that doesn’t include any negative

periods. The first two years are close to break even, which is related to the investment to expand the

mineral processing plant and the development in waste to prepare the mining areas. The Marmato UZ

is projected to produce a pre-tax cash flow of US$319.8M over its life. Table 22-9 and Figure 22-7

present the LoM cash flow metrics and the UZ mining area’s cash flow profile respectively.



Table 22-9: Marmato UZ LoM Cash Flow Metrics

LoM

Marmato UZ

Gold Revenue (US$) 838,514,585

Silver Revenue (US$) 14,395,253

Doré Refining Charges (US$) (3,821,231)

Net Revenue 849,088,606

Mining Operating Costs (249,251,102)

Processing Operating Costs (71,282,947)

Other Operating Costs (71,086,437)

Total Operating Costs (391,620,486)

Royalties (78,116,152)

Sustaining Capital (59,546,192)

Working Capital -

Pre-Tax Cash Flow 319,805,777

Source: SRK, 2020

Source: SRK, 2020

Figure 22-7: Marmato UZ Cash Flow Profile

The cash flow prepared for the MDZ indicates that this mining area also projects good profitability. The

MDZ requires significant capital investment between the years of 2021 and 2023. Commercial

production starts in 2024 and this year already shows a positive cash flow, all subsequent periods also

present positive cash flows. Payback is projected to happen sometime in the year 2027 and the area

is projected to generate a pre-tax cash flow of US$381.4M through its life. Table 22-10 and Figure 22-8

present the LoM cash flow metrics and the MDZ’s cash flow profile respectively.



Table 22-10: MDZ LoM Cash Flow Metrics

LoM

MDZ

Gold Revenue (US$) 1,772,617,350

Silver Revenue (US$) 12,233,353

Doré Refining Charges (US$) (8,078,070)

Net Revenue 1,776,772,632

Mining Operating Costs (512,288,429)

Processing Operating Costs (199,113,126)

Other Operating Costs (119,771,142)

Total Operating Costs (831,172,697)

Royalties (163,463,082)

Initial Capital (269,394,005)

Sustaining Capital (131,299,895)

Working Capital -

Pre-Tax Cash Flow 381,442,953

Source: SRK, 2020

Source: SRK, 2020

Figure 22-8: MDZ Cash Flow Profile

When both mining areas are combined to calculate the operation’s cash flow the results indicate an

after-tax IRR of 19.5% and an after-tax NPV of approximately US$256.1 million, based on a 5%

discount rate and gold and silver prices of US$1,400/oz and US$17.00/oz respectively. The cash flow

profile also shows a shorter payback for the investment required by the MDZ, bringing it back about a

year to 2026. The operation is projected to have negative cash flows between the years 2020 and

2023, when the MDZ is installed, with payback for the expansion expected by 2026. The annual free

cash flow profile of the Project is presented in Figure 22-9. The full annual technical economic model

(TEM) is in Appendix D.



Source: SRK, 2020

Figure 22-9: Marmato After-Tax Free Cash Flow, Capital and Metal Production

Indicative economic results are presented in Table 22-11. The Project can be considered a gold

operation with a sub-product of silver, where gold represents 99% of the total projected revenue and

silver the remaining 1%. The underground mining cost is the heaviest burden on the operation

representing 62% of the operating cost, as presented in Figure 22-10.

Source: SRK, 2020

Figure 22-10: Marmato Operating Cost Break-Down (Combined UZ and MDZ)



Table 22-11: Marmato Indicative Economic Results (Combined UZ and MDZ)





G&A Cost USD (190,857,579)






Before Tax


NPV @ 5% USD 396,654,830

NPV @ 8% USD 279,571,263

NPV @ 10% USD 219,652,793

IRR % 26%

After Tax


NPV @ 5% USD 256,075,253

NPV @ 8% USD 167,009,205

NPV @ 10% USD 121,855,455

IRR % 19.5%

Payback Year 2026

Source: SRK, 2020

Table 22-12 shows annual production and revenue forecasts for the life of the Project. All production

forecasts, material grades, plant recoveries and other productivity measures were developed by SRK

and CGM.

Table 22-12: Marmato LoM Annual Production and Revenues

Year Ore Tonnes

(t) Au Head

Grade (g/t) Ag Head

Grade (g/t) Recovered Gold

(oz) Recovered Silver

(oz) Pre-Tax FCF

(US$) After-Tax FCF

(US$)

2020 286,171 3.66 16.83 29,338 51,450 (3,281,096) (3,281,096)

2021 437,485 3.99 14.51 48,865 67,774 (98,299,718) (102,103,013)

2022 490,050 3.95 13.06 54,277 68,375 (92,164,816) (99,585,569)

2023 734,015 3.72 11.95 78,237 95,819 (42,309,756) (50,319,350)

2024 1,802,201 3.37 7.65 180,798 159,908 89,076,640 84,113,531

2025 1,850,108 3.51 7.10 193,793 154,484 126,352,678 102,380,468

2026 1,912,679 3.62 7.43 206,617 167,150 137,823,352 107,759,629

2027 1,870,718 3.25 6.88 181,023 151,934 95,751,085 61,799,556

2028 1,817,042 3.08 6.05 167,073 130,324 94,527,940 70,071,282

2029 1,847,704 2.85 5.27 156,776 113,104 71,792,907 52,619,644

2030 1,849,274 2.72 5.49 149,502 118,354 70,195,721 55,471,500

2031 1,905,225 2.95 6.38 166,768 139,387 87,200,108 75,778,198

2032 1,555,563 2.87 4.31 135,586 81,863 78,614,272 60,602,811

2033 1,342,374 2.84 3.85 116,441 66,464 85,969,414 76,398,172

2034 - - - - - - (831,650)

Total 19,700,609 3.19 6.87 1,865,094 1,566,389 701,248,730 490,874,111

Source: SRK, 2020









G&A USD/Au-oz 102





AISC USD/Au-oz 880


22.5 Sensitivity Analysis

A sensitivity analysis on variation of Project costs, both capital and operating, and metal prices

indicated that cash generation is most sensitive to reduction in metal prices, or possibly loss on metal

recovery, and secondly to an increase in operating costs.

Source: SRK, 2020

Figure 22-11: Marmato NPV Sensitivity



23 Adjacent Properties SRK highlights to the reader as discussed in Section 14.12 that all Mineral Resources within

#CHG_081 (yellow) and upper areas of #RPP_357 (above 1,300 m MSCW) have been excluded from

the Mineral Resource statement as they were not transferred to CGM and therefore are excluded from

the Mineral Resources. These Mineral Resources are currently held by Gran Colombia which is listed

on the TSX.



24 Other Relevant Data and Information

24.1 Project Execution Plan

This PFS project execution plan focuses on the construction of the MDZ Project. It is not intended to

discuss the UZ project ongoing operations. The development of the MDZ Project will include all

activities to support the permitting, design, procurement, and construction of the MDZ project. The

project will include a new MDZ site access, plant site and processing facility, a new MDZ mine access

to a new mining area below the existing UZ mine, MDZ tailings storage facilities, and support facilities

including a new camp and administrative area. Caldas has engaged Pathfinder, LLC to develop the

detailed project execution plan.

24.1.1 Project Objectives

The new MDZ facilities will meet all Colombian federal, provincial, state, and district design,

construction, and operating requirements, and be compliant with the Company's design standards.

The Project objectives in order of priority are:

• Achieve the highest safety, health, and environmental standards during construction and start-

up with no fatalities, no lost-time injuries, and a total recordable incident rate (TRIR) that is

less than 0.20

• Minimize the total Project investment cost

• Achieve high operational reliability for the overall facility

• Achieve the Project schedule start-up target of 2nd quarter 2023

• Apply Project development and execution best practices to achieve the Project objectives

24.1.2 General Project Description

The MDZ project is envisioned as an underground gold operation accessed by decline with

underground crushing and conveyance to the surface that will mine and process 1.46 Mt/a of ore

through a carbon-in-pulp plant to produce doré bars on-site. Tailings are disposed of in a new dry stack

facility near the process plant. Project life is currently expected to be slightly over 10 years.

The general layout of the project is presented in Figure 18-2. The new mine facilities are described in

detail in Sections 16.5, 17.3, and 18.3-18.15.

The overall MDZ Mine Project scope is generally described by the summary below:

• Exploration Drilling

• Mine Development

• Mining Equipment

• Surface Facilities and Equipment (including the new Standard Processing Facility - MDZP)

• Underground Facilities and Equipment

• Power Supply

• Access Road

• Camp

• Other Supporting Infrastructure



• Mineral Processing Plant

• DSTF

The Marmato Process Plant (MDZP), 4,000 t/d, is standard to the industry and includes:

• Primary crushing

• SAG/ball mill grinding


• CIP cyanidation of the gravity tailings

• Gold elution and electrowinning

• Intensive cyanide leaching of the gravity concentrate followed by electrowinning

• Detoxification of the cyanide leach residues

• Smelting of cathode precipitates to produce final doré products

The MDZ site infrastructure includes:

• The new access road to the MDZ plant site

• The security access point at the main road

• Parking and administrative (office) area

• Plant area with crusher, crushed ore storage, reagent storage, and processing plant

• Mine truck shop, fueling system, and change house with ROM pile

• The new portal, mine waste rock storage, and mine access decline

• Paste backfill plant

• Power substation and distribution system

• Water supply from Cauca River supplementing mine dewatering and tailings water recycle

• Pump station and pipeline for tailings management

• DSTF and press filter system

24.1.3 Site Preparation and Infrastructure

Site Preparation

The project is a greenfield development and will require full development access and earthworks for

the processing site, portal location, and administrative camp area. Site prep will include clearing and

grubbing, site grading, pad preparation, and access road construction. Topsoil will be segregated and

stockpiled for closure. The site will be fenced and controlled access through a guard gate will be

established. Early works include providing access to the portal location so that the portal can be

prepared for construction of the decline to the MDZ. The process plant area will be graded to five

cascading benches following the natural topography as described in section 8.2:

• Bench 1: Filtration plant

• Bench 2: Process plant

• Bench 3: Reclaim tunnel and future pebble crusher

• Bench 4: Secondary Crusher

• Bench 5: Mine Portal

The pads will be mainly cut in west of each pad and filled in east of each pad. The transition between

each pad will be constructed of 1V:1.5H sloped grade, soil nailed stabilized slopes and Mechanically



Stabilized Earth (MSE) retaining wall as required, following the required footprint for process plant

layout.

The pads are accessed via in plant roads at the south edge of the process plant pads. The in-plant

roads have a slope of 10% to 14% max.

Additionally, the DSTF construction prep work and initial cell construction will be conducted with the

process site preparation.

Offsite Infrastructure

The project will include a new overhead 115kV powerline to be installed by a third party from the

Salamina substation. The powerline will be financed by the third party and costs recovered through

fees during operations.

A new pumping station near the Cauca River is being designed by the Owner team and will be

constructed to provide water to both the existing mine/plant (UZ) and the MDZ project.

Onsite Facilities and Infrastructure

The site will be fenced and controlled access through a guard gatehouse with a weigh-scale will be

established. Temporary power will be provided by diesel generator supplied by contractors until line

power is available.

A second new overhead 115kV powerline will be constructed from the local utility substation to a new

MDZ substation that will be constructed at the process plant site. The MDZ substation will feed the

mine portal substation and the underground. The substation will also feed the process plant facilities

as well as the new administrative and camp area. Construction power will be provided by the individual

contractors until line power is available.

The mine pad at the portal will also provide staging for construction of the portal and decline.

Temporary fans will be installed for the mine development. Two second accesses for ventilation will

be developed near the UZ plant site and will be developed at the same time as the MDZ decline.

A pad will also be developed near the portal pad to provide future siting of the underground backfill

plant and shotcrete plant. This location could be considered for cement/concrete batch plan during the

construction development.

Water will be supplied by truck during the construction phase. A permanent feedwater tank will be

constructed at the paste backfill pad to support the mine and backfill plant. The final water supply will

be from the new water pumping station to a water tank at the process plant that will then feed the paste

plant/mine water tank.

The tailings area will be constructed using waste rock from the mine portal development and will be

concurrent with decline development.

The administrative area development will include an access road, administrative buildings and camp.

The administrative area will provide a location for staging and will provide housing during the

construction effort.

Other facilities are included in the plant discussion in Section 17 of this report.



24.1.4 Underground Mine and Supporting Infrastructure

Once the mine bench (Bench 5) portal area is established, the MDZ decline will be developed. Mine

access will occur via a single main decline ramp that houses the conveyor from the underground

crusher to the plant. The portal is located next to the plant site, and the main conveyor decline

intersects the orebody at the midpoint of the expected mining area, dividing the orebody into an upper

and lower block. Maintenance shops, the paste backfill plant booster pump area, miscellaneous

underground support facilities, and the crusher cavity will be located on or near this level.

Two ventilation declines as the main mine access will also be constructed during the same period.

These declines are located near the existing Marmato mine.

The stope accesses are connected to a level access, which is offset approximately 20 m away from

the ends of the stopes. Each stope access typically connects to the level access, except in cases

where stopes are small and long development is required to reach the stope. In those instances, a

connection from an adjacent stope is included in the design. The level accesses connect to the main

ramp, which is offset at least 75 m from stoping into the footwall. On the northeast side of each level,

the level access connects to an intake air ventilation raise. On the southeast side, the level access

connects to an exhaust air raise.

The construction effort will build out the underground infrastructure and tie the ventilation drifts to the

MDZ mine workings. The effort will include installation of services for power, water, paste backfill,

communications, and ventilation.

24.1.5 Process Plant

The processing facilities will be constructed on multiple pads downgrade from the portal location. The

proposed flow sheet uses standard processes for:


• Gravity/Leach/Adsorption


• Cyanide Detoxification

• Tailings Thickening/Filtration

The crusher will be constructed underground and will feed a conveyor that will move the RoM material

to the surface. At the surface the material is deposited into a surge bin with feeders that feed the

secondary cone crusher. The secondary crusher discharge is conveyed to the 24-hour open air

stockpile. Crushed ore from the stockpile will feed the SAG mill via conveyor. The SAG mill material

discharges to a cyclone feed hopper and is then fed to a ball mill. The ball mill discharges to a trommel

screen that allows the undersize to feed the classification cyclone feed pumps. Cyclone underflow

slurry, after screening, is then gravity fed to an aeriation tank prior the leach circuits. A portion of the

cyclone underflow reports to a gravity concentrate and intensive leach circuit. The pregnant leach

solution is sent to the gold room for electrowinning and smelting. The other portion of the underflow

reports to the leach and adsorption circuit that then discharges to the CIP tanks. The slurry is eventually

fed to the cyanide detoxification tanks. A carbon acid wash, elution, and the regeneration circuit

remove the gold from the carbon and regenerates the barren carbon. Gold is recovered in the

electrowinning room and then smelted to produce doré.



Tailings material is processed through a tailings thickener where the underflow is moved by pump to

a filter plant where the tailings paste is filtered in two horizontal plate and frame pressure filters to

product a cake for deposit in the dry stack tailings facility. Slurry tailings also can be pumped to the

paste backfill plant and combined with cement to create backfill used in the underground mine. Filtered

tailings are removed from the loadout bunker by loader to trucks that place the material in the DSTF.

Additional facilities at the plant site include:

• Reagent storage area

• Compressed air system

• Diesel storage and fueling facility

• Propane storage

• Water systems including fresh water, process water, gland water, and firewater systems

• Maintenance shops for the surface and underground equipment

• Laboratory

• Water treatment facilities and water collection ponds

• Truck wash facility

• Explosives storage

• Mill administrative office and first aid station

• Warehouse and storage yard

• Wastewater treatment plan

24.1.6 Project Delivery Approach

EPCM/EPC Approach

The Project will be delivered through the CGM Owner’s team and an EPCM contractor that will manage

the development of a final design exercise that will develop a final cost estimate and schedule that will

lead to an EPC construction phase.

The Owner’s team will manage the offsite third-party transmission line and the water supply system

scopes.

The EPCM contractor will provide a FEL3/FEED to prepare a +/-10% cost estimate and a level 3

resource loaded project schedule for an EPC construction phase. The EPCM project will include the

mine, mill, and associated infrastructure including the DSTF.

The EPCM contractor will further develop the following: detailed design deliverables; procurement

activities; subcontractor identification and selection with Owner; develop construction contracts,

request for proposals, contract with qualified bidders, conduct bid evaluations, negotiate contracts,

and recommend contractors/contracts for Owner’s review and approval.

The EPCM contractor will also: oversee, manage, coordinate, and report on all supervised contractors

throughout the EPC phase. The EPCM contractor will also support the Owner in determining whether

each contractor is performing in accordance with the terms of its respective contract, and work to guard

the Company against defects and deficiencies in the work performed.

The EPCM contractor will be responsible for placing all material/equipment orders and management

for those third parties engaged during the EP stage of the project. As described above, the EPCM

contractor will lead the contracting program to screen construction (supervised) contractors, evaluate



and condition bids, and prepare an award recommendation for the Company. However, the

construction (supervised) contract(s) will be placed on the Company's paper, in Company's name, and

paid directly by the Company, after the EPCM contractor's review/verification/endorsement of the

construction (supervised) contractor's invoices. The EPCM contractor will be responsible to manage

and coordinate the construction (supervised) contractor on the Company's behalf through Project

completion and facility start-up.

Project Schedule

The Project completion date is projected to be in the second quarter of 2023. The high level schedule

is shown in Figure 24-1.

Source: CGM, 2020

Figure 24-1: Project Execution Schedule

WBS Structure

A work breakdown structure (WBS) was developed by CGM and is used for the project. The WBS is

summarized at the highest level in Table 24-1.



Table 24-1: Project Work Breakdown Structure – Level 1 and Level 2

WBS Description

0000 Mine & Infrastructure

0100 Exploration - Capitalized

0200 Above-ground Mining / Geology

0300 On-Site Infrastructure - Plant

0400 Off-Site Infrastructure - Excludes Plant & Mine

0500 Heap Leach

0600 Underground Mining / Geology

1000 Process Plant

1100 Crushing / Conveying

1200 Milling

1300 Flotation / Concentrate Handling

1400 Leach / Ore Separation / Sulfide Ore Processing

1500 CN Destruction

1600 Recovery

1700 Solvent Extraction / Electrowinning

1800 Reagents

1900 Plant Services

2000 Residue Management/Tailings

3000 Power

4000 Port

5000 Airport

6000 Water Management

7000 Indirect Cost

8000 Other Project Cost

9000 Owner's Cost

Source: CGM, 2020

24.1.7 Project Team Organization

The CGM Project Team will include approximately 50 individuals. The team will report through a Senior

Project Manager that reports directly to the President of the Company. The team has a number of

people in place today with the staffing schedule established. The team includes personnel for the key

functions of engineering, construction management, plant development, mine development,

commissioning, operational readiness, procurement, project controls, and health and safety. The

corporate team includes geology, environmental, security, community relations, finance, and IT. The

overall organization chart for CGM is presented in Figure 24-2. The EPCM team will report to the CGM

Senior Project Manager.



Source: CGM, 2020

Figure 24-2: MDZ Project Team

The Owner’s team with their EPCM supplements and target staffing dates is provided in Table 24-2.



Table 24-2: Owner’s Project Team with EPCM Supplements

Position Qty Owner Team (OT)/EPCM

Appearance Date in Org Chart

Senior Project Manager 1 OT 8/29/19

Engineer Manager 1 OT 8/29/19

Surface Infrastructure Coordinator 1 OT 6/24/20

High Voltage Coordinator - Connection 1 OT 6/24/20

Water Works Coordinator 1 OT 5/17/20

Civil Works Coordinator 1 OT 5/17/20

SME - Senior Mechanical Engineer 1 OT 5/17/20

SME - Senior Electrical Engineer 1 OT 6/24/20

SME - Senior Instrumentation and Control Engineer 1 OT 6/24/20

QA/QC Superintendent 1 OT 6/24/20

Permits 1 OT 5/17/20

Junior Engineer 2 OT 6/24/20

Construction Manager 1 OT 6/24/20

Plant Superintendent 1 OT 8/29/19

Bulk Earthworks/Tailings 1 EPCM 6/24/20

Mechanical Engineer 1 EPCM 6/24/20

Electrical Engineer 1 EPCM 6/24/20

Instrumentation and Control Engineer 1 EPCM 6/24/20

Civil Works Coordinator 1 EPCM 6/24/20

Plant Manager 1 OT 8/29/19

Metallurgical Superintendent 1 OT 6/24/20

Structural Mechanical Piping 1 OT 6/24/20

Mining Manager 1 OT 8/29/19

Horizontal Development Coordinator 1 OT 8/29/19

Vertical Development Coordinator 1 OT 6/24/20

Underground Infrastructure Coordinator 1 OT 8/29/19

Mechanical Engineer (Conveyor) 1 OT 6/24/20

Electrical Engineer (Substation) 1 OT 6/24/20

Commissioning Manager 1 EPCM 6/24/20

Readiness Manager 1 OT 6/24/20

Procurement Manager 1 OT 8/29/19

Purchases 2 OT 5/17/20

Site Administrator 1 OT 6/24/20

Logistics 1 OT 8/29/19

Project Controls Manager 1 OT 8/29/19

Document Controls 1 OT 8/29/19

Scheduler 1 OT 8/29/19

Costs Control 1 OT 8/29/19

Estimator 1 OT 5/17/20

Contracts 1 OT 8/29/19

Geology/Geotech 1 OT 8/29/19

Environmental 1 OT 8/29/19

OH&S 1 OT 8/29/19

Finance 1 OT 8/29/19

Community Relations 1 OT 8/29/19

Security Superintendent 1 OT 6/24/20

Security Coordinator 1 OT 6/24/20

Labor Total 49

Source: CGM, 2020



24.1.8 Project Execution Supporting Plans

As the project is advanced, the EPCM and Owner’s team will further develop plans, consistent with

CGM existing corporate structure and organization for the overall control and management of the

project that will include:

• Project Governance

• Health, Safety and Environment

• Social Responsibility

• Permitting and Regulatory

• Project Management

• Quality Management

• Constructability Management

• Human Resources Management

• Project Controls

• Document Controls and Data Management

• Engineering

• Supply Chain Management

• Construction Management

• Operational Readiness

• Commissioning

• Ramp-Up

• Project Closeout



25 Interpretation and Conclusions

25.1 Property Description and Ownership

SRK notes within the transfer of licenses from the previous owner, a gap between the existing licenses

for #014-89m and #RPP_057 was identified and CGM applied to the Colombian government for formal

approval to continue mining in the identified gap. SRK has reviewed the application within the

government website and notes that the status is defined as “in progress”, which has been the reported

status since September 30, 2009. The Company has been taking steps to get the approval finalized.

SRK understands that at the time of writing the issue has been resolved, with the government

determining that there is no gap and that the area falls within the license for Zona Baja (#014-89m).

SRK has not completed sufficient work to confirm this but highlights that it should be resolved and

enable additional material to be used in mine plans for future studies.

In 2017 CGM began the process and submitted to the government the application for the license

extension to the current operation and future exploration for license #014-89m, with the original license

currently held to October 2021. The process is expected to be completed in Q4 2020.

25.2 Geology and Mineralization

SRK produced an updated 3D geological model for the Marmato deposit as part of the current study.

SRK considers this to have increased the confidence in the spatial location of the various geological

units. CGM geologists as part of the on-going exploration continue to develop the geological

knowledge on the project and have supplied additional fault information which should be integrated

into further lithological models. SRK does not consider these faults to have a material impact on the

current mineral resource estimate but notes that it may impact future underground infrastructure (such

as a decline). SRK therefore recommends that the geological model be updated to reflect the impact

of these in future models and prior to construction.

25.3 Status of Exploration, Development and Operations

SRK has been supplied with electronic databases covering the sampling at the Project, all of which

have been validated by the Company. The databases comprise of a combination of historical and

recent diamond core and underground channel samples. In total, there are some 1,317 diamond

drillholes for a combined length of 266,390 m, plus 24,824 individual underground channel samples,

inclusive of current mine sampling contained in the databases. Isolated historical channel samples in

the upper portion of the mines have a degree of uncertainty on spatial location and quality as they

have not been independently verified by SRK during site visits. SRK has excluded these samples from

estimation of the porphyry units but has used them to guide the geological interpretation of the veins

at higher elevations.

Historic underground channel samples are typically taken across the width of the vein, with limited

sampling into the hanging-wall and footwall where possible. CGM geologists have made great efforts

to position the sampling as accurately as possible, but SRK has noted some limitations still exist.

These samples have been allocated separate geological codes in the modelling process as to not

influence the geological model.



SRK is of the opinion that the exploration and assay data is sufficiently reliable to support evaluation

and classification of Mineral Resources in accordance with generally accepted CIM Estimation of

Mineral Resource and Mineral Reserve Best Practices Guidelines (2014).

SRK notes that CGM exploration continues at the project throughout 2020 and SRK has reviewed the

2020/2021 drilling plan. The drilling is targeting mineralization in the hangingwall of the current

estimate (named the New Zone), which may impact on current mining infrastructure if further

mineralization is located, which may require modifications to the current mine design. SRK therefore

recommends that the geological model and mineral resource be updated to reflect the new drilling

upon completion as the impact of these in future models may impact the design prior to construction.

25.4 Mineral Processing and Metallurgical Testing

Native gold is the predominant gold carrier and over 99% of the gold particles occurred within mineral

structures that would be readily accessible by leaching solutions.

The PFS metallurgical program optimized process parameters required to recover gold and silver

values from MDZ ore using a process flowsheet that includes gravity concentration followed by

cyanidation of the gravity tailing.

Comminution tests demonstrated that the MDZ ore is classified as hard with regard to impact breakage

and grinding characteristics.

Overall gold recovery is estimated at 95% and overall silver recovery is estimated at 51%. This is very

similar to the results from the PEA metallurgical program in which gold recovery was estimated at 95%

and silver recovery was 47%. There is little difference in reported gold recoveries for the master and

variability composites and gold recovery appears to be independent of ore grade over the range tested.

Cyanide destruction tests demonstrated that weak acid dissociable cyanide (CNWAD) could be reduced

to less than 10 mg/L with the SO2/air process. However, CNWAD levels will further attenuate to less

than 1 mg/L with time.

Pressure filtration will be required to dewater thickened tailings in order to achieve less than 15%

moisture content required for disposal in a DSTF.

25.5 Mineral Resource Estimate

The mineralization occurs in parallel, sheeted and anastomosing veins (vein domain), all of which

follow a regional structural control, with minor veins forming splays of the main structures (splays)

which often have limited strike or dip extent. The vein domain intersects broader zones of intense

veinlet mineralization (termed “porphyry domain” for the purpose of this report) and is hosted by a

lower grade mineralized porphyry. In addition, a discrete, relatively high-grade core, or feeder zone, to

the main mineralization (MDZ), has been identified at depth by CGM geologists.

The lowest levels of the mine (level 21) have currently intersected a combination of the porphyry

domain which is where the gold is associated with veinlets with pyrite, and the MDZ where gold (Au)

is associated with pyrrhotite. There is a small transition between the two domains, which is observed

to some extent in the current mine workings but is not clearly defined from the current drilling.

Underground mining remains focused on the vein structures located in the central portion (Zona Baja)

of the Marmato deposit.



Stillitoe (2019) concluded the only geological parameter than can be used to constrain the grade model

is veinlet intensity, although the presence of visible native gold also acts as a useful grade indicator.

SRK has used this assumption as the basis for the mineralization model, by using an indicator gold

(1.7 g/t) grade to act as a proxy to higher grades of vein density (which have been logged consistently

in older holes in the area).


Leapfrog™Geo of wireframe models for the fault networks, veins, definition of resource domains (high-

grade sub-domains), data conditioning (compositing and capping) for statistical analysis, geostatistical

analysis, variography, block modelling and grade interpolation followed by validation. Grade estimation

for the veins has been based on block dimensions of 5 m by 10 m by 10 m for the porphyry and MDZ

units. Sub-blocking to 0.5 m by 1 m by 1 m has been allowed to reflect the narrow nature of the

geological model. The block size reflects the relatively close-spaced underground channel sampling

and spacing within veins compared to the wider drilling spacing, with the narrower block size used in

the MDZ at depth to reflect the proposed geometry of the mineralization (steeply dipping feeder zone).

SRK is of the opinion that the MRE has been conducted in a manner consistent with industry best

practices and that the data and information supporting the stated mineral resources is sufficient for

declaration of Measured, Indicated and Inferred classifications of resources. SRK considers currently

the veins (including splays) and the MDZ to be of sufficient confidence for use in the PFS but

recommends further work on the short scale variability within the porphyry be completed to confirm

the current interpretation within areas of the existing mining infrastructure prior to completing a FS.

The exclusive structural control of the MDZ orebody implies that additional examples could exist

elsewhere within the P1 stock and that they represent a priority exploration target, but further

exploration will be required to test this theory and there is no guarantee of exploration success.

25.6 Mining & Reserves

UZ Mine Design

CAF is the current mining method used for the veins and is appropriate for the deposit geometry. A

modified longhole stoping method is envisioned for the Transition zone to take advantage of the bulk

characteristics of the deposit.

Optimizations were run using a minimum cut-off of 2.23 g/t Au for the Veins and 1.91 g/t Au for the

Transition zone.

Access to the Veins is already established. The primary haulage level is 18 and material from levels

above is brought down via existing orepasses. Material below level 18 is transported up an incline or

in the apiques. The main production apique is at level 22, a secondary production apique is at level 20

and will extend down to level 22.

The Transition zone is accessed via level 21 and level 22. A ramp will also connect the two levels as

a secondary egress and ventilation exhaust.

Tonnage and grades presented in the reserve include dilution and recovery. Productivities are based

on the current mine productivities




1,500 t/d with a gradual ramp up. There is also a 2 Mt/y limit for total moved material, which will limit

the production from the UZ.

MDZ Mine Design

Longhole stoping is an appropriate mining method for the deposit geometry. Stopes are sized to be

large enough to take advantage of bulk mining methods, yet small enough to minimize dilution.



increase/decrease in CoG has a material impact on the material available for design. A minimum cut-

off of 1.61 g/t Au was used for design/reserve. Higher grade stopes using 3.5 g/t stope optimization

results were designed as a first pass, with the lower grade stopes added as separate stopes. This

allowed for scheduling of higher grade stopes first.

The MDZ is accessed through a decline drift with conveyor. Tonnage and grades presented in the

reserve include dilution and recovery and are benchmarked to other similar operations. Productivities

were generated from first principles with inputs from mining contractors, blasting suppliers, and

equipment vendors where appropriate. The productivities were also benchmarked to similar

operations. Equipment used in this study is standard equipment used world-wide with only standard

package/automation features.


4,000 t/d.

Geotechnical

From the PFS geotechnical investigation, SRK concludes:

• The geotechnical investigation, laboratory tests and design are suitable for a PFS. The

proposed design parameters are acceptable for a PFS study only and should not be fully

implemented before the FS is completed.

• The proposed stope design consists of maximum stope dimensions of 30 m high, 30 m long

and 10 m wide to maintain stability. Empirical charts suggest that the side walls are located in

unsupported transition zones, which could require some spot ground support for potential

wedge formations depending on discontinuity persistence/continuity.

• The empirical chart for estimating the open stope stand-up time was accepted for the PFS.

The results indicate that a 10 m span stope can likely be open for one to six months without

ground support.

• Dilution was estimated using the empirical Clark and Pakalnis (1997) method. The thickness

of external dilution is estimated as ELOS. The ELOS charts indicate significant dilution is

unlikely due to the good rock mass quality. Wall damage would likely be associated with

blasting overbreak. SRK recommends that CGM conduct a blasting study during the FS to

evaluate the degree of overbreak.

To estimate the backfill strength requirements, SRK applied the Mitchell et al, 1982 analytic solution.

The analytic solution results for the case when secondary stopes are open (after mucking), suggests

that:



• A backfill UCS strength of 1 MPa will be adequate to maintain backfill stability and prevent

backfill from sloughing into the open stope.

• Negligible wall sloughing is anticipated.

SRK estimated a sill pillar height equal to 9.5 m, considering an FoS of 1.5 based on the empirical

method proposed by Potvin el al.,1989. A simple 2D numerical simulation indicates that the average

distance between stopes and the crusher station (approximately 40 m) does not affect the stability of

the crusher station. At PFS level, the crusher dimensions are acceptable. However, more detailed

studies should be implemented in the FS.

In terms of the underground workshop infrastructure, the stability assessment was conducted using a

tributary area method. The method assumed that the workshop station is about 750 m deep and

requires a FoS of 1.5, which resulted in a 9 m pillar width and a 7 m pillar height.

Assuming a maximum bay width of 7.5 m, SRK anticipates needing systematic bolting of 2 m long

bolts, 25 mm in diameter and spaced 1.2 m apart. 150 by 150 mm steel welded wire mesh (5 mm

diameter) with 5 cm fiber reinforced shotcrete is also anticipated.

The decline route selection was considered a key part of the PFS during design. High-level geological,

geotechnical, hydrological, hydrogeological and structural factors were taken into consideration to

select the suitable decline route. Special attention was paid to the effect of the modeled major faults

on the drift stability. SRK considered the effect of the major fault location and the RMQ using the

following criteria:

• Reduce the exposure of the decline to major faults

• Decline trajectory should cross perpendicular to major faults

• Avoid faults shear zones

• Avoid crossing highly clayed materials

The full SRK geotechnical report (SRK, 2020) has detailed conclusions regarding the rock mass fabrics

and geotechnical domains

Hydrogeological

The 3D groundwater flow model for the Marmato project was developed, reasonably calibrated to

available measured water level and groundwater flow data, and used to make predictive simulations

of:

• Passive inflow to the existing and planned deep underground mines.

• Propagation of drawdown during proposed dewatering during mining.

• Changes in groundwater discharge to rivers and creeks during mining.

The model predicts that:


159 L/s) is expected from the upper levels above 730 m, where elevated hydraulic conductivity

values of bedrock groundwater system were measured.

• Mine inflow to the MDZ planned mine below 730 m is predicted to be lower (15 L/s with upper

limit of 34 L/s) due to reduced measured hydraulic conductivity with depth.

• The total maximum planned mine discharge is predicted to be up to 88 L/s, with a possible

range from 61 to 167 L/s.




is predicted to be up to 111 L/s, with a possible range from 89 to 168 L/s.

• Major sources of mine inflow are depletion of groundwater storage and capturing of

groundwater discharge to surface water bodies (i.e., streams). The model does not predict




• Lowering of the water table in the mine area of up to 140 m and drawdown propagation of up

to 2 km away from the mine, assuming a 10-m drawdown extent

In SRK’s opinion, the completed predictions are conservative, given the following:

• The model is based on extrapolation of the measured hydraulic conductivity values in mine

area for entire model domain, including topographic highs areas outside of the mine area,

where measured water levels are high and hydraulic conductivity values are most likely lower

than in the mine area.

• The model uses of high recharge from precipitation to calibrate the model to measured water

levels, combined with geomean hydraulic conductivity values in discrete depth intervals that

are derived from measured hydraulic conductivity values in the mine area.

• The model uses calibrated conductance values that reproduce measured inflow to the existing,

relatively shallow mine for simulation of groundwater inflow to the deep underground

developments of the planned mine.

• The model simulates no restriction of groundwater inflow to the backfilled stopes for Base

Case and Maximum Inflow scenarios.


indicate that several uncertainties remain in understanding of the hydrogeology, including

hydrogeological role of the faults, hydraulic properties of bedrock outside of the mine area, recharge

estimates, spatial and vertical distribution of groundwater inflow to the current mine, water table

elevation, and water level changes due passive mine dewatering and seasonal changes in

precipitation.

To reduce these uncertainties, SRK recommends CGM complete the following additional

hydrogeological investigations/analyses for the FS:











o Drawdown propagation as a result of mine dewatering

o Seasonal variation as a result of precipitation










predictability:





25.7 Recovery Methods

An ore processing plant has been designed to process MDZ ore at the rate of 4,000 t/d using

conventional processes that are standard to the industry including: primary and secondary crushing,

SAG/ball mill grinding, gravity concentration, agitated cyanide leaching, carbon-in-pulp (CIP), gold

elution, electrowinning and smelting to produce a final doré product.

25.8 Project Infrastructure

The existing infrastructure for the UZ project is established and meets the operational requirements.

The addition of the water supply pumping system from the Cauca River will address potential water

sourcing issues during drought seasons.

The new MDZ infrastructure includes the required access, power supply, water supply, tailings storage,

and support facilities to support the production of 4,000 t/d from the new plant and mine.

A full understanding of the mine water and DSTF water requirements and runoff will allow for

optimization of the site runoff pond and water treatment capacities.

25.8.1 Water Supply

The water balance indicates that adequate water supply is available from the underground dewatering

flows and additional water supply will be available from contact water runoff and seepage flows from

the planned DSTF. Additionally, short term pumping from the UZ is being implemented to support

current mine activities. During development of the MDZ, contact water supply from the planned DSTF

will be erratic and uncertain and should not be considered a steady water supply, and the underground

dewatering flows represent a single source for a reliable water supply, thus a backup supply is

recommended. CGM is currently designing a new Cauca River pumping system to provide separate

water supplies for both the existing UZ and the MDZ. This system will provide a redundant water supply

for both the UZ and MDZ in addition to any water that will be available from existing DSTF or planned

DSTF recovery.

The water balance of the Project indicated that both contact water from the planned DSTF and

underground dewatering flows are likely to exceed the makeup requirements at the processing plant

at certain times in the LoM. Discharges from both water sources are expected and should be



addressed with appropriate discharge structures as part of general infrastructure, including a water

treatment plant expected to be completed in 2021/2022. Additionally, discharges of contact water

should be monitored to ensure environmental compliance is maintained.

25.8.2 Tailings Management Facility

SRK advanced the conceptual designs of DSTF 2 and DSTF 1 to a level sufficient for cost estimating.

The designs include consideration of the following specific elements:

• Subgrade preparation include topsoil salvaging, removal of unsuitable material and excavation

of stability benches and embankment keys

• Construction of rockfill starter embankments using a combination of imported and on-site

borrow

• Construction of underdrain network and underdrain flow management

• Construction of seepage collection drains on dry stack benches and seepage management

systems

• Construction of stormwater diversion and control channels

• Management of contact stormwater on dry stack top deck and return to process

• Access and haul roads between plant and DSTF 2 and DSTF 1

• Temporary storage area for filtered tailings

• Temporary holding pond for non-filtered tailings

• Topsoil and unsuitable soil stockpile area with underdrainage system

Currently identified risks and opportunities with respect to the costs developed for the PFS have been

identified in relation to the following:

• The inability to characterize the foundation conditions beneath the conceptual DSTF

footprints.

• Ongoing geochemical characterization of both waste rock and ore/tailings indicating some of

the waste rock and tailings may be acid generating and therefore require special management

considerations.

• Immediate characterization and analysis of Cascabels 1 and 2 to demonstrate compliance

with internationally accepted standards of practice and provide for tailings management

through commissioning of a new DSTF.

• More extensive testing of tailings to confirm tailings geotechnical characteristics and cement

addition requirements.

25.9 Environmental Studies and Permitting

The following interpretations and conclusions have been drawn with respect to the currently available

information provided for the Marmato Project:

• Environmental Studies: Baseline studies have been completed or are currently underway

with respect to the existing facilities (additional tailings storage capacity request) and MDZ

proposed expansion. These resource studies will be used for impact analysis and the

development of mitigation actions and environmental management planning.



• Environmental and Social Management: Environmental and social issues are currently

managed in accordance with the approved PMA and will likely need to be updated and/or

modified for the proposed MDZ expansion project.

• Monitoring: Routine monitoring is currently conducted on seven domestic wastewater

discharges and three non-domestic (industrial) wastewater discharges. Air quality emissions

from the metallurgical laboratory and smelter are also monitored for: particulate matter (PM),

sulphur dioxide (SO2) nitrogen oxides (NOx) and lead (Pb). The tailings are infrequently

monitored for hazard classification purposes through a Corrosive, Reactive, Explosive, Toxic,

Inflammable, Pathogen [biological] (CRETIP) program. The results of the monitoring are

provided to Corpocaldas. This monitoring program will require significant modification to

include the facilities for the proposed MDZ expansion project, and to bring it up to international

best practice standards.

• Geochemistry: Acid-generating sulfide minerals identified in the deposit include pyrite,

arsenopyrite, iron-bearing sphalerite, pyrrhotite, and chalcopyrite (SRK, 2017). Samples of

groundwater discharging into the underground are predominantly acidic. The underground

water samples contain elevated metal(loid) concentrations. While the tailings will be

discharged with a neutral to alkaline supernatant, the tailings themselves will be PAG with the

potential to eventually overwhelm the alkaline supernatant and produce acid drainage in the

long term. SRK’s waste rock characterization program is in progress and will be reported in a

separate report. A waste rock analytical program completed in 2012 in support of an open pit

mine design indicated that a significant fraction of waste rock could be potentially acid

generating (KP, 2012).

• Permitting: Operations are permitted through the posting of an Environmental Management

Plan (PMA) and secondary permits for use of water abstraction, forest use, air emissions,

discharges and river course (channel) construction. The PMA for the current operations was

originally approved in 2001. Minor modification of the PMA (including and environmental

impact analysis) is currently underway as part of the request for additional tailings storage

areas. Major modification of the PMA will be required for the MDZ expansion project.

• Stakeholder Engagement: CGM has conducted extensive stakeholder identification and

analysis programs and has set stakeholder engagement objectives and goals to develop

communications plans with government, community, media and small miners but the

Company does not currently have a formal stakeholder engagement plan.

• Closure Costs: The reclamation and closure cost estimate provided for the current operations

is approximately US$6.1 million, though there is considerable uncertainty surrounding the

basis for this estimate. An additional US$3.0 million is estimated for the MDZ expansion

facilities (assuming concurrent tailings reclamation), for a total of US$9.1 million. A

requirement for long-term post-closure water treatment, if any, could significantly increase this

estimate.

There do not appear to be any other known environmental issues that could materially impact CGM’s

ability to conduct mining and milling activities at the site. Preliminary mitigation strategies have been

developed to reduce environmental impacts to meet regulatory requirements and the conditions of the

PMA.



25.10 Capital and Operating Costs

Marmato UZ is a currently operating underground mine, and the estimate of capital includes some



projected production schedule. The estimate prepared for this study indicates that the Project requires

sustaining capital of US$59.5 million to support the projected production schedule throughout the LoM.



2021 and 2023. The cost estimate indicates that the expansion will require an investment of US$269.4

million, this includes an estimated capital of US$237.2 million plus 13.6% contingency of US$32.2

million.

Ausenco prepared a detailed cost estimate for MDZ mineral processing facility and other mine

infrastructure but did not prepare an expenditure estimate for this capital.

SRK, Ausenco and CGM prepared the estimate of operating costs for the PFS’s production schedule.

The estimated operating cost for Marmato UZ is US$76.12/t-ore and for MDZ is US$57.10/t-ore







G&A USD/Au-oz 102





AISC USD/Au-oz 880


25.11 Economic Analysis


after-tax NPV of approximately US$256.1 million, based on a 5% discount rate and gold and silver

prices of US$1,400/oz and US$17.00/oz respectively. The cash flow profile also shows a shorter

payback when comparing to a stand-alone MDZ operation to the combined operations, which present

a payback within the year of 2026, while a stand-alone MDZ operations would present a payback in

the year of 2027. The operation is projected to have negative cash flows between 2020 and 2023,

when the MDZ is installed, with payback for the expansion expected by 2026. LoM is projected to end

in 2033 resulting in a total production of 1.87 Moz of gold and 1.57 Moz of silver in the form of doré

bars containing both precious metals. Indicative economic results are presented in Table 25-2.








G&A Cost USD (190,857,579)






Before Tax


NPV @ 5% USD 396,654,830

NPV @ 8% USD 279,571,263

NPV @ 10% USD 219,652,793

IRR % 26%

After Tax


NPV @ 5% USD 256,075,253

NPV @ 8% USD 167,009,205

NPV @ 10% USD 121,855,455

IRR % 19.5%

Payback Year 2026

Source: SRK, 2020

The Project is a gold operation with a sub-product of silver, where gold represents 99% of the total

projected revenue and silver the remaining 1%. The underground mining cost is the heaviest burden

on the operation representing 62% of the operating cost, while processing costs represent 22% and

G&A costs the remaining 16%.



26 Recommendations

26.1 Recommended Work Programs

26.1.1 Property Description and Ownership

CGM has two key aspects of the property ownership that are currently in progress. CGM has submitted

the relevant documentation to the Colombian Government to extend the current 30-year term on the

#014-089m (Zona Baja) license.

Follow-up on the gap highlighted between the licenses needs clarification prior to mining. It is SRK’s

understanding that initial feedback from the Government has been received at the time of writing but

has not been accounted for in the PFS. Clarification on the criteria will likely result in upside of

additional material being available for future mine planning, within the existing infrastructure.

26.1.2 Geology and Mineral Resources

Additional ongoing recommendations for the Mineral Resource studies on this project, to be done prior

to the FS, should include:

• A detailed review of the geological model with CGM geologists to ensure geological continuity

is suitably modelled. This should include incorporation of a number of additional faults which

have been identified by CGM as having potential impact and controls on the system. These

faults also should be considered prior to the final underground decline design as it may have

geotechnical implications

• It is recommended that CGM develop a system to flag channel samples in the database taken

from the working stopes (vein channels), compared to more detailed exploration channel

samples taken from cross-cuts and exploration development where possible. Any updates in

the database will need to be completed prior to the FS update, and SRK will work with the

geological team to ensure the best solution can be found in the time available. The aim will

be to identify areas for potential mining targets to provide additional feed to the current

operation and plant

• Continual monitoring of the MDZ drilling program with regular updates on the Leapfrog Model

• Richard H. Sillitoe concluded that the 500 m long, west-northwest-striking MDZ orebody is

entirely controlled by a veinlet array developed during dextral transpression raises the

possibility that one or more look-alike deposits could be present elsewhere within the 6 km by

5 km P1 stock. They could be exposed at lower elevations than Marmato (such as less than

900 masl) as a result of greater degrees of erosion or, as in the case of the MDZ itself, remain

concealed beneath and partly overprinted by a swarm of ‘epithermal’ massive base-metal

sulfide veins. Either possibility would represent an attractive exploration opportunity, which it

is considered worth pursuing. SRK considers this an important consideration for future

exploration upon completion of the current drilling program.

SRK is currently working with CGM’s geologists to optimize the remainder of the 2020 drilling program.

The 2020 drilling program will have three main areas of focus (Figure 26-1) which includes:

• Phase A: Drilling the hangingwall of the MDZ to test for extensions to the northern limb of the

MDZ, which potentially hosts mineralization within the areas of the current mine design



presented in this technical report. This should be considered high priority as it may require

changes to the underground infrastructure locations moving them further to the north.

• Phase B: Increasing the confidence in the MDZ at depth to potentially add additional years to

the life of mine. The aim is to infill the drilling spacing to a 50 by 50 m grid in the upper portions

of the MDZ, and potentially increase the Inferred Resources at the end. CGM has planned to

make use of directional drilling from a series of mother-holes to complete the deep drilling

while access to favourable intersection angles from the current underground locations is

limited.

• Phase C: Exploration into the hangingwall to test for similar structures located below other

major features.

Source: SRK, 2020

Figure 26-1: 2020 Exploration Plan Showing Phases A through C (Left to Right)

26.1.3 Mineral Processing and Metallurgical Testing

Confirmatory metallurgical testwork should be conducted on material from new ore zones as they are

identified.

Additional testwork should be conducted to further assess cyanide destruction process parameters

required to reduce weak acid dissociable cyanide (CNWAD) in the leach residue to <1 ppm CNWAD.

26.1.4 Mining & Reserves

UZ Mine

It is recommended that CGM work to improve grade control practices and minimize dilution in the UZ

mine. SRK further recommends that CGM transition to using 3D mine design for better planning and

scheduling. (US$270,000)

For the Transition zone, additional geotechnical work is needed to recover the 7 m pillar between Level

21 and Level 22. (US$75,000)

??



MDZ

Continue to monitor costs and CoG. Small changes in CoG can have a material impact to the mine

design. Similarly continue to look at optimizing the sequence to mine higher grade material earlier in

the mine life in the next level of study. Develop the mine design to FS level design. (US$350,000

includes the mine design through faults)

The design crosses through several known faults, however, little is understood about these faults in

the location of the development. Additional drilling/testwork is necessary to understand this prior to

development. (Included in geotechnical estimate)

In the following paragraphs, recommended actions to be completed for the FS are provided. SRK

notes that some of the recommended activities have started.

Update the hydrogeologic information available and revisit the pumping system design to optimize the

system to the updated hydro information. Refine the pump sizing and consider an updated risk profile

to match the pump system sizing to actual expected inflows. This evaluation could lead to a reduced

pump size and lower power requirements including sizing of substation and electrical infrastructure.

(US$30,000 for FS Pump Design Optimization)

Evaluate the ventilation standard applied with respect to diesel dilution to consider whether a variance

to NA standards would allow a more optimized ventilation fan sizing that would potentially reduce

ventilation capital cost, operating cost, power system distribution size and infrastructure dimension.

(US$50,000 for ventilation standard evaluation)

Geotechnical

• Complete a major fault model update.

• Specific geotechnical drill holes to characterize the rock mass parameters around the critical

underground infrastructure should be drilled.

• Geotechnical core logging and televiewer data in specific exploration drill holes should be

collected and analyzed.

• Complete specific geotechnical drill holes to characterize the rock mass parameters around

the conveyor tunnel.

• Conduct pre-mining situ stress measurements.

• Collect tiltmeter measurements to confirm that there is minimal subsidence above the

transition zone.

• Perform mine scale stress analyses of the planned stoping sequence.

• A mine scale hydrogeological pore pressure model should be developed that considers

locations and hydraulic conductivity of specific fault structures as they intersect drifts and

stopes.

• Long term access to critical infrastructures should be evaluated, such as the crusher station

and workshops. Specific geotechnical drill holes and numerical simulations need to be

considered for the FS.

• 3D numerical modeling to determine the mining sequence effect on the mine stability

SRK estimates that the full FS geotechnical program will cost about US$400,000, which includes

laboratory testing, stress measurements, numerical simulations and geological engineering works.

The total cost does not include drilling activities.



Hydrogeology

To reduce these uncertainties, SRK recommends CGM complete the following additional

hydrogeological investigations/analyses for the FS:











o Drawdown propagation as result of mine dewatering

o Seasonal variation as result of precipitation








predictability:





26.1.5 Recovery Methods

During the next phase of study additional process design work should be conducted in order to achieve

a definitive capital cost estimate with an accuracy of +/-15%.

26.1.6 Project Infrastructure

UZ Project

The UZ project electrical system should be further reviewed during the next phase of study to confirm

the impacts of adding the MDZ fan loads to the system and optimize backup generation. Evaluate the

interconnection of the UZ electrical system to the MDZ system to establish a loop for reliability

purposes. (US$25,000)

Refine and further develop the water supply system to detailed engineering level and confirm in more

detail the costs. (US$50,000)

Develop a full understanding of the mine water and DSTF water requirements and runoff will allow for

optimization of the site runoff pond and water treatment capacities



MDZ Project

Due to lack of site-specific geotechnical investigations at the process plant site, several assumptions

have been made on ground bearing capacity, ground stability, depth of bed rock, suitability and

availability of general fill and granular material, borrow source for granular material. These

assumptions should be verified in the next phase of the project. (Work in progress by CGM)

Further develop the water supply system to feasibility level and review the overall site water balance

to confirm sizing requirements. (US$25,000)

Carry out an overall review of all the electrical loads in the mine and mill areas once additional precision

is available with all equipment loads to assess if main transformers can be reduced. (US$50,000)

Update the paste backfill plant design and costing to feasibility level. (US$350,000)

Tailings Management Facilities

For the next phase of study, SRK makes the following recommendations related to DSTF design and

costing:

• Detailed geotechnical site investigations should be completed at each proposed DSTF site

and stockpile locations, a thorough program of foundation, tailings and cement amendment

geotechnical testing should be completed and the conceptual designs, stability analyses and

costs updated to reflect the results. The geotechnical investigation should also include a

trenching study within DTSF-2’s footprint to assess the activity of known faults in the area.

(US$700,000)

• Geochemical characterization of waste rock and ore/tailings should be completed and the

conceptual DSTF designs and associated borrow source, operating, and closure requirements

should be updated accordingly, including the potential additional requirement of water

treatment prior to discharge. (US$100,000)

• The detailed characterization of Cascabel 1 and 2 recommended by Dynami in 2020 should

be expedited and the results incorporated into a stability analysis prepared in accordance with

internationally accepted standards of practice. If the facility cannot be shown to be sufficiently

stable, or if mitigating measures or design revisions cannot be identified to bring the facility up

to those standards, CGM should identify an alternative option for tailings storage until a new

DSTF can be constructed and operational. (US$2.6 million)

• A more thorough evaluation of DSTF 6 acquisition, permitting, development, operating and

closure requirements should be completed based on more favorable topography and

stormwater management requirements than either DSTF-1 or DSTF-2. (US$150,000)

26.1.7 Environmental Studies and Permitting

The following recommendations are made with respect to environmental, permitting and social issues

regarding the Marmato Project:

• Prepare a more detailed site-wide closure plan for the existing Marmato facilities, including

building plans and equipment inventories) from which a more accurate final closure cost

estimate can be developed.

• Continue work on groundwater hydrogeology and surface water to better define the risk

associated with potential groundwater contamination and underground dewatering impacts. A



detailed evaluation, including a groundwater model, could provide information that would

assist in forecasts of post-closure mine water discharge and possible long-term water

treatment requirements. Such an investigation could also provide vital information on

underground geotechnical stability, both during operations and post closure.

• Characterization work should be completed on artisanal tailings and waste rock to understand

their ARDML potential and devise a long-term management plan.

• A comprehensive baseline surface and groundwater sampling program will be important to

establish the baseline condition and try to quantify the contributions from artisanal or pre-

mining conditions, especially with respect to mercury from artisanal mining.

Substantial financial resources and technical specialist support will be required to implement the

environmental monitoring and mitigation measures likely to be presented in the updated PMA for the

expansion project.

26.1.8 Capital, Operating Costs and Economic Analysis

The following recommendations are made with respect to capital and operating cost and economic

evaluation of the Marmato Project:

• Prepare first principles estimate of capital and operating costs with enough accuracy to

support a future FS study of the project, including:

o Prepare cash flow model based on shorter periods of production.

o Prepare an expenditure curve for MDZ mineral processing and site infrastructure

construction costs.

o Further detail site-specific operating cost data and cost models to include fixed and

variable nature of costs and detail cost models to include breakdown by area and function.

o Improve cost models to include currencies used to estimate each cost and prepare

sensitivity to currencies variability.

• The schedule prepared for Marmato UZ doesn’t fully utilize its mineral processing capacity for

several years of the LoM. Investigate the possibility to expand the total mine movement permit

to allow Marmato UZ to process its run of mine using its plant full capacity, as this will very

likely improve the overall project economics.

26.2 Recommended Work Program Costs

Table 26-1 summarizes the costs for recommended work programs.



Table 26-1: Summary of Costs for Recommended Work

Discipline Program Description Cost (US$) No Further Work is Recommended Reason:

Property Description and Ownership N/A

Geology and Mineralization Captured in next line item.

Status of Exploration, Development and Operations Complete Infill Drilling $6,500,000

Mineral Processing and Metallurgical Testing Conduct confirmatory testing on material from new ore zones as they are developed

$200,000

Mineral Resource Estimate

Mineral Reserve Estimate

Mining Methods As Detailed in Section 26 $775,000

Recovery Methods FS engineering during next phase of study to bring process plant capex to a +/-15% level of accuracy

$750,000

Project Infrastructure As Detailed in Section 26 $4,550,000

Geotechnical As described in Section 26.1.4 $400,000

Exploration team will continue conducting geotechnical core logging as part of its in fill drilling programs, This information will be useful for further rock mechanics model update

Hydrogeology As Detailed in Section 26 Drilling and equipment included.

1,600,000

Environmental Studies and Permitting

Site Wide Detailed Closure Plan $50,000

Forecast Post-Closure Mine Water Discharges

$75,000

Characterize Artisanal Mine Wastes $100,000

Capital and Operating Costs Updated analysis resulting from new information

US$200,000

Economic Analysis Updated analysis resulting from new information

US$25,000

Total US$ $15,225,000

Source: SRK, 2020



27 References Barton, N. (1974). A review of the Shear Strength of Filled Discontinuities in Rock. Norwegian

Geotechnical Institute publication no. 105.

CIM (2014). Canadian Institute of Mining, Metallurgy and Petroleum Standards on Mineral Resources

and Reserves: Definitions and Guidelines, May 10, 2014.

Clark and Pakalnis (1997). An empirical design approach for estimating unplanned dilution from open

stope hangingwalls and footwalls. 99th CIM Annual General Meeting, Vancouver, British

Columbia.

Dynami (Dynami, 2020a). TSF 1 and TSF 2 Hydrologic and Hydraulic conceptual analyses

Memorandum, July 24 2020, Colombia.

Dynami (Dynami, 2020b). TSF 1 and TSF 2 Slope Stability Analyses Memorandum, August 10 2020,

Colombia.

Dynami (Dynami, 2020c). Analisis De Estabilidad Relaveras Cascabel 1 & 2, Caldas Gold Marmato,

Proyecto Marmato, May 2020, Colombia.

Telluris Consulting Ltd. (2010). Structural study of the Marmato District, Colombia. Unpublished report

for Medoro Resources/Minerales Andinos del Occidente S.A., 22p.

Gran Colombia Gold (Caldas). Lead Isotopic Compositions of the Gold Mineralization of Marmato,

Colombia: Characterization of the Transition Domain in Epithermal – Porphyry Systems.

Grimstad, E & Barton, N (1993), ‘Updating the Q-system for NMT’, in C Kompen, SL Opsahl & SL Berg

(eds), Proceedings of the International Symposium on Sprayed Concrete, Norwegian

Concrete Association, 21 p.

IDEAM, (2013). Aguas Subterraneas en Colombia: Una Vision General, Instituto de Hidrología,

Meteorología y Estudios Ambientales de Colombia, Bogota D.C., 2013.

Knight Piésold Consulting, (2012a). Marmato Mine, Pre-Feasibility Study Hydrogeology Report, April

12, 2012.Hatch, 2012.

Mitchel et. al (1982) Stability Analysis of Paste Fill as Stope Wall using Analytical Method and

Numerical Modeling in The Kencana Underground Gold Mining with Long Hole Stope Method.

Pinzón, F. D. & Tassinari, C. C. G., (2003), Ages and Sources of the Gold Mineralizations from

Marmato Mining District, NW Colombia: Based on Radiogenic Isotope Evidences. IV South

American Symposium on Isotope Geology. IRD (Institut de recherche pour le development).

Salvador, Bahia, Brazil, 24–27 August 2003, p. 758 to 761.

Potvin and Milne (1992). A Critical Review of the Stability Graph Method for Open Stope Design.

Proceedings MassMin 2012, Sudbury, Ontario, Canada.

Rocscience (2016). DIPS v 7.0. – program for the projection of joint and discontinuity data on a

sterographic projection. Version 7.0, Rocscience Inc., Toronto.

Rocscience (2019). UnWedge, 2019. – program for underground excavation stability assessment

Rocscience Inc., Toronto.



Rossetti, P. & Colombo, F. (1999). Adularia-sericite gold deposits of Marmato (Caldas, Colombia):

field and petrographical data. Geological Society, London, Special Publications 155, 167-182.

Sillitoe, R. H., Jaramillo, L., Damon, P. E., Shafiqullah, M. & Escovar, R., (1982), Setting,

Characteristics, and Age of the Andean Porphyry Copper Belt in Colombia. Economic

Geology, 77, p. 1837 to 1850.

SRK (2020). Pre-Feasibility Geotechnical Study Marmato Deep Zone, Colombia.

SRK (SRK, 2020a). Marmato DSTF Siting Study Memorandum, May 21 2020, Reno Nevada.

Vinasco, C. J., (2001). A utilização da metodologia 40Ar-39Ar para o estudo de reativações tectônicas

em zonas de cisalhamento. Paradigma: O Falhamento de Romeral nos Andes Centrais da

Colômbia. Masters dissertation. Institute of Geosciences. University of Sao Paulo.



28 Glossary The Mineral Resources and Mineral Reserves have been classified according to CIM (CIM, 2014).

Accordingly, the Resources have been classified as Measured, Indicated or Inferred, the Reserves

have been classified as Proven and Probable based on the Measured and Indicated Resources as

defined below.

28.1 Mineral Resources

A Mineral Resource is a concentration or occurrence of solid material of economic interest in or on

the Earth’s crust in such form, grade or quality and quantity that there are reasonable prospects for

eventual economic extraction. The location, quantity, grade or quality, continuity and other geological

characteristics of a Mineral Resource are known, estimated or interpreted from specific geological

evidence and knowledge, including sampling.

An Inferred Mineral Resource is that part of a Mineral Resource for which quantity and grade or

quality are estimated on the basis of limited geological evidence and sampling. Geological evidence

is sufficient to imply but not verify geological and grade or quality continuity. An Inferred Mineral

Resource has a lower level of confidence than that applying to an Indicated Mineral Resource and

must not be converted to a Mineral Reserve. It is reasonably expected that the majority of Inferred

Mineral Resources could be upgraded to Indicated Mineral Resources with continued exploration.

An Indicated Mineral Resource is that part of a Mineral Resource for which quantity, grade or quality,

densities, shape and physical characteristics are estimated with sufficient confidence to allow the

application of Modifying Factors in sufficient detail to support mine planning and evaluation of the

economic viability of the deposit. Geological evidence is derived from adequately detailed and reliable

exploration, sampling and testing and is sufficient to assume geological and grade or quality continuity

between points of observation. An Indicated Mineral Resource has a lower level of confidence than

that applying to a Measured Mineral Resource and may only be converted to a Probable Mineral

Reserve.

A Measured Mineral Resource is that part of a Mineral Resource for which quantity, grade or quality,

densities, shape, and physical characteristics are estimated with confidence sufficient to allow the

application of Modifying Factors to support detailed mine planning and final evaluation of the economic

viability of the deposit. Geological evidence is derived from detailed and reliable exploration, sampling

and testing and is sufficient to confirm geological and grade or quality continuity between points of

observation. A Measured Mineral Resource has a higher level of confidence than that applying to

either an Indicated Mineral Resource or an Inferred Mineral Resource. It may be converted to a Proven

Mineral Reserve or to a Probable Mineral Reserve.

28.2 Mineral Reserves

A Mineral Reserve is the economically mineable part of a Measured and/or Indicated Mineral

Resource. It includes diluting materials and allowances for losses, which may occur when the material

is mined or extracted and is defined by studies at Pre-Feasibility or Feasibility level as appropriate that

include application of Modifying Factors. Such studies demonstrate that, at the time of reporting,

extraction could reasonably be justified.



The reference point at which Mineral Reserves are defined, usually the point where the ore is delivered

to the processing plant, must be stated. It is important that, in all situations where the reference point

is different, such as for a saleable product, a clarifying statement is included to ensure that the reader

is fully informed as to what is being reported. The public disclosure of a Mineral Reserve must be

demonstrated by a Pre-Feasibility Study or Feasibility Study.

A Probable Mineral Reserve is the economically mineable part of an Indicated, and in some

circumstances, a Measured Mineral Resource. The confidence in the Modifying Factors applying to a

Probable Mineral Reserve is lower than that applying to a Proven Mineral Reserve.

A Proven Mineral Reserve is the economically mineable part of a Measured Mineral Resource. A

Proven Mineral Reserve implies a high degree of confidence in the Modifying Factors.

28.3 Definition of Terms

The following general mining terms may be used in this report.

Table 28-1: Definition of Terms

Term Definition

Assay The chemical analysis of mineral samples to determine the metal content.

Capital Expenditure All other expenditures not classified as operating costs.

Composite Combining more than one sample result to give an average result over a larger distance.

Concentrate A metal-rich product resulting from a mineral enrichment process such as gravity concentration or flotation, in which most of the desired mineral has been separated from the waste material in the ore.

Crushing Initial process of reducing ore particle size to render it more amenable for further processing.

Cut-off Grade (CoG) The grade of mineralized rock, which determines as to whether or not it is economic to recover its gold content by further concentration.

Dilution Waste, which is unavoidably mined with ore.

Dip Angle of inclination of a geological feature/rock from the horizontal.

Fault The surface of a fracture along which movement has occurred.

Footwall The underlying side of an orebody or stope.

Gangue Non-valuable components of the ore.

Grade The measure of concentration of gold within mineralized rock.

Hangingwall The overlying side of an orebody or slope.

Haulage A horizontal underground excavation which is used to transport mined ore.

Hydrocyclone A process whereby material is graded according to size by exploiting centrifugal forces of particulate materials.

Igneous Primary crystalline rock formed by the solidification of magma.

Kriging An interpolation method of assigning values from samples to blocks that minimizes the estimation error.

Level Horizontal tunnel the primary purpose is the transportation of personnel and materials.

Lithological Geological description pertaining to different rock types.

LoM Plans Life-of-Mine plans.

LRP Long Range Plan.

Material Properties Mine properties.

Milling A general term used to describe the process in which the ore is crushed and ground and subjected to physical or chemical treatment to extract the valuable metals to a concentrate or finished product.

Mineral/Mining Lease A lease area for which mineral rights are held.

CGM Mining Assets The Material Properties and Significant Exploration Properties.

Ongoing Capital Capital estimates of a routine nature, which is necessary for sustaining operations.

Ore Reserve See Mineral Reserve.

Pillar Rock left behind to help support the excavations in an underground mine.



Term Definition

RoM Run-of-Mine.

Sedimentary Pertaining to rocks formed by the accumulation of sediments, formed by the erosion of other rocks.

Shaft An opening cut downwards from the surface for transporting personnel, equipment, supplies, ore and waste.

Sill A thin, tabular, horizontal to sub-horizontal body of igneous rock formed by the injection of magma into planar zones of weakness.

Smelting A high temperature pyrometallurgical operation conducted in a furnace, in which the valuable metal is collected to a molten matte or doré phase and separated from the gangue components that accumulate in a less dense molten slag phase.

Stope Underground void created by mining.

Stratigraphy The study of stratified rocks in terms of time and space.

Strike Direction of line formed by the intersection of strata surfaces with the horizontal plane, always perpendicular to the dip direction.

Sulfide A sulfur bearing mineral.

Tailings Finely ground waste rock from which valuable minerals or metals have been extracted.

Thickening The process of concentrating solid particles in suspension.

Total Expenditure All expenditures including those of an operating and capital nature.

Variogram A statistical representation of the characteristics (usually grade).

28.4 Abbreviations

The following abbreviations may be used in this report.

Table 28-2: Abbreviations

Abbreviation Unit or Term

A Ampere

AA atomic absorption

A/m2 amperes per square meter

ANFO ammonium nitrate fuel oil

Ag Silver

Au Gold

AuEq gold equivalent grade

°C degrees Centigrade

CCD counter-current decantation

CIL carbon-in-leach

CoG cut-off grade

Cm centimeter

cm2 square centimeter

cm3 cubic centimeter

Cfm cubic feet per minute

ConfC confidence code

CRec core recovery

CSS closed-side setting

CTW calculated true width

° degree (degrees)

dia. diameter

EIS Environmental Impact Statement

EMP Environmental Management Plan

FA fire assay

Ft foot (feet)

ft2 square foot (feet)

ft3 cubic foot (feet)

g Gram

Gal Gallon




g/L gram per liter

g-mol gram-mole

Gpm gallons per minute

g/t grams per tonne

Ha hectares

HDPE Height Density Polyethylene

hp horsepower

HTW horizontal true width

ICP induced couple plasma

ID2 inverse-distance squared

ID3 inverse-distance cubed

IFC International Finance Corporation

ILS Intermediate Leach Solution

kA kiloamperes

kg kilograms

km kilometer

km2 square kilometer

koz thousand troy ounce

kt thousand tonnes

kt/d thousand tonnes per day

kt/y thousand tonnes per year

kV Kilovolt

kW Kilowatt

kWh kilowatt-hour

kWh/t kilowatt-hour per metric tonne

Liter

L/sec liters per second

L/sec/m liters per second per meter

Lb Pound

LHD Long-Haul Dump truck

LLDDP Linear Low Density Polyethylene Plastic

LOI Loss On Ignition

LoM Life-of-Mine

m Meter

m2 square meter

m3 cubic meter

masl meters above sea level

mg/L milligrams/liter

mm millimeter

mm2 square millimeter

mm3 cubic millimeter

MME Mine & Mill Engineering

Moz million troy ounces

Mt million tonnes

MTW measured true width

MW million watts

m.y. million years

NGO non-governmental organization

NI 43-101 Canadian National Instrument 43-101

OSC Ontario Securities Commission

oz troy ounce

% Percent

PLC Programmable Logic Controller

PLS Pregnant Leach Solution

PMF probable maximum flood

ppb parts per billion

ppm parts per million

QA/QC Quality Assurance/Quality Control

RC rotary circulation drilling




RoM Run-of-Mine

RQD Rock Quality Description

SEC U.S. Securities & Exchange Commission

sec Second

SG specific gravity

SPT standard penetration testing

st short ton (2,000 pounds)

t tonne (metric ton) (2,204.6 pounds)

t/h tonnes per hour

t/d tonnes per day

t/y tonnes per year

TSF tailings storage facility

TSP total suspended particulates

µm micron or microns

V Volts

VFD variable frequency drive

W Watt

XRD x-ray diffraction

y Year

SRK Consulting (U.S.), Inc. NI 43-101 Technical Report – Marmato Appendices

MMS/KD August 2020

Appendices


MMS/KD August 2020

Appendix A: Certificates of Qualified Persons


Suite 600

1125 Seventeenth Street

Denver, CO 80202

T: 303.985.1333

F: 303.985.9947

[email protected]

www.srk.com

U.S. Offices:

Anchorage 907.677.3520

Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:

Saskatoon 306.955.4778

Sudbury 705.682.3270

Toronto 416.601.1445

Vancouver 604.681.4196

Yellowknife 867.873.8670

Group Offices:

Africa

Asia

Australia

Europe

North America

South America

CERTIFICATE OF QUALIFIED PERSON

I, David Bird, MSc., PG, RM-SME, do hereby certify that:

1. I am an Associate Principal Consultant (Geochemistry) of SRK Consulting (U.S.), Inc., 1125 Seventeenth Street, Suite 600, Denver, CO, USA, 80202.

2. This certificate applies to the technical report titled “Revised NI 43-101 Technical Report Pre-Feasibility Study Marmato Project Colombia” with an Effective Date of March 17, 2020 (the “Technical Report”).

3. I graduated with Bachelor’s Degrees in Geology and Business Administration Management from Oregon State University in 1983. In addition, I obtained a Master’s Degree in Geochemistry/Hydrogeology from the University of Nevada-Reno in 1993. I am a Registered Member of the Society for Mining, Metallurgy, and Exploration (SME). I am a certified Professional Geologist in the State of Oregon (G1438). I have worked full time as a Geologist and Geochemist for a total of 32 years. My relevant experience includes design, execution, and interpretation of mine waste geochemical characterization programs in support of open pit and underground mine planning and environmental impact assessments, design and supervision of water quality sampling and monitoring programs, geochemical modeling, and management of the geochemistry portion of numerous PEA, PFS and FS-level mine projects in the US and abroad.

4. I have read the definition of “qualified person” set out in National Instrument 43-101 (NI 43-101) and certify that by reason of my education, affiliation with a professional association (as defined in NI 43-101) and past relevant work experience, I fulfill the requirements to be a “qualified person” for the purposes of NI 43-101.

5. I have not visited the Marmato property.

6. I am responsible for Geochemistry Section 20.1.3, and portions of Sections 1.10.1, 25 and 26.

7. I am independent of the issuer applying all of the tests in section 1.5 of NI 43-101.

8. I have had prior involvement with the property that is the subject of the Technical Report. The nature of my prior involvement is acting as QP on Bluenose Gold Corporation’s Marmato Project report titled “NI 43-101 Technical Report Preliminary Economic Assessment Marmato Project Colombia” with an effective date of July 31, 2019 and a report date of December 19, 2019.

9. I have read NI 43-101 and Form 43-101F1 and the sections of the Technical Report I am responsible for have been prepared in compliance with that instrument and form.

10. As of the aforementioned Effective Date, to the best of my knowledge, information and belief, the sections of the Technical Report I am responsible for contains all scientific and technical information that is required to be disclosed to make the Technical Report not misleading.

Dated this 18th Day of September, 2020.

_______Signed_________________________ Sealed

David Bird, MSc, PG, RM-SME

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America


5250 Neil Road, Suite 300

Reno, Nevada 89502

T: (775) 828-6800

F: (775) 828-6820

[email protected]

www.srk.com


I, R. Breese Burnley, PE do hereby certify that:

1. I am Practice Leader/Principal Engineer of SRK Consulting (U.S.), Inc., 5250 Neil Road, Reno, Nevada 89502.


3. I graduated with a B.Sc. degree in in Geology in 1991 from the University of Nevada, Reno. In addition, I obtained an M.Sc. in Geological Engineering in 1993, also from the University of Nevada, Reno.

4. I am a registered Professional Engineer in the State of Nevada (PE No. 16225). I have worked as an engineer for a total of 27 years since my graduation from university. My relevant experience includes site investigations, conceptual and detailed design, construction supervision, management and operational assessments, mine reclamation permitting and closure design and permitting at numerous industrial and mining properties throughout the western United States and South and Central America.


6. I have visited the Marmato property. I visited the Marmato property on January 28, 2020 for 1 day.

7. I am responsible for Tailings Section 18.15, and the tailings portions of Section 21, and portions of Sections 1, 24, 25 and 26.


9. I have not had prior involvement with the property that is the subject of the Technical Report.




_______Signed_________________________ Sealed

R. Breese Burnley, PE


Suite 600


Denver, CO 80202

T: 303.985.1333

F: 303.985.9947

[email protected]

www.srk.com

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America


I, Fredy Henriquez, MSc Eng, SME, ISRM do hereby certify that:

1. I am Principal Consultant (Rock Mechanics) of SRK Consulting (U.S.), Inc., 1125 Seventeenth Street, Suite 600, Denver, CO, USA, 80202.


3. I graduated with a degree in Civil Mine Engineer from University of Santiago, Chile in 2000. In addition, I have obtained a Masters degree (MSc) in Engineering (Rock Mechanics) from WASM, Curtin University, Australia (2011). I am a Registered Member of the Society for Mining, Metallurgy, and Exploration (SME, register number 4196405RM). I have worked as a geotechnical engineer for a total of 25 years since my graduation from university. My relevant experience includes civil and mining geotechnical projects ranging from conceptual through feasibility design levels and operations support. I am skilled in both soil and rock mechanics engineering and specialize in the design and management of mine excavations. My primary areas of expertise include mine operations, mine planning, hard rock and soft rock characterization, underground and open pit stability analysis, database management, geotechnical data collection, probabilistic analysis, risk assessment, slope monitoring, modeling and pit wall pore pressure reductions. I have undertaken and managed large geotechnical projects for the mining industry throughout North, Central, South America, Australia and South Africa.


5. I visited the Marmato property on January 8, 2020 for 4 days and July 16, 2019 for 3 days.

6. I am responsible for geotechnical Section 16.4 of the Technical Report.






_______Signed_________________________ Sealed

Fredy Henriquez, MSc Eng, SME, ISRM Principal Consultant (Rock Mechanics)


Suite 600


Denver, CO 80202

T: 303.985.1333

F: 303.985.9947

[email protected]

www.srk.com

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America


I, David Hoekstra, BSc Civil Engineering, P.E, do hereby certify that:

1. I am Principal Consultant (Civil Engineer) of SRK Consulting (U.S.), Inc., 1125 Seventeenth Street, Suite 600, Denver, CO, USA, 80202.


3. I graduated with a degree in Civil Engineering from Colorado State University in 1986. I am a Professional Engineer of the States of Alaska, Colorado, Montana, South Carolina, and Wyoming. I have worked as an Engineer for a total of 33 years since my graduation from university. My relevant experience includes the design and implementation of mine water management systems and storm water controls for numerous PEA, PFS, FS-level and operating mine projects in the US and abroad.



6. I am responsible for Section 18.14, Hydrology Section 20.2.5, and portions of Sections 1, 24, 25 and 26.


8. I have had prior involvement with the property that is the subject of the Technical Report. The nature of my prior involvement is acting as QP on the Bluenose Gold Corporation’s Marmato Project report titled “NI 43-101 Technical Report Preliminary Economic Assessment Marmato Project Colombia” with an effective date of July 31, 2019 and a report date of December 19, 2019.




_______Signed_________________________ Sealed

David Hoekstra, BS, PE, NCEES, SME-RM


Suite 600


Denver, CO 80202

T: 303.985.1333

F: 303.985.9947

[email protected]

www.srk.com

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America


I, Eric Olin, MSc, MBA, RM-SME do hereby certify that:

1. I am a Principal Process Metallurgist of SRK Consulting (U.S.), Inc., 1125 Seventeenth Street, Suite 600, Denver, CO, USA, 80202.


3. I graduated with a Master of Science degree in Metallurgical Engineering from the Colorado School of Mines in 1976. I am a Registered Member of The Society for Mining, Metallurgy and Exploration, Inc. I have worked as a Metallurgist for a total of 40 years since my graduation from the Colorado School of Mines. My relevant experience includes extensive consulting, plant operations, process development, project management and research & development experience with base metals, precious metals, ferrous metals and industrial minerals. I have served as the plant superintendent for several gold and base metal mining operations. Additionally, I have been involved with numerous third-party due diligence audits, and preparation of project conceptual, pre-feasibility and full-feasibility studies.


5. I have visited the Marmato property. I visited the property on December 17, 2019 for 2 days.

6. I am responsible for the preparation of Metallurgy Sections 13, 17.1 17.2 and Upper Zone processing portion of Section 21, portions of Sections 1, 24, 25 and 26 of the Technical Report.






_______Signed_________________________ Sealed

Eric Olin, MSc, MBA, RM-SME


Suite 600


Denver, CO 80202

T: 303.985.1333

F: 303.985.9947

[email protected]

www.srk.com

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America


I, Jeff Osborn, BEng Mining, MMSAQP do hereby certify that:

1. I am a Principal Consultant (Mining Engineer) of SRK Consulting (U.S.), Inc., 1125 Seventeenth, Suite 600, Denver, CO, USA, 80202.


3. I graduated with a Bachelor of Science Mining Engineering degree from the Colorado School of Mines in 1986. I am a Qualified Professional (QP) Member of the Mining and Metallurgical Society of America. I have worked as a Mining Engineer for a total of 32 years since my graduation from university. My relevant experience includes responsibilities in operations, maintenance, engineering, management, and construction activities.


5. I visited the Marmato property on July 16, 2019 for 3 days and on August 22, 2017 for 2 days.

6. I am responsible for Infrastructure and portions of the Cost Estimation Sections 18.1, 18.2, 18.13,18.16, and 21 (excluding processing and tailings portions of Section 21), and portions of Sections 1, 24, 25 and 26.






_______Signed_________________________ Sealed

Jeff Osborn, BEng Mining, MMSAQP [01458QP] Principal Consultant (Mining Engineer)


Suite 600


Denver, CO 80202

T: 303.985.1333

F: 303.985.9947

[email protected]

www.srk.com

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America


I, Benjamin Parsons, MSc, MAusIMM (CP) do hereby certify that:

1. I am a Principal Consultant (Resource Geology) of SRK Consulting (U.S.), Inc., 1125 Seventeenth Street, Suite 600, Denver, CO, USA, 80202.


3. I graduated with a degree in Exploration Geology from Cardiff University, UK in 1999. In addition, I have obtained a Masters degree (MSc) in Mineral Resources from Cardiff University, UK in 2000 and have worked as a geologist for a total of 19 years since my graduation from university. I am a member of the Australian Institution of Materials Mining and Metallurgy (Membership Number 222568) and I am a Chartered Professional.


5. I visited the Marmato property on June 11, 2020 for 3 days, August 17, 2017 for 1 day and March 12, 2012 for three days.

6. I am responsible for Sections 2 through 12 (except 4.4), 14, 23 and portions of Sections 1, 24, 25 and 26.






_______Signed_________________________ Sealed

Benjamin Parsons, MSc, MAusIMM Principal Consultant (Resource Geology)


Suite 600


Denver, CO 80202

T: 303.985.1333

F: 303.985.9947

[email protected]

www.srk.com

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America


I, Cristian A. Pereira Farias, SME-RM, do hereby certify that:

1. I am Principal Consultant (Hydrogeologist) of SRK Consulting (U.S.), Inc., 1125 Seventeenth Street, Suite 600, Denver, CO, USA, 80202.


3. I graduated with a degree in Bachelors of Science in Geology from Universidad de Chile in 1999. I am a registered member of the Society for Mining, Metallurgy, and Exploration. I have worked as a hydrogeologist for a total of 19 years since my graduation from university. My relevant experience includes the developing conceptual and numerical hydrogeological models, the evaluation of groundwater resources, mine dewatering requirements, environmental impacts of mining, pit lake infilling, brine extraction, and pore pressure analyses.


5. I visited the Marmato property on August 12, 2019 for 2 days.

6. I am responsible for Hydrogeology Sections 16.3, and portions of Sections 1, 24, 25 and 26.






_______Signed_________________________ Sealed

Cristian A. Pereira Farias, SME-RM


Suite 600


Denver, CO 80202

T: 303.985.1333

F: 303.985.9947

[email protected]

www.srk.com

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America


I, Joanna Poeck, BEng Mining, SME-RM, MMSAQP, do hereby certify that:

1. I am a Principal Mining Engineer of SRK Consulting (U.S.), Inc., 1125 Seventeenth Street, Suite 600, Denver, CO, USA, 80202.


3. I graduated with a degree in Mining Engineering from Colorado School of Mines in 2003. I am a Registered Member of the Society of Mining, Metallurgy & Exploration Geology. I am a QP member of the Mining & Metallurgical Society of America. I have worked as a Mining Engineer for a total of 15 years since my graduation from university. My relevant experience includes open pit and underground design, mine scheduling, pit optimization and truck productivity analysis.



6. I am responsible for the opening statement in Section 15, Section 15.1.5 through 15.1.8, portions of Sections 15.2 and 15.3 pertaining to the MDZ, Section 16.5, portions of 16.6 pertaining to the MDZ, and portions of Sections 1, 24, 25 and 26.






_______Signed_________________________ Sealed

Joanna Poeck, BEng Mining, SME-RM[4131289RM], MMSAQP[01387QP] Principal Consultant (Mining Engineer)


Suite 600


Denver, CO 80202

T: 303.985.1333

F: 303.985.9947

[email protected]

www.srk.com

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America

QP_Cert_Rodrigues.docx


I, Fernando Rodrigues, BS Mining, MBA, MMSAQP do hereby certify that:

1. I am Practice Leader and Principal Consultant (Mining Engineer) of SRK Consulting (U.S.), Inc., 1125 Seventeenth Street, Suite 600, Denver, CO, USA, 80202.


3. I graduated with a Bachelors of Science degree in Mining Engineering from South Dakota School of Mines and Technology in 1999. I am a QP member of the MMSA. I have worked as a Mining Engineer for a total of 21 years since my graduation from South Dakota School of Mines and Technology in 1999. My relevant experience includes mine design and implementation, short term mine design, dump design, haulage studies, blast design, ore control, grade estimation, database management.


5. I visited the Marmato property on August 22, 2017 for 2 days, January 7, 2020 for 4 days, February 2, 2020 for 4 days.

6. I am responsible for Upper Zone Mining and Economics and related portions of Section 15.1.1 through 15.1.4, portions of Sections 15.2 and 15.3 pertaining to the Upper Zone, the opening statement in Section 16, Sections 16.1, 16.4, portions of 16.6 pertaining to the Upper Zone, 19 and 22, and portions of Sections 1, 24, 25 and 26 summarized therefrom, of this Technical Report.






_______Signed_________________________ Sealed

Fernando Rodrigues, BS Mining, MBA, MMSAQP [01405QP]

U.S. Offices:


Clovis 559.452.0182

Denver 303.985.1333

Elko 775.753.4151

Reno 775.828.6800

Tucson 520.544.3688

Canadian Offices:


Sudbury 705.682.3270

Toronto 416.601.1445



Group Offices:

Africa

Asia

Australia

Europe

North America

South America


5250 Neil Road, Suite 300

Reno, Nevada 89502

T: (775) 828-6800

F: (775) 828-6820

[email protected]

www.srk.com


I, Mark Allan Willow, MSc, CEM, SME-RM do hereby certify that:

1. I am Practice Leader/Principal Environmental Scientist of SRK Consulting (U.S.), Inc., 5250 Neil Road, Reno, Nevada 89502.


3. I graduated with Bachelor's degree in Fisheries and Wildlife Management from the University of Missouri in 1987 and a Master's degree in Environmental Science and Engineering from the Colorado School of Mines in 1995. I have worked as Biologist/Environmental Scientist for over 25 years since my graduation from university. My relevant experience includes environmental due diligence/competent persons evaluations of developmental phase and operational phase mines through the world, including small gold mining projects in Panama, Senegal, Peru, Ecuador, Philippines, and Colombia; open pit and underground coal mines in Russia; large copper and iron mines and processing facilities in Mexico and Brazil; bauxite operations in Jamaica; and a coal mine/coking operation in the People's Republic of China. My Project Manager experience includes several site characterization and mine closure projects. I work closely with the U.S. Forest Service and U.S. Bureau of Land Management on permitting and mine closure projects to develop uniquely successful and cost-effective closure alternatives for the abandoned mining operations. Finally, I draw upon this diverse background for knowledge and experience as a human health and ecological risk assessor with respect to potential environmental impacts associated with operating and closing mining properties and have experience in the development of Preliminary Remediation Goals and hazard/risk calculations for site remedial action plans under Superfund activities according to current U.S. EPA risk assessment guidance.

4. I am a Certified Environmental Manager (CEM) in the State of Nevada (#1832) in accordance with Nevada Administrative Code 459.970 through 459.9729. Before any person consults for a fee in matters concerning: the management of hazardous waste; the investigation of a release or potential release of a hazardous substance; the sampling of any media to determine the release of a hazardous substance; the response to a release or cleanup of a hazardous substance; or the remediation soil or water contaminated with a hazardous substance, they must be certified by the Nevada Division of Environmental Protection, Bureau of Corrective Action;

5. I am a Registered Member (No. 4104492) of the Society for Mining, Metallurgy & Exploration Inc. (SME).


7. I visited the Marmato property on December 1, 2016 for 1 day.

8. I am responsible for Section 4.4, Environmental Section 20 (except section 20.1.3), and portions of Sections 1, 24, 25 and 26.


10. I have had prior involvement with the property that is the subject of the Technical Report. The nature of my prior involvement is acting as QP on Bluenose Gold Corporation’s Marmato Project report titled “NI 43-101 Technical Report Preliminary Economic Assessment Marmato Project Colombia” with an effective date of July 31, 2019 and a report date of December 19, 2019.

SRK Consulting Page 2

QP_Cert_Willow.docx




_______Signed_________________________ Sealed

Mark Allan Willow, MSc, CEM, SME-RM SME-RM# 4104492


Tommaso Roberto Raponi, P. Eng.

To Accompany the Report entitled, “Revised NI 43-101 Technical Report Pre-Feasibility Study

Marmato Project Colombia” prepared for Caldas Gold Corp. effective date March 17, 2020 and

dated September 18, 2020.

I, Tommaso Roberto Raponi, P. Eng., do hereby certify:

1. I am a Principal Metallurgist at Ausenco Engineering Canada Inc., 11 King St West, Suite

1550, Toronto, ON, M5H 4C7.

2. I hold a Bachelor's degree in Geological Engineering from University of Toronto, Toronto,

Ontario, Canada.

3. I am registered as a Professional Engineer in Ontario and British Columbia. I have worked for

more than 36 years in the mining industry in various positions continuously since my

graduation from university. I have worked as an independent consultant since 2016.

4. I have read the definition of "qualified person" set out in National Instrument 43‐101 (NI 43‐

101) and certify that by reason of my education, affiliation with a professional association (as

defined by NI 43‐101) and past relevant work experience, I fulfill the requirements to be a

"qualified person" for the purposes of NI 43‐101.

5. I have not visited site.

6. I am responsible for the MDZ process plant and infrastructure engineering and related

portions of Sections 17.3, 18.3 through 18.12, MDZ processing and infrastructure portions of

21.1.2 and 21.3.2 and portions of Sections 1, 24, 25 and 26 of the Technical Report.


8. I am independent of the issuer applying all of the tests in section 1.5 of NI 43‐101.

9. I have read NI 43‐101 and Form 43‐101F1 and the sections of the Technical Report I am

responsible for have been prepared in compliance with that instrument and form.

10. As of the aforementioned Effective Date, to the best of my knowledge, information and belief,

the sections of the Technical Report I am responsible for contains all scientific and technical

information that is required to be disclosed to make the Technical Report not misleading.

Dated this 18th day of September 2020

“Signed” “Sealed” Tommaso Roberto Raponi, P. Eng.

tommasoraponi

Cross-Out


MMS/KD August 2020

Appendix B: MDZ Tailings Drawings

DSTF 6(SITE 6)

ROCK STARTER EMBANKMENT

ROCKSTARTER

EMBANKMENT

DSTF 1

DSTF 2

CAUCA RIVER



STORMWATERDIVERSION CHANNELS

PROCESS PLANT

PORTAL

EXISTING TSF

HIGH VOLTAGEPOWERLINES (230Kv)

LOW VOLTAGEPOWERLINES

HIGH VOLTAGE POWERTOWER (TYP.)

MARMATO RIVER

CAMP

DSTF 1 ACCESS/ HAUL ROAD

FILTER PLANT

MARMATO MUNICIPAL ROAD

NATIONAL HIGHWAY

STORMWATERDIVERSIONCHANNELS

EXISTING PLANT AND PORTAL

EL LLANO TOWNSHIP

TEMPORARYSTOCKPILE AREA

CONCEPTUAL TAILINGS FACILITY SITE PLAN

A

N

FILE NAME: 100-SitePlan_544400.020_20200729.dwg

DRAWING TITLE:

consultingPROJECT:

PREPARED BY:

CALDAS GOLD CORPORATION

544400.070-800

DRAWING NO.DATE:

SRK JOB NO.:

C:\Users\jsames\Documents\SRK\GCG\544400-020_Marmato_PEA_PFS\040_Drafting\Task_810_PFS_Design\100-SitePlan_544400.020_20200729.dwg

8.12.2020IF THE ABOVE BARDOES NOT MEASURE 20mm,

THE DRAWING SCALE IS ALTERED

DESIGN: JS/ RBBDRAWN: JSREVIEWED: RBB

APPROVED: RBBPROJECT:

REV. NO.:

MARMATO PRE-FEASIBILITY STUDY

1000m

METERS

100m 200m 300m

(SCALE - 1:100)


DRAFT


STORMWATER DIVERSIONCHANNEL AROUND PLANT

(REF. DETAIL 7-104)

FILTER PLANT

A'

AB'

B

PORTAL

STILLING BASIN

PRECAST CONCRETE BRIDGE(REF. DETAIL 5-104)

DSTF 1 ACCESS/ HAUL ROADCENTERLINE (REF. DETAIL 6-104)

MARMATO MUNICIPALROAD

STORMWATER DIVERSIONCHANNEL ABOVE DSTF

(REF. DETAIL 7-104)

5%

STORMWATER DITCH (TYP.)(REF. DETAIL 7-104)

BENCH DETAIL (TYP.)(REF. DETAIL 2-103)TEMPORARY TOP

DECK STORMWATERMANAGEMENT PONDS

TRANSMISSION TOWER (TYP.)

HIGH VOLTAGEPOWER LINE (TYP.)

PRECAST CONCRETEBRIDGE (REF. DETAIL 8-104)

5%

DSTF 2

2.5:12.5:1

2.5:1

2:1

2:1

2.5:12.5:1

1000

1010

980

990950

960

970

920930

940

1010

1010

980

9901000

UNDERDRAIN (TYP.)(REF. DETAIL 4-103)

COLLECTION TANKS(REF. DETAIL 5-104)

EXISTING MUNICIPAL ROAD

EXISTING STRUCTURE (TYP.)

TRAFFIC CONTROLS

PROCESSPLANT

1100

1125

1150

1175

1200

1050 975

1000

1025

1050

1075

850

875

900

925

950

975

1000

1025

975

1000

10251050

10751100

1100

1075

1100

10251050

1200

1225

1250

1275

1300

1075

1100

1125

1150

900

925

950

90092

5

900

925

950

970

980990

980

ACCESS ROAD TOPORTAL BOUNDARY(REF. DETAIL 6-104)

CAMPPADSCAMP ACCESS ROADBY OTHERS

Elev

atio

n (m

, am

sl)

Station (m)

850

900

950

1000

1050

1100

1150

850

900

950

1000

1050

1100

1150

-0+100 0+000 0+100 0+200 0+300 0+400 0+500 0+600 0+700

Elev

atio

n (m

, am

sl)

Station (m)

900

950

1000

1050

1100

1150

1200

900

950

1000

1050

1100

1150

1200

-0+100 0+000 0+100 0+200 0+300 0+400 0+500

STORMWATER DIVERSION CHANNEL ABOVE DSTF(REF DETAIL 7-104)

DSTF 2: COMPACTED CEMENTAMENDED FILTERED TAILINGS


EXISTING GROUND

BENCH (TYP).(REF. DETAIL 2-103)

12.5

5%

COLLECTION TANK(REF. DETAIL 5-104)

KEY SUBGRADE FOUNDATIONWITH BENCHES

(REF. DETAIL 1-103)

LINED TOP DECK STORMWATER MANAGEMENT POND



LINED TOP DECK STORMWATERMANAGEMENT POND


(REF. DETAIL 1-103)UNDERDRAIN (TYP).(REF. DETAIL 4-103)

12.5

PLANT

EXISTINGGROUND

5%

CONCEPTUAL DSTF PLAN VIEW AND SECTIONS

A

N

FILE NAME: 101-DSTSF-2_544400.020_20200729.dwg

DRAWING TITLE:

consultingPROJECT:

PREPARED BY:


544400.070-800

DRAWING NO.DATE:

SRK JOB NO.:

C:\Users\jsames\Documents\SRK\GCG\544400-020_Marmato_PEA_PFS\040_Drafting\Task_810_PFS_Design\101-DSTSF-2_544400.020_20200729.dwg





REV. NO.:


1010m

METERS

50m 100m 150m

DSTF 2 PLAN VIEW AND SECTIONS(SCALE - 1:5)

DSTF 2: SECTION A(SCALE - 1:5)

DSTF 2: SECTION B(SCALE - 1:5)

CAU

CA

RIV

ER

MARMATO RIVERNATIONAL HIGHWAY

C'

CD'


RIPRAP APRON

DSTF-1 ACCESS ROADCENTERLINE (REF. DETAIL 6-104)

STORMWATER DIVERSION CHANNELSOUTH (REF. DETAIL 7-103)

STORMWATER DITCH (TYP.)(REF. DETAIL 7-104)

BENCH DETAIL (TYP.)(REF DETAIL 2-103)

TRANSMISSION TOWER (TYP.)

HIGH VOLTAGE POWER LINE (TYP.)

UNDERDRAIN (TYP.)(REF DETAIL 4-103)


DTEMPORARY TOP DECK

STORMWATERMANAGEMENT POND

STORMWATER DIVERSIONCHANNEL NORTH (REF. DETAIL

7-104)

DSTF 1

4x1m CULVERTSTORMWATERCROSSING

RIPRAP APRON

EL LLANOTOWNSHIP

2:1

2.5:1

2.5:1

2:12:1

2:1

-5%

-5%

-5%

86087

0870

83084

0850

810

820

86087

0

83084

0850

800810820

870

900925950

775

800

825

700

725

750 750

750

775

800

800

825

850

875

900

900

825

850

875

900

900

925

850

875

850

875

900

725

750775800

825

700

725750

775

800

Elev

atio

n (m

, am

sl)

Station (m)

700

750

800

850

900

950

1000

700

750

800

850

900

950

1000

0+000 0+100 0+200 0+300 0+400 0+500 0+600 0+700 0+750

Elev

atio

n (m

, am

sl)

Station (m)

700

750

800

850

900

950

1000

1050

700

750

800

850

900

950

1000

1050

-0+100 0+000 0+100 0+200 0+300 0+400 0+500

STORMWATER DIVERSION CHANNEL SOUTH(REF DETAIL 7-104)


ROCK STARTER EMBANKMENTEXISTING GROUND


12

5%



(REF. DETAIL 1-103)

LINED TOP DECK STORMWATER MANAGEMENT POND

STORMWATER DIVERSION CHANNEL NORTH(REF DETAIL 7-104)

EXISTING GROUND

12.5

12

12.5




(REF. DETAIL 1-103)

UNDERDRAIN (TYP).(REF. DETAIL 4-103)

CONCEPTUAL DSTF 1 PLAN VIEW AND SECTIONS

A

N

FILE NAME: 101-DSTSF-2_544400.020_20200729.dwg

DRAWING TITLE:

consultingPROJECT:

PREPARED BY:


544400.070-800

DRAWING NO.DATE:

SRK JOB NO.:

C:\Users\jsames\Documents\SRK\GCG\544400-020_Marmato_PEA_PFS\040_Drafting\Task_810_PFS_Design\101-DSTSF-2_544400.020_20200729.dwg





REV. NO.:


1020m

METERS

50m 100m 150m

DSTF 1 PLAN VIEW AND SECTIONS(SCALE - 1:5)

DSTF 1: SECTION C(SCALE - 1:5)

DSTF 1: SECTION D(SCALE - 1:5)

TOP DECK 5% MIN. SLOPE

BENCH (TYP.)(REF. DETAIL 2-103)


HORIZ. DRAIN 5% MIN SLOPE

UNDER DRAIN

EXISTING GROUND

SHEAR KEY


EMERGENCYOVERFLOW

TOP DECK CONTACT WATER STORAGE

FILTERED TAILINGS

SUBGRADE BENCHING

10.0m

FILTEREDTAILINGS

5.0m

VARIES

CLOSURE COVER

HORIZONTAL DRAIN EVERY 10m VERTICAL

100mm PERF. COLLECTION DRAIN

SUBGRADE BENCHING EXISTING GROUND

GRADE BENCH TO NEARESTUNDERDRAIN AT 2% MIN.

4.0m

12

12

1.0m

5% MIN.

1m CLOSURE COVER

FILTERED TAILINGS

6.5m

STORMWATERDITCH

(REF. DETAIL 7)

1m Min.

1.5m Min.

MIN. AREA 7m2FILTER SAND COVER

EXISTING GROUND

FILTER FABRIC

600mm PERF. PIPE

100mm MINUSDRAIN ROCK

0.25m0.25m

0.25m

DRAIN ROCK FILTER FABRIC

FILTER TAILINGS

100mm PERFORATED PIPE

FILTER SAND COVER

CONCEPTUAL DSTF DETAIL SHEET 1

A103

FILE NAME: 103-Detials-2_544400.020_20200729.dwg

DRAWING TITLE:

consultingPROJECT:

PREPARED BY:


544400.070-800

DRAWING NO.DATE:

SRK JOB NO.:

C:\Users\jsames\Documents\SRK\GCG\544400-020_Marmato_PEA_PFS\040_Drafting\Task_810_PFS_Design\103-Detials-2_544400.020_20200729.dwg





REV. NO.:


SCALE - 1:2000DSTF TYPICAL SECTION1

SCALE - 1:100DSTF BENCH TYPICAL SECTION2

SCALE - N.T.SHORIZONTAL DRAIN - TYPICAL SECTIONS3

SCALE - N.T.SUNDERDRAIN TYPICAL SECTION4

12

12

10.0m

2.9mVOL = 40,000 LITERS

INFLOW

EMERGENCYOUTFLOW

PUMP TO PLANT

INFLOW

EMERGENCYOUTFLOW

PUMP TO PLANT

PRECAST CONCRETE TANK

CL

SAFETY BERM15.5m

STORMWATERDITCH

EXISTING GROUND

1.5

1

10.5

300mm ROAD WEARING LAYER

SLOPE STABILITY ANCHOR(ASSUMED 15 OF ROAD)

Xm

Hm

CL

EXISTING GROUND

CONCRETE LINED

1

1

0.5

1

Xm

Hm

CL

EXISTING GROUNDRIPRAP LINED

1

0.5

1

0.51

0.5

1

0.5

CL

CONCEPTUAL DSTF DETAIL SHEET 2

A104

FILE NAME: 103-Detials-2_544400.020_20200729.dwg

DRAWING TITLE:

consultingPROJECT:

PREPARED BY:


544400.070-800

DRAWING NO.DATE:

SRK JOB NO.:

C:\Users\jsames\Documents\SRK\GCG\544400-020_Marmato_PEA_PFS\040_Drafting\Task_810_PFS_Design\103-Detials-2_544400.020_20200729.dwg





REV. NO.:


SCALE - 1:100CONTACT WATER TRANSFER TANK5

SCALE - 1:200ACCESS/ HAUL ROAD TYPICAL SECTION6

SCALE - 1:100STORMWATER MANAGMENT CHANNELS/ DITCHES7

SCALE - N.T.S.PRE-CAST CONCRETE SPAN BRIDGE8

CONCRETE LINED DIVERSION CHANNEL(REF. DETAIL 7-104)

WINGWALLSPAN

CONCRETE FOOTING


MMS/KD August 2020

Appendix C: Economic Model Snapshots

Period Start 1‐Jan‐20 1‐Jan‐21 1‐Jan‐22 1‐Jan‐23 1‐Jan‐24 1‐Jan‐25 1‐Jan‐26 1‐Jan‐27 1‐Jan‐28Period End 31‐Dec‐19 31‐Dec‐20 31‐Dec‐21 31‐Dec‐22 31‐Dec‐23 31‐Dec‐24 31‐Dec‐25 31‐Dec‐26 31‐Dec‐27 31‐Dec‐28

Total0 4 8 12 16 20 21 22 234 4 4 4 4 1 1 1 1

Gold Price US$/oz 1,400 1,400 1,400 1,400 1,400 1,400 1,400 1,400 1,400 Silver Price US$/oz 17.00 17.00 17.00 17.00 17.00 17.00 17.00 17.00 17.00

Upper ZoneTonnes milled tonnes 5,144,663 286,171 437,485 490,050 525,001 491,534 388,538 452,030 409,596 351,864 Head grade (g/t Au) g/t 4.16 3.66 3.99 3.95 3.97 3.96 4.39 4.30 4.34 4.37 Contained gold (ozs) ounces 687,339 33,668 56,077 62,288 67,074 62,579 54,875 62,529 57,171 49,475 Gold Recovery % 87% 87% 87% 87% 87% 87% 87% 87% 87% 87%Gold produced (ozs) ounces 598,939 29,338 48,865 54,277 58,448 54,531 47,817 54,487 49,818 43,112 Silver produced (ozs) ounces 846,780 51,450 67,774 68,375 83,310 85,300 70,465 76,045 66,625 54,200

Revenue (net of refining) US$ 849,088,606 41,760,769 69,251,505 76,804,266 82,870,375 77,444,895 67,836,735 77,227,101 70,560,392 61,002,941 Opex US$ (391,620,486) (24,798,781) (36,247,543) (39,831,170) (40,016,749) (34,589,723) (28,758,672) (32,310,055) (29,933,224) (26,755,887) Royalties US$ (78,116,152) (3,841,991) (6,371,138) (7,065,993) (7,624,075) (7,124,930) (6,240,980) (7,104,893) (6,491,556) (5,612,271) Sustaining Capital US$ (59,546,192) (13,276,174) (15,170,406) (9,448,862) (7,303,250) (4,526,700) (617,550) (742,950) (463,550) (463,550) Working capital adjustments US$ ‐ (2,037,295) (646,058) (598,695) (262,565) (251,342) 584,058 (479,916) 352,593 532,084 Pre‐tax Cashflow US$ 319,805,777 (2,193,471) 10,816,361 19,859,547 27,663,737 30,952,201 32,803,591 36,589,286 34,024,655 28,703,317

AISC US$/oz 866

MDZTonnes milled tonnes 14,555,946 ‐ ‐ ‐ 209,014 1,310,667 1,461,570 1,460,649 1,461,122 1,465,178 Head grade (g/t Au) g/t 2.85 ‐ ‐ ‐ 6.05 12.63 3.27 3.41 2.94 2.77 Contained gold (ozs) ounces 1,332,795 ‐ ‐ ‐ 20,831 132,913 153,659 160,137 138,110 130,485 Gold Recovery % 10 ‐ ‐ ‐ 95% 95% 95% 95% 95% 95% Gold produced (ozs) ounces 1,266,155 ‐ ‐ ‐ 19,789 126,268 145,976 152,130 131,204 123,961 Silver produced (ozs) ounces 719,609 ‐ ‐ ‐ 12,509 74,608 84,019 91,104 85,309 76,124

Revenue (net of refining) US$ 1,776,772,632 ‐ ‐ ‐ 27,791,633 177,237,280 204,863,671 213,560,206 184,299,357 174,048,543 Opex US$ (831,172,697) ‐ ‐ ‐ (25,988,296) (82,587,852) (83,444,458) (82,447,740) (82,834,207) (83,068,605) Royalties US$ (163,463,082) ‐ ‐ ‐ (2,556,830) (16,305,830) (18,847,458) (19,647,539) (16,955,541) (16,012,466) Initial Capital US$ (269,394,005) (1,087,625) (109,116,079) (112,024,363) (47,165,937) ‐ ‐ ‐ ‐ ‐ Sustaining Capital US$ (131,299,895) ‐ ‐ ‐ (19,492,087) (13,352,799) (8,471,341) (9,434,155) (25,219,945) (10,025,079) Working capital adjustments US$ ‐ ‐ ‐ ‐ (2,561,975) (6,866,359) (551,328) (796,706) 2,436,766 882,229 Pre‐tax Cashflow US$ 381,442,953 (1,087,625) (109,116,079) (112,024,363) (69,973,493) 58,124,440 93,549,086 101,234,066 61,726,430 65,824,623

AISC US$/oz 886

Combined

Tonnes milled tonnes 19,700,609 286,171 437,485 490,050 734,015 1,802,201 1,850,108 1,912,679 1,870,718 1,817,042 Head grade (g/t Au) g/t 3.19 3.66 3.99 3.95 3.72 3.37 3.51 3.62 3.25 3.08 Contained gold (ozs) ounces 2,020,134 33,668 56,077 62,288 87,905 195,492 208,534 222,666 195,281 179,960 Gold Recovery % 13 87% 87% 87% 89% 92% 93% 93% 93% 93%Gold produced (ozs) ounces 1,865,094 29,338 48,865 54,277 78,237 180,798 193,793 206,617 181,023 167,073 Silver produced (ozs) ounces 1,566,389 51,450 67,774 68,375 95,819 159,908 154,484 167,150 151,934 130,324

Cash cost/oz US$/ozAISC/oz US$/oz

Revenue (net of refining) US$ 2,625,861,238 41,760,769 69,251,505 76,804,266 110,662,008 254,682,175 272,700,406 290,787,307 254,859,749 235,051,483 Opex US$ (1,222,793,183) (24,798,781) (36,247,543) (39,831,170) (66,005,045) (117,177,575) (112,203,130) (114,757,795) (112,767,430) (109,824,492) Royalties US$ (241,579,234) (3,841,991) (6,371,138) (7,065,993) (10,180,905) (23,430,760) (25,088,437) (26,752,432) (23,447,097) (21,624,736) Income taxes paid US$ (210,374,619) ‐ (3,803,295) (7,420,753) (8,009,594) (4,963,110) (23,972,210) (30,063,723) (33,951,529) (24,456,658)

Period Start 1‐Jan‐20 1‐Jan‐21 1‐Jan‐22 1‐Jan‐23 1‐Jan‐24 1‐Jan‐25 1‐Jan‐26 1‐Jan‐27 1‐Jan‐28Period End 31‐Dec‐19 31‐Dec‐20 31‐Dec‐21 31‐Dec‐22 31‐Dec‐23 31‐Dec‐24 31‐Dec‐25 31‐Dec‐26 31‐Dec‐27 31‐Dec‐28

TotalWorking capital adjustments US$ ‐ (2,037,295) (646,058) (598,695) (2,824,539) (7,117,701) 32,730 (1,276,622) 2,789,358 1,414,313 Operating cash flow US$ 951,114,202 11,082,703 22,183,472 21,887,657 23,641,924 101,993,029 111,469,359 117,936,734 87,483,052 80,559,910 Sustaining Capital US$ (190,846,086) (13,276,174) (15,170,406) (9,448,862) (26,795,337) (17,879,499) (9,088,891) (10,177,105) (25,683,495) (10,488,629) Free cash flow US$ 760,268,116 (2,193,471) 7,013,066 12,438,795 (3,153,413) 84,113,531 102,380,468 107,759,629 61,799,556 70,071,282

495,563,815 Expansion capex US$ (269,394,005) (1,087,625) (109,116,079) (112,024,363) (47,165,937) ‐ ‐ ‐ ‐ ‐

Project cash flow US$ 490,874,111 (3,281,096) (102,103,013) (99,585,569) (50,319,350) 84,113,531 102,380,468 107,759,629 61,799,556 70,071,282

Project NPV @ 5% US$ 256,075,253

Project IRR % 19.5%

Cash Cost per Ounce US$/oz 777 953 855 849 959 769 701 678 745 780 AISC US$/oz 880 1,405 1,165 1,023 1,302 868 748 727 886 843

Period StartPeriod End

Gold Price US$/ozSilver Price US$/oz

Upper ZoneTonnes milled tonnesHead grade (g/t Au) g/tContained gold (ozs) ouncesGold Recovery %Gold produced (ozs) ouncesSilver produced (ozs) ounces

Revenue (net of refining) US$Opex US$Royalties US$Sustaining Capital US$Working capital adjustments US$Pre‐tax Cashflow US$

AISC US$/oz

MDZTonnes milled tonnesHead grade (g/t Au) g/tContained gold (ozs) ouncesGold Recovery %Gold produced (ozs) ouncesSilver produced (ozs) ounces

Revenue (net of refining) US$Opex US$Royalties US$Initial Capital US$Sustaining Capital US$Working capital adjustments US$Pre‐tax Cashflow US$

AISC US$/oz

Combined

Tonnes milled tonnesHead grade (g/t Au) g/tContained gold (ozs) ouncesGold Recovery %Gold produced (ozs) ouncesSilver produced (ozs) ounces

Cash cost/oz US$/ozAISC/oz US$/oz

Revenue (net of refining) US$Opex US$Royalties US$Income taxes paid US$

1‐Jan‐29 1‐Jan‐30 1‐Jan‐31 1‐Jan‐32 1‐Jan‐33 1‐Jan‐3431‐Dec‐29 31‐Dec‐30 31‐Dec‐31 31‐Dec‐32 31‐Dec‐33 31‐Dec‐34

24 25 26 27 28 291 1 1 1 1 1

1,400 1,400 1,400 1,400 1,400 1,400 17.00 17.00 17.00 17.00 17.00 17.00

386,605 388,892 445,184 91,713 ‐ ‐ 4.30 4.18 4.43 4.22 ‐ ‐

53,418 52,299 63,442 12,443 ‐ ‐ 87% 87% 87% 87% 0% 0%

46,548 45,573 55,282 10,843 ‐ ‐ 59,176 59,945 82,869 21,244 ‐ ‐

65,876,165 64,529,962 78,451,441 15,472,058 ‐ ‐ (28,549,711) (28,657,425) (30,512,417) (10,659,130) ‐ ‐ (6,060,607) (5,936,757) (7,217,533) (1,423,429) ‐ ‐ (463,550) (463,550) (463,550) (6,142,550) ‐ ‐ (260,792) 119,500 (991,766) 3,940,194 ‐ ‐

30,541,505 29,591,730 39,266,176 1,187,143 ‐ ‐

1,461,099 1,460,382 1,460,041 1,463,850 1,342,374 ‐ 2.47 2.33 2.50 2.79 2.84 ‐

116,029 109,399 117,354 131,308 122,570 ‐ 95% 95% 95% 95% 95% ‐

110,228 103,929 111,486 124,743 116,441 ‐ 53,928 58,409 56,517 60,618 66,464 ‐

154,532,483 145,830,586 156,329,699 174,874,555 163,404,620 ‐ (81,760,362) (82,536,536) (81,375,837) (78,072,993) (67,055,810) ‐ (14,216,988) (13,416,414) (14,382,332) (16,088,459) (15,033,225) ‐

‐ ‐ ‐ ‐ ‐ ‐ (18,779,831) (10,052,664) (11,679,257) (1,512,011) (3,280,726) ‐ 1,476,102 779,020 (958,341) (1,773,963) 7,934,554 ‐

41,251,403 40,603,991 47,933,932 77,427,129 85,969,414 ‐

1,847,704 1,849,274 1,905,225 1,555,563 1,342,374 ‐ 2.85 2.72 2.95 2.87 2.84 ‐

169,447 161,698 180,795 143,751 122,570 ‐ 93% 92% 92% 94% 95% 0%

156,776 149,502 166,768 135,586 116,441 ‐ 113,104 118,354 139,387 81,863 66,464 ‐

220,408,648 210,360,548 234,781,140 190,346,613 163,404,620 ‐ (110,310,074) (111,193,962) (111,888,253) (88,732,123) (67,055,810) ‐ (20,277,596) (19,353,170) (21,599,865) (17,511,888) (15,033,225) ‐ (19,173,263) (14,724,222) (11,421,910) (18,011,461) (9,571,242) (831,650)

Period StartPeriod End

Working capital adjustments US$Operating cash flow US$Sustaining Capital US$Free cash flow US$

Expansion capex US$

Project cash flow US$

Project NPV @ 5% US$

Project IRR %

Cash Cost per Ounce US$/ozAISC US$/oz

1‐Jan‐29 1‐Jan‐30 1‐Jan‐31 1‐Jan‐32 1‐Jan‐33 1‐Jan‐3431‐Dec‐29 31‐Dec‐30 31‐Dec‐31 31‐Dec‐32 31‐Dec‐33 31‐Dec‐34

1,215,310 898,520 (1,950,107) 2,166,231 7,934,554 ‐ 71,863,025 65,987,713 87,921,005 68,257,372 79,678,898 (831,650) (19,243,381) (10,516,214) (12,142,807) (7,654,561) (3,280,726) ‐ 52,619,644 55,471,500 75,778,198 60,602,811 76,398,172 (831,650)

‐ ‐ ‐ ‐ ‐ ‐

52,619,644 55,471,500 75,778,198 60,602,811 76,398,172 (831,650)

827 866 793 780 702 ‐ 950 936 865 836 730 ‐

Revised NI 43-101 Technical Report Pre-Feasibility Study ...

Documents

Transcript of Revised NI 43-101 Technical Report Pre-Feasibility Study ...