sim HOJEN

Hydrogeological and HydrogeochemicalInvestigations in Boreholes—Final reportof the phase I geochemical investigationsof the Stripa groundwatersD.K. Nordstrom, US Geological Survey, USAJ.N. Andrews, University of Bath, United KingdomL Carlsson, Swedish Geological Co, SwedenJ-C. Fontes, Universite Paris-Sud, FranceP. Fritz, University of Waterloo, CanadaH. Moser. GesellschaftfurStrahlen-undUmweltforschung, West GermanyT. Olsson, Geosystem AB, Sweden

July 1985

TECHNICAL REPORTAn OECD/NEA International project managed by:SWEDISH NUCLEAR FUEL AND WASTE MANAGEMENT CODivision of Research and Development

Mailing address:Box 5864, S-102 48 Stockholm, Telephone: 08-67 95 40

HYDROGEOLOGICAL AND HYDROGEOCHEMICAL INVESTIGATIONS IN BOREHO-LES - Final Report of the Phase I Geochemical Investigations ofthe Strlpa Groundwaters

D.K. Nordstrom, editorUS Geological Survey, USA

J.N. AndrewsUniversity of Bath, United Kingdom

L. CarlssonSwedish Geological AB, Sweden

J.-C. FontesUniversite Paris-Sud, France

P. FritzUniversity of Waterloo, Canada

H. MoserGesellschaft fur Strahlen- und Umweltforschung, West Germany

T. OlssonGeosystem AB, Sweden

July 1985

This report concerns a study which was conducted for the StripaProject* The conclusions and viewpoints presented in the reportare those of the authors and do not necessarily coincide withthose of the client.

A list of other reports published in this series is attached atthe end of the report* Information or previous reports is avail-able through SKB.

ABSTRACT

The hydrogeochemical investigations of Phase I of the Stripa Pro-ject (1980-64) have been completed, and the results are presentedin this final report» All chemical and isotopic data on thegroundvaters from the beginning of the Stripa Project to the pre-sent (1977-84) are tabulated and used in the final interpreta-tions» The background geology and hydrology is summarized and up-dated along with new analyses of the Stripa granite* Water-rock in-teractions form a basic framework for the changes in major-ele-ment chemistry with depth, Including carbonate geochemistry, thefluid-inclusion hypothesis, redox processes, and mineral precipi-tation* The irregular distribution of chloride suggests channel-ling is occurring and the effect of thermomechanical perturba-tions on the groundwater chemistry is documented* Stable and ra-dioactive isotopes provide information on the origin and evol-ution of the groundwater itself and of several elements withinthe groundwater* Subsurface production of radionuclldes is docu-mented in these investigations, and a general picture of uraniumtransformations during weathering is presented. One of the prima-ry conclusions reached in these studies is that different dissol-ved constituents will provide different residence times becausethey have different origins and different evolutionary historiesthat may or may not be related to the overall evolution of thegroundwater itself.

Ii

CONTENTS

Page

ABSTRACT i

SUMMARY vii

1 INTRODUCTION 1:11.1 General 1:11.2 Test sites 1:2

2 GEOLOGY 2:12.1 Introduction 2:12.2 Major lithologic units 2:12.3 The Stripa ore 2:22.4 Petrology 2:42.5 Chemical composition of the Stripa granite 2:92.6 Fracture minerals 2:112.7 Fluid Inclusions 2:142.8 Radiogeology 2:172.9 Structure 2:20

3 HYDROLOGY 3:13.1 Hydraulic units 3:13.2 Porosity of the intact rock material 3:23.3 Hydraulic conductivity of the rock mass 3:43.3.1 General 3:43.3.2 Testing techniques 3:53.3.3 Results 3:73.4 Hydraulic head 3:83.5 Model calculations 3:93.6 Dewatering of the granite 3:12

4 GROUNDWATER CHEMISTRY 4:14.1 Introduction 4:14.2 Methods of sample collection and preservation 4:14.3 Methods of analysis 4:24.4 Accuracy and precision 4:34.5 Chemical analyses 4:204.6 Saline groundwaters In central Sweden: Regional program 4:204.7 Distribution of salinity with location and time 4:21

lii

Page

4.8 Ion ratios and classification of groundwater 4:284.8.1 Seawater 4:284.8.2 Baltic seavater 4:294.8.3 Yoldia seawater 4:294.8.4 Jurassic-Cretaceous sedimentary basin brines of Skloe 4:304.8.5 Permian evaporates 4:314.8.6 Stripa groundwaters 4:314.8.7 Dissolved organics and microbial life 4:32

5 WATER-ROCK INTERACTIONS 5:15.1 Introduction 5:15.2 Chloride correlations and mineral reactions 5:15.2.1 Alkali metals: Na, K, Li 5:15.2.2 Alkali-earth metals (Ca, Mg, Sr, Ba) and 5;5

carbonate geochemistry

5.2.3 Halogens (F, Cl, Br, I) and fluorite solubility 5:125.2.4 Alumlnosilicate reactions 5:145.2.5 Iron chemistry and redox potencials 5:15

5.2.6 Sulfate 5:175.3 Chemical geothermometers 5:185.3.1 Ca/Mg and Mg/Cl ratios 5:195.3.2 Na-K-Ca and s i l i ca geothermometers 5:225.3.3 The Mg/Li geothermometer 5:235.3.4 Conclusions from geothernometry 5:245.4 The fluid-inclusion hypothesis and rock-leaching studies 5:2^5.4.1 Introduction 5:245.4.2 Fluid-inclusion measurements and volumetric 5;26

considerations5.4.3 Preliminary fluid-inclusion leaching study 5; 275.4.4 Microfractures: porosity and salinity measurements 5;275.4.5 Literature survey of halogens in granites 5;29

6 OXYGEN-18 AND DEUTERIUM CONTENTS 6 : 16.1 Introduction 6:16.2 General considerations 6:16.2.1 Conservative stable isotope contents 6:16.2.2 Non-conservative stable isotope contents 6:76.3 Discutslon 6:86.4 Regioi.-jl saline groundwater survey 6:146.5 Conclusions 6:16

iv

Page

7 STABLE ISOTOPE GEOCHEMISTRY OF SULPHUR COMPOUNDS 7:17.1 Introduction 7:17.2 Variations in the stable isotope contents of sulphur 7:1

compounds7.2.1 Redox systems 7:47.2.2 Sulphide and H2S oxidation 7:47.2.3 Oxygen isotope equilibrium between sulphate and water 7:57.3 Sampling and analyses 7:67.4 Results and discussion 7:77.4.1 Aqueous sulphate in subsurface waters 7:77.4.2 Aqueous sulphate in deep waters 7:97.4.3 Discussion of the various possible origins 7:10

of the aqueous sulphate7.4.4 In-situ reactions involving sulphur compounds: 7:13

V 2, N 1, M 3 and R 17.4.5 Regional survey of saline groundwaters 7:14

8 THE CARBON AND OXYGEN ISOTOPIC COMPOSITIONS OF 8:1AQUEOUS CARBONATE AND CALCITES

8.1 Introduction 8:18.2 Sampling and analyses 8:18.2.1 Aqueous carbonate 8:18.2.2 Fracture calcite 8:28.3 Aqueous carbon 8:28.3.1 Shallow groundwaters 8:38.3.2 Deep groundwaters 8:58.4 Carbon-14 measurements 8:78.4.1 Shallow groundwaters 8:98.4.2 Deep groundwaters 8:108.5 Fracture calcites 8:128.6 Conclusions 8:20

9 THE IN-SITU PRODUCTION OF RADIOISOTOPES AND THE 9:1^ AND 36C1 CONTENTS OF THE GROUNDWATERS

9.1 The in-situ neutron flux in the Stripa granite 9:19.1.1 Neutron production due to U spontaneous 9:1

fission and (a, n) reactions9.1.2 Neutron flux measurements with a BF3 neutron counter 9:39.1.3 Results from neutron flux measurements 9:79.2 In-sltu production of 3H and 36C1 in the Stripa granite 9:79.2.1 3H production 9:79.2.2 36C1 production 9:9

Page

9.3 Tritium contents in groundwaters at Stripa 9:9

9.3.1 Introduction 9:99.3.2 Measuring techniques 9:119.3.2.1 Liquid scintillation counting 9:119.3.2.2 Gas counting 9:119.3.2.3 Electrolytic enrichment 9:139.3.3 Results 9:139.3.4 Discussion 9:169.3.A.I Surface waters, shallow waters and groundwaters 9:16

from southern Sweden9.3 .4 .2 Stripa nine waters from different boreholes 9:179.3.5 Conclusions 9:189.4 Chlorine-36 in the Stripa groundwaters 9:219.4.1 Atmospheric sources of *6C1 9:219.4.2 36C1 in groundwaters due to cosmogenic 36C1 fallout 9:22

9.4.3 Possible use of 36C1 for groundwater studies 9:229.4.4 36C1 contents of the Stripa groundwaters 9:23

9.4.5 Implications for groundwater residence tine 9:26

10 RADIOELEMENTS IN THE STRIPA GRANITE AND GROUNDWATERS 10:110.1 Radioelement content of the Stripa granite 10:110.1.1 Uranium series equilibria in the Stripa granite iQ:210.2 Uranium and thorium solution by groundwaters 10:410.2.1 Analytical method for U and Th isotopes in solution 10:610.2.2 U-solution and 234U/238U activity ratio in the 10:7

shallow groundwaters10.2.3 Uranium chemistry in the deeper groundwaters 10:910.2.4 Thorium in solution 10:1310.3 Radon and radium solution in groundwaters 10:1510.3.1 Analytical method for 222Rn and 226Ra determinations 10:1710.3 .1 .1 222Rn 10:1710 .3 .1 .2 2 2 6Ra 10:1810.3 .2 2 2 2Rn contents of Stripa groundwaters 10:1810.3.3 2 2 6Ra contents of Stripa groundwaters 10:21

11 ATMOSPHERIC AND RADIOGENIC GASES IN SOLUTION 11:111.1 Atmosphere derived gases 11:111.2 Radiogenic helium 11:111.3 Radiogenic argon 11:211.4 Biogenlc gases 11:311.5 Analytical methods 11:311.5.1 Sampling for inert gas analyses 11:311.5 .2 Isotope d i lut ion analys is of dissolved inert gases 11:511.6 Radiogenic helium 11:611.7 A 0Ar/3 6Ar rat ios 11:JO11.8 Recharge temperatures 11:1211.9 Nj/Ar rat ios 11:1311.10 l 5 N/ 1 4 N rat ios 11:14

vi

Page

12 CONCLUSIONS 12:1

13 ACKNOWLEDGEMENTS 13:1

14 REFERENCES 14:1

vii

SUMMARY

The geochemistry of the groundwaters, bedrock and fracture mi-neralogy at the Stripa test site is being investigated to under-stand the origin and evolution of groundwaters in a granitic bed-rock. Geochemical parameters provide an important constraint tothe hydrogeologic properties of groundwater flow and c an pieraentphysical investigations such as geophysical measurements of frac-tures, hydraulic testing and tracer migration studies. Geochemi-cal studies provide the only measure of long-term migration ofsolutes and water in the subsurface environment» These investiga-tions have contributed substantially to our understanding of geo-logically both old and modern water-rock-gas interactions occur-ring within crystalline bedrock.

Several lines of evidence strongly suggest that the groundwatersystem at Stripa has evolved from fresh meteoric waters, typicalof central Sweden, interacting with the Proterozoic crystallinebedrock composed dominantly of feldspars and quartz and fracture-fill minerals, such as calcite, chlorite, epidote, sericite, py-rite, fluorite, and hematite. Several hypotheses may account forthe source of the Na-Ca-Cl type water found at depth, (a) salt as-sociated with the crystalline rock itself, i.e., fluid inclu-sions and associated grain boundary salts or salty fluids, (b)intrusion of old seawater and (c) leaching of salts of sedimenta-ry origin. All of these hypotheses are given careful considera-tion. Many of the water-rock interactions can be related to weat-hering processes and solubility equilibria such that a firmbasis for predicting the effect of perturbations, like radioacti-ve waste storage, can be made with greater reliability. For ex-ample, thermal stress will clearly affect the water chemistryand could actually Increase chloride concentrations significant-ly in the near-field by extruding saline fluids from the micropo-res and/or microfracturing fluid inclusions» Increased salt con-centrations can both increase and decrease the solubility of va-rious minerals, depending on the mineral, the temperature, andthe composition of the salt components» Changes in solubilitycan, In turn, affect the permeability of the bedrock.

Identification of active processes, such as calcite, fluorite,ferric hydroxide, and possibly barite precipitation, provides fa-vorable conditions for radionuclide retardation in the far-fieldby coprecipitation or adsorption. When these processes are link-ed to other processes, such as silica dissolution and reprecipi-tation through a temperature gradient in the near-field, possib-le clay mineral formation, and the absorbing properties of thebackfill, then the outlook for long-term radioactive waste stor-age looks even more favorable.

1:1

INTRODUCTION

1.1 General

The program "Hydrogeologlcal and Hydrogeochemlcal Investigations

in Boreholes" within the Stripa Project has the following objec-

tives

- Methodology development for hydrogeological and hydrogeochemi-cal investigations in subsurface horizontal and vertical bore-holes.

- Instrumentation and equipment development In subsurface hori-

zontal and vertical boreholes.

- Hydraulic, chemical and isotopic characterization of the Stri-

pa granite and groundwaters.

The work was carried out according to a defined program (Carls-son and Olsson, 1981) which was slightly revised during its per-formance. The program included several activities that describedthe hydrogeological characteristics of the Stripa granite andthe hydrogeochemical properties of the Stripa groundwaters.These activities included

- Drilling of two vertical boreholes (VI and V2) and two horizon-tal boreholes (Nl and El).

- Core logging with respect to rock type, fracturing, fracture

minerals and fracture orientation.

- Core Investigation for porosity and density.

- Geophysical logging for deviation, radiation, temperature, nor-mal and lateral resistivity, differential resistance, caliperand density.

- Geophysical cross-hole measurements for fracture zone tracing.

- Hydraulic pressure build-up and fall-off-tests.

- Water Injections tests.

- Hydraulic interference tests»

- Hydraulic head monitoring.

1:2

- Hydrogeochemical sampling and analysis with respect to majorand minor constituents, trace elements, stable and unstableisotopes and noble gases»

In the vertical holes, priority was given to the hydrogeochemi-cal studies and only a minor program for the hydraulic testingwas carried out* The program for hydrogeochemical sampling, ana-lyses and evaluation was slightly revised as a second deep bore-hole was included. The Hydrogeochemical Advisory Group (HAG),consisting of experts in several areas of hydrogeochemistry, wasformed to advise and to make analyses and interpretations on thehydrogeochemistry. The members in this group are

John Andrews UKErik Eriksson SwedenTad Florkowski IAEA

Jean-Charles Fontes FrancePeter Fritz CanadaHeinz Loosli Switzerland

Heribert Moser West GermanyKirk Nordstrom USA, chairman

This document is the final report of the hydrogeochemical inves-tigations carried out within the Stripa Project during 1980-84.All analytical data collected during the previous investigations(under the auspices of the Swedish-American Cooperative Program,SAC, 1977-80) have also been Included so that all the data andInterpretations could be found in a single document.

1.2 Test sites

Boreholes at two selected sites in the Stripa mine have beenused to make measurements and collect water samples for the pro-gram. At the SGU-site, which is the main site located at the 360m mine level, three boreholes were drilled, one vertical (VI)and two subhorizontal boreholes (Nl and El). The fourth borehole(V2) is an extension of the old borehole Dbh VI that was made du-ring the SAC program. It was deepened by drilling to a totaldepth of 822 m (1230 m below ground surface). The location ofthe sites and the boreholes are shown in Figure 1-1. Data on theboreholes are given in Table 1-1. In addition to these bore-holes, a number of other holes in the SAC-area were used for minortests and water sampling. Water sampling was also conducted inprivate wells at the surface and from surface water schemes.

1:3

Figure 1-1. The investigation areas in the Stripa Mine withthe boreholes used for water sampling.

Table 1-1. Data on the main boreholes included in the hydro-

geological program.

Bh no

VIV2NlEl

Diameter

USD

76567676

Collar

X

336.B270342.2338.4

coordinates

Y

1195.710751194.61199.7

Z

356.407.355.355.

7757

Length

m

505.9822.0300300

2:1

GEOLOGY

2.1 Introduction

Field investigations for the International Stripa Project arecarried out at the Stripa Mine, located in Örebro County, south-central 9weden, (59C'43'N, 15°5'30"W), 250 tan west-northwest ofStockholm. The Stripa Mine lies within the Central Swedish OreProvince ("Mellansverige") consisting of more than a hundrediron-manganese and copper-zinc-lead mines. Mining at Stripabegan around the year 1450 for iron ore (mostly specular hemati-te). The mining was intermittent but continued until 1976 whennegotiations were made to keep the mine open as a research site.

The topography is hilly with the elevation at Stripa about 140 mabove sea level. The Stripa area was slightly inundated by seawa-ter during the Holocene since the highest paleo-shoreline isfound at 170 m above sea level. Glacial sediments and tillscover most of the region but bedrock is exposed on the tops ofhills. To the northwest, the elevation increases steadily by se-veral hundred meters. Maps of the bedrock geology, structuralgeology, and a magnetic survey have been published on a scale of1:50,000 (Koark and Lundström, 1979). Recent geological investi-gations have been reported by Olkiewicz, et al. (1978, 1979) andWollenberg, et al. (1980). This section summarizes these previo-us studies and contains some additional geological data obtainedin the course of the hydrogeochemical investigations.

2.2 Major lithologic units

The bedrock geology in the Stripa region consists of highly fol-ded and deformed Precambrian rocks; primarily metasediments andmetavolcanics intruded by several granitic bodies. The metamor-phic rocks occur in a northeast-southwest-trending direction andare dominated by high-grade, silica-rich schists and gneisses(Figure 2-1). They also include a metamorphosed carbonate sec-tion (calcareous and dolomitic marble) occurring about 3 km fromthe Stripa site and varying in outcrop width from 50 m up to 1.7km. Granitic intrusions range in size from less than 100 n to se-veral tens of kilometers. Their compositions range from a truegranite to granodiorite, although aplites, pegmatites, small gab-broic and amphibollte dikes all occur in the region (Koark andLundström, 1979). Their occurrence is somewhat erratic and irre-gular. The granitic bodies appear greyish white, grey, reddish-grey, and red in color, and are commonly medium grained in tex-ture.

2:2

PEGMATITE ( Serorogemc )

STRIPA GRANITE ( Serorogenic )

METABASJT£ LMK£

GRANITE ( Synorogen:c)

SEDIMENTARY OR VOLCANIC -METAS0MAT1C ROCKS I Svecokorelion )

CRYSTALLINE LIMESTONEI Svecokarelian )

METAVOLCANJTES ( LEPTITE )( Svecokorelian I

Figure 2-1. Geologic features of the Stripa area.

2.3 The Stripa ore

Stripa ore was, according to old registrations, first mined du-ring 1448-1470 and 1551-1578. After years of no activity miningwas again reestablished 1634-1771 (Paulsryd, 1941). In the 1780smining started in larger scale, but it was first during the pre-sent century that the mining went deep underground and as lateas 1930-1960 the deeper parts were mined.

2:3

The ore deposits occured mainly as two bodies called the Mainand the Parallel ore body, respectively. The Main body with amaximum thickness of 16 to 17 meters was folded Into a sync linewith an undulating eastward pitch. Both the ore bodies were sur-rounded by leptite.

The iron ore is mainly a quartz-banded hematite, but magnetiteoccurs also. The skarn minerals associated with the ore aremainly actinolite, diopside and epidote. Pyrite occurs locallyand is evidently of secondary origin (Geijer, 1938). The iron-ore in the Main ore body has a higher content of Fe than the orein the Parallel body. In Table 2-1 the general composition ofthe two ore bodies is given.

Table 2-1. General composition of the two ore bodies at Stri-pa (Paulsryd, 1941).

Body Fe P S

Main 50 0.007 0.016Parallel 41 0.009 0.037

The hematite of the ore bands developed as grains of 0.2 - 0.6mm diameter, slightly elongated in the plane of bedding. Theporphyroblastic magnetite can reach up to 15 mm, but the normalsize is from 0.5 to 5 mm. Most of the magnetite ore In Stripaseems to have occured in connection with a secondary enrichmentprocess that resulted in a very rich, coarsely crystalline magne-tite ore (Geijer, 1938). Most of the high-grade magnetite occurswithin a marked syncline, from which fault zones diverge stri-king ENE.

The type of folding and associated faulting movements that is re-presented by the Stripa deposit is typical of what could be enco-untered in large portions of the ore-bearing region of CentralSweden. On the whole the folding of the Stripa deposit has beencomparatively small. There wa3 little interior deformation with-in the ores, in spite of the normally incompetent character ofthe quartz-banded ores.

The faulting movements that apparently accompanied the foldingexhibited a variety of types. One type appeared to be accentua-tions of folds, through slipping along contact planes. Echelondisplacements in the different ore bodies belonged to this typealso. Usually these displacement zones striked E, that is Inslight angle to the fold-axis. Another related type is interpre-

2:4

ted as due to shearing movements. Movements along well-defin-ed, steeply dipping fault planes were represented by most dis-turbances of this group in the main ore body. They usually wereparallel to the fold axis. Movements along these faults appearedto have been in the form of overthrusts and there were, in somecases, indications that the horizontal component was greaterthan the vertical one (Geijer, 1938).

Later sets of faults, separated in age from those described, areassociated by suc'i geological events as the intrusions of basicrocks and of granite aplite. These faults were usually steep dip-ping and their orientations differed. However, the most commonorientation was almost perpendicular to the fold axis and also to-wards NE. Usually the displacements had a larger vertical thanhorizontal component. According to Geijer (1938), the verticalcomponent was usually net exceeding a few meters.

2.4 Petrology

The target rock for the investigations in the Stripa Mine is asmall intrusive body of granite - Stripa granite, that predomi-nantly is a grey to reddish, medium-grained rock of Precambrianage. The Stripa Granite occurs at the surface in a belt of oldersupracrustal metamorphic rocks. The largely concordant nature ofthe granite is not uncommon. Many postorogenetic granites in theStripa region have been mapped as elongated intrusions parallelto the structures of the supracrustal belts (Koark and Lund-ström, 1979).

Leptite, a strongly metamorphosed sedimentary rock, normally of

volcanic origin, is the dominant rock type in the supracrustal

formation. The regional distribution of the rock types is shown

in Figure 2-1.

The main features of the configuration of the contact betweenthe leptite syncline and the granite is illustrated in Figure 2-2, based on the data obtained from the mine workings and investi-gations in the SAC-program. The contact between the leptites andthe granite is transected by the access drift to the hydrogeolo-gical test site at the 360 m level, approximately 300 m SSE fromthe ventilation shaft. The granite at the contact occurs partlyas inclusions or dikes in the leptite. The granite surrounds theleptites in the Stripa syncline in the northeastern part of themine. The limits of the subsurface extension of the granite tothe SE is partly shown by the prospecting boreholes Pjt 3 andPjt 3B as shown in the vertical section in Figure 2-2. This sec-tion is taken perpendicular to the contact, i.e. in a northwest-southeast direction. The location of the section is indicated inFigure 2-3.

2:5

-10001200

0 200 400 600

1 - f | Monzogranite

l ^ ^ l Mico gneiss

Kg^'3 l«P"'« (quartjose metovol-Vs^y.'A camtc with iron are)

Boreholes

.A. (riveted m the section plane

^ . projected on the section plane

1200 UOOm

Geological boundary

defined

uncertain

Figure 2-2. Vertical section through the investigation area.The location of the section is shown in Figure 2-3.

The petrology of the Stripa granite was studied by Olkiewicz, £tal. (1978, 1979), Koark and Lundström (1979) and Wollenberg etal. (1980). In these reports the granitic intrusion is namedquartz monzonite, monzogranite or granite. The technically cor-rect name for this intrusion is "granite" based on both the mi-neralogy and the chemical composition.

Igneous rocks are classified petrographically where possible andchemically where not possible by petrographic techniques. Classi-fication of high-silica, wholly-crystalline igneous rocks isbased on the proportions of the three essential minerals:quartz, alkali feldspar and plagioclase feldspar. The Stripa gra-nite contains 30-40% quartz, 25-35% plagloclase and 18-34% mic-rocline. An additional 5-10% Is muscovite and chlorite (alteredbiotite). The average of 6 modal analyses from Wollenberg, £tal. (1980) and one sample by R Donahoe, US Geological Survey,is shown In Table 2-2. Plotting this data on the ternary dia-gram in Figure 2-4 shows that the rock falls well within the gra-nite field according to the Streckeisen classification systemwhich has been adopted by the International Union of GeologicalSciences (Streckeisen, 1973, 1976). It has been adopted also bymost modern workers in igneous petrology (Barker, 1981). Further

2:6

o oo o o

O 700 400 »00 m

STR1PA GRANITE

SEDIMENTARY OR VOLCANIC-METAS0MAT1C ROCKS

METAV0LCAN1TE(Leptite witti iron ort t

AREAS Of EXPOSED BEDROCKWITH PREDOMINATINGFRACTURE PATTERN

j?' SYNCLINAL AXIS OF THE*" LEPTITE SYNFORM

* ? LINE Of VERTICAL' ' SECTION

• 200 >. MINE COORDINATES

MORPHOLOGICALLINEAMENT

DISLOCATION (AT MOM LEVEL

VENTILATION SHAFT

OTTES DISLOCATION AT 175 M LEVEL

Figure 2-3. Major structures In the Strips area.

chemical evidence for this classification is given in the nextsection.

Optical and X-ray studies by R Donahoe (pers. comm.) and Wollen-berg, et al. (1980) indicate that the plagioclase is oligoclase(An2y) in composition. The plagioclase grains are sericitlzedand the microcline grains are frequently perthitic or micropert-hitic. The microcline is commonly interstitial to the quartz andplagioclase. The quartz appears optically biaxial due to strainand are highly fractured. Hematite is In some places dispersedas fine dust within the feldspar grains, particularly plagiocla-

2:7

Table 2-2. Analyses from 7 samples of Strida granite*.

QuartzPlagloclase feldspar

K-feldsparMuscovlte/sericiteChlorite

A1

35.630.025.75.03.0

i?

353225

3

•Accessory minerals (opaques, carbonates, epidote, fluorite, zir-con, garnet and apatite) are typically <1%. Biotite has mostlybeen altered to chlorite. Column A is the average of 6 modal ana-lyses from the data of Wollenberg, et al. (1980). Column B is asingle modal analysis of VI core, interval 149.2 - 149.5 m obtain-ed by R. Donahoe, US Geological Survey.

se, or along grain boundaries and cracks within grains. The redcolor in many of the granite samples is due to the occurrence ofhematite (Wollenberg, et al., 1980).

The chlorite appears in two varieties - one is strongly pleo-chroic, dark green-black to light brown and the other is weaklypleochroic, light to dark green. The former variety appears tobe incompletely chloritized biotite whereas the other occursalong grain boundaries and as microfracture veinlets, suggestingmore complete alteration.

Fluid inclusion measurements are discussed in Section 2.7.

The Stripa granite typically shows an abundance of fractures,both continous and discontinous on a microscopic scale. Even inrelatively unfractured rock samples fine, discontinuous crackswithin primary grains or along grain boundaries are common.These cracks are filled with intergrown chlorite and sericite orby quartz and feldspars and they frequently originate among pri-mary grains of the same minerals as those filling the cracks.This suggests that the crack fillings have not crystallized fromfluids introduced from extraneous sources, but are due rather toreroobilizatlon and »deposition of primary components of therock matrix (Wollenberg, et al», 1980). Veins or dikes of peg-matites and aplites are common in the granite.

Another distinctive feature of the Stripa granite is the preva-lence of cataclastic textures. There are evidences of movementsalong fracture surfaces or breccia zones. Slickensides and frac-tures are filled with a microscopically irresolvable clay-richfault gouge and contain rounded fragments of granitic rock (Wol-lenberg, et al., 1980).

2:8

.90

Granites -//•»•>• ; ; ; fe-V^i\, Granodiorites

20// pQuartz monzonite* Quartz monzodiorites

Figure 2-4. Classification of silica rich plutonic rocks accor-ding to the Streckeisen system. Compositions arenormalized to A (alkali-feldspar) + P (plagiocla-se) + Q (quartz) = 100% by volume assuming thatmafic minerals are less than 90%. The range of themodal analyses of the Stripa pluton fall withinthe range for granites.

Samples of Stripa granite and neighboring granitic rocks (Gussel-by and Kloten massifs) were dated by the potassium-argon methodat the University of California, Berkeley (Wollenberg, et al.,1980). The dates obtained were in millions of years 1691+16,1604+14 and 1640+44 for Stripa, Gusselby and Kloten respective-ly. The leptite is considered to have a similar origin and is be-lieved to be slightly older than the granite but no dates havebeen obtained.

The leptite is usually a gray, red or grey-green to black, fine-grained foliated metamorphic rock (microschist) cut by white orlight green fractures. Mineralogically it is similar to the Stri-pa granite. Texturally, however, it does not resemble the grani-te, as it is finer, more even-grained and homogeneous. In de-tail, the leptite generally consists of a fine, even-grained mosa-ic of equant quartz with fewer plagioclase and microcline gra-ins. Darker leptites generally contain more chlorite, at the ex-pense of muscovite, than the lighter leptites, or they containfewer porphyroblasts.

The contacts of leptite with Stripa granite was studied by Wol-lenberg, et al. (1980), in thin sections. The contacts are gene-rally sharp, possibly fault contacts. They show little sign ofalteration such as might be expected if the granite had intrudedthe leptite. The growths of sericite, or epidote-chlorite-filledfractures occur at the contact, in some places also opaque grainsiich in uranium are associated with the contact.

2:9

Another variety of metamorphic rock at Stripa is a dark green fi-ne-grained foliated rock rich in blue-green prismatic amphibole,usually logged as greenstone in core-logs. Texturally, it is amicroschist like the more abundant leptite. However, its foliationis determined by amphibole prisms instead of chlorite and rausco-vite laths.

2.5 Chemical composition of the Stripa granite

Three samples were analyzed (ore in duplicate) by X-ray fluores-cence and reported by Wollenberg, et al. (1980). Seven sampleswere analyzed (all in duplicate) at the U.S. Geological Surveyusing X-ray fluorescence, emission spectroscopy and ion-selec-tive electrodes depending on the constituent. Of the latter sev-en samples, four were taken from the VI borehole at depths of107, 408 and 445 m from the borehole origin (355 m below sur-face) and three were taken from the V2 borehole at depths of456, 471 and 760 m. The chemical analyses of the seven boreholesamples are for fresh unfractured rock and are reported in Tab-le 2-3. They are in excellent agreement with those reported byWollenberg, et al- (1980), except for silica which is about 3%higher. Since the total sum of all constituents is also 3% hig-her (102%) than the sum in Wollenberg's report (99%) then it islikely that the silica values reported in Table 2-3 are biased abit too high. In any case, silica is always greater than 72% andwith K2O falling in the range of 4-5%, the rock must be classi-fied as a granite (Barker, 1981). The CIPW normative calculationused by Wollenberg, et al. (1980), is useful for basalts and ex-perimental systems but not for natural granites.

Based on the chemical data given in Table 2-3 the Stripa granitecan be further classified as a peraluminous granite by the crite-ria AI2O3 > Na20 + K2O + CaO (Barker, 1981). In an effort to cor-relate the composition of granites to their mode of origin, re-cent emphasis has been placed on those derived by partial mel-ting of sedimentary rocks, the S type, versus those derived frompre-existing igneous rocks, the I type. The S types are biotite-musrovite-bearing and strongly peraluminous whereas I types areblotite or biotite-hornblende bearing and metaluminous (Barker,1981). The Stripa granite is clearly an S type which is consis-tent with the concept of partial melting of volcanic sediments inthe leptite.

Certain minor and trace elements (Cl, F, S, Li and B) were determin-ed to assist in the interpretation of fluid inclusion measure-ments, sources of sulfur and neutron flux studies. There is no-thing unusual in the concentrations of these elements comparedto most granite.-;.

2:10

T a b l e 2 ~ 3 U.S.C.S. Analyse* of Strip* Granite.*

S102 J

A12O3 J

F.2O3 J

MgO

CaO

N.JO i

K2O :

T10 2 J

P2°5 'MnO J

C l )

T :

S(total) !

U(pp»)B(ppa)

V2-*56

[ 76.3

C 14.0E 1 . 5

E .24

C 1.0

[ 4.1

E 4.9

; .08

E .09

E .05

E .018

E .052

E .03

1.8

3

V2-471

76.9

13.7

1.2

.26

.72

3.9

3.0

.08

.09

.05

.013

.074

.02

2.8

3

V2-760

76.3

14.0

1.4

.25

.87

4.0

4.6

.09

.09

.05

.018

.061

<.O1

28

3.5

Vl-107

75.9

14.0

1.2

.26

.80

4.0

4.6

.08

.08

.07

.015

.036

<.O!

8.4

5

Vl-408

76.6

13.9

1.2

.24

.40

4.0

4.6

.09

.09

.05

.012

.019

<.O1

4.5

4

Vl-408.03

78.1

14.0

1.4

.28

.43

4.0

4.7

.08

.10

.05

.020

.023

<.O1

4.6

4

Vl-445

77.5

14.2

1.4

.30

.76

4.4

4.1

.09

.10

.06

.018

.024

<.O1

5.5t

Average

76.8 + .75

U.O + .21

1.33 + .IB

.26 + .02

.72 + .22

4.07 + .22

4.62 + .29

.08 + .

.09 + .01

.06 + .01

.016 +.004

.044 +.021

£.037.9 + 8.73.8 + 0.9

* Analysti: J Cllllson, H. Rait, J. Fletcher, R. Johnson. Results froa 7 sample» done In duplles-te. Averages sre for 14 determinations of each constituent with one standard deviation. Notethat rock samples arc froa clean, unfractured sections of core, and the analyses represent freshgranite. Higher concentrations of sosie constituents will be encountered In highly fracturedrocks, especially where fracture-itll ulnerals occur.

Rare earth elements were determined on eight granite samplesat Pierre Siie Laboratory in Saclay, France, by instrumental neut-ron activation analysis (NAA). The countings were performed us-ing high resolution Ge-Li detectors, and low energy photon acti-vities were measured by an intrinsic germanium detector, espe-cially used for the Gd determination. The data are reported inTable 2-4 with duplicate analyses for each depth in VI and V2boreholes. Analytical precision is about 57. for each measurement.

T a b l e 2—4 Rare earth element concentrations (ppm) In Strips granite.

Sample

VI 259

VI 318

VI 373

V2 804

.4 •

.4 a

.3 a

.6 a

U

2925

3433

2538

3534

.5

.0

.5

.5

.5

.0

.5

.5

Ce

72.058.0

70.069.0

56.581.0

72.871.0

H6

49.535.3

47.048.0

38.557.5

49.748.5

Sa

14.:11. .

13.14.

11. '17.

15.?16.

Eu

0.39i 0.25

Z 0.23S 0.39

0.30i 0.59

1 0.452 0.25

Cd

11.89.5

11.513.5

9.511.5

13.512.0

Tb

2.01.8

2.02.5

1.52.0

3.02.5

Yb 1

7.06.5

7.58 . 0

5.67.5

8.58.0

»u

.2

• '

.3

.4

.0

.3

.3,4

Total

188149

1R9190

150215

200194

Rare earth elements plus Ba, Co, Cr, Cs, Hf, Rb, Sb, Ta, Th, U,Zn, Zr and Sc were determined on 14 granite samples by NAA atthe U.S. Geological Survey. These results are shown in Table 2-5.

2 :11

T a b l e 2 - 5 NAA Analyse* of Strip» Granite

Element

(ppn)

&a

Co

Cr

Ca

Hf

Rb

Sb

Ta

Th

U

Zn

Sc

La

Ce

Nd

Sm

Eu

fid

Tb

Tm

Yb

LJ

V2-456

511

0.67

<1

4.3

3.7

261

0.17

7.9

31

31

19

4.5

27

70

35

7.B

0.45

13

2.1

1.1

8.0

1.2

V2-471

659

0.64

1.8

4.3

3-7

307

0.13

8.0

31

33

11

4.5

26

66

35

7.4

0.41

13

2.1

1.0

8.2

1.2

V2-760

571

0.76

1.0

9.1

4 .0

282

0.23

6.8

36

35

22

5.5

31

80

42

9.1

0.50

14

2.4

1.0

9.1

1.4

Vl-107

513

0.71

1.6

3.1

3.7

271

0.27

8.8

32

32

23

4.4

26

67

33

7.6

0.42

13

2.1

1.0

8.1

1.2

Vl-408

474

0.64

1.2

2.6

3.8

245

0.12

7.8

34

26

16

4.7

28

73

36

8.4

0.44

12

2.1

1.0

8.3

1.2

Vl-408.03

525

0.73

2.7

3.1

3.8

249

0.17

7.9

34

31

21

5.2

29

75

39

9.1

0.46

14

2.3

1.0

8.8

1.2

Vl-445

565

0.81

1.7

3.4

3.9

265

0.22

7.8

34

33

19

4.9

32

80

44

9.4

0.47

15

2.3

1.2

8.6

1.3

Average

545+60

0.71+0.06

1.7+0.6

4.3+2.2

3.8+0.12

269+21

O.19+.05

7.9+0.6

33+1.9

32+2.8

19+4.1

4.8+0.4

28+2.4

73+5.7

38+4.1

8.4+0.8

0.45+0.03

13+1.0

2.2+0.13

1.0+0.08

8.4+0.4

1.2+0.08

Rare earth elements distributions in geological samples areoften described by normalized patterns. In igneous rocks, the re-ference sample is a chondritic ratio. The chondritic-normalizedpattern of the granite is characterized by a light-REE enrich-ment, and by a strong Eu depletion (Figure 2-5). According to

the alkali content of the granite (Wollenberg et al.., 1980),

these REE distribution curves are characteristic of similar gra-nites (Henderson, 1984).

2.6 Fracture minerals

All boreholes within the program show similar characteristics re-garding the existing fractures. Detailed fracture logs are givenin previous reports on the core logs (Carlsson, et al., 1981;Carlsson, et al., 1982a and 1982b). The recorded fractures maybe classified into one of five different groups;

1 Fractures with fresh, uneven surfaces

2 Open or sealed fractures

2:12

l I

Lo C« Pr Nd Er T» Yb Lo

Figure 2-5. Rare earth abundance of the Stripa granite relati-ve to the normalized chondrite values (see Haskinand Haskin, 196fi; Haskin et al., 1966).

3 Small-scale shear zones

4 Brecciation and granulation of the granitic matrix

5 Quartz veins

Except for the fractures included in the first group, all othersare characterized by the existence of coating minerals or weat-hering indications on the fracture surfaces. The fracture fill-ing minerals were megascopically classified on the basis of col-our, hardness and appearence of carbonates. As pointed out byWollenberg, et al. (1980), it is normal that different fracturefilling minerals are intergrown in varying combinations whichmakes the megascopical classification somewhat uncertain. Anextra check was therefore made by using X-ray diffraction onfive samples taken from the V2 core.

The result of this test showed that the plagioclase mixing withepidote was much more common than expected. The conclusion whichmay be drawn from the result is that plagioclase in general havebeen underestimated.

2:13

Chlorite, which is the most common fracture filling mineral, isvery dark, almost black, and much harder than normal due to mix-ing with epidote and plagioclase. Also, the epidote shows manytypes of colour variations in the green colour spectrum whenmixed with plagioclase.

Sericite, commonly intergrown with chlorite, is nearly as commonas chlorite. Next to chlorite anH sericite, calcite is the mostcommon fracture filling mineral. Its occurrence ranges from fil-lings of hairline cracks and thin coatings and intergrowns withother minerals to coarse crystals grown in the spaclngs of largefractures.

Epidote occurs commonly in fractures, veins and shear zones,which are sealed in the core, associated with quartz, chloriteand sericite. Other fracture filling minerals identified in thecores include pyrlte, chalcopyrite, fluorite, iron oxides andzinc sulphide. The great majority of the fracture infillings areless than 1 mm in width.

Borehole V2 penetrates the most deep-seated rock mass and itwas therefore of Interest to study the variation in mineral coat-ings versus depth. The result of this study is summarized in Fi-gure 2-6, where it is seen that the coating of chlorite and chlo-rite/calcite-mixing are about constant throughout the full lengthof the borehole. The most striking change with depth is that thecalcite shows a marked decrease at 250 - 450 m depth with a si-multaneous increase in epidote. Each of the chlorite, calcite andepidote coatings make up about 25 - 30 per cent of all coatedsurfaces. The group of other minerals form a complex group withgreat variety. Pyrlte, fluorite, iron oxides, zinc sulphide andclay are examples of coatings within this group.

The mentioned conditions are generally in agreement with the re-sults reported by Wollenberg, et al. (1980), which Is based onmegascopical classifications, X-ray diffraction analyses and ana-lyses of thin sections. An additional observation is the rarebut noteworthy occurrence of asphaltite as fracture fill in theStripa granite (Wollenberg, et al., 1980). Nothing is knownabout the origin of this organic material and it was found tocontain high concentrations of uranium.

Additional studies on fracture minerals were reported by T. D.Reimer as part of the geochemical investigations during the SACprogram (Fritz, et al.t 1980). Chalcopyrite was found in onesample and some analyses of muscovite, sericite, Ca-mica, chlo-rite, feldspar, pyrite, chalcopyrite, calcite and dolomite(?) are reported. The compositional data for chlorites indicatethat there are two varieties: an Iron-rich member (FeO/MgO • 7)and a raagnesium-rlch member (FeO/MgO • 0.9). This observationis consistent with the optical information on chlorites in therock matrix that indicates two species of different composition.

2:14

Figure 2-6. Distribution of coating minerals in V2 versus depth.

2.7 Fluid Inclusions

Fluid inclusions are microscopic droplets of aqueous solution,silicate or sulfide melt, gas or organic material, that occur inmost rocks and minerals. They commonly range in size between 1and 10 micrometers (observed range is 20 nanometers up to a volu-me of 100 milliliters). The number of inclusions per unit volumevaries inversely with the size of the Inclusion, but commonlylies in the range of 10 inclusion/cm (maximum observed is 10in a sample of white quartz). They typically contain 10 wt. %dissolved salts with NaCl being most abundant, although concen-trations range from 0 to 50%. Many careful and detailed measure-ments have been made on fluid inclusions in the last several de-cades because of their usefulness in interpreting the origin ofore deposits. Most measurements are made on quartz, fluorite,calcite and sulfide ore minerals. Freezing point, melting point,and homogenization temperatures can be measured on the inclusionfluids, and, under the appropriate conditions, these can indica-te the composition and temperature of the fluid in contact withthe rock when it was formed (Roedder, 1984).

2:15

Fluid inclusions may affect the deep groundwater chemistry ofcrystalline rocks (Garrels, 1967; Gambell and Fisher, 1966;Jacks, 1978; Nordstrom, 1983), and this suggestion, supported bygroundwater chemistry data, prompted an investigation of fluid in-clusions in the Stripa granite.

Lindblom (1984) completed a study of fluid inclusions on bothfractured and megascopically unfractured rocks from VI and V2cores. Freezing point, melting point, homogenization point tempe-ratures were measured, and the population density of inclusionsin quartz grains were measured. The main conclusions are:

o o

1 The number of inclusions per cm varies from 0.5-9 x 10 .

2 The total volume of inclusions averages 14 pL per cm ofquartz in unfractured rock and 20 yL/cnr (20 L/nr) in fractur-ed rock.

3 The salinity of the fluid inclusions in the unfractured rocksections gives a mode of 4 wt. % NaCl and in the fracturedrock sections, 1.7 wt. % NaCl. Unfractured rocks have aslightly lower volume of more saline inclusion fluid thanthe fractured rocks.

4 Homogenization temperatures vary between 90°C and 270°C andrepresent the temperatures of rehealing of fractures inquartz.

5 A modal homogenization temperature of 130°C represents anearlier event in unfractured rock sections. A later reheal-ing event is represented by another modal horaogenizationtemperature at 190°C in fractured rock sections.

6 The rock may have been flushed by deep-circulating meteoricwaters at a possible late date (something younger than seve-ral million years).

7 Total salt content in quartz inclusion fluids is 5.6 x 10~*g/cro for unfractured rocks and 3.4 x 10"^ g/cm (340 g/m )for fractured rocks.

8 If the same fluid-inclusion content for the whole rock is as-sumed, then the percentage of fluid inclusions needed to mixwith fresh groundwater to achieve 650 mg/L is only 1-2%, as-suming static, or closed system, conditions. If the worst as-sumptions are made (e.g., that inclusions only occur inquartz, etc»), then the maximum of fluid inclusions neededmight be as high as 10%.

9 The concentration of water-soluble chloride is estimated tobe .0132 wt. % for unfractured rock sections, and .0078 wt.1 for fractured rock sections, assuming that feldspars havethe same number and type of inclusions as the quartz grains.

2:16

From the total Cl analyses shown, this represents 50 - 80%of the total Cl (although the uncertainty is high on the es-timates of water-soluble chloride because the measurementgives "equivalent wt. % NaCl" not a direct reading).

10 Fluid inclusions in the Stripa granite appear to be dominant-ly or completely secondary and cover a wide range of tempera-ture and compositional conditions. These data indicate thatnumerous thermal and tectonic events took place since theoriginal emplacement of the granite.

The results of Lindblom's study shows that there is morethan sufficient fluid-inclusion salt in the Stripa graniteto account for the salinity of the groundwater, assuming sta-tic or near-static groundwater conditions and a porosity of1%.

An independent measurement of chloride concentrations inthe inclusion fluids was made by a published procedure forextracting fluid inclusions (Roedder, 1958; Roedder, et^al. , 1963). Samples were fragmented, cleaned, evacuated ina vacuum line for several days and then crushed under vacu-um. The extracted water was measured and its H content ana-lyzed (see Table 2-6). Then the samples were sequentially ex-tracted with a small volume of distilled water for 5 or 6times until the Cl concentration was no longer detectable.The calculated Cl concentrations in the inclusion fluidsare shown in Table 2-7. The average value of 41 g/L Cl isgreater than the estimate made by measuring freezing/mel-ting temperatures, but the range of values overlaps. If therock leaching value is taken for the average Cl concentra-tion of the fluid inclusions in quartz only, then the water-soluble Cl becomes 50 - 77% of the total. The leaching esti-mate of fluid-inclusion chloride is suspected of being bia-sed too high because fluid inclusions are likely to occur inthe feldspars (Roedder, 1972), and because some of the valu-es give a higher rock chloride concentration than found bytotal rock analysis (Table 2-3). Therefore, the estimate offluid-inclusion chloride taken from the freezing/meltingtemperatures is considered more reliable.

Table 2-6. H data on fluid inclusions from the Stripa granite.

Drillhole

VIVIVIV2V2V2V2

Dep

408445445363404455471

Depth (meters) 2H °/oo, SMOW

-58.8-68.5-86.5-79.8-73.7-52.5-74.5

2:17

Table 2-7. Chloride concentrations of fluid inclusions by di-

rect analysis of leachates.

Cl(g/L)

Vl(445)VI(445)V2(404)V1(5O5)VI(408)V2(455)V2(363)V2(471)Average

445447341646671841

2.8 Radiogeology

The abundance of radioelements in the rocks was measured by Wol-lenberg, jet aL. (1980). The fission-track radiographic methodwas used to determine the location and abundance of uranium inuncovered thin sections.

The Stripa granite is rather unique in its radioelement content,both in the abundance of elements and their ratios. Table 2-8 in-dicates the relatively high uranium and thorium contents of thegranite, compared with other plutons in the region.

The measurements indicate that uranium is depleted in surface ex-posures of granite and leptite at Stripa, relative to its abun-dances in the same rock units underground.

In the Stripa granite, uranium is roost highly concentrated intiny opaque grains on the order of 50 micrometer in diameter,generally euhedral and In some places square in cross-section.These grains are usually found in chlorite, but also in muscovi-te-chlorite-sericite filled fractures, and even in cracks withinquartz or feldspar» Usually the grains contain up to 5% uranium,bi't concentrations up to 10-15% have been observed (Wollenberg,et al., 1980). Another locus of uranium concentration was obser-ved in opaque grains with both a quartz-epidote-sericite-filledfracture on a contact between granite and leptite, and with finecarbonate-sericite stringers intersecting that contact on thegranite side. Although the concentration of uranium is lower inthese grains, on the order of 2% U, the absolute abundance ofuranium contained in them is greater.

Uranium is also found, in lower concentrations, dispersed alongchlorite-filled fractures without associated discrete grains.

2:18

T a b l e 2 - 8 Radloeleaent content* (after WoUenberg «• »1. . 1980).

Rock type

Stripa granite

SurfaceUnderground

Lcpttte

SurfaceUnderground

Regional rock*

Cranlte*M*ta«orphic

No.

934

59

75

UrantuBPP»

26.9+5.537.4+6.2

3.3+0.75.4+3.1

17.6+15.46.1+1.5

TtiorluaPP"

33.0+5.29.2+3.

11.9+2.17.9+1.

26*6+4*14.6+8.

78

94

67

FotaeduaX

4.6+0.73.9*0.3

3.1+0.62.8+0.5

5.2+1.52.5+1.1

Th/U

1.1+0.10.8+0.1

3.6+0.43.9+1.2

2.4+1.22.6+1.9

Table 2-9. Radiogenic heat production of the rock in the Stri-pa region (Wollenberg, etal., 1980).

Rock type No. Heat produc-tion uW/m3

Stripa graniteSurfaceUnderground

Leptite

SurfaceUnderground

Regional rocks

GranitesMetamorphic

934

59

75

9.5

11.9

2.02.9

6.8 to 7.12.8

Concentrations are generally about 0.5% or lower, but occasional-ly range up to 1.0% uranium.

The Stripa leptite contains no appreciable discrete concentra-tion of uranium either in the matrix or in a coarse epidote-fil-led fracture cutting it (Wollenberg, et al., 1980). Uranium mine-rals were observed in chlorite-filled fractures cutting the ironore at Stripa (Welin, 1964).

The heat production was calculated from the radioelements of theStripa pluton by Wollenberg, et al. (1980). In Table 2-9 the ra-diogenic heat production of the various rocks In the Stripa re-gion are listed.

2:19

The Strtpa granite averages 11.9 microWatts per cubic meter.This should be compared with 2.8, considered to be the mean forgranitic rocks (Heier and Rogers 1963), 6.9 uW/m^ for the Bohusgranite of southwestern Sweden (Landström, et al., in prep) and7.1 pW/m for the Malingsbo granite just north of Stripa (Malm-qvist, et al., 1983). The radiogenic heat production of the Stripagranite is four times that of the neighboring leptite and nearlytwice the heat production of other plutons in the region.

Geophysical well logging played an important part in the currentprogram for hydraulic, structural, stability and chemical purpo-ses. A standard geophysical well-logging program was set up andperformed in the four main boreholes. This program compiled thefollowing logs:

Type of log Purpose

Deviation log

Natural gamma

Single point resistivity

Resistivity logs

Temperature log

Drillhole fluid

Self potential log

Direction and deviation of the

borehole

Rock type, dykes, veins andfracture indications

Resistivity of the rock in theborehole wall i.e. conductingminerals and fractures

Fracture indications

In- and outflow zones in the

hole

In- and outflow zones in thehole resistivity

Measures anomalies which indi-

cate fracture zones

Table 2-10. Radiation level in the boreholes El, Nl, VI andV2.

Average

VIV2ElNl

65100117250

microR/hmicroR/hmicroR/hraicroR/h

Peak

406 microR/h250 microR/h

430 microR/h

2:20

200 400 600 800

DEPTH BELOW 360 M MINE LEVt .1000

Figure 2-7. Temperature versus depth In boreholes VI and V2.

The results of natural gamma logs presented in Table 2-10 showlarge differences between the holes. The high level in Nl isdue to the radon content in combination with the very low wateroutflow. The radiation level in the granite Is most accuratelygiven by the values from VI where the water outflow is high.

The temperature of the water in the boreholes VI and V2 is givenin Figure 2-7. The influence on the temperature from the driftcan be seen down about 100 m in V2. From there the temperaturegradient is about 17 degreesC/km down 480 m where it decreasesto around 15 degreesC/km. From 610 m the gradient Increases to18 degreesC/km down to the bottom of the hole. In VI the in-fluence of the drift is not recognized. Instead there is a hig-her temperature than in V2 due to the outflowing water emanatingfrom the borehole below 460 m depth. At the bottom of VI the tem-perature is 19.1 degreesC and at the corresponding level in V2the temperature is 19.6 degreesC.

2.9 Structure

In Figure 2-3, the major structures in the Stripa area are visua-lized. The location of the section (Figure 2-2 above) Is also In-cluded together with existing boreholes made from the ground sur-face. As regards the lineaments In the granite, it f.s seen thattheir direction generally is parallel to the syncline axis ofthe supracrustal formation.

The greater morphological lineaments diverges in direction fromthe syncline geometry, but may, however, also be governed by theconfiguration of the supracrustal rocks. Two dominant directionsoccur for the lineaments, i.e. ENE-WSW and NW-SE.

Based on both surface boreholes (SBH1 and SBH2) and subsurfaceholes (VI and V2), the variation in fracture frequency versusdepth was studied. The variation with depth is shown in Figure 2-8. It must be noted that since VI and V2 both are vertical, me-

2:21

Fracture frequency(fr/m)

0 5 10 ££>:>01—»—•—»—r«—N—•—•—±—i—•-

200

Figure 2-8. Fracture frequency versus depth based on core-logsfrom SBHl, SBH2, VI and V2. The frequency is assum-ed to be log-normally distributed.

dium-steep and steep fractures will be underestimated. The frac-ture frequencies obtained in Nl and El are also included in thefigure. These boreholes are more accurate measures of the steep-ly dipping fractures at the 360 m level.

A number of zones of fractured or crushed rock also exists inthe granite. Normally these zones are thin, not exceeding 1 m inthe cores, but a few zones are of several meters in thickness. Amore extensive zone was found in the lowermost part of VI. Tecto-nically less disturbed granite in the upper part of the boreholeextends down to the 466 m depth and contains more widely spacedfracture zones and crushed zones usually less than 1 m in width.Fracturing tends to be more intense towards the bottom of thissection, with a prominent Increase in number of subvertlcal frac-tures.

A detailed compilation of fracturing is impractical for thestrongly crushed part of the borehole (466 m down to the bottomof the borehole at 505 o). Totally 7.7 u of this section is dis-connected or crushed to rubbles. The number of the fractures inthe crushed zone is partly based on an estimation (38 per centfrom totally 510 fractures within this 40 m wide zone) and theirdipping were not possible to establish. The fracture frequency

2:22

Figure 2-9. Vertical section through the investigation area.

was 12.9 fr/m in the zone to be compared to 1.5 fr/m for therock mass above the zone.

The fracture zone in VI has a high water inflow. The hydraulicconductivity is high in comparison to the rock mass and the zoneis assumed to be of crucial importance for the groundwater systemin the granite. However, the extension and orientation of thezone could be interpreted according to different possibilities.

One of the possible interpretations of this zone is Indicated inFigure 2-2. However, the zone was not found in V2 when this bore-hole was deepened down to a final depth of 820 m (1 210 m belowground surface). A cross-hole electrical measurement was made be-tween VI and V2 and the result indicated that the major zonefound in VI was connected through three (possible four) minorzones intersected by V2. None of these zones showed, however, afracturing in accordance with that found In VI.

Figure 2-9 shows a profile through the rock mass with the bore-holes VI and V2 in relation to the ore body. In this section afracture zone found during the ore mapping is included. Itsorientation Is well defined in and around the ore body while Itsextension through VI and V2 is hypothetical. The assumed exten-

2:23

Table 2-12. Fracture frequency in VI, V2, Nl and El.

Borehole Fracturefrequency

VI (above the crushed zone) 1.5VI (crushed zone) 12.9V2 2.1Nl 1.6El 4.7

sion is, however, indicated by the intense fracturing in the low-ermost part of VI, but also of a somewhat more intense fractur-ing in the uppermost part of V2.

This gives three probable explanations of the geometry of thefracture zone found in VI, none of which is more reliable thanthe others.

1 A zone striking N70E and dipping 60SE as indicated in Figure2-2.

2 A zone striking NW-SE steeply dipping to NE as indicated in

Figure 2-9.

3 The major zone is connected to V2 by a number of minor zones

as indicated by geophysical measurements.

The actual interpretation may also be a combination between any

of the mentioned possibilities.

The mean fracture frequencies for the boreholes included in thehydrogeological program are given in Table 2-11.

Figure 2-10 shows a cumulative fracture diagram for V2 with re-gard to the dipping of the fractures. It is seen that mediumsteep fractures dominate while steeply dipping fractures have alow fracture frequency. Flat-lying fractures are in an intermed-iate position. This is in full agreement with the result obtain-ed in VI (Carlsson, et al., 1981). It must be stressed that thevertical borehole V2 tends to underestimate vertical or steeplydipping fractures while sub-horizontal or flat-lying fracturesare recorded with their actual frequency. With this in mind, itis clearly seen from Figure 2-10 that the steeply dipping fractur-es dominate and the relative frequency of these fractures in-creases with depth with a simultaneous decrease in flat-lyingfractures. This is even more pronounced than illustrated in the

o

2:24

Z FracturesO 1 0 0 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 700 800

200-

300-

400-

500-

600-

800-Oepth

(m)

00 300 400 500 600i i i i i

\\

\1\

": \

\

V2\

IIv

700-. . . . . 0--30- -.

3i*-60* •: ,61* - 90* •". I

\\

Figure 2-10. Cumulative fracture diagram for V2 with regard todipping of the recorded fractures.

figure owing to the variation in estimation depending on the dip-ping. The flat-lying fractures show a low frequency below ap-proximately 400 m depth with 0.3 fr/m which decreases down to0.1 fr/m in the lowermost 230 m. This condition indicates thatmedium steep or steep fractures dominate the fracture system atdepth and the horizontal fracturing become more sparse.

The fracture pattern which predominates in the rock mass at theSGU-site may be established by the fracture orientation datafrom the boreholes. The information from the boreholes Nl and Elgives a rock mass dominated by steeply dipping fractures inN3OE. Other fracture sets of importance are N30W and N1OE, bothsteeply dipping. However, both of these boreholes are nearly ho-rizontal, which indicates that flat-lying fractures will not bepenetrated by the boreholes and consequently they will be under-estimated. The vertical boreholes may serve as a tool to evalua-te the existence of flat-lying fractures.

2:25

V 1 , N1 AND ElSAC-AREA

w

Figure 2-11. Fracture sets obtained at the SGU-site and fromthe SAC-area (lined parts of the diagram). Semis-pherical projection, Schmidt net - lower hemisphe-

re.

Steep fractures dominate clearly the fracture pattern and makeup as much as 40 percent of all fractures in the rock mass adja-cent to the SGU-slte. Medium steep fractures makes up 31 percentand the remaining 20 percent are attributed to flat lying fractu-res. Thus it is possible to distinguish the following sets offractures at the SGU-site.

1 N1OE;8OE2 N3OE;85E

3 N30W-.90

4 Sub-horizontal;25

These sets are shown in the semispherical projection in Figure 2-11.

The obtained orientations from the SGU-area could be comparedwith the orientations found in the huge stock of fracture datawhich ex i s t s from the SAC-area (Wollenberg, et a l . , 1980, Olkie-wicz, et a l . , 1979). In that area the following fracture setswere found:

1 NNW-SSE;60N2 NW - SE;85NE3 N65W;50SW4 Horizontal

2:26

These sets are also included in Figure 2-11. As seen in the fi-gure there is a difference between the fracturing at the SGU-site and at the SAC-area, but some resemblance may be found. Thedifference may be an effect of the sadimentary structures whichcould have effected the fracturing of the granite. This is alsoindicated in Figure 2-3. There seems to be a change in orienta-tion of the fracture system which probably is governed by theconfiguration of the leptite syncline. Closer to the contact bet-ween granite and leptite the fracturing is affected by thesyncline, while at farther distances it seems to be more indepen-dent with increased upright and orthogonal fracturing of the gra-nite.

3:1

HYDROLOGY

3.1 Hydraulic units

The hydraulic properties of a crystalline rock mass such as theStripa granite is characterized by fractures, faults and otherdiscontinuities which transect the rock. The granitic rock matrixis, from a practical point of view, almost impervious and themain flow paths are constituted by the fracture system, zones offractured or crushed rock and other structural discontinuities.As shown in previous sections, there exists a number of disconti-nuities, some of which are associated with the synclinal struc-ture of the sedimentary sequence and others more independent ofit. However, as the dominant tectonization took place before orimmediately at the intrusion of the pluton, the granite is inter-sected only by a few larger fracture zones.

The dominant ruptural deformation is concentrated in the superfi-cial part of the rock, which shows a rather high fracture frequen-cy and a high hydraulic conductivity. This more fractured partof the rock mass extends down to about 250 m depth. Below thislevel the rock becomes more sparsely fractured, with fractureswhich are sealed to a great extent. The fracturing continues todecrease and reaches its lowest frequency below the 1 100 m le-vel (c.f. Figure 2-8).

In the deep-seated rock mass the water flow seems to be channelledin a few zones of fractured rock, where the zone found in the low-ermost part of VI is an extreme example of these flow paths. Atthese deep levels it is probable that the discrete fracture flowis of minor importance.

The mine itself is one of the most important structures govern-ing the water flow in the area. It acts as a drain, with a drai-nage threshold which was successively lowered as the mining con-tinued. During the SAC-program, efforts were put into the estab-lishment of the draining effect of the mine. As reported by Gale(1982) piezometric recordings taken at different levels in SBH-1, SBH-2, SBH-3 and DbhVl show that there is a downward gradientabove the excavations. Around the test areas, the groundwatergradients are directed towards the excavations.

3:2

Figure 3-1. Schematic representation of the geometric rela-tionship between the different porosities in arock mass. The arrows denote an arbitrarily chosendirection of water flow within fracfures formingthe kinematic porosity. The smaller fractures andpores represent the diffusion and the residual pc-rosity (Norton and Knapp, 1977).

3.2 Porosity of the Intact rock material

According to Norton and Knapp (1977) the total porosity in afractured medium, 0», may be expressed as

0T °Dwhere C^ - effective flow porosity or kinematic porosity (1),

Op • diffusion porosity (1),QR • residual porosity (1).

The kinematic porosity represents the fractures through whichthe dominant fluid flow proceeds, while the diffusion and residu-al porosities refer to fractures or pores in which no or very li-mited flow occurs. The fractures making up the residual porosity

3:3

are not connected with those included in either the kinematic orthe diffusion porosity. Figure 3-1 illustrates the geometric re-lationship between the different porosities.

In a fractured rock mass the interconnected fractures comprisethe kinematic porosity which in a two-dimensional section ofunit area can be defined as

With equal fracture apertures and spacing, equation (11) becomes

0K = ne,

and with three sets of water-bearing joints with equal properti-es the kinematic porosity will be

G = 3ne.

Calculations of the porosity based on measurements in drill-holes will result in different porosity values depending on thepenetration of the drill-hole in different sets. Moreover, theexisting joint sets in fact have different fracture properties,a point discussed by Parson (1972) and A. Carlsson (1979).

Snow (1968) reports a study of fracture spacing, fracture apertur-es and kinematic porosity, and presents a method of determiningthe kinematic porosity from water-pressure tests. With equal pene-tration in three equal joint sets, the following relation was ob-tained

0R - 2.4 ne.

This equation, together with those above, demonstrates a rela-tionship between the kinematic porosity, fracture spacing andfracture apertures.

The rock matrix has a very low hydraulic conductivity where themicrofractures constitute the majority of flow paths. The conduc-tivity is estimated to be in the range 10"1-* - 10 m/s, depen-ding on the degree of microfracturing. Although the conductivityis low, the storativity is fairly high, the total porosity accor-ding to laboratory tests are in average 0.47%. The determina-tions are summarized in Table 3-1. The porosity measurements aremade on 12 samples from Nl and 6 samples from VI. The resultsare very consistent, with a minimum value of 0.36% and a maximumof 0.61%.

Most of the pores and fractures included in the total porosityare not actively contributing to the flow porosity. The flow po-rosity is estimated to be In order 10"' - 10 which is only 2-20 permille of the total porosity.

3:4

Table 3-1. Porosity obtained by laboratory tests on drill cor-

es from the boreholes.

Borehole

ElNlVIV2

Number ofsamples

108

Mean porosity Std deviation

4.6 104.7 • 10

-3,-3

+0.840.5

1010

-3-3

However, in the heavily fractured part of VI, the flow porosityis higher and estimated from the hydraulic tests to be in therange 1-3 »10 .

3.3 Hydraulic conductivity of the rock mass

3.3.1 General

A naturally fractured formation is in general represented by atight matrix broken up by fractures of secondary origin. The frac-tures vary considerably in size from voids and interconnectedchannels to fine cracks. Some of the fractures are assumed to becontinuously throughout the formation and to represent the pathsof principal hydraulic conductivity. The rock matrix consistingof the fine disinterconnected cracks has a lower hydraulic con-ductivity but generally a higher primary porosity.

The transient behaviour of groundwater pressure versus time de-pends on the hydraulic conditions around the tested section. Ingeneral the flow situation may be described as linear, radial orspherical. A radial flow is usually prevailing when the condi-tions around the tested section are not hydraulically favouredby a fracture along the section. Instead the flow is radially outfrom or in to the section in one or more fractures perpendicularor inclined to the borehole. In the radial flow case, no flow isassumed along the direction of the borehole. Thus two imaginaryno flow boundaries are assumed at the ends of the test sectionand perpendicular to the section.

In the radial flow case in natural fractured formations, the mat-rix has a "delayed" response to pressure changes that occur inthe surrounding fractures. Such nonconcurrent responses cause

3:5

pressure depletion or inpletion of the fracture relative to thematrix which in turn induces matrix-to-fracture crossflow. Theresponse of the fracture fluid to pressure changes is almost in-stantaneous, whereas that in the rock matrix is much slower. Theperiod of transient crossflow takes place immediately after thefracture pressure response and before the matrix and the fractu-re pressures equilibrate, after which the formatici acts as auniform medium with composite properties (Streltsova-Adams 1978;Streltsova and McKinley 1984).

The spherical flow case may occur after a longer time when the in-fluence has reached longer distances from the test section» Theshorter the test section, the better are the conditions for sphe-rical flow.

Linear flow is a one dimensional flow which exists wheneverthere is a fracture of high hydraulic conductivity along thetest section. In the case of a fracture parallel to by not inter-sected by the test section, the linear flow behaviour mightoccur during certain conditions regarding distance and contrastin hydraulic conductivity between fracture and rock matrix.

3.3.2 J.6!1!11 . .techniques

The groundwater system at the Stripa Mine has successively beenaffected by the mining activities. As the mine was sunk, new flowpaths were activated and the drainage threshold was successivelylowered. The groundwater system was almost continously in balan-ce with the drainage from the underground drifts, i.e. thegroundwater system was in a steady state condition. In 1976 the mi-ning was terminated, but the drainage pumping continued. Afterthe mining, only minor additional impacts affected the system.This gave a hydraulic situation which was well suited for hydro-geological studies underground, any controlled disturbanceshould take place in an affected but steady groundwater system.

A number of techniques may be applied to the underground hydrau-lic testing. However, requirements and demands from other activi-ties and research programs make some of the probable techniquesless suitable* In order to obtain accurate water sampling andanalytical results, the groundwater system should be contaminatedas little as possible with external water and other chemical com-pounds. This condition calls for a testing technique where thegroundwater should be extracted rather than injected. Other testprograms within the project, as for instance, the Buffer MassTest, is strongly dependent on a undisturbed supply of groundwa-ter and pressure build-up, which calls for a minor extractionand disturbance on the water head around the mine.

However, as the hydraulic testing takes place deep underground,in the potential sink made up by the mine, it was found conveni-

3:6

ent to utilize the existing potential field for the testing,i.e. to use the natural drainage for water extraction as themain tool and to measure the pressure build-up after shut in andthe fall-off after release. By this technique, no foreign wateris introduced into the groundwater system, and the disturbanceson the head should be in a natural sense. This technique wasused as the main tool both in single hole tests and in interfe-rence tests between different boreholes. However, as a test ef-fort, water injection tests were carried out in order to comparethe results from different techniques.

Thus, the following techniques are used for the hydraulic tes-ting included in the program:

Single hole tests

- Build-up tests

- Fall-off tests- Single packer fall-off tests- Double packer injection tests

Multiple hole tests

- Build-up tests with selected transmitter and receiver holes- Fall-off tests with selected transmitter and receiver holes

The procedure for build-up and fall-off tests are normally sing-le hole tests where the pressure change in a sealed off sectionis monitored. The tests are carried out only for selected sec-tions of the boreholes (Carlsson and Olsson, 1985a).

The packer system is lowered into the borehole to the actualdepth. When the system is in the correct position, the packersare inflated with nitrogen gas. The flow from the innermost testsection, between the inner packer and the bottom of the hole, isthus packed off and the pressure in the section begins to increa-se. The main test section, between the packers, continues in freeflowing conditions and the flow rate is recorded. After a fewhours, the downhole valve is closed and the actual build-up testsection continues for about five days after which the valve isreopened. During this stage, the pressure build-up in the maintest section as well as in the inner section is monitored.

The test cycle ends with deflation of the packers and thus, apressure relief in the inner section. By this procedure, a com-pleted test at one level requires one week including installa-tion, free flowing and pressure build-up and fall-off. The tech-nique has been used for 10 test sections in borehole Nl, 13 inEl and 8 in V2.

3:7

As a complement to the build-up tests, water Injection testswere carried out in boreholes VI, Nl and El. These tests weremade as hydraulic loggings of the entire boreholes in 10-metersections (Carlsson and Olsson, 1985b).

The tests were initiated with a short build-up period afterwhich the water injection was started and continued for 2 hours.A 2 hour or longer period of fall-off monitoring completed thetest cycle. In order to identify the pressure tranclence onwhich the injection was superposed, the information from the ini-tial stage and the fall-off period was analysed with respect tothe pressure build-up.

The total testing time for a ten meter section thus become 5-6

hours including installation and testing.

Three different interference tests were also conducted during theprogram, i.e.:

- Interference test between VI and V2- Interference test between VI, V2, Nl and El

- Interference test between Nl and the BMT-area

In each of these tests, a specific test section in one of theboreholes was used as a source hole where the pressure disturbancewas to be introduced. The other boreholes acted as receiverholes where the resulting pressure change was recorded.

3.3.3 Results

A great number of hydraulic tests have produced values on the hy-draulic conductivity of the Stripa granite. Tests exist from thesurface boreholes as well as from subsurface holes in differenttest sites, from the large scale ventilation test and from thelarge scale injection test. This huge stock of values provides agood base for determinations of the water flow in the graniticrock mass around the mine. Table 3-2 summarizes the range in re-sults from the SAC-program.

Thus, it is seen that in the surface boreholes the conductivityis at its maximum, about 5 E-8 m/s, while it is 1 E-9 m/s orlower in the tests made down in the mine. The large scale tests,ventilation and injection tests, which both are measures of thegross conductivity gave low values. 1 E-l 1 and 4 E—11 respective-ly. Those latter values are probably representative for the rockmass Including minor zones of fractured rock.

The hydraulic tests carried out in the current program, were allmade as pressure build-up tests, where the natural water flowinto the boreholes was used for the build-up. The tests were ana-lysed according to conventional Interpretation techniques. An ex-ample of a test is shown in Figure 3-2.

3:8

Table 3-2. Hydraulic conductivity values of the Stripa gra-nite obtained during the SAC- and the current pro-gram.

Test type

Injection surface holesInjection ventilation driftVentilation testLarge scale injection

VI, fracture zoneVI, rock mass

V2, rock massEl

Conductivity rangem/s

5 E-ll -1 E-12 -1 E-ll4 E-ll7 E-8

5 E-ll1 E-105 E-12 •

- 5 E-8- 1 E-9

- 4 E-8

Two zones have been found with relatively high conductivity, one40 m wide zone in VI (40 m along the borehole) with a conductivi-ty of 7 E-8 m/s and one 2 m wide zone in El with 4 E-8 m/s. Besi-de these zones, the obtained conductivity is lower than E-9 m/s.

3.4 Hydraulic head

The hydraulic head in the rock is determined by geological, hydro-meteorological and topographical factors. In the current situa-tion it is also, to a very high degree, dependent on the geometri-cal configuration of the mine.

The hydrometeorological conditions in the Stripa area and on anannual basis can be described by a mean precipitation of 780 mm,an annual evapotranspiration of 480 mm and a run-off of 300 mm(9 1/s sq.km). The climatic conditions are humid and in the run-off term both the recharge and the discharge of groundwater areincluded.

The geological factor, which determines the hydraulic conductivi-ty and thus the rate of the groundwater flow in the bedrock,points to a rather low conductivity and consequently a low ground-water flow even at high hydraulic gradients. In the upper partof the bedrock the groundwater level in general follows the topo-graphy.

The hydraulic head was measured in boreholes both from the mineand from the surface. Measurements of the head are normally madein short sections (2-10 m) in boreholes tightly sealed off bypackers. When starting such measurements the head is usually in

3:9

L06 HERD CHRNGC <•>

..,7

R/

S ' N1 7 4 - 7 6

0 1 2 3LOG TIME Cain)

HERD <•>

.125

100

75

60

25 /

1

j/

9

i

HERO <«O

125

100

SO

25

\

•

1 2LOG TlnE (mln)

0 .02 .04 .06 .061/SOUMtE ROOT OF TinE (« ln)

Figure S-2. Graphs of test section 74-76 m in borehole Nl. Itis seen that the test is dominated by well borestorage (WBS) in the Initial stage. After this pe-riod the flow becomes radial as shown by the log-time versus head plot. At the end the flow enters aspherical regime as seen in the third diagram.

a transient state and the monitoring has to be carried out under

a longer period. As regards the procedure, it is described in

Tale (1982) and in Carlsson and Olsson (1985).

Registrations of hydraulic head in boreholes have been carriedout earlier by Olkiewicz, et al, (1979), Witherspoon, et al,(1980) and Gale (1982).

3.5 Model calculations

Preliminary and rough calculations were made of the groundwaterconditions and the groundwater inflow to the mine. The calcula-tions were based on available data on hydraulic conductivity andtopographical conditions. The calculations were performed for avertical plane laid out from the center of Lake Rosvalen,through the mine and further on about 4 km towards NNW. In total

3:10

the section was 7 km in length and 2.6 km in depth» The mine wasillustrated as two horizontal drifts, each 1 000 n in length, atthe levels 410 m and 290 m, respectively, in the mine system. Theheight of the drifts was taken as 70 m.

The calculations were carried out using a finite-element programand assuming two-dimensional flow at steady state. The lower andvertical boundaries of the studied plane were set as no flowboundaries. The groundwater head at the upper boundary was givenas the ground-surface. At the mine the head was set as the datumlevel.

As results of the calculations the head distribution around themine was given together with the inflow to the mine. The calcula-tions were performed in a vertical plane and the total inflow tothe mine was estimated by assuring the same inflow per m of minealong the whole mine.

The hydraulic conductivity of the rock mass was gi^en differentvalues to illustrate different possible conditions in the rock.The conductivity distribution versus depth is given in Figu-re 3-3. The results of the calculation given as distance ofinfluence on the head and the water inflow are summarized inTable 3-3.

The actual inflow to the mine for the period Jan. 1983 - Sept.1984 was recorded to be about 470 1/min on the average.

The result from the calculations based on decreasing conductivi-ty values versus depth (Case B) is more reliable as it is basedon actual test results from the area. The groundwater head forthis distribution is given in Figure 3-4, where the impact ofthe mine is clearly visualized.

It should be noted that the calculated inflow value doesn't takethe flow in fractured zones into account. These zones, as theone in VI, are probably the cause for the major part of the actu-al inflow to the mine.

The interference tests made in the mine show that the flow infractured zones may originate from great distances. The resultsshow that there is a clear Interconnection between boreholes VIand V2 and between Nl and the BMT-area. On the other hand, it isalso clear that no interconnection was obtained between V2 andNl or between El and Nl. A slight influence was noted betweenthe fractured zone in VI and the innermost part of Nl.

SMUUTfoD

500

3:11

HYORAULC CONDUCTIVITY (tn/sl

10" 10'" IQ* 10+

C»SE 61000

1500-

2000

2500

3000-(m) Depth

107 10*

C»5E A

Figure 3-3. Hydraulic conductivity versus depth used in the

model calculations»

Table 3-3. Results of numerical calculations of the distanceof influence and the groundwater inflow to theStripa mine.

Assumptions maderegarding theK-valueof the rock mass

Horizontal distancefor 50 m influenceat the mine levelof 400 m, in km

Case A Horn, conditionwith K-l.E-9 m/s 0.8

Case B Horn, conditionwith decreasing 0.45K-value from 2.E-8down to 3.E-11 m/s

Groundwaterto the minein 1/min

73

96

3:12

0.5 1.5 2.5 3km

icor -

T505

200C -

2S0C

in)

Figure 3-4. Groundwater head around the mine calculated by nu-merical method, case B.

3.6 Dewatering of the granite

The pressure sink made up by the mine has significantly affectedthe groundwater conditions in the granite. Over the years a con-tinous water inflow has taken place and as the mine was sunk thedrainage threshold was lowered. After the mining activities wereterminated in 1976, the drainage pumping continued and the gro-undwater recharge from the surface and from adjacent areas balan-ced the groundwater discharge in the mine. After the mining,only minor impacts on the groundwater system was introduced, andthen mainly from different boreholes which also acted as drain-age structures in the rock mass.

The total effect from the drainage on the groundwater system isa shorter residence time for the groundwater. The recharge fromthe surface is more quickly flowing down to deeper levels in therock mass, i.e. those levels which are of interest for water samp-ling and analysis. Also the groundwater flow from adjacent areas

3:13

is faster which leads to a more complex mixing of differentgroundwaters. These conditions are of great significance for the in-terpretation of the hydrogeochemical data; a foreign groundwatermay be sampled in the granite.

The total discharge from the mine amounted to about 470 1/rain asan average value for the period January 1983 to December 1984.This discharge gives a total discharged volume of water from themine since the start of the SAC program to the end of 1984 of al-most 2 million cu. m. Thus, a most significant discharge ofgroundwater. Also, the boreholes made for the present program givesa most significant discharge, especially for the borehole V2,which was drilled as early as 1978 and presently deepened to itspresent depth. This borehole has been free-flowing for most ofthe time, and it is only in connection to various tests thatthis borehole has been packed off. Borehole VI contributes great-ly to the total discharge due to its high yield. As regards bore-hole Nl it has been packed off during most of the time as a con-sequence of its influence on the buffer mass test. In Table 3-4,the total estimated discharge from the main boreholes are summa-rized. The values given represent the total discharge since thedrilling of the boreholes till the end of 1984. The figures arebased on the recorded discharge in relation to the borehole his-tory, and are therefore only rough estimates. In total, the er-rors on the given figures are in the range of 10 per cent.

Table 3-4. Total discharge in the boreholes since the dril-

ling of the holes till the end of 1984.

Borehole Total dischargecu. m

Nl 567El 432VI 11.312V2 2.756

4:1

GROUNDWATER CHEMISTRY

4.1 Introduction

Numerous chemical analyses of Stripa groundwaters have been ob-tained since 1977. Both major and trace constituents have beendetermined and presented in three separate reports (Fritz, et^al., 1979, 1980; Nordstrom, 1983a) along with preliminary inter-pretations. Additional discussions of the groundwater chemistryhave been reported by Nordstrom (1982, 1983b) and Fritz, et al.(1983). The initial findings can be summarized as follows:

1. Some of the deeper groundwaters (>700 m) have unusually ele-vated salinity, up to 700 mg C1/L.

2. The more saline water at depth has a markedly differentwater chemistry than the shallow groundwaters; i.e., it isan Na-Ca-Cl- SO^ type water.

3. The pH increases to the range of 9-10 with depth.

4. Dissolved inorganic carbon becomes very low with increasing

salinity, reaching 9 mg HCO-j/L.

These findings, along with the isotopic data, have led to muchdiscussion and controversy regarding the origin and evolution ofthe Stripa groundwaters. This chapter presents the chemical dataand selected element ratios for comparison with known types ofsaline fluids such as seawater.

4.2 Methods of sample collection and preservation

Beginning in June, 1981, groundwater samples have been collectedfrom packed-off zones by pumping water from tubing connected tothe packer sample tube at the borehole. A peristaltic pump trans-ported the groundwater to a flow cell arrangement closed to theatmosphere where on-site measurements of pH, EMF (electromotiveforce), temperature, and specific conductance are obtained» Mea-surements of pH were made with a glass electrode calibrated atsample temperature with pH 7 and 9 buffers» As an accuracycheck, a pH 10 buffer was occasionally measured after the samp-le; and if the deviation was greater than about 0.05 pH units,then the electrode was recalibrated and the sample reroeasured. Aplatinum electrode with a calomel reference was used for EMF mea-surements and corrections made for the reference potential from

4:2

the data of Ives and Janz (1961), to derive the Eh value.ZoBell's solution provides a check on the EMF measurements (Nord-strom, 1977).

All samples were f i l tered by pumping the water from the same c los -ed l ine system with the per i s ta l t ic pump directly to a preclean-ed plast ic plate f i l t e r f i t ted with an 0.1 micrometer Milliporemembrane. The membrane was preleached with at least 1 l i t e r ofgroundwater before collecting samples for analysis. One sampleset (81WA202, Nl) was unfiltered due to a broken membrane, andwhen this problem was noticed a second f i l tered sample was co l l e c t -ed. The unfiltered sample was saved for a comparison of the ef-fect of f i l t ra t ion . Sample preservation techniques variedaccording to the particular constituents being analyzed. Samplesfor determination of major cations and trace elements were c o l -lected in teflon or polyethylene bottles and acidified with u l -trapure ni tr ic acid to a pH <1.5.

Samples collected for Fe ( I I , III) and As ( III , V) were ac id i -fied with ultrapure hydrochloric acid to a pH <1.5. Mercury waspreserved with the addition of potassium permanganate solution(Avotins and Jenne, 1975), and zinc acetate was added for thepreservation of sul f ide .

A duplicate set of water samples was collected at each s i t e , du-ring the June f ie ld tr ip , for an interlaboratory comparison ofanalyses. One set was analyzed by the U.S.G.S. and the other setby S.G.U.. Each set consisted of a 250-ml teflon bott le for ca-tions and trace metals, a 250-ml polyethylene bottle for n i tra-te , n i t r i t e , phosphate and ammonia, a 250-ml polyethylene bott lefor iron ( II , III) and arsenic ( I I I , V), a 1-l i ter glass bott lefor mercury, a 50-ml glass bottle for bromide and iodide, a 250-ml polyethylene bottle for anions, a 50-ml glass bottle for d i s -solved organic carbon (D.O.C.), and a 2- l i ter or 1-gallon poly-ethylene bottle for sulf ide.

4.3 Methods of analysis

Most metals were analyzed by plasma emission spectrophotometryusing both DCP (direct-current plasma) and ICP (inductively-coupled plasma) sources. Selected samples were also analyzed forNa, K, and Li by AAS (atomic absorption spectrophotometry) for acheck on accuracy. Rb and Cs were analyzed by flame emissionspectroscopy, Hg was analyzed by flameless atomic absorptionspectrophotoraetry, and Iron was done by the ferrozlne colorimet-ric method (Glbbs, 1976) in addition to ICP. Anions were doneby IC (ion chromatography) except for nitrogen and phosphorusspecies which were analyzed by colorimetry on an autoanalyzer.Selected samples were additionally analyzed for fluoride by ISE(ion-select ive electrode) , and H2S was analyzed by ISE only.

4:3

Cho

BoIa

20 .

ie

-IB .

-28 .

tee 388 488

Chi or Ida, mg/L

638 708

Figure 4-1 Charge balance errors as a function of Cl concen-tration.

4.4 Accuracy and precision

Precision was occasionally checked by determining the reproduci-

bility on selected samples, and was usually found to be less

than 5%. Accuracy was determined by (1) interlaboratory compari-

son, (2) analysis by an alternate method, and (3) charge balance

errors.

The interlaboratory comparison is described in detail in Nord-strom (1983a). The results showed that discrepancies were usual-ly within 5% after cross-checking and revising the initial valu-es. Charge balances were also exceptionally good (<4%) for thisone group of samples. Charge balance error was calculated fromthe following equation:

neq./L, cations - meq./L, anions

meq./L, cations +meq./L, anlons

For all of the analyses given in Table 4-1, the charge balancewas calculated and plotted as a function of chloride concentra-tion as shown in Figure 4-1.

Nearly all the values fall in the range from +102 to -10% with aslight decrease in the dispersion at higher chloride concentra-tions. These results suggest the determinations for major consti-tuents are accurate. Most analytical determinations, other thanthose in the interlab comparison, can be considered accurate to

4:4

about 102. Greater errors are occasionally found In the anion

data, especially Cl and SO^. The reasons for these discrepancies

are not known.

Table 4-1. Stripanoted).

water analyses (In rag/L except where otherwise

Drillhole/Site

Sample Code No.Stapling Interval (a)Date CollectedTemperature (°C)pHCond (uS/cn)Eh(nV)Total Alkalinity

(BJ/L, HCO3)Charge Balance (')

Species

CaH»NaKS04

FClBrIP04SiO2BNOjNO3NH4AlFe (total)re2+HnCuZnPbCoNlCrVMoLiSrCaRbBeBaDOC»

Stream

10....77091212.96.325....

6.1-13.*

3.50.51.70.47.5....2.9........<0.013.6........0.41........

0.23............

....

....

....

....

....

....

....

....

....

....

....

....4.9

Stream

11....7709129.36.7331

8.6-6.6

4.51.01 1r.69.0....3.3

....

<0.014.4........0.48........

0.39................................................................4.9

SW1

61WA217....S1100610.16.96

31.5466

91.77

3.51.22.00.636.10.512.30.0140.0030.055.35<0.005<0.0051.30.030.055.1830.1740.015<0.0050.018....<0.005............

<0.005<.0050.011....0.005....

0.011....

T.P.

9....77120913.27.57

152....

48.67.39

265.56.91.915....11........<0.016.2........

1.66........

0.43............

....

....

....

....

....

....

....

• •••....

....5.0

T.P.

....

....63100310.07.42

130474

740.9

204.35.31.88.10.796.3....<.OO5<.025<.005<.005<.l.04

<.01.12

.....03

<.001.002

<.01.011

<.001<.005«.01<.01.006.065

.019<.003.042....

PW1

18-1

7709279.57.85

263

206.2-5.06

371224I.I13....16........

0.208.6........

....

....<0.01........................................• ••-................• •••

0.8

* Diaiolvtd Organic Carbon

4:5

Table 4-1. Stripa water analyses (in mg/L except where otherwise

noted).

Drlllhole/Slte

S»mple Code No.Date CollectedTemperature (°C)pHCond (uS/cm)Eh («V)Total Alkalinity

(ng/L, HCO3)Charge Balance

Specie*

CaMgNaKS0A

ClP04

S102

NO3Fc (total)DOC

PW5

21-3771006

7.77.35

96

83.64 .5

194.54 .51.4

135.2

<P.O18.82.272.455.2

PU5

21-10771007

7.76 .6

104. . . .

61.03 .5

IS4 .54 .91.4

144 .70.037.82.476.255.2

PW5

21-11/25771010-12

7.66.85

94112

78.710.3

23.64.84 .51.1

10.14 .4

9.9. . . .. . . .. . . .

PW2

23-10/20771024-26

7.07.6

105. . . .

102.5-1.0

21.25.74 . 31.92 .81.5

13.6. . . .. . . .. . . .

PU3

20-3771026

7.2245. . . .

1972.46

6944 .32 .48.4

13<0.0111.40.743.40.7

PU3

20-4771026

7.2245

19710.3

663.74.63.49.0

10.7

12.0. . . .. . . .. . . .

4:6

Table 4-1. Stripa water analyses (in mg/L except where otherwisenoted).

Drillhole/Site

Sample Code No.Sampling Interval (n)Date CollectedTemperature (°C)pHCond (uS/cm)Eh (mV)Total Alkal ini ty

<mg/L, HCO3)Charge Balance

Species

CaMgNaKS0 4

FClBrIP04

S102

BN02

NO 3NH;,AlFe (total)Fe2+.InCuZnCoMoLiSrRbBaDOC

UT3

70-1. . . .

7905016.06.62

63. . . .

34.15.0

•

8.91.584.691.477.4

2.3

. . . ..9.4

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

PW5

21-29. . . .790515

5.86.2

so263

17.3-0.8

5.71.52.71.1

13.1

2 .3. . . .

. . . .10.0. . . .

<0.4. . . .

1.73. . . .. . . .. . . .. . . .. . . .. . . .. . . .

. . . .

. . . .

. . . .

PW3

20-6. . . .790516

7.57.58

230370

2101.22

64.54.345 .81.8

10.4

3 . 2. . . .. . . .

12.2. . . .

<0.4. - - .

<0.05. . . .. . . .. . . .. . . .. . . .. . . .. . . .

. . . .

. . . .

. . . .

PW1

81WA2150-80

81100811.2

5.1150

463

12-9.29

4 . 01.13 .71.1

100.304 . 50.0140.002

<0.0213.5<0.005<0.0050.800.040 11V. Li

.068

.0310.0270.0510.12

<0.005<0.005<0.0050.016

<0.0050.018

. . . .

PU2

81UA21G0-80

81100810.2

6.1382

490

20-3 .28

132.52.51.7

120.22

140.0230.003

<0.0210.5<0.005<0.005

4 .3<0.02n non

.037

.0310.010

<0.0050.027

<0.005<0.005<0.0050.0330.0080.022

drlllwater

25-1

77110710.0

8.1

. . . .

11811.5

46118.02.1

28

18. . . .

<0.0112. . . .

12.6. . . .

0.03

. . . .

. . . . .

. . . .

. . . .

. . . .

. . . .3.0

4:7

Table 4-1. Stripa water analyses (In tng/L except where otherwise

noted).

Drillhole/Site

Sanple Code No.Stapling Interval (m)[li.e CollectedTemperature (°C)pHCond (US/CB)Eh («V)Total Alkalinity

(ag/L, HCO3)Charge Balance

Speclea

Ca«gNaKS0 4

FClBrP04

SiO2NO 3Fe (total)HnDOC

drlpwater

26-3340-360771117

8.35

129-2.5

523.5

380.6

89. . . .28

<0.019.6

.1.120.05

. . . .1.7

dripwater

27-4360-410771117. . . .

8.1. . . .. . . .

104-2.4

816

452.9

147. . . .

26. . . .<0.01

9.4750.04

. . . .2.0

SBH3

85-1509-104790327

8.07.89

162. . . .

142.51.0

33.94 .5

12.51.78.8

. . . .3.7

. . . .

. . . .12.5. . . .. . . .. . . .. . . .

Rl

53-70-60

78111711.9

9.0202158

97.70 .9

14.10.29

49.00.202.3

. . . .42.1. . . .. . . .11.7. . . .<0.05<0.05. . . .

Rl

53-200-60

79050211.76.89

19937

97.43.0

15.40.27

44.70.173.0

. . . .35.4. . . .

11.2

. . . .

. . . .

. . . .

Rl

53-290-60

79051710.08.95

208131

101.5-13.1

14.80.31

43.50.193 .14 . 3

38.3<0.1. . . .11.9. . . .<0.05. . . .

4:8

Table 4-1. Stripa water analyses (in tng/L except where otherwise

noted).

Drillhole/Site Rl:4 R9 H-2 H3 H3 M3

Sample Code Mo.Sampling Interval (n)Date Col lec tedTcaperature (°C)pHCond (uS/cm)Eh («V)Total A l k a l i n i t y

(•fc/L, HCO3)Charge Balance

81WA2040-60

61060412.08.96

21557

951.9

79-5. . . .

790522. . . .

. . . .

. . . .

68.45.4

38-2

78060115.0

8.2. . . .. . . .

81.1-8.3

16-53-10

77092610. S

. . . .97

63.01-7.6

16-23A3-10

771019

8.7210

92

78.78.32

42-23-10

780607158.93

. . . .

74.40.52

Species

Ca

MSNaKS04

FClBrIP04

S102BNO2

NO3

NH4

AlFe (total)Fe*+

MnCuZnCoHoLiSrKbBaDOC

150.18

490.103.24.7340.30.014<0.1W0.15<0.0050.5<0.010.0050.046.016

<0.005<0.003<0.005<0.0050.0380.0200.12<0.0020.0064.0

18.00.2540.60.131.3

49.6

12.1

<0.05

11.80.3251.50.354.0

60.2

11.8

15<0.5430.32.1

52

<0.0111.6

0.24

0.07

14.10.350.80.21.4

49.3

12.0

14.20.3154.50.291.0

65.8

12.0

0.7

4:9

Table 4-1. Stripa water analyses (in mg/L except where otherwise

noted)•

Drillhole/Site

Staple Code No.Sanpling Interval (m)Date CollectedTemperature (°C)pHCond (uS/cn)Eh (»V)Total Alkalinity

(mg/L, HCO3)Charge Balance

Species

CaMgNaKS04

H2SFClBrIP04

S102BNO 2

N03

NHj,

AlFe ( to ta l )Fe2+

HnCuZnCdHgPbCoNlCrVHoLISrC(RbBeBaDOC

H3

42-353-10

790517138.98

207

86-8.91

14.70.25

47.80.243.6

5.048.1<0.1. . . .. . . .It . 4. . . .. . . .. . . .. . . .. . . .<0.05. . . .. . . .. . . .. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

H3

81UA2053-10

810604128.84

210. . . .

86-3.9

140.23

470.124 .90.0265.0

38.50.30.013

<0.1120.18

<0.0050.6

<0.01.0055.008.005

<0.005<0.003<0.005. . . .. . . .. . . .<0.005. . . .

. . . .0.0560.0200.140.047

<0.002

<0.003. . . .

M3

. . . .3-10

831109149.14

235136

92-8.2

140.22

400.316.5

5.4360.160.011

<0.2120.16<.005< . l<.02<.01

.01

.001<.001<.001. . . .. . . .<.01<.005$.001

.004

.01

.07

.0320.15

<.005<.003

.004. . . .

M3

. . . .3-10

840223159.04

23598

93-6.7

140.22

420.305 .1

5.136<50.021

<0.2110.23<.005< . l<.02<.001<.005<.01<.001<.001<.001. . . .<.01<.01<.OO5i.OOi,<.001<.0050.110.0240.13. . . .<.005<.003<.005. . . .

El

81WA218003-300811111

10.5

175-1

8422.6

200.32

450.806.8

3 .125

.160.006

<0.0211.80.10

<0.0050.10

<0.02<.00S

.071

.00550.005

<0.0050.073

. . . .

. . . .

. . . .<.005

. . . .

0.0490.0190.13

<.005

0.0031.3

El

. . . .003-300820323

10.49.2

180

103-4.0

220.81

302.3

10

2.222

.078

<0.0213.3

.041<0.005<0.1<0.020.06

.009

.05<0.005<0.01<0.005

<.01<0.005<0.005

<.005<.005<.005

.0210.17

<.005<.0030.027. . . .

4:10

Table 4-1. Stripa water analyses (in mg/L except where otherwise

noted).

Drillhole/Site

Saaple Code No.Sampling Interval (m)Date CollectedTemperature (°C)pHCond (uS/cm)Eh (mV)Total Alkal in i ty

(ng/L, HCO3)Charge Balance

Species

CaMgNaKSO4H2SFClBr1?OU

SiOjBNO2NO3

NH4

AlFe ( total )Fe2+MnCuZnCdPbCoNiCrVMO

LiSrCJ

RbBeBaDOC

El

. . . .003-300840306

10.38.9

183110

87-2.8

180.23

360.178.7

3.4220.Z10.005<.02

130.049<.005<. l<.02

.017<.O1<.010.004

<0.001<.001

. . . .<.01<.OO50.007<.001<.0050.0400.0170.12

. . . .<.005<.0030.008

. . . .

El

. . . .127-129820323

. . . .

. . . .

13

2 .023

.078. . . .. . . .. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

Nl

81UA209003-300810819

10.48.91

190-39

816.44

230.57

370.610.45

3 .5400.420.014

<0.02U . I0.07

<0.005<0.2

<.020.080

.15

.003<0.01<.005<.010

. . . .

<.005. . . .. . . .. . . .

.050

.020

.15. . . .

<.005. . . .

.0051.2

Nl

81WA214003-300611006

11.58.93

215-4

72-11 .1

210.21

330.380.30

3 .3510.500.016<.02

11.8.064

<.005.40.02.033.07.035

<.005<.005<.005

. . . .

. . . ..010

. . . .

. . . .

. . . ..046.022

0.16. . . .<.005

. . . .<.003

. . . .

Nl

. . . .003-30082060310.08.71

195-185

75-3.9

210.77

302.21.0

3.440

.38. . . .

<.0213.5

.045<.0051.0<.02

.007

.033. . . .

<.01<.oos<.01<.005<.01<.OO5<.005

.005<.005<.005

.0230.16

. . . .<.005<.003

. . . .

. . . .

Nl

81WA202010-300610603

10.18.85

19013

77- 1 . 2

Filtered

230.12

35.50.120.60.026

.07045.5

0.50.015

<0.1120.094<.0050.7<.01

.059

.017

.009<.005<.003<.O05

. . . .

. . . .<.005

. . . .

. . . .

. . . ..049.019

0.160.0350.003

. . . .0.003

. . . .

Unfiltered

22.50.22

35.50.19

. . . .

. . . .

. . . .

. . . .150.10

. . . .

. . . .

. . . .0.560.16

.018<.OO50.0430.024. . . .. . . .<.005. . . .. . . .. . . .

.054

.0170.170.0660.003

0.004. . . .

4:11

Table 4-1. Stripa water analyses (In mg/L except where otherwise

noted)•

Drillhole/Site

Sample Code No.Saapling Interval (a)Date CollectedTemperature (°C)pHCond (uS/cm)Eh (aV)Total Alkalinity

(ag/L, HCO3)Charge Balance

Specie»

CaMgNaKSO*FClBrIPO4SiO2BH02

N03

AlFe (total)

NnCuZnCdPbCoMlCrVHoLISrRbBeBa

Nl:2

151-25184012610.38.99

21545

87-6.8

220.18290.250.673.3533

.32

.015<.02

• 120.12<.005<. 1<.O2<.001<.01

.004<.001<.001<.001<.01<.005.007

<.001.005.040.022

0.13<.005<.003<.005

Nl:l

252-30084012610.58.89

20039

81-11.6

230.20290.190.463.35

43.38.017

<.O2120.12<.005<.l<.O2<.001<.01

.003<.001<.001<.001<.01<.005.008

<.001.005.047.022

0.15<.OO5<.003<.005

Nl

123-12582083010.28.73

355-243

39-3.25

290.45600.270.93.1

1221.5.038.03

10.3.054

<.OO5<.l<.020.09.15

<0.01<.005<.01<.005<.01

<.OO5<.OO5<.005<.005<.005.033

0.22<.005<.O030.16

Nl

203-205820906108.43

250-146

552.3

240.42350.260.13.358....

.021<.02

12.2.063

<.005<.l.06.05.006

" " Ö l<.005<.01<.005<.01<.005<.OO5.007

<.005<.005.026

0.19<.005<.0O3.19

Nl

271-27382091410.58.85

205-108

75-15.8

220.29290.452.73.1

480.57.012

<.0210.3

.047<.OO5<.l<.02<.01.007

"<"oi<.OO5<.01<.005<.O1<.005<.OO5<.005<.005<.OO5.024

0.16<.OO50.047.10

Nl

274-27682092310.48.79

221-64

63.028

260.09300.21.50

3.050.64.017

<.0210.3

.036<.005<. 1.03

<.01.021

" " Ö l<.005<.01<.005<.01<.005<.01<.005<.005<.005.033

0.19<.0050.010.065

4:12

Table 4-1. Stripa water analyses (in mg/L except where otherwise

noted).

Dril lhole/Site

Sample Code No.Sanpllog Interval (a)Date CollectedTemperature (*C)pHCond (uS/cm)Eh(aV)Total Alkalinity

(ag/L, HCO3)Charge Balance

Species

CaHgNaKSO*FClBrIP0<,S10 2

BH02

N03

AlFe (total)Fe^+

HnCuZnPbCc.NiCryHoLiSrCiRb8eBaDOC

VI

000-506810409

. . . .

122.1

160. 6 4

2801.6

914.2

620

. . . .<.3

. . . .

* <.0056.1<.01

<.005. . . .

.007<.OO5<.005<.010<.005<.005<.005

. . . .

. . . .2 .4

. . . .<.003

.038. . . .

VI

61WA212004-506810921

15.59.10

12-0.29

1460.55

2482.4

904 .5

5605.90.14<.02

130.21<.0056.4<.02

.02

.0053

.0046<.01<.005

.051. . . .

<.005. . . .

.0100.151.7

<.005. . . .

0.025. . . .

VI

S1UA213004-506811006

14.59.11

1650135

11-0.67

1460.57

2462 .4

904 . 1

5605.80.14<.02

13.50.20<.0057.8<.O2

.009

.004

.002<.005<.005

.020. . . .

<.O05

.0130.151.7

.005. . . .

0.027. . . .

VI

. . . .006-506810326

. . . .

. . . .

10-1.0

143.93

2572.2

893.7

580. . . .. . . .

<.3. . . .. . . .

<.0056.5<.01

. . . .<.OO5

. . . .<.005<.005

. 0 2 /<.010<.005<.005

.010

. . . .

. . . .1.9

. . . .

. . . .<.003

.024. . . .

VI

81UA206000-406810604

10.39.73

51022

182.5

27<.005

1160.24

11.54.6

1902.00.061

<0.1140.24<.0052.5

<0.010.032

.019

.010

.008<.003<.005

. . . .<.005

. . . .

.050

.020

.20

.078

.009. . . .

<.0031.1

VI

. . . .100-505831003

13.28.95

1540162

11-3.0

1380.56

2171.9

854.5

5005.50.10<.02

13.023

<.0056.4<.O2<.01<.01

. . . .<.01<.001<.001<.01<.005<.001<.005

<!oi0.201.3

. . . ..035

<.OO3.025

. . . .

4:13

Table 4-1. Stripa water analyses (in mg/L except where otherwisenoted).

Drillhole/Site

Sample Code No.Sampling Interval (m)Date CollectedTemperature COpHCond (uS/cm)Eh(mV)Total Alka l in i ty

(mg/L, HC03)Charge Balance

Species

CaHgNaKS0<,H2SFClBrIP04

S102

BMO 2NO 3NH4

AlFt (total)Fe*+MnCuZnCdHgPbCoMlCrVHoLiSrC»RbBeBaDOC

VI

100-505831019

12.19.15

1625225

127.4

1440.51

1952.8

85

4.5500

5.60.16<.02

.17.021

<.005< . l<.02<.01

.02. . . .

.002<.001

.002. . . .

<.01<.005«.001<.OO5<.01<.010.211.2

. . . ..017

<.003.026

VI

100-505831105

11.59.25

1420126

14-6.1

1380.46

2082.9

88

4 .5500

5.50.13<.02

130.24<.O05< . l<.02<.012.00.12

.002<.001

.001. . . .

<.01<.005«.001

.002.01

<.010.191.2

. . . .<.005<.003

.025. . . .

VI

. . . .100-505831207

12.29.2

1460127

112.93

1490.48

2251.5

85

4.6517

J.4.080.28

10.076

<.OO5<0.1<.O2<.001<.01

. . . .<.001<.001

.005<.001

<.01<.005

.005<.001<.005

.0110.141.3

. . . .<.005<.003

.022. . . .

VI

. . . .100-505840111

12.69.24

1580102

11-1 .93

1460.47

2121.9

89

4 . 5516

5.6.14

<.O212

.23<.005

<0.1<.02<.001<.01

<-001<.001<.001<.001

<.01<.005<.001<.001<.005

.050

.0751.3

. . . .<.005<.OO3

.024

VI

. . . .100-505840208

12.29.30

1540120

17-1 .53

152O.':3

2082.4

86.5

4 .6518

5 .4.22

<.0213

.24<.005

<0.l..?2. .01.004

<.O1.003

<.001<.001<.001

<.O1<.005<.001<.001<.005

.090

.141.4

. . . .<.005<.003

.025. . . .

VI

81UA203410-506810603

10.69.31

1420-56

9.251.1

1720.19

2771.2

102.0026

4.6630

6.50.16< . l

13.00.25<.0057.0<.01

.024

.004

.0007<.005<.003<.005

. . . .<.005

<.005. . . .. . . .. . . .

.027

.0851.7

.074

.032. . . .

.0354.2

4:14

Table 4-1. Stripa water analyses (in mg/L except where otherwise

noted).

Dill lhole/Sitc

Saople Code No.Stapling Interval (m)Date CollectedTemperature (*C)pHCond (US/CD)Eh(mV)Total Alkalinity

<ng/L, HCO3)Charge Balance

Specie!

CaHgNaKSOt,TClBrIPO4

SiO2

BH02

MO 3NH4

Alre (total)Fe2+

HnCuZnCoNiCrHoLiSrRbBeBaDOC

VI

S1UA208410-506810819

10.69.27

1340129

13-2.34

1700.32

3042.3

954.5

7006.20.11<.O2

13.70.24<.0057.0<.02<.OO5

.006

.002<.01<.005<.010<.005

. . . .

. . . ..010

0.181.8<.OO5

. . . ..030

4.S

VI

61WA210410-J0681090810.59.28

142079

16-.021

1670.27

2902.5

1054 .2

6506.60.16<.02

13.00.22<.005<.2<.O2

.01

.013

.006<.00S<.O05<.010<.005

. . . .

. . . ..020

0.152.0

.005. . . .

.0301.0

VI

B1WA211410-506810911

15.59.17

1580

19-.039

1460.81

2602.2

924.4

5755.90.14<.O2

13.00.21<.0051.4<.O2.008.010.004

<.005<.005<.010<.005

. . . .

. . . ..020

0.151.8.005

. . . ..030

. . . .

VI

81UA207410-505810713

10.59.54

160055

113.21

1520.60

2702.6

1104 .5

5706.60.16<.O3

13.70.25<.005<.10

.03<.005<.01<.01<.01<.005<.005

.009. . . .. . . .

.0100.192.0

.005. . . .

.0401.6

VI

. . . .092-094810114

. . . .

. . . .

65-14

364.8

743.09.62.2

174. . . .. . . .

<.3. . . .. . . .

.0372.3

.20. . . .

<.01. . . .

<.01.029.078

<.005<.005<.005

. . . .

. . . ..147

. . . .<.003

. . . .

VI

. . . .092-094810116

. . . .

172.2

32.25

1131.11.13.2

220. . . .. . . .

<.3

.0372.6

.10. . . .

<.01. . . .

<.01.033.020

<.005<.010<.005

. . . .

. . . ..192

. . . .<.003

. . . .

4:15

Table 4-1. Stripa water analyses (in mg/L except where otherwisenoted).

Drillhole/Site V2 V2 V2 ¥2 V2 V2

Sample Code No.Saapling Interval (•)Date CollectedTemperature C OpHCond (uS/cm)Eh(»V)Total Alkalinity

(ag/L, HCO3)Charge Balance

. . . .OOe-622820421

9.29.37

710-146

36-6.96

24-2000-471771110

7.69.25

30829

45.80.66

000-471810409

. . . .

. . . .

18.6

6-5152.3-471

7709089 .39.38

439122

38.64.27

6-28152.3-471

7709148.29.54

46472

31.51.72

6-30152.3-471770914

8.29.54

46472

31.55.7

Speclei

CaNgNaK

rci?°4SiOjBN02

N03

ÅlFe (total)HnCuZnCdPbCoNiCrVNoLiSrRh

K»nor

490.72

1«22.6

2412

270<.02

17.097

<.OO5*3.4

<.02.17

0.02<.01<.005<.01<.005<.01<.005<.005<.005

23<.5

645.43 .6

114<.01

11.2. . . .. . . .

.15. . . .. . . .

.05

. . . .

. . . .

. . . .

. . . .

. . . .

34.32

711.1

125.1

120. . . .. . . .. . . .. . . .

7 .6

. . . .<.005

.OOS

.OOS.010

. . . .<.010<.005<.005<.005

350 .5

960.26.9

. . . .182

< .0 l12.6

. . . .

. . . ..15

. . . .

. . . ..19

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

36.8.09

1O00.378 .0

192. . . .

13.1. . . .. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

37<.5

1000.26.2

. . . .185

.0112.8

. . . .

. . . ..15

. . . .

. . . ..03

. . . .

. . . .

. . . *

. . . .

. . . .

. . . .

. . . .<.005<.005

.0570.11

<.nni.014

.34

<.<vn.017

n.7 I.I 0.4

4:16

Table 4-1. Stripa water analyses (in mg/L except where otherwise

noted).

Dtlllholc/Slte

Staple Code No.Stapling Interval (a)Dite CollectedTeaptrtture C OpHCond (US/CB)Eh(aV)Total Alkalinity

(ag/L, HC03)Charge Balance

Speciei

CaMiN«KS0 4

H2SFClB iIP04

sio2BN02

NO3NH4

AlFe (total)Fe2+HnCuZn»iPbCoMiCrVHoLiSrCiRbBeBaDOC

V2:l

. . . .562-822831129

8.29.94

121544

25-2.1

146<.005

1950.82

35

4.6540

5.10.56-<.02180.39<.005<. l<.02

.034<.01

<.001<.001<.001

<.01<.005«.001<.001<.005<.0050.111.0

. . . .<.005<.003.019

V2

15-4285-471770915

8.09.49

4637

29.04.7

37<0.5

1000.27.2

. . . .

189. . . .. . . .

.0112.8

. . . .

. . . .0.14

. . . .

. . . ..02

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .0.4

V2

81WA201356-471810603

8.09.53

96020 to 79

11.50.18

1040,11

160.0.37

44.5.00023

3.9410

4.00.11

<0.1120.20<.0054.8<.01

.01

.088

.008

.009<.003<.005

.023. . . .

<.005. . . .

. . . .

. . . ..051.038.99.046.015

. . . ..024

4 .0

W2

81WA219356-471811119

8 .09.17

. . . .

. . . .

28-7 .15

1010.29

1701.0

45. . . .

3 . 4400

4 . 20.110.02

11.80.21<.0053.7O . U

.015

.085

.016<.005<.005

.069. . . .. . . .

<.OO5. . . .

. . . .

. . . ..071.073

1.0

.010. . . .

.022. . . .

V2

29-3376.5-471

7801307.59.75

600169

15.4-2.4

590.5

1250.4

19. . . .. . . .283. . . .. . . .

<.0111.2

. . . .

. . . .0.28

. . . .

. . . .<.O2

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .0 .8

V2

29-34376.5-471

7802227.09.7

600144

12.3-11.7

550.1

1200 .4

19. . . .3.7290

. . . .<.01

17. . . .

<0.0010.050.01

. . . .0.08

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

4:17

Table 4-1. Stripanoted).

wat^r analyses (in mg/L except where otherwise

Drillhole/Site

Sample Code No.Sampling Interval (m)Date CollectedTemperature (*C)pHCond (uS/cm)Eh(mV)Total Alkalinity

<mg/L, HCO3)Charge Balance

Species

Ca"gNaKSO*FClBrIP°4SiO2

BNO 2

NO 3NH4AlFe (total)

HnCuZnFbCoNiCrVHoLiSrRbBeBaDOC

V2

29-39376.5-471780223

7.09.7

600144

12.3-9.6

61<0.5

1200.8

18

285. . . .

<.0112.8

. . . .0.19

12.8. . . .

<0.02

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .0.6

V2:3

. . . .424-499

8311289.0

10.041210

52

20-3.7

106<.005

2310.46

674 .6

5005 .30.84<.O2

160.21<.O05<. 1<.02

.063<.01

<.00l0.04<.001<.01<.0054.001<.001<.0050.034

.0861.0<.005<.OO3

.007. . . .

V2:2

. . . .500-561

8402288.6

10.061090

45

18-1 .6

940.03

2180.44

575.3

4604 . 50.14<.02

180.34<.005< . l<.02

.009<.005

<.002.007

<.001<.01<.005«.001<.001<.005

.060.087

0.89<.005<.003

.009. . . .

V2

17-326.3-50771003

8.08.83

262- 1

53.80.4

18<0.556

0.43 .3

. . . .84. . . .. . . .

<.0111.2

. . . ..20

. . . .

. . . .0.15

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .0.6

V2:4

. . . .382-423

8402288.09.51

114028

110 .4

1450.12

1980.38

603 .3

5005 .00.11<.O2

120.13<.OO5< . l<.02

.013

.005

<.001.003

<.001<.01<.0054.001

.007<.005 •

.0510.101.1<.005<.003

.028. . . .

V2

43-309-40780606

7.59.04

-87

57.139.9

9.80.26

93.82

78.5

. . . .

. . . .

. . . .

. . . ..035

. . . .

. . . .. . . .. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

4:18

Table 4-1. Stripanoted).

water analyses (in mg/L except where otherwise

Drillhole/Site V2 V2 V2 V2 V2 V2

Sample Code No.Sampling Interval (n)Date CollectedTemperature (*C)pHCond (uS/ca)Eh(aV)Total Alkalinity

(mg/L, HCO3)Charge Balance

43-708-407811168.29.16

234-57

34.4

59-3401-4287811208.09.6

925105

9.4-4.81

69-1401-4287904067.09.4

830

-1.23

69-14401-4287905178.09.48

1000248

8.72.15

....406-41062112410.09.36

1410219

13-10.2

413-4178212148.39.20

1200-120

9-0.52

Sptcica

Ca

HgNaKso*rciBrI

SiO2BN02

N03

Alr« (total)MnCuZnCdPbCoHiCrVNoLISrRbBtBa

19.60.29640.245.4...90.5......

11.3

<0.05<0.05...

1100.18

1770.5555.4....449............10.2

....

....

....<0.05<0.05....

1080.17

176.00.54

42.3....437.0....

....10.4

....

....

....

....

....

1090.17

1690.5341.22.4

435.0<0.1........11.4

....

....

....

<o.os........

1560.35

2181.377.52.75

6087.0.50<.029.6.049

<0.005

<.02<0.01.005

<0.01<.005

1350.37

1961.6572.5

4905.1.14<.O2IB0.69<0.005

<.O2<0.01

.025<0.01<.OO5

<.005

<.005

<.005<.005<.005

.111.5<.005

.003

.099

<.005

<.005<.OO5

.161.3<.005<.003

.031

4:19

Table 4-1. Stripa water analyses (in mg/L except where otherwise

noted).

Dril lhole/Site

Sampling Interval (n)Date CollectedTemperature C OpHCotH (uS/cm)Eh(nV)Total Alkalinity

(mg/L, HCO3)Charge Balance

Species

CaMgNaKS0 4

rClBrIP0 4

S102

BN02

N03

NH4

AlFe (total)HnCuZoCdFbCoHiCrVHoLISrRbBeBa

V2

490-494830119

9.39.85

1220-5

IS-2.98

890.39

2240.73

554.2

4605.10.24<.O2

150.74<.O05<. l<.020.10

.004<.01<.005<.O1<.005<.01<.01<.01<,005<.O05<.O10.131.0<.005<.003.024

V2

549-553830207

9.49.84

1040-100

18-4.67

610.34

1980.69

515.6

3754 . 80.17<.O2

131.3<.005< . l<.02<.01

.27<.O1<.005<.O1<.005<.O1<.01<.O1<.005<.005<.O10.120.74<.005'.003

.006

V2

584-588830307

9.49.89

960-120

191.00

72<.005

1751.7

486.2

3254 .40.49<.02

160.47<.005<. l0.020.11

.015<.01.010

<.0l

<.01<.01<.005<.OO5

. . . ..070

0.74. . . .

<.003. . . .

4:20

4.5 Chemical analyses

Table 4-1 contains chemical analyses of Stripa groundwaters andnearby wells and streams for the period 1977-84, including boththe data from the SAC program (Fritz, et al., 1979, 1980) andthe international program. The analyses have been organized se-quentially in a hierarchy as follows:

Borehole or site designation

Size of borehole interval

Chronological sequence

The borehole or site sequence follows the order: surface waters,lakes, ponds, private wells, shallow boreholes, Rl, M3, El, Nl,VI, V2. The size sequence proceeds from largest to smallest, andthe chronological sequence proceeds from oldest sampling toyoungest. The determinations of Fe have been included for thefirst time because the sampling and preservation procedures havebeen found to be reliable.

Also, the following abbreviations are used: SW = surface water,T.P. - tailings pond, PW = private well, WT * water table and SB =shallow borehole.

A few analyses from the SAC program were not included. Thesewere samples that were either duplicates from the same locationand had identical chemistries, or they were samples clearly con-taminated by cementing material as indicated by a radical changein chemistry (high pH and Ca).

4.6 Saline groundwaters in central Sweden: Regional program

Southern and central Sweden have been invaded by seawater on twoseparate occasions during the Holocene. Approximately 10,000 -9,000 B.P. the Yoldia sea covered a large area directly acrosscentral Sweden, reaching levels of 150 - 170 m.a.s.l. on the pre-sent land surface* Stripa is located just within these limits(Figure 4-2), and thus, it Is possible that Yoldia seawatercould have infiltrated the Stripa granite as suggested by Fritz,et al. (1983). The second invasion came through Öresund, andonly affected coastal areas (mostly southern and western Swe-den), reaching levels of 45-55 m.a.s.l. (Engqvist, 1981). Thisinvasion of the Litorina sea could not have affected the Stripaarea.

There are many known occurrences of saline groundwaters (over

800 wells: Nordberg, 1981; Lindewald, 1981; Engqvist, 1981) in

central Sweden believed to be the result of entrapped Yoldia/Li-

torina seawater. A dozen locations were found and sampled to bet-

4:21

Figure 4-2. Southern and central Sweden showing sampling loca-tions, Stripa and the limits of Yoldia/Litorinasea invasion.

ter identify its chemical characteristics for comparison withthe Stripa groundwaters. The sampling locations are given in Fi-gure 4-2, and the analyses are provided in Table 4-2.

4.7 Distribution of salinity with location and time

A more definitive statement can now be made about changes in thewater chemistry with respect to location and time. The sugges-tion has been made that there is a gradual increase in chlorideconcentrations with depth (Fritz, et al., 1979; Nordstrom,1983a). Further sampling indicated an abrupt change in chloridebelow 770 m (Fritz, et al., 1980). With approximately 100 watersamples collected and analyzed to date, this question can be ad-dressed more confidently and in more detail.

4:22

Table 4-2. Chemical analyses of saline groundwaters in central Sweden(concentrations in mg/L).

SampleDate CollectedTemp (°C)Field pHCond (uS/cm)Eh <BV)AlkalinityCharge Balance (X)

Species

CaHgNaKSO,,FClBrIP04

SiO 2

BNO 2N03

NH<,AlFe (total)Fe2 +

HnCuZnCdPbCoNlCrVHoBeLiSrRbBaDOC

STENE8304225.88.31280224139

-0.8

556.63263.564.31.75012.350.011<0.02120.41<0.005<1.00.02<0.010.110.027<0.01<0.0050.01<0.005<0.01<0.01<0.01<0.0050.010<0.01<0.0030.111.2<0.010.0434

HÄSSELBY8304207.88.011750223262-5.4

100164165.4842.07002.430.016<0.02110.41<0.005<1.00.07<0.010.100.0330.07<0.0050.04<0.005<0.01<0.01<0.01<0.0050.057<0.01<0.0030.0381.5<0.010.060• -•-

ÄKER83050510.07.6462001179311.0

57761527951351.521007.700.560.15211.0<0.005<1.04.5<0.011.370.750.12<0.O050.02<0.005<0.01<0.01<0.01<0.0050.18<0.01<0.0030.130.98<0.010.04514

SMEDTOFTA8305046.77.052850....581-1.2

30268034741.511003.330.194.0190.56<0.005<1.09.40.255.65.20.44<0.0050.01<0.005<0.01<0.01<0.01<0.0050.071<0.01<0.0030.0270.33<0.010.05721

ROCKACARDEN8305059.57.782600180736

-0.9

484070040911.68352.020.160.08260.480.015<l.O1.70.010.500.240.13<0.0050.15<0.005<0.01<0.01<0.01<0.0050.093<0.01<0.0030.0850.88<0.010.05410

HAHHARÖ8305038.27.5-7.81710-2000202254-41

2722270161801.6540

0.030.03270.11<0.005<1.00.540.011.750.790.56<0.005<0.01<0.005<0.01<0.010.02<0.0050.039<0.01<0.0030.0423.9<0.010.1112

4:23

Table 4-2. Chemical analyses of saline groundwaters in central Sweden(concentrations in rag/L).

SampleDate CollectedTenp (°C)Field pHCond (uS/cn)Eh (aV)AlkalinityCharge Balance (£)

Species

CaMgNaKS04

FClBrIP0Asio2BNO 2NO 3NH4AlFe (total)Fe2+MnCuZnCdPbCoNlCrVHoBeLiSrRbBaDOC

ST. SUNDBY6305028.78.3719202831661.2

1086.34636.41071.87402.030.040.03200.29<0.005<1.0<0.02<0.010.0200.0010.02<0.0050.02<0.005<0.01<0.01<0.01<0.0050.011<0.01<0.0030.0691.2<0.010.0285

SKOFTEBY8305047.47.933750214879

-0.85

5358112059663.015005.060.523.2181.3<0.005<1.03.2<".010.0760.0310.04<0.0050.01<0.005<0.01<0.01<0.01<0.0050.13<0.01<0.0030.O741.6<0.Cl0.1618

OSMO-JURSTA830422

7.343354512142.2

758.36.31115.1810....<0.0053.5130.021<0.005450.04<0.01<0.01<0.01<0.010.0360.11<0.005<0.01<0.01<0.01<0.0050.020<0.01<0.003<0.0030.16<0.010.022....

HANCLÖSA83050412.58.0219402666901.7

, 2511.554022812.84501.780.150.0615.50.800.008<1.00.68<0.01.069.057.OV.03.03<0.005<0.01<.005<.004<0.005<.OO5<0.01<0.0030.054.44<0.01.4312

KACA83042111.77.382990252164-12

400123802133.8140014.30.080<0.02110.300.055<1.00.03<0.010.220.021.20<0.0050.27<0.005<0.01<0.005<0.004<0.005.012<0.01<0.0030.148.5<0.013.53.8

UILHELMSLUND83042110.57.88362019653

-3.1

J664.0587121.041.8163018.30.16<0.026.60.440.030<1.00.72<0.010.290.0460.230.0120.09<0.005<0.01<0.01<0.010.023<0.005<0.01<0.0030.378.1<0.010.92....

4:24

Fractures Flow Depth (m) Chloride Concentrations (mg/U(ml mm)

400 r

500 -

600

700

— .20

— 70— 12

800

— 300 90°

1000

1100

1200

1300 L

T841

85

117

185

190

88

285

287

500:460

460

= 375 -:325

540

Figure 4-3. Chloride concentrations of sampled intervals in V2borehole from oldest to youngest (far right).

Several short interval samples have been collected which are in-dicative of narrow fracture zones, eliminating the possibilityof mixing of different zones within the borehole. These narrowintervals can be compared to large intervals to gain an apprecia-tion for chemical variability over different sections of a sing-le borehole. The two best examples are from boreholes V2 and Nl.Figure 4-3 shows some of the fracture zones, flows, depth andchloride concentrations that have been measured in the V2 bore-hole. The chloride concentrations are in chronological sequencefrom left to right (right-hand side is the most recent in time).

The first, most obvious, observation is that several water samp-les have been taken over large intervals of the borehole, andthe larger the borehole interval the lower the chloride concent-ration tends to be. In these instances there is a considerableamount of mixing of waters from different fracture zones in thesame borehole. Where narrow intervals of the borehole (<50 m)have been sampled, there Is considerable variation In chloride

4:25

1 I IDate:

•35 | ' 42—H 84-01-26

122 58 48

If 82-08-03/09-23

40 H 82-06-03

51 | 81-10-06

40 | 81-08-19

46 | 81-06-03

J I I I0 50 100 150 200 25& 300

N1 BOREHOLE INTERVAL (meters)

Figure 4-4. Chloride concentrations (in mg/L) of sampled inter-vals in Nl borehole from oldest to youngest (top).

concentration from 84 mg/L to 600 mg/L. One zone giving 88 mg/Lchloride is only 47 m away from another zone having 600 mg/Lchloride. There is no clear trend with depth, although the hig-hest salinity tends to be around 800-900 m depth; and samplestaken below 800 m are consistently higher in salinity than samp-les taken from above 800 m. Figure 4-4 shows chloride concentra-tions at different intervals of the Nl borehole. The samplingdates are shown on the right-hand side. Again it can be seenthat when narrow intervals are sampled, zones of distinctly hig-her salinity can be found. The water chemistry is a very sensiti-ve measure of the location of different types of permeable frac-ture zones.

The conclusion is that the salinity is quite variable from onefracture zone to the next, the existence of the mine may havehad a large influence on diluting the salinity above 800 m, andthe geocheraical evolution of these groundwaters is closely asso-ciated with the fractures of the local rock. This also indicatesthat there is not enough interconnection between the fracturesto allow sufficient mixing to give a more homogeneous water che-mistry or a clear pattern of mixing with depth. Mixing in thiscontext simply refers to mixing of groundwater between fracturezones or between water-bearing conduits in the granite. This sa-linity-depth profile will be called "heterogeneous," in contrastto the type of "homogeneous" profile which has been observed inSweden and Finland where Baltic seawater (or Yoldia/Litorlna se-awater) has infiltrated crystalline bedrock. An example of thelatter can be seen in Figure 4-5 (Snellman, 1982).

The results from Stripa lead to the conclusion that there is nota continuous increase in salinity with depth, nor is there a

4:26

Surface

1000 2000 3000 4000

CHLORIDE (mg/L)

5000 6000

Figure 4-5. Chloride vs. depth profile for Yl borehole, Lovi-isa, Finland.

sudden break in salinity below 770 m. Instead, there is an irre-gular increase in chloride with depth (the heterogeneous profi-le). In Nl, a 2-m interval is discharging water of 120 rag/L chlo-ride at a depth of about 360 m. At 92-94 m in VI borehole (about460 m depth), a discharging water contains about 200 mg/L chlori-de. These and other irregularities indicate that there is noclear break in the salinity with depth, but rather an irregularincrease much like the depth profile for hydraulic conductivity.

Detailed mapping of fracture-zone water chemistry shows promiseas a tool in finding flow paths in crystalline rock. It might bepossible to make permeable connections between boreholes basedon similarities in water chemistry.

The heterogeneous profile seen at Stripa suggests an originother than simple intrusion of saline water from an external so-urce. However, the general chemical characteristics of these wa-ters are very much the same from one fracture zone to another,i.e., independent of salinity (see chapter *>), which impliesthere is a single type of source or a single type of process ac-counting for the salty components. This heterogeneous profilemay be typical of groundwater in crystalline rocks that have notbeen intruded by saline waters such as seawater.

4:27

1977 1978 1979 1980 1981 1982 1983 1984

Figure 4-6. Chloride concentrations as a function of time forthe M3 borehole.

Chemical variations with time are not exactly clear. The datashown in Figure 4-3 for the V2 borehole suffers from two prob-lems: (1) rarely has the same borehole interval been sampledmore than once, and (2) it is known that the discharges from thevarious fracture zones change with time; and if they don't allchange proportionately the same, then the change in chemistrycould simply reflect different borehole mixtures obtained by va-rying hydrologic flow conditions. Comparing the interval 690-880m in V2, the increase in chloride from 190 to 287 mg/L may bea change in a single fracture system, or it may be due to aslightly different positioning of the packers, or it may be dueto different proportions of flows from individual fractureszones.

The M3 borehole has been monitored over a long period of timeand shows a recognizable pattern (Figure 4-6). Background chlori-de concentrations were initially about 52 mg/L. On June 1, 1978,the heater experiments were begun, and one week later chloridevalues jumped to 66 mg/L. Following that initial increase, thechloride values follow a decay curve leveling off at around 36mg/L. These data show that a pulse of chloride was released as aresult of the start-up of the heater experiments, and with timethe chloride decreased to constant values that are significantlylower than the original background. These results could be inter-preted in either of two ways. First, if the chloride originatesfrom the rock (such as fluid inclusions), then the thermal gradi-ents could have caused an increase in diffusion of chloridealong micro-cracks, causing a pulse of chloride to come out; orit might even have caused some mechanical raicrofracturing in wa-

4:28

ter-bearing zones, which would immediately release a greater amo-unt of chloride from inclusions. Thermal stress experiments onnatural quartz grains have shown that water from fluid inclu-sions Is released by microf racturing at less than 100°C up to550°C (Barker and Robinson, 1984). Second, the thermal stressmay have changed the flow direction in fractures so that waterat higher chloride concentrations was temporarily rerouted tothe M3 borehole. The flow direction may have been altered by theaffect of thermal convection on flow, or by changes in porosityand flow channels caused by the heating. Distinguishing these ef-fects would be very difficult without additional information.Further discussion on the effects of the heater experiments onthe water chemistry from nearby boreholes can be found in Fritz,et al. (1980).

Other boreholes, such as Rl and VI, do not seem to show any sig-

nificant chemical variability with time.

4.8 Ion ratios and classification of groundwater

Ion ratios can provide a useful means of distinguishing diffe-rent water types, sources of water, and their classification(e.g., White, et al., 1963). Although there are limitations tothe interpretation based only on ion ratios, these ratios do pro-vide a way of initially characterizing the water chemistry. Chlo-ride is useful for normalizing because It is one of the most con-servative common ions in natural waters, and it provides a muchclearer picture of depicting the water chemistry than comparingions with depth or specific conductance. Ratios of ions to chlo-ride can also provide some Information on the evolution of sali-ne waters. A good example is provided by Eugster (1980), who des-cribes the origin and evolution of brines at Lake Magadi, Kenya,using chloride as a normalizing factor. The main purpose of thissection will be to compare the Stripa groundwater chemistry withother possible sources of salinity, such as seawater, Yoldia seawa-ter, Baltic water, and the sedimentary basin brines in the Creta-ceous-Jurassic strata of southernmost Sweden. These four typesof saline water are the only ones known to occur in Sweden.Table 4-3 summarizes several ion ratios for these four types ofwater and for several locations at Stripa.

Permian evaporates transported to Stripa have also beensuggested as a source of salinity.

4.8.1 jJeawater

Surface seawater has a very constant composition for all majorconstituents and most minor and trace constituents regardless oflocation. The occurrence of modern seawater into Swedish ground-

4:29

Table 4-3. Ion ratios (by weight) for saline waters and Stri-pa waters.

lUNIII

••itu MIM

Mi l l HHttir:Itt Ilat tt

ttia* krlM»

VI ana V]

III

t i

i i u l i w «*1U

Surface «at«ra

•r/Cl

0.001*7

0.00)42

0.001*20.0107

0.00»}

0.0107

0.0101

0.00*7

0.00J*

0.0077

<I/Cl)xlO6

1.1

16177

*O-JJ

«*•

1)»

112

110

uoo

Ca/M»

0.12

0.32

* .7•5

22

*JI

10*

M

».7

*.»

H«/C1

0.047

0.M7

0.0)20.00)5

0.012

0.00107

O.OMt

0.01»

O.«*

O.*O

Ci/Cl

0.021

0.022

0.0720.2»

0.20

0.23

o.»a

O.al

7.1

1.77

•a/Cl

0.J»

O.JJ

0.7*0.J2

O.JO

O.*7

O.t»

l . M

1.17

0.»)

t/ei

0.020»

0.020}

0.0)0»0.011

0.011

0.00*0

O.flll

0.0*1

0.1*

0.21

IO4/C1

o.u

0.1*

0.12}0.012

0.0021

o.u

0.01»

0.1)*

2.1

2.1

waters is limited to coastal areas, especially near highly popu-

lated areas, such as Göteborg and Skåne, where high pumpage

rates encourages the intrusion of seawater (Nordberg, 1981).

Unaltered modern seawater clearly cannot be entering the Stripa

granite.

4.8.2 Baltic seawater

The Baltic sea is a mixture of seawater and fresh waters ente-ring the Baltic region. The dilution factor averages 6.5 for sur-face Baltic water, but decreases considerably with depth becausethe denser seawater enters the Öresund (the sound between Den-mark and Sweden) along the sea bottom. Since the Baltic is a re-latively simple mixture, many of the element ratios should bethe same as those of seawater. Extensive investigations by Krem-ling (1969, 1970, 1972) at 21 stations from the Öresund to theBothnian Bay and the Gulf of Finland show that even with conside-rable dilution the ion ratios deviate very little from seawater(Table 4-3). Coastal Baltic seawater intrusion also occurs insome coastal aquifers along the east coast (Sund and Bergman,1981), but direct intrusion from the modern Baltic cannot be thesource of saline components in the Stripa groundwaters becauseion ratios are incompatible.

4.8.3 Yoldia seawater

As previously mentioned, the Yoldia sea invaded central Swedensome 9,000 years ago and in some places became entrapped in Qua-ternary clay deposits of low permeability. These waters have che-mical compositions which have undergone modification from theoriginal seawater composition by such processes as clay forma-tion, ion exchange, and mineral precipitation (Jacks, 1973,

4:30

1978; Agerstrand, et al. 1981). The analyses for these watersshown In Table 4-2 can be grouped Into two sets, based on Ion ra-tios. Set I comprises eight samples which have nearly IdenticalBr/Cl ratios averaging 0.00342 (+0.00071) which Is Indistinguis-hable from modern seawater and Ca/Mg ratios averaging 4.74(+5.8). Set II comprises two samples (Kaga and Wilhelmslund)which average 0.0107 for Br/Cl and 65 for Ca/Mg, considerablyhigher than all of the other samples. Of the remaining two samp-les, one was found to be entirely diluted with fresh water, andthe other was unavailable for Br analysis. Set I is consideredto be typical of Yoldia seawater altered by water-rock and biolo-gically-mediated reactions. The ion ratios for this set fall inthe same range as those from coastal groundwaters in Finland,which are assumed to be Yoldia or Litorina seawater (e.g., Hyyp-pä, 1984). Set II has ion ratios that are more comparable tothose found at Stripa, especially the Br/Cl, Ca/Mg and Mg/Cl ra-tios.

Thus, the Stripa groundwaters are not unique in their chemistry,and they are not comparable to Yoldia/Litorina entrapped seawa-ter. Elimination of Holocene seawater as a possible source ofthe saline components at Stripa means that these components (andprobably the groundwater itself) must be older than 10,000 yearsB.P.

4.8.4 J_ur_assjLcHCre£a£eou£ £edimetitary_ basl£ ]»rine_s_of_ Skåne

Deep (1 km) subsurface brines are known to occur at Skåne in Ju-rassic and Cretaceous sedimentary rocks (Brotzen and Assarson,1951). Chloride concentrations can be as high as 154,000 mg/L,and the general chemical characteristics are similar to otherbrines in deep sedimentary basins of the world» Ion ratios forthese brines (shown in Table 4-3) appear to be roughly interme-diate between the altered Yoldia seawater ratios and those ofStripa. Assuming that the saline components at Stripa had a mari-ne origin, then it is possible to interpret them as the extremecase of modified seawater in which the Yoldia waters and theSkåne brines are intermediate steps. The progression from modernseawater ion ratios to the Stripa ion ratios may represent theeffect of time, or the effect of temperature, or both. The tempe-rature effect may be important because the Skåne brines areslightly elevated in temperature (up to at least 80°C), and therocks in the Stripa region have undergone much higher temperatu-re alterations. Further discussion of a possible temperature sig-nature in the Stripa groundwater chemistry is discussed In Chap-ter 5.

4:31

4.8.5

4.8.6

Extensive deposits of Permian evaporite beds are found in largeareas throuhout Germany, Russia and Poland and the hypothesis isdeveloped in Chapter 7 that some of these may have been erodedand subsequently transported to central Sweden. This suggestionis based on the stable isotope data on dissolved sulfate whichare similar to the values for Permian marine sulfates. However,there are no known Permian deposits in Sweden so that transportof dissolved salts over long distances is required. Such trans-port would involve dissolution and probably redox processeswhich would tend to destroy and/or dilute the Permian isotopicsignature. Furthermore, erosion of Permian deposits such as theZechstein evaporates is confined to just a few diapiric intru-sions composed dominantly of halite and smaller amounts of gyp-sym, anhydrite and carnallite. Hence the Br/Cl ratio would be do-minantly reflected by the ratio found in halite which is conside-rably lower than seawater and an order of magnitude or morelower than the ratio in the Stripa groundwaters. Finally, theFennoscandian shield has been an area of uplift and erosionsince the Devonian and has not experienced any depositional envi-ronments until the Holocene (Brinkmann, 1969). It is very diffi-cult to imagine a scenario that involves erosion, transport andinfiltration at some unidentifiable period of time of evaporatesso that the subsequent saline water contains the Permian sulfatestable isotope signature but most of the rest of the chemistryhas changed.

The ion ratios listed in Table 4-3 for Stripa groundwaters showseveral anomalous trends which must be accounted for in any geo-chemical interpretation:

1. Br/Cl and I/Cl ratios are markedly higher than seawater andmodified seawater.

2. Ca/Mg ratios are 1-2 orders of magnitude higher than seawa-

ter and modified seawater.

3. Mg/Cl are markedly lower than seawater and modified seawa-ter.

These anomalies cannot be fully accounted for by modern, Holoce-ne, Paleozoic or Mesozoic seawater. An alternative hypothesis,based on fluid inclusion leakage, is presented in Chapter 5.

Another important aspect of the water chemistry is the compari-son of ion ratios between various boreholes. From Table 4-3 itwould appear that if VI and V2 are characteristic of one type of

4.8.7

4:32

water, then boreholes Nl and El are a mixture of VI/V2 type withtypical surface waters or shallow groundwaters. El is dilutedmuch more with freshwater than Nl because the Br/Cl ratio of Elis more like a shallow groundwater, and the total chloride con-centration is less than Nl. A regular trend in nearly all theion ratios can also be seen in progressing from V1/V2 to the sur-face waters that indicates a mixture of two sources of dissolvedions. These trends are examined in more detail in Chapter 5.

A study completed by J.L. Means (1981) showed that the dissolvedorganics in a water sample collected from an unspecified locali-ty underground at Stripa contained low molecular weight fulvicacid fraction (<700 MW). Much more research Is needed to under-stand the role of organic compounds at Stripa.

More Interesting results came from a study by J.M. West (pers.comm.) on the microbiology of 2 samples of borehole water, M3and the 100 - 505 m interval in VI. The M3 borehole sample wasdominated by aerobic heterotrophs and contained smaller concen-trations of anaerobic heterotrophs and denitrifying bacteria.These populations would indicate that some small amount of oxy-gen is present in M3, consistent with the high tritium valuesand the implication of recent recharge. The water sample from VI,however, contains almost no aerobic heterotrophs and a high con-centration of anaerobic heterotrophs with a moderate concentra-tion of denitrifying bacteria. VI is clearly a more reducedwater that can be considered anoxic.

5:1

WATER-ROCK INTERACTIONS

5.1 Introduction

Preliminary interpretations of water-mineral reactions have beendiscussed in a qualitative fashion in the earlier reports ofFritz, et al. (1979, 1980) and Nordstrom (1983a, b). In thischapter a more quantitative picture is described based on a muchmore extensive set of water analyses that have undergone qualitycontrol review. There is no question that reactions of thegroundwater with the bedrock have taken place to a considerable ex-tent, altering its chemistry, and making interpretations diffi-cult. Several of these reactions can be identified using chlori-de correlation plots, chemical equilibrium computations, andfluid-inclusion data. Equilibrium computations were done withthe WATEQ3 program (Ball, et al., 1980, 1981). The results ofthese investigations indicate that it is possible to explain theorigin and evolutionary processes of the groundwater chemistryby interactions between infiltrating meteoric freshwaters withthe crystalline bedrock, such as the granite and the leptite.

5. 2 Chloride correlations and mineral reactions

In most hydrogeochemical processes, chloride acts in a stronglyconservative manner. It enters into precipitation-dissolutiononly at high-salt (brine) concentrations, and it rarely entersinto oxidation-reduction reactions or adsorption reactions (Feth,1981). It is, therefore, frequently used as a conservative tra-cer in deciphering hydrochemical processes. The groundwater che-mistry data at Stripa covers a large range of chloride concentra-tions, and depth profiles are not meaningful due to their hetero-geneity. Hence, chloride can be used as a tracer to define theamount of loss (or gain) of other constituents, as well as a nor-malizing factor to interpret possible sources of saline constitu-ents.

5. 2.1 Al]cali_m£ta_ls :_Na,JCj_ Li

Sodium is frequently conserved in hydrochemical processes inwhich It is strongly associated with chloride. For example, sodi-um is preferred over chloride as a reference element for study-ing fractionation of elements in atmospheric processes (Duce andHoffman, 1976), and Eugater (1980) has shown that over a largerange of evaporative concentration, saline lakes can be strongly

5:2

388

2S0 .

zee .

158 .

100 i o /9

oo

oo

58O/O.

i i i i r i i

lee 280 300 488 500 600 788

Chloride. mg/L

Figure 5-1. Plot of sodium against chloride concentrations-Narrow interval samples representing single water-bearing fracture zones are shown by closed circlesin this and following plots.

conservative with respect to both sodium and chloride. There is

also evidence for the conservative nature of sodium in natural

water during weathering (Graustein, 1981) under certain condi-

tions-

Figure 5-1 plots sodium against chloride for the Stripa waters,

and a fairly constant increase Is apparent that could be inter-

preted as a mixing line (showing a small loss of sodium compared

to the seawater dilution line). Many of the borehole intervals

were so large that mixing did occur between different water-be-

aring fracture zones in the borehole. To determine whether sing-

le fracture zones deviate from this apparent mixing line, the

narrow intervals are shown as closed circles. From this compari-

son (and from several other elements) there is no significant de-

viation from the mixing line for the individual water-bearing

zones within the analytical uncertainty of the data. Hence,

there are no major losses of sodium during mixing. Comparison of

the Na:Cl correlation with the ratio for seawater shows a : .: ; :

that suggests a loss of sodium or gain of chloride compared .:-

seawater streamwater mixtures. If this loss is real, it must

have occurred before invasion of the saline waters into the Stri--

pa granite, or there would have been a change in slope. The only

change in slope that can be demonstrated is shown in the semi—

log Na:Cl plot of Figure 5-2. At around 40 mg/L chloride, the so-

dium values intersect from a steep slope at low concentrations

to a lower slope at higher concentrations. This slope change at

5:3

,e2.5.

A

,e1.5.

1001

zeei i

CHI or I da. mg/L

T

see1

eee

Figure 5-2. Semi-log plot of sodium against chloride concentra-tions.

40 mg/L chloride is consistent with a few other constituents.For example, the Br/Cl ratio becomes constant above 40 mg/L chlo-ride. This point may mark the change from shallow weathering pro-cesses to deep seated processes that are characterized by a dif-ferent set of reactions. The lower slope that correlates withchloride means that the main source of sodium at higher concen-trations is a salty component, and it Is not primarily derivedfrom feldspar weathering.

Lithium also shows a similar increase with chloride as sodium,again suggesting a mixing line (Figure 5-3). The lithium concen-trations do not cover a very large range, but they are stronglyenhanced over the seawa_er dilution line. Lithium also tends tobe conservative in dilute waters, and this Is suggested by theStripa data.

Potassium shows an entirely different trend. There may be aslight increase In potassium at the very highest chloride values(Figure 5-4), but most of the values are fairly low and constantregardless of chloride. This strongly nonconservative trend sug-gests that the source of the chloride has little or no potassiumassociated with it, and that either a steady-state weatheringrate or a mineral solubility equilibrium is controlling the po-tassium concentrations. VATEQ computations show that these wa-ters are both undersaturated and supersaturated with respect tollllte. However, it Is felt that the uncertainties in (a) the so-lubility product constant for illite, (b) some of the aluminumdeterminations, and (c) some of the potassium determinations are

5:4

0 2

8. 175 .

0 15 .

h 0 125 .Iu

* a. i .

e 8 875 .

L0.05 .

0 025 .

SEAWATER DILUTION LINE

t"" 7 "I T ~1 ~1 I100 208 380 480 580 680 708

Chloride, mg/L

Figure 5-3. Linear plot of lithium against chloride concentra-

tions.

3 .

2 .

I .

e too 400 500Ch I or i Of», fflg/'u

600 700

Figure 5-4. Plot of potassiui against chloride concentrations.

5:5

0/L

0 f-100 200 300 -»00

Chloride.

500 600 700

Figure 5-5. Plot of calcium against chloride concentrations-

such that it is difficult to claim equilibrium saturation for

this mineral. This solubility control appears likely, but is not

proven.

5. 2. 2

The correlation plot for calcium against chloride, shown in Figu-re 5-5, has several significant features. The general trend withdepth (see Figure 5-7) shows increasing calcium concentrationsto a depth of about 100 m where the concentrations peak, andthen they decrease with increasing chloride to about 50 mg/Lchloride (strikingly close to the change in slope for the sodiumvs. chloride plot). Thereafter a steady, almost linear, increaseis observed that corresponds to depths of about 360 m (N 1 bore-hole) and deeper. The Ca/Cl slope is strongly enhanced above theBeawater dilution line. The initial increase in calcium at lessthan 100 m depth suggests calcite dissolution up to saturationwhere the peak in the curve occurs. Having reached calcite satu-ration, the groundwater then precipitates calcite at increasingchloride concentrations. This evolutionary sequence is similarto that outlined by Fritz, et al. (1979) and Jacks (1978).

The pH of groundwaters is often regulated by equilibria in theCaO-COj-I^O system, i.e., by calcite solubility, CO2 solubility,and the aqueous carbonate equilibria. The general feature-? ofthis equilibria in groundwater have been described by Langmuir(1971) and summarized by Freeze and Cherry (1979). A schematicevolutionary diagram applicable to Stripa is shown in Figure 5-6.

5:6

§8

Figure 5-6. Schematic evolutionary diagram of bicarbonate, pHand Pco during calcite dissolution.

Shallow dilute groundwaters have pH values near 6, and carbondioxide partial pressures (PCQ ) about 10"2 bars. Under theseconditions, calcite is unstable and tends to dissolve under par-tially closed conditions to reach the high pH values of around 8at saturation. If calcite then begins precipitating, the pH sho-uld be maintained close to 8. The deep groundwaters, however,have pH values of 9-10, and the calcium concentrations have in-creased considerably. This unusual situation is displayed moreclearly in Figure 5-7 where the plots for calcium, bicarbonate,pH and calcite saturation index (S.I.) are shown as a functionof chloride concentration. The S.I. is defined as the log of theion activity product, AP, divided by the solubility product con-stant, K (Drever, 1982). The bicarbonate alkalinity mimics thefirst part of the calcium curve, reaching a peak at around 200mg/L and then steadily decreasing at higher salinities, sugges-ting calcite precipitation is active over the remaining range ofsalinity. This interpretation is confirmed by the S.I. for calci-te which shows undersaturation for the most dilute groundwatersless than 100 m depth, reaching saturation at a pH of 8 (coinci-dent with the peaks in the bicarbonate and calcium curves). TheS.I. continues to increase to one-half an order of magnitude su-persaturated where it remains constant in spite of increasingcalcium concentrations. This behavior can be explained if the bi-carbonate concentrations are decreasing due to calcite precipita-tion which in turn is being driven ty calcium input from someother source. Calcium input may be derived from the weatheringof plagioclase feldspar or from the same saline source as thechloride. The strong correlation of calcium with chloride sug-gests the latter is the dominant process. It Is important tonote, however, that feldspar weathering must play a role to main-tain the high pH values and part of the calcite supersaturation.The rise to pH • 8 can be accounted for by closed-system calcitedissolution, but once saturation is reached, calcite precipita-tion will not allow the pH to increase any further. The fact

5:7

S1

SUPERS»TUR»"ECUNDERS1TURATED

St»W»TER DILUTION LINE

Chlorirt», «g/L

Figure 5-7. Trends in carbonate geochemistry with Cl concentra-tions.

5:8

12

te 4

\oo

T-o • o

Ttea 280 300 400 588

Chlorld», mg/L

688

O

—r~780

Figure 5-8. Plot of magnesium against chloride concentrations.

that the pH continues to increase to almost 10 suggests that si-licate (feldspar) hydrolysis is the primary factor behind calci-te precipitation until the calcium concentrations begin to inc-rease again. At that point the saline input may be just as impor-tant as feldspar dissolution In maintaining calcite supersatura-tion.

The degree of calcite supersaturation shown in Figure 5-7 is pro-bably a real supersaturatlon because errors in the thermodynamicdata (Plummer and Busenberg, 1982), and the analytical data(shown by the scatter in the points) are less than this amount.The combined error from these two sources does not exceed about0.25 log units, which is about half of the degree of supersatura-tion.

This outline of the carbonate geochemistry of deep groundwatersat Strlpa complements and extends the earlier discussions ofJacks (1978), and provides a general picture that should be ap-plicable to most deep groundwaters in crystalline rocks of simi-lar salinity.

Magnesium concentrations range from 2-12 mg/L in the shallow sub-surface, and generally reflect chemical weathering of magnesiumsilicates by carbonic acid (Keller et al., 1963). As the carbona-te alkalinity decreases below about 50 mg/L, the magnesium isfixed below 1 mg/L. This change in magnesium concentration corre-sponds t c about 40 mg/L chloride, shown in Figure 5-8, whichalso corresponds to the change in slope of the sodium: chlorideplot. The strong lack of correlation with chloride suggests a

5:9

2 5

2 .

V 1.S .

1 .

8.5 .

100 280 300 400 500 600 700eChloride, mg/L

Figure 5-9. Plot of strontium against chloride concentrations.

greatly reduced weathering rate of magnesium silicates at depthsgreater than about 100 m and a possible solubility control.WATEQ computations for the S.I. values of sepiolite, Mg^Si^O^OH^'oHjO, and pure Mg-chlorite (magnesium-clinochlore,both 7A and 14A), Mg5Al(Si3Al)O10(OH)8, indicate a range of afew orders of magnitude covering both undersaturation and super-saturation. The range of S.I. primarily reflects the complexityof the mineral formula and, secondarily, the uncertainties inthe analytical determinations of magnesium and aluminum whichare larger than the uncertainties in most other constituents.Even with improved analytical data no firm conclusions could bedrawn because the range of Fe-Mg, Al-Fe, and Al-Si substitutionin chlorite produces such a range of solubility (as well as in-creased uncertainties) as to be obviously ambiguous. S.I. computa-tions for magnesite, MgCO-j, and dolomite, CaMg(CO3)2 are all un-dersaturated for the whole range of chloride concentrations.

The plot in Figure 5-8 does make an important point, however,that at chloride concentrations up to 700 mg/L there is no corres-ponding increase in the magnesium concentrations. Sodium andcalcium both increase in spite of calcite precipitation, but mag-nesium doesn't increase, not even where the magnesium silicatesaturation values are 1-2 orders of magnitude undersaturated.This observation suggests that little or no magnesium is beingcontributed along with sodium, calcium, and chloride from the sa-line source. The magnesium content is orders of magnitude lowerthan the calcium content, and can't be simply explained by ionexchange phenomena. This result contrasts sharply with the inter-

5:10

SUPERSATURATED

UNDERSATURATED

-i

*T--3 4

-4 J

-5

"T

-6 iI

-7 J9 100 200 300 400 500 600 700

CMortde, WQ/L

Figure 5-10. Plot of the S.I. for strontianite against chloride

concentrations.

pretation of Jacks (1978), and will be discussed further in Sec-tion 5.3.

Strontium in natural waters might be regulated by the solubilityof celestite, SrSO^, or strontianite, SrCO-j. The chloride corre-lation plot in Figure 5-9 shows an increase in strontium compa-rable to the plots for sodium and calcium. Strontium is an impor-tant component of the saline source. All of the Stripa watersare undersaturated with respect to celestite because of the rela-tively low salinity of these waters, and because celestite ismore soluble than strontianite. The S.I. values for strontianiteare plotted in Figure 5-10, and demonstrate a constant undersatu-ration of about one order of magnitude for all the water samp-les. Other than a small amount of substitution into calcitethere is no solubility control on strontium concentrations. Theconstant amount of undersaturation most likely reflects the cons-tant S.I. of calcite and the removal of carbonate alkalinity bycalcite precipitation. The carbonate content is kept so low thatstrontianite saturation is simply never reached.

Barium does not correlate with chloride as shown in Figure 5-11.The highest barium concentrations are found in the four short in-terval samples of the N 1 borehole. N 1 borehole waters also con-tain some of the highest concentrations of HoS (up to 0.03mg/L), and these results may indicate enough sulfate reductionis occurring in N 1 so that barlte might dissolve. The S.I. valu-es for barite shown in Figure 5-12 indicate barite undersatura-tion in all but the deepest (V 1 and V 2) boreholes where satura-

0 2

a 175 _,

e is ,

r

um

m

/I

0

0

125 .

0 1 -

075 .

•

•

0.05 ..

0 025 _ o

5:11

~l

,0 O O

r

0 108 200 300 400 500 600 700

Chloride, mg/L

Figure 5-11. Plot of barium against chloride concentrations.

Lo0

AP

B 5 .

0 SU

-0 5 .

-I .

-I .5 .

-2 .

-2 5

SUPERSATURATEDUNDERSATURATED

d _ _ —

• 0 0

1 I I T I I 1180 200 300 400 500 600 700

Chi or Ida, me/t

Figure 5-12. Plot of the S.I. for barite against chloride con-centrations.

5:12

12

6 H

4 .

2 .

O

00

SE»1

0 180 280I I

300 488

Chi or I da. mg/L

T

500 680I

700

Figure 5-13. Plot of bromide against chloride concentrations.

tion is reached. This saturation equilibrium would explain thelow and constant concentration of barium at the higher chlorideconcentrations.

5.2.3

Both bromide and iodide correlate very strongly with chloride,and Br/Cl and I/Cl ratios are strongly enhanced relative to sea-water (Figures 5-13 and 5-14). The seawater dilution curve foriodide is congruent with the horizontal axis for the scale shownin Figure 5-14. These linear correlations are strong evidencefor a simple mixing of a saline source with fresh groundwaters*The saline source might be the intrusion of a saline aquiferfrom somewhere originally external to the Proteroi.oic bedrock,or it might be internal to it, i.e., it might be salt from with-in the granite and leptite being leached. This latter possibili-ty does not mean that two water bodies are mixing, but ratherthat the bulk groundwater is mixing with a source of salt whosewater volume is negligible relative to the bulk groundwater. Theformer possibility should show distinct trends in D and 0 iso-tope content with salinity, and the latter possibility wouldn't.In any event, two-component mixing is indicated, and bromide andiodide are associated with the saline source.

Fluoride shows no correlation with chloride and strongly indicat-es a mineral reaction-controlling process (Figure 5-15). The mi-neral fluorite is fairly common as a fracture-filling, and wouldprovide an obvious solubility control on fluoride concentra-

5:13

odid

Q

L

0.175 .

0 15 .

0 125 .

0 1 .

0.075 .

0.05 .

0 025 .

0

<*>

1 l

0 0 0

00

00 0

1 1 1 1

100 200 300

ChI or I d«. mg/L

400 500 600 700

Figure 5-14. Plot of iodide against chloride concentrations.

14

12

10 J

"T T T"200 300 400 500

Chloride, mg/L

600 700

Figure 5-15. Plot of fluoride against chloride concentrations.

5:14

1 .

0 .

-I .

-2 .

-3

SUPERSATURATED

UNDERSATURATED

tee 280—r~380

T400

CM or Id», mg/L

—r~500

600 700

Figure 5-16. Plot of the fluorite S.I. against chloride concent-

rations.

tions. S.I. values for fluorite are shown in Figure 5-16 whereup to an order of magnitude supersaturation exists, although thevalues are very constant with increasing chloride. The K usedfor fluorite is believed to be about 0.3 log units too insolublebased on more recent data (see citations by Ball, et al.,(1979). If this correction is made, the values would only be su-persaturated by about one-half an order of magnitude just likethe calcite S.I. values. It is difficult to say just how realthis amount of supersaturation is, but it can be concluded thatfluorite solubility controls the fluoride concentrations in theStripa groundwaters, and that modern-day precipitation of fluo-rite in the fractures seems likely.

5.2.4

Although it has been mentioned that there are some large uncer-tainties in the aluminum determinations, partly due to contamina-tion problems and partly due to analytical errors, it is never-theless possible to make some constructive statements abcut aluml-nosilicate reactions.

Samples collected from V I , V 2, R 1, and M 3 boreholes betweenJune and November, 1982, have more reliable aluminum values thanthe rest of the samples analyzed. These particular samples havebeen analyzed for aluminum by three different analysts and bytwo, three, or sometimes four independent methods. The valuesare generally in agreement within about 10%. WATEQ computations

5.2.5

5:15

on these sample results show consistent undersaturation with res-pect to gibbsite, and random under- and oversaturation with res-pect to kaolinite. It is safe to say that gibbsite solubilitydoes not play a role in regulating aluminum concentrations, andthat some more complex aluminosilicate phase probably is. Kaoli-nite, chlorite, illite, smectite, or a combination of these, maybe important as solubility controls, but neither the thermodyna-mic data nor the precision on the aluminum, potassium, or magne-sium concentrations allow such statements to be made with muchcertainty.

A noteworthy observation that came out of the WATEQ computationsis the consistent slight supersaturation with respect to prehniteand laumontite in all of the V 1 and V 2 waters. These two mine-rals are very common fracture-fill minerals in granitic rocks(e.g., Tullborg and Larson, 1982; Kerrich, 1984), and are common-ly found in geothermal mineral assemblages of 100-350°C (Bird,et al., 1984) along with wairakite, epidote, and chlorite. TheWATEQ results indicate these minerals are also present in theStripa granite (and leptite) fractures, although they haven'tyet been identified. The presence of lauraontite generally indica-tes low-temperature metamorphism up to about 200°C, and prehniteand epidote indicate temperatures of 2OO-35O°C (Zen and Thomp-son, 1974; Bird and Helgeson, 1981; Bird, et_al., 1984). Both inprogressive burial metamorphism and in geothermal alteration,the appearance of large amounts of chlorite usually occurs attemperatures of 150°C and higher (Boles and Franks, 1979; Zenand Thompson, 1974). In fact, prehnite, wairakite, laumontite,and epidote are important "zone markers" or "index minerals" forthe intensity of alteration by temperature and pressure gradi-ents (Schiffman, et al., 1984; Zen and Thompson, 1974). Epidoteand chlorite are common in the Stripa fractures, and the occur-rence of laumontite and prehnite should be expected from theabove discussion. Further fracture mineralogical studies shouldbe done to find if they occur.

Nineteen of the Stripa groundwater samples have had Fe(II) andFe(III) determinations made, and fourteen of these also have Ehmeasurements so that a direct comparison can be made between theredox potential measured with the platinum electrode, Eh(Pt), andthe Fe(lI/III) redox potential, Eh(Fe). The Fe(ll/lII) potentialis calculated after correcting for the distribution of speciesend activity coefficients with the WATEQ program in the same man-ner as Nordstrom, et al. (1979). The comparison in Figure 5-17shows that there is a rather poor correlation. The iron concent-rations (nearly always below 0.060 mg/L) are simply too low toadequately equilibrate at the electrode surface. A few samplescompare rather well, and this may be due to waiting a longer pe-

5:16

(bLJ-

LU

0.75

9.56-

0.25-

0.00-

-0.25 •

-0.50-0.50 -0.25 0.00 0.25 0.50 0.75

Eh (Pt)

Figure 5-17. Comparison of the platinum electrode potential,Eh(Pt), against the ferrous-ferric reJox potenti-al, Eh(Fe) in volts. The straight line representsthe 1:1 correlation.

riod of time for the electrode measurement to reach a steadyvalue. It is very common for the electrode to slowly drift to-ward lower potentials, and the drift rate seems to decrease withtime as it approaches the calculated Fe(II/III) couple. More ana-lyses and more careful electrode measurements might improve thecorrelation, but they won't change the conclusion that these wa-ters are very poorly poised. With respect to radioactive wastecontainment, this means that leaking radionuclides are likely todominate the redox reactions, and reaction rates will be more im-portant than equilibrium processes. It also means that there isvery little dissolved iron to react with and coprecipitate radio-nuclides.

Since both Fe(aq) and ^faq) can be measured in these groundwa-ters, and their activities calculated, it is possible to ex-amine directly the saturation state of both reduced and oxidizediron minerals. It is noteworthy that in many samples the concent-ration of Fe(gq) is as large as, or larger than, the concentra-tion of Fe^ a q). In Figure 5-18 the saturation indices for ferri-

5:17

a>

288. 488. 688. 888.

Chloride, mg/LFigure 5-18. Plot of the ferrihydrite S.I. against chloride con-

centrations.

hydrite are plotted against chloride concentrations. Ferrihydri-te is a poorly crystalline form of ferric hydroxide with a pK of37-39 (Schwertmann and Taylor, 1977). This mineral phase has com-monly been confused with amorphous ferric hydroxide, and the re-search of Schwertmann (1979) indicates that "amorphous" ferrichydroxide usually turns out to be ferrihydrite. The WATEQ S.I.computations use the most soluble pK for ferrihydrite, and accor-ding to Figure 5-18, the water samples are either saturated orsupersaturated by up to two orders of magnitude. This amount ofsupersaturation seems unreasonable, and it is concluded thatsome of the samples must be contaminated with iron, either fromdrilling or sampling. Three important observations indicate thatactive precipitation of ferric hydroxides within the boreholesis governing the iron chemistry and the iron redox potentials:(1) ferrihydrite supersaturation is common; (2) the lowest ironredox potentials tend to correlate with the lowest Fe2+ % concent-rations, as they should; and (3) some of the more recent plati-num electrode potentials are as low in Eh as the lowest ironredox potentials shown in Figure 5-17. Hence, the iron chemistrymay be dominated by active precipitation of ferrihydrite, andonly careful long-tern monitoring of a continuously flowing bore-hole might provide true background values of ferrous and ferriciron.

5.2.6 Sulfate

The data for sulfate shows slightly different trends compared to

other ions that may be due to either a different source of sulfa-

5:18

o/

120 .

100 -

80 .

60 .

40 .

0 .

O

O

4•

fy. *^ I 1 1

*** s

1

0oo

1 °°

1 1

o

. - "

1

0 100 200 300 400 S00 600 780

Chlorld», I»Q/L

Figure 5-19. Plot of sulfate against chloride concentrations.

te or the effect of redox processes, or a combination of both.The chloride correlation plot for sulfate is given in Figure 5-19.

Clearly sulfate increases with increasing chloride, but it seems

to increase in an almost exponential fashion, rather than linear-

ly-

This observation may be more apparent than real, but the highersalinity waters do have a higher SO^/Cl ratio than the lower sali-nity waters, and the lowest ratios are encountered in the N 1 bore-hole. This may indicate that sulfate reduction is more impor-tant in the N 1 borehole than for the deeper borehole samples,assuming that a single source of salinity with a distinct SO,/C1ratio exists. This suggestion finds some corroboration in thesulfate stable isotope data (Chapter 7).

5.3 Chemical geothermometers

It was mentioned in Chapter A that the Br/Cl, Ca/Mg, and Mg/Clratios of the deep groundwaters are anomalous compared to typi-cal Swedish groundwaters and marine-derived water. The possibili-ty of a temperature signature in the water chemistry was alsomentioned. These anomalies will now be further explored.

5:19

5.3.1 £a/Hg £nd Mg/C1 ratios

The magnesium concentrations in the Stripa groundwaters are ex-tremely low, resulting in very high Ca/Mg ratios and very lowMg/Cl ratios. These ratios tend to he 1-2 orders of magnitudedifferent than those for any other type of low-temperature gro-undwater regardless of origin or age. That is to say, given anytype of original water chemistry and an unlimited amount oftime, groundwaters do not achieve these anomalous ion ratiosseen in the Stripa groundwaters, with one exceptional group.From this observation we can infer that ion exchange and clayformation do not alter the Ca/Mg and Mg/Cl ratios to the extremevalues found at Stripa. The exceptional group is geothermal wa-ters. As a group, geothermal waters are the only ones which havesimilar ratios. This can be explained and demonstrated in threeways: (1) by empirical field observation, (2) by experimental ob-servation, and (3) by theoretical calculation.

Geothermal chemists have pointed out fo»- many years that highCa/Mg ratios are indicators of high temperatures. White (1970)pointed out that high Ca/Mg ratios are characteristic of high-temperature systems because Mg strongly favors solid phases suchas chlorite and smectite at high temperatures, whereas the abili-ty of Mg to react at low temperatures decreases greatly. Hence,seawater, oilfield brines of moderate temperatures, and water incontact with ultramafics have much higher Mg content than anyhigh-temperature groundwater. Another process that affects Ca/Mgand Mg/Cl ratios is dolomitization. However, the change in theseratios during dolomitization do not approach the Stripa values,and Hitchon et al., (1971) pointed out that the statistical rela-tionship between Ca and Mg was not simple, suggesting that dolo-mitization was not the only process occurring in the western Ca-nadian plains.

Graphical demonstration of the effect of temperature on Ca/Mg andMg/Cl ratios was first shown by Fournier and Potter (1979) for50 well waters, and can be seen more quantitatively in Figures 5-20 and 5-21. Approximately 255 groundwater analyses of all typesthat have in situ temperature measurements (or silica geotherrao-meter temperatures) are plotted as log(Ca/Mg) (weight ratio) andlog(Mg/Cl) against 1000/T where T is the Kelvin temperature.These are in the form of Arrhenius plots for convenience and in-creased linearity. The groundwater samples are taken from the pu-blished literature, and include both low- and high- temperaturegroundwaters, carbonate and silicate aquifers, freshwater andbrine compositions, and very old and very young waters. The sour-ces of data are given in Table 5-1. In spite of the very largerange of groundwater age and composition, it is clear that tempe-rature is the overriding factor affecting the Ca/Mg and Mg/Cl ra-tios. Note the distinct increase in Ca/Mg and decrease in Mg/Clabove about 100°C. These plots support the observation that Mgis a very temperature-sensitive parameter.

5:20

3

2

t

0

- 1

-2

350 300 250] 1 r

1 o1 o1 °1 %> 8\ "8 °

<=> f •• w 0

\ °|\ °°oo\

o o

o* o

__ oO Q

0

o

-

p

—

-

1 ,

200

—1

o

o0

N^

150— r _

o

>o

0o

o „

•>> ° !>

1001

o

O O "°

0o o oo OA

OJ»?OOO%

>v •*" °\ °

N

° >

SO 25 0 °C

1 T 1

I M OKOONOWATtdS (T ? I00°CI

90 \ inclution limn

8o

*

• o» o" <**?<&> o§

' o gjy | iifi On O

SP o %

\ V •N .v

\ °\ °\ °i

0

o

1 i 1

toppTIKT

Figure 5-20. Plot of log (Ca/Mg) against 1000/T.

MU jbo ?oi) i«i *) n o "c

" "'I'

90% incluiKM limit ^ ° S

• / ©

° y o o Q °°

o '7,°' «, o°«»

A*. /

i .. _ L

Tim

Figure 5-21. Plot of log (Mg/Cl) against 1000/T.

5:21

Table 5-1. Literature sources for data plotted in Figure 5-20and 5-21.

Arnomon, S., BJornsson, A., Cttlason, C. and Cudmundsson, C. (1975),lo Proc. Second United Nation» Symposium en the Development and Useof Geothermal Resources 2, 853-864.

Aroorason, S., Cronvold, K. and Slgurdson, S. (1978) Geochlm. Cotmochtm.Acta 4£, 523-536.

Back, V. (1966) U.S. Ceol. Survey Prof. Paper 498-A, 42 pp.

Baroea, I., Downes, C. J. and Huatoa, J. R. (1978) Am. J. Scl. 226, 1412-1427.

Brown, D. U. and Gulbrandaen, R. A. (1973) J. Res. U.S. Ceol. Survey j_, 105-111.

Capuano, R. M. and Cole, D. R. (1982) Geoehlm. Cosmichim. Acta Ub_, 1353-1366.

Feth, J. H., Roberson, C. E. and Pohir, W. L. (1964) U.S. Ceol. Survey Water-Supply Paper 1535-1, 170 pp.

Fuslllo, T. V. and Voronln, L. M. (1981) U.S. Ceol. Survey Open-File Report81-814, 38 pp.

Helteton, H. C. (1968) Aa. J. Scl. lbb_, 129-166.

Hen, J. D. (1970) U.S. Ceol. Survey Uater-Supply Paper 1473, 363 pp.

fcltchoo, B. and Friedman, I. (1969) Ceochlm. Cosmochim. Acta 2?., 1321-1349.

Hltchoo, B., BlUlB(t, C. K. and Klovan, J. £. (1971) Ceochlm. Cosmochim.Acta 21. 567-598.

Kharaka, Y. K., Callender, E. and Carothen, V. V. (1977), £n Proc. ThirdCeopretsured-Geothermal Energy Conference, _> 121-165.

Kharaka, Y, K., Carother», W. W. and Brown, P. M. (1978) Soc. Petroleum Eng.AIME 7505, 5 pp.

Kharaka, Y. K., Lico, H. S., Wright, V. A. and Carothers, V. W. (1979), _l_nProc. Fourth United States Culf Coast Geopressured-Ccothermal EnergyConference: Research and Developoent, 168-193.

Langoulr, 0. (1971) Ceochim. Cosmochim. Acta 25. 1023-1045.

Miller, T. P., Barnes, I. and Patton, W. U., Jr. (1975) J. Res. U.S. Ceol.Survey 3., 149-162.

Robcrtton, E. C , Fournler, R. 0. and Strong, C. P. (1975), In Proc. SecondUnited Nations Symposium on the Development and Use of GeothermalResources, l_, 553-561.

Ryt, R. 0. and Haffty, J. (1969) Econ. Ceol. 6,4, 629-643.

Trucsdell, A. H., Thompson, J. M., Coplcn, T. B., Nehring, N. L. and Janlk,C. J. (1981) Geothcrsilct J£, 225-238.

White, A. F., Claasscn, H. C. and Benton, L. V. (1980) U.S. Geol. SurveyUtter-Supply Piper 1535-Q, 34 pp.

White, D. E. (1968) Econ. Ctol. £), 301-335.

White, D. E. (1981) Econ. Ceol. 75th Ann. Vol., 392-423.

5.3.2

5:22

Mineralogical evidence from field studies in both geothermalareas associated with magmat ism and/or tectonics and geothermalareas associated with deep basin burial metamorphism indicatethat large quantities of chlorite/montmorillonite are producedas the temperature increases above 100-150°C (Boles and Franks,1979; McDowell and Elders, 1983; Ellis and Mahon, 1977). Chlori-te formation at high temperatures removes large quantities ofmagnesium from the fluid phase.

Several experimental water-rock investigations have shown thatlarge depletions in dissolved magnesium result from chlorite for-mation (Hawkins and Roy, 1963; Ellis and Mahon, 1967; Bischoffand Dickson, 1975; Dickson, 1977; Mottl and Holland, 1978; Sey-fried and Bischoff, 1979; Seyfried and Mottl, 1982) regardlessof rock type. In all of these experiments at 150°C or higher,magnesium was strongly depleted due to the growth of a magnesiumsilicate phase.

Thermodynaraic equilibrium computations also support the strongtemperature dependence of the Ca/Mg and Mg/Cl ratios. A notableexample is the work of Ryzhenko, et al. (1981) in which the af-fect of pressure, temperature, COj and rock/water ratio were va-ried in equilibrium computations involving a rock of graniticand another of basaltic composition. One of the results was thatmagnesium was strongly depleted under nearly all conditions whenthe temperature increased and when the rock/water ratio increas-ed.

These investigations demonstrate that orders of magnitude varia-tions in Ca/Mg and Mg/Cl ratios, especially when compared to sea-water, can only be accounted for by high-temperature formationof magnesium silicates, generally chlorites, and cannot be repre-sentative of low-temperature clay formation or ion exchange pro-cesses.

The classical Na/K, Na-K-Ca, and silica geothermometers are goodtemperature indicators for geothermal waters (Ellis and Mahon,1977). Unfortunately, low-temperature re-equilibration preventsthe use of the silica geothermometer, and re-equilibration mayeasily affect the Na/K and Na-K-Ca geothermometers, given suffi-cient time. It is noteworthy that when the Ca correction to theNa/K geothermometer is applied, the calculated temperature in-creases by at least several tens of degrees. However, the calcula-ted temperatures fall in the range of 10-28°C, only slightly hig-her than the actual neasured temperatures. This geothermometer,if reliable, reflects a low temperature origin.

5:23

tee

MaL

T

tee 300 400 see

Chi or Id., mg/L

Figure 5-22. Mg/Li geothermometer temperature against chlorideconcentrations.

5.3.3 The MsZ.Ll. &.e2.thermometer

A new geothermometer has been described by Kharaka, et al.(1985) that has been found to be reliable over the range of 40-350°C and often gives the best results for temperatures of 40-70°C. This geothennometer takes advantage of the fact that Litends to increase with temperature, whereas Mg decreases withtemperature, according to the empirical equation:

t(°C) 1900

4.67 + log(/Mg/Li)-273

where the magnesium and lithium concentrations are in mg/L. Aplot of the Mg/Li temperatures against chloride concentrationsis shown in Figure 5-22 for the Stripa groundwaters. A consis-tent trend is apparent, indicating temperatures of 80-95°C at thehighest chloride concentrations. These temperatures overlap sig-nificantly with some of the fluid-inclusion homogenization tempe-ratures in the Stripa granite. Although the correlation in Figu-re 5-22 reflects a mixing line, the saline end-member appears tocarry a high-temperature signature.

5:24

5.3.4 Conclusions _from_geothennome£r^

Several ion ratios are anomalous in the Stripa groundwater chemist-ry* These ratios, such as Ca/Mg, and Mg/Cl, are difficult toexplain by low-temperature processes such as clay-mineral forma-tion. They are consistent, however, with processes that occur athigh temperatures. Since high temperatures have not occurred atStripa for several millions of years, these anomalies might be ex-plained by the leakage of fluid inclusions from the Stripa grani-te (and/or the leptite). Fluid inclusions incorporated in thegranite would carry a high-temperature signature and, if theyleaked into the groundwater regime, would explain a high-tempera-ture signature in a low-temperature environment. One of the mostdirect approaches to test such a hypothesis is to leach fluidinclusions from the granite and compare the Br/Cl ratios of theleachate with those in the groundwater.

5.4 The fluid-inclusion hypothesis and rock-leaching studies

5.4.1

Fluid inclusions are common in igneous and metamorphic rocks,but they are rarely larger than 1 mm and frequently go unnotic-ed. They can account for more than 50% of the chloride contentof granitic rocks (Fuge, 1979), and they have the potential forplaying a role In water-rock interactions when the rock/waterratio becomes very high (i.e., very low porosity). In two compre-hensive reviews Roedder (1972, 1984) has discussed nearly allaspects of fluid inclusions, and these references should be con-sulted for further details.

If fluid inclusions might enter Into a groundwater system, thenseveral general questions must be addressed, such as: Can theyleak? How? How fast? Are there enough fluid inclusions? Do theycontain sufficient salt content?

Although no one has specifically addressed the question of whet-her fluid inclusions can leak significantly into groundwater sys-tems, Roedder and Skinner (1968) did address the general ques-tion as to whether they leak during sample preparation or experi-mentation. They concluded that if adequate care was used in samp-le preparation, leakage was rare, but leakage due to surficialfractures formed during inadequate sample preparation was respon-sible for earlier reports of widespread leakage. Sawing, grin-ding, and polishing rock samples can produce surficial microfrac-tures which penetrate the sample by more than 1 mm.

Fracturing the granite can easily lead to exposure of large quan-tities of fluid inclusions, both directly and by slow leakage,

5:25

from microfractures. Leakage may occur either by diffusion or bymass transport. Diffusion through solid mineral grains is tooslow of a process to be important. Diffusion along microcrackswould be driven by the concentration gradient if the inclusionswere more saline than the groundwater. Microfractures, likely tobe abundant near major fractured zones, might be large enoughfor mass movement of the fluid. If a small volume of groundwaterflows by a large surface area of rock (a typical situation indeep crystalline rocks) containing abundant inclusions, then theaccumulation of some salinity from this source seems quite pro-bable.

Another important observation is the common occurrence of fluidinclusions as planar structures in crystalline rock. These fluid-inclusion planes are usually secondary features which reflectthe microfracturing and rehealing of the rock. Such fluid-inclu-sion planes cut across mineral grain boundaries, and are parallelto and coincident with microjoints . Wise (1964) found evidencefor younger, smaller planes of fluid inclusions reopening olderfluid-inclusion planes. These older planes represent Precambrianjoint systeras, and they controlled the orientation of subsequentfracturing during Laramide orogeny (late Cretaceous to early Ter-tiary). Dale (1923) found numerous examples of fluid-inclusionplanes that were parallel and coincident to microfractures ingranites and gneisses from New England. He provides some stri-king examples of fluid-inclusion planes becoming microfracturesof the same strike and dip. Large numbers of very small inclu-sions have been found on mineral grain boundaries of fracturedmineral surfaces by electron microscopy (Sella and Deicha,1962a,b, 1963).

Microcracks and micropores are common in igneous rocks, and havebeen investigated for their effect on the physical properties ofthe rock, especially electrical resistivity, compressibility,and permeability (Brace, et al., 1972). Relatively few micro-cracks are induced by drilling (Nur and Simmons, 1970), and theeffects of surface grinding and polishing can be removed by ionthinning (Brace, et al., 1972; Sprunt and Brace, 1974). Thesemicrocavities have been found both along grain boundaries andwithin mineral grains (Sprunt and Brace, 1974), they have beenfound to connect, to form larger microcracks when placed undersubcritical stress (Tapponler and Brace, 1976), they are morecommon in rocks with a higher water content than those that are"drier" (Montgomery and Brace, 1975), and they are more abundantin sodic feldspar than in quartz or potassium feldspar (Spruntand Brace, 1974; Montgomery and Brace, 1975). This last observa-tion is particularly important because many of these microcaviti-es are identical in description to fluid inclusions, and are

Microjoints are not necessarily microscopic in character, butrather a distinct size class of jointing on a small scale.

5:26

thought to be fluid inclusions by Sprunt and Brace (1974) and Mont-gomery and Brace (1975); and yet little is known about fluid in-clusions in feldspars. Most fluid-inclusion studies are done onrelatively transparent minerals, such as quartz, fluorite, andcalcite. Feldspars are notoriously difficult to study for fluidinclusions because of their opacity, and nothing is really knownabout the distribution of inclusions between quartz and felds-pars in a granite. If the results of Montgomery and Brace (1975)are valid for most granites, then the density of fluid inclu-sions in sodic feldspar must be considered greater than in coex-isting quartz grains, and the volume contained by fluid inclu-sions could take up to 2% of the feldspar. Fluid inclusions may be,in fact, the reason for the turbid, opaque appearance of feld-spars (Folk, 1955) and the milky white appearance of quartz (Roed-der, 1972, 1984).

Thus, fluid-inclusion planes can be zones of weakness where frac-turing has a higher probability of occurring, and fracturedgroundwater zones would have direct access to the highest densityof fluid inclusions. This close relationship between fluid-inclu-sion planes and microfractures/microjoints could make the acces-sibility of fluid inclusions to the groundwater nearly instantane-ous in response to tectonically- or anthropogenically-derivedchanges in rock stress. It may not be necessary for fluid inclu-sions to be transported any significant distances in order forthem to mix with the groundwaters; only microfracturing needs tooccur. The microfracturing would depend on tectonic events andman-made disturbances. The amount of inclusion fluid in thegroundwater may depend upon the rate of microfracturing relative tothe groundwater flow rate, in addition to the length of the flowpath and the rock/water ratio.

Investigations of fluid inclusions in the Stripa granite have fol-lowed along three complementary lines: (1) direct measurementsof fluid inclusions by standard techniques, (2) chemical analys-is of inclusions by leaching studies, and (3) microcavity poro-sity and salinity measurements.

5.4.2 FJL£id_- nc_lusi_qn me£8 iremen£8_arid_yolume_t£i£ £qnsjLde_ra teoris

The fluid-inclusion study by Lindblom (1984) that is summarizedin Chapter 2, has demonstrated that there should be more than suf-ficient fluid-inclusion salt in the Stripa granite to accountfor the salinity of the groundwater, assuming near-static hydrolo-gic conditions. There is, on the average, 17 liters of inclusionfluid per cubic meter of granite, assuming that there is a compa-rable amount of inclusions in feldspars as in quartz. This fluidaverages around 3 wt. % NaCl or about 0.0105 wt. % Cl for thegranite. In other words, there is about 278 grams of chlorideper cubic meter of granite. The porosity of the granite is about1%, i.e. about 100 times more rock than water. Hence, there are

5:27

278 grams of chloride per 10,000 cubic centimeters of ground*ra-ter. Assuming static groundwater conditions, that would producea concentration of 28 g/L chloride if all the fluid inclusionswere opened. The maximum observed concentration is 700 mg/L,which only requires about 2.5% of the total fluid-inclusion chlo-ride. Therefore, on volumetric grounds it is certainly feasiblefor fluid inclusions to be the primary source of chloride in theStripa groundwaters. A simple test of this hypothesis is to com-pare the Br/Cl and I/Cl ratios from fluid-inclusion leachateswith those in the groundwater.

5.A. 3 Pf&lipipilTZ. fluidi-lncluBlpn^ l_eaching_study

Nine samples of V 1 and V 2 drillcore were leached according tothe procedures of Roedder, et al. (1963) and Hall and Friedman(1963). Rock fragments about 1 cm across were cleaned in a sonicbath (same samples were electrolytically cleaned until it wasfound that no further contaminants were removed by this additio-nal step) and then vacuum dried, crushed in an iron pipe undervacuum, the water extracted, and the remaining fragments subjectto a one-minute leach with doubly-distilled water. The leachatewas filtered after one minute, and then another one-minute leachwas obtained. Five or six successive leaches were made until thechloride concentrations were no longer detectable. Deterruina-tions of F, Cl, Br, and SO^ were performed by ion chroraatograp-hy, and I was determined by the standard colorimetric methodbased on iodide-catalyzed cerium reduction (Whittemore, pers.comu.).

The results and the average values for Br/Cl and I/Cl are shownin Table 5-2. The ratios for Br/Cl and I/Cl are consistentlyhigh and identical to the groundwater ratios. This comparisonprovides the strongest evidence yet for the association betweenfluid inclusions and the groundwater salinity.

5.4. A Mi£rofr a£tures !

A cooperative investigation is underway (involving D.K.Nordstrom, I. Neretnieks and K. Skagius) to determine the porosi-ty of microfractures and the salinity of the microfracture fluidin the Stripa granite. Cores taken from V 1 and V 2 are coredagain along the length about every 3-4 cm. These smaller cores(60-90 cm3) are leached in 7-10 ml of distilled water for twoweeks, and then the solution is analyzed for Cl, Br, I, F, andSO^. The core is then blotted dry, weighed, put under vacuum,then heated at 90°C for three days, and reweighed. The weightloss gives an estimate of the water content, and when combinedwith the chloride concentrations, they give an estimate of thechloride concentrations in the microfractures. Next, the core isground up in a mortar and pestle to approximately 50-400 mesh

5:28

Table 5-2. Anion analyses on fluid-inclusion leachates ofStripa granite in preliminary study (in mg/L).

VI-445Al««ch #1 .

2.3 .4 .5.6.

VI-445Bleach #1 .

2.3 .4 .5.

V1-445Cleach tl.

2.3 .4 .5.6.

Cl

9.953.601.420.31

<0.12<0.12

7.272.860.57

It.It!

2.110.36<0.3<0.3<0.3

Br

0.0850.037

0.0580.027<0.01<0.01<0.01

0.050<0.01<0.01<0.01<0.01<0.01

0.00165

0.0021

0.350.490.330.170.160.18

0.310.340.330.430.27

0.210.150.110.090.110.04

SO.

5129113.62.11.2

0.170.10

1.61.10.230.15

Br/Cl I/Cl x 10*

.0085

.0103

0.00800.0094

0.0112

2.2

4.7

Vl-505leach #1.

2.3.4.5.

Vl-408leach *1.

2.3.4.5.

V2-363leach #1.

2.3.4.5.

V2-404leach #1.

2.3.4.5.

6.182.450.770.380.28

2.793.110.500.500.31

8.783.420.700.340.10

5.001.870.580.220.24

0.0820.031<0.01<0.01<0.01

0.027<0.01<0.01<0.01<0.01

0.09G0.031

0.0450.017<0.0l<0.01<0.01

0.0050

0.001

0.0032

0.0019

0.400.130.0b

0.18

0.580.370.310.290.24

0.450.220.130.110.12

0.531.2

0.270.400.26

0.700.150.110.071<0.05

1.850.980.370.290.18

0.530.24<0.05<0.05<0.05

0.01330.0127

0.0097

8.1

3.6

0.01030.0091

0.00900.0091

3.6

3.8

V2-445leach #1.

2.3.4.5.

V2-471leach #1.

2.3.4.5.

4.701.440.28

3.691.510.560.260.20

0.0490.016<0.01<0.01<0.01

0.035<0.01<0.01<0.01<0.01

0.0013

C.0020

0.400.0850.400.100.28

0.690.450.200.200.20

Leachatca:Average Br/Cl - 0.010K+0.0015)Average 1/C1 - 4.28 x TV* (+1.8 x 10"*)

Croundwatcri:Average Br/Cl - 0.0107 (+0.001)Average I/Cl - 4.46 x 1CT*(+4.1 x 10**)

nn

0.44<0.05<0.05<0.05<0.05

2.540.760.270.290.09

15

2524

0.01040.0111

0.0095

2.8

5.4

5:29

Table

Vl l / lVll /3V12/1V12/3V13/1V13/3V21/1V21/3V22/1V22/3V23/1V23/3

5-3. Preliminarycrushed and

Cl(l i ( / | of rock)

UncTUihcd

2.61.31.30.610.736.B1.73.82 .63 ,23 .32 . 3

data on Stripauncrushed samples

CKuj/t of rock)

Cruihed

11.220.615.010.611.69 .2

18.917.519.917.613.714.8

granite leachates f<•

Mlcrofracture salinity

C:(>E/D

3680862356492518

5390141030302030205047303060

(0.3-0.03 mm diameter) and leached again in distilled water for15 minutes, filtered, leached again in distilled water for 15 mi-nutes, and filtered again. These measurements have been done onthe same samples that were used for fluid-inclusion analysis bySten Lindblom. Preliminary results, shown in Tables 5-3 and 5-4demonstrate that the microfracture chloride concentrations aresometimes equal to, but usually greater than, the groundwaterchloride concentrations, and less than the fluid-inclusion con-centrations. These concentrations may not be the same as thosethat exist in situ, but this experiment does show that fluid in-clusions can leak out of the rock matrix, and that fresh groundwa-ter can enter raicrofractures and mix with the fluid inclusions*This premise Is further verified by tha halogen ratios summari-zed in Table 5-5. The Br/Cl and I/Cl ratios are nearly identicalin the crushed and uncrushed samples, and nearly the same as thegroundwater ratios. Thus, we have demonstrated an association be-tween the salt in the inclusion fluid, in the microf racturefluid, and in the groundwater. It is very noteworthy that theI/Cl ratios that we have determined are higher than any recordedfrom the published literature on natural waters.

5.4.5 I ltej: at u£e_surv ey_ j>f_hal<>g£ns_ .in_graiTit e£

Relatively little information is available on the halogen con-tent and halogen ratios in granitic rocks. The lack of data ispartly due to the very low concentrations of Br and I and the in-herent difficulties in analyzing for them. Another reason may belack of interest. A literature survey, however, has revealedsome fascinating trends in halogen content and ratios.

Typical average halogen concentrations and rrnges of measured va-lues for granites are shown in Table 5-6. The water-soluble por-tion of the halogens is, of course, more important from a ground-

5:30

Table 5-4. Analyses of F, Cl, Br and SO^ (in mg/L) for Stripa

Granite Leachates. (a) is first leach, (b) is se-

cond 1'iach. Number in parentheses refer to one

standard deviation.

SAMPLE

V I , 1 1V I , l l aV I , l i b

V I , 1 3V I , 1 3 aV I , U b

V I , 2 1V I , 2 1 aV I , 2 1 b

V I , 2 3V I , 23aV I , 2 3 b

V I , 3 1V I , 3 1 aV I , 3 1 b

V I , 3 3V I , 3 3 aV I , 3 3 b

V 2 . l lV 2 , l l aV 2 , l l b

V 2 . 1 3V2.13»V2 ,13b

V 2 . 2 1V2 ,21aV2 ,21b

V2.23V2,23aV2,23b

V2.31V2,31aV2,31b

V2.33VI,33*V2,33b

Mt»n, uncrushtd

Ncai), cruthcd(a)

!Uaa, crushtd(b)

0.030.080.17

2.100.741.31

2.221.111.45

2.1S1.621.59

3.001.321.99

2.271.211.62

1.350.600,61

1.J90.781.04

1.590.681.19

1.540.S61.12

0.0610.140.31

0.090.650.21

1.501

.7911

1.051

Cl

2.6011.25.98

1.2620.65.73

1.5515.06.92

0.6110.42.65

0.7311.62.95

6.809.193.81

1.7318.98.82

3.8417.510.1

2.5819.94.44

3.2417.613.7

3.2613.735.6

2.2914.8S.44

Br

0.126P. 048

0.0160.2400.126

0.0140.1900.076

0.013O.lHb0.055

0.0240.2180.067

0.0220.1590.094

0.0220.2340.078

0.0330.1260.220

0.0330.2690.049

0.0330.1400.220

0.0110.2130.153

0.0070.1640.184

S 0 4

l.5O(.975) 2.54(1.»8) .O21O(.OO9)

.791(.427) 15.0(3.88) .189(.O49)

l.05(.*00) 9.10(8.94) . l l*( .O63)

.0112

.0226

.0255

.0094

.0393

.094

.0044

.067

.250

.0042

.192

.0117

.0054

.0206

.107

.0056

.0225

.0164

.0057

.0246

.213

.0053

.304

.071

.0053

.0364

.0182

.0077

.0284

.286

.0059

.0497

.0046

.0310

,00623(.00214)

,0734(.0910)

.102M01)

0.3181.491.41

0.7011.842.002

0.3911.101.54

0.5931.371.33

0.4B31.231.54

0.3781.371.41

1.022.401.95

0.8601.952.20

0.5301.051.54

0.5461.822.20

0.3622.172.79

0.3530.4272.33

.S45(.219)

1.52<.547)

1.851.461)

Table 5-5. Weight Ratios for Stripa Granite Leachates.

SAMPLE

V I , 1 1

vi. IUV I , l i b

VI.13VI.13aVI,13b

VI,21VI,21aVI,21b

VI,23VI,23aVI,23b

VI,31VI.31aVI,31b

VI,33VI,33aVI,33b

V 2 , UV 2 , l l aV 2 , l l b

V2.13V2,13aV2,13b

V2,21V2,21aV2,21b

V2.23V2.23»V2,23b

V2.31V2,31aV2,31b

V2.33V2,33aV2,33b

Htan, uncruthid ltachatc

Ntao, cruahcd ltachatt(a)

Hcan, cruihtd lcachatc(b)

SOJCl Br/Cl 1/Br

.122

.133

.236

.556

.089

.349

.252

.073

.222

.974

.131

.502

.661

.106

.522

.056

.149

.370

.590

.127

.221

.224

.111

.216

.205

.053

.347

.168

.105

.161

.111

.158

.078

.154

.029

.276

.339(.29)

,105(.039)

.27K.15)

.0112

.0080

.0127

.0116

.0220

.0090

.0126

.0110

.0213

.0178

.0208

.0328

.0188

.0227

.0032

.0173

.0247

.0127

.0124

.0088

.008''

.0072

.0218

.0128

.0135

.0110

.0102

.0079

.0161

.0034

.0155

.0043

.0031

.0111

.0218

.O116(.O0B8)

.013K.0037)

.0160(.0071)

4.312.014.26

7.461.9016.4

2.844.4536.1

6.9018.44.42

7.391.7736.3

.8242.454.30

3.291.3024.1

1.3617.47.03

2.051.834.10

2.381.6120.9

1.813.63

2.01

3.67

3.55(2.4)

5.16(6.4)

14.7(12.9)

.179

.531

.588

.164

.746

.314

.3533.29

.3231.0J.213

.225

.0945

.491

.254

.142

.174

.259

.1052.73

.1612.41.323

.161

.135

.371

.233

.2031.30

.536

.233

.657

.168

.337C.174)

.494(.739)

.940(1.08)

5:32

Table 5-6.

Cl

Br

I

Table 5-7.

Cl

Br

i

Average halogen concentrations andfor granites.

Average (ppn)

200

O.S

0.2

Range(ppm)

10-1,180

0.13-5.01

0.04-0.72

Water-soluble halogens in granites.

X of Total Halogen

20-100

(20-100)

60-100

Reference

range of vali

Reference

Fuge(1974)

Fuge(1974)

Fuge(1974,1978)

Faber(1941), Behne(1953), Fuge(1979)

Behne(1953)

Fuge(1978)

water viewpoint, and from Table 5-7 it's clear that a largefraction, usually most, of the halogen content is readily leachedby water. This amount represents the fluid-inclusion salt of therock. If we assume that 50% of the chloride in a granite iswater soluble, then 0.010% Cl is the average amount of fluid-inc-lusion chloride in a granite. From this we can calculate thatthere are 260 grams of leachable chloride in 1 m of granite.This value is very similar to the value obtained for the Stripagranite.

Additional verification of the soluble nature of chloride in ig-neous rocks has been demonstrated in at least three experimentalstudies. Ellis and Mahon (1967) showed that anywhere from 3 to1000 ppm Cl could be leached from igneous rocks subject to tempe-ratures of 250-600°C. The maximum amount extracted was 50-100%of the rotal chloride in the rock (for a water/rock ratio of1:1). Moore, et al. (1983) forced distilled water through cylin-ders of two different granites at temperatures of ambient to300°C, and collected and analyzed water samples at regular Inter-vals. At room temperatures the Cl concentrations ranged from 20to 132 tng/L, and the last samples collected after 30-60 ml ofwater flow still had 20-50 mg/L Cl coming through with no signof a decrease (Moore, pers. corntn.). In one run a total of 2.3 ragCl had been extracted with 42 ml of distilled water. Since thegranite cylinders were about 1025 g each (assuming a density of2.6), with an estimated leachable Cl content of .01%, then thetotal leachable chloride would be about 103 mg. If about 50times more water had been passed through, then all of the leac-hable chloride should have been extracted. This amount would beequivalent to a total water/rock ratio of 5:1. Alternatively,the data Is amenable to a rough estimation of the time necessaryto wash out all the soluble chloride. 2.3 tug Cl was extracted inabout 7 days at a pore pressure of about 100 bars* Assuming a

5:33

constant extraction rate, it would take nearly a year to washout all the chloride from this volume of rock.

The third study is in progress on the Carnmenellis granite (Ed-munds, et al., 1983, 1984). Groundwaters in this granite havehigh salinities (up to 19300 mg/L), and it has been demonstratedthat significant quantities of chloride can be leached from thegranite. However, the investigators have stated that the salinityis not derived from fluid inclusions, but rather leached fromprimary minerals (Kay, 1984) having the same Br/Cl ratio as sea-water (Edmunds, et al., 1984).

There are precious few determinations of the Br and I content ofgranites, and even fewer published data where the ratios Br/Cland I/Cl are available. The best data seems to be that of Fuge(1974, 1978), who suggests that the average Br concentration forgranites is 1 ppm, and the average I concentration is 0.2 ppm.Combining these values with an average Cl content of 0.02% forgranites, the ratios become Br/Cl - 0.005 and I/Cl - 0.001 whichare noticeably higher than seawater values. Fuge (1978, 1979)also noted that close to 100% of the total iodine content was wa-ter-soluble, whereas lower proportions of chlorine were water-so-luble (30-70%). Assuming that 50% of the Cl in the average grani-te is water-soluble, and 100% of the Br and I is water-soluble,then the water-soluble leachates would have Br/Cl = 0.010 andI/Cl • 0.002. These values are nearly identical to those at Stri-pa, and are considerably higher than values found in most othertypes of rock, mineral, or natural water. They are, however,very consistent with the studies by Behne (1953), who found aBr/Cl ratio of 0.0125 for 14 granites; Vinogradov (1944), whofound a Br/Cl ratio of 0.010 for crystalline rocks; and Kozlow-ski and Karwowski (1974), who found a Br/Cl ratio of 0.015(+0.013) for 34 f! uid-inclusion extracts from hydrothermalquartz in several massifs from lower Silesia. This consistencysuggests that a high-temperature process may be responsible forenriching the Br/Cl ratio relative to seawater. The only known ex-planation is the partitioning of halogens into the hydroxyl siteof micas and amphiboles. At high temperatures a larger amount ofchloride can substitute for this site, but little or no bromideor iodide can substitute because their ionic size is too great.Hence, there should be an enrichment of Br/Cl and I/Cl in the re-sidual fluid and any fluid inclusions remaining from that pro-cess. The degree of enrichment probably depends very heavily ontemperature and residence time of the fluid at that temperature.At temperatures of 2OO-3OO°C in active geothermal areas that con-tain modern-day seawater, there is no change in the Br/Cl ratio(Truesdell, et al. 1981). Thus, longer residence times and hig-her temperatures are probably required.

6:1

OXYGEN-18 AND DEUTERIUM CONTENTS

6.1 Introduction

Oxygen-18 and deuterium measurements have been performed o theStripa groundwaters since the beginning of the LBL-KBS project.Basic principles and a first discussion of results can be foundin Fritz, et al., 1979, 1980. All data obtained within the frame-work of the entire Stripa project by several laboratories (Ge-sellschaft fiir Strahlen- und Umweltforschung mbH Miinchen(GSF), International Atomic Energy Agency (IAEA), Université deParis-Sud (UPS), University of Waterloo (UW)) are presented inTable 6-1.

i Q n

All samples for 0 and H determinations in water were collec-ted in tightly sealing bottles and prepared according to stan-dard procedures which include COj-equilibration at constant tem-perature. Results are then expressed as per mille (°/oo) devia-tions (6-values) from the SMOW reference (Standard Mean OceanWater). A 618O = +10 °/oo then signifies that the samples has 10°/oo more 0 than the standard.

Analytical reproducibility in each laboratory is better than+0.2 °/oo for * 0 and +1.0 °/oo for deuterium. However, betweenlaboratories minor differences do exist. This is shown in an in-tercomparison of 0 and H data between different laboratories(Table 6-2) as well as the data listed in Table 6-1. These diffe-rences do not affect the interpretation of the results.

6. 2 General considerations

Stable isotope analyses are used to distinguish waters from dif-ferent origins or they may be used to study reactions betweenrocks and minerals. The first case assumes that H and 0 con-centrations are conservative whereas modifications of isotopiccompositions may occur in the second case.

6.2.1 C_on8ery_ative_s ab_le ,lsot£pe contents

Conservative stable Isotope contents are characteristic for many"normal" groundwater systems In whi .h H and 0 contents are de-fined by recharge processes and remain unchanged during groundwa-ter flow. This natural labelling has recently been reviewed inIAEA publications 1981, 1983a, 1983b. For practical purposes,two observations are of particular importance

6:2

T a b l e 6—la Su»a>ary of ' 0 and 2H Data on Surface Waters and Shallow Crnundwatera at theStrlpa Teat Site. AH analyses were done at University of Waterloo (UW) unleasotherwise indicated.

NAME STRIPA NO.KBS-LBL

DATE « l 8 0 ° / o o SHOW 6 }H ° / o o SHOW ANAL.

PONDS AND LAKES

Tailing pond 33

ti ti 3

" •* 3

STREAMS

Herr gärd Sauna

Danshuttegard Bridge

Into tailing! pond

WATERTABLE WELLS AND SEEPS

WT - 2WT - 3UT - 7seep at SBH :3

PRIVATE WATER SUPPLY WELLS

Private

PrivateH

11

Privatet i

"

Privatei i

Private"•i

••

H

w e l l

w e l l"I I

w e l l

w e l l

w e l l

I I

"

•i

11122223224455555

SBII-3

89-104

1

9

4

5

72

71705473

18B3

1923

20

SO

21

B5

77-09-0778-06-1678-11-2377-09-12

77-09-0777-09-0 7

78-11-2382-06-1083-03-2379-05-06

79-05-5-1879-05-0178-08-2479-05-06

77-09-2779-05-188J-O6-1O77-09-2877-10-2477-10-2577-10-2777-09-2877-10-2679-05-1679-05-1782-06-0972-10-0677-10-0777-10-1177-10-1279-05-15

79-05-23/2479-05-2579-05-2679-05-27

- 9.2-10.8- 9.3

- 9.1- 9.1- 9.6-10.7-11. i-12.7

-14.2-12.5-11.8-12.1

-10.8- 9.6-11.56- I I . 0-11.0-10.8-11.0-10.6-10.9-13.4-11.4-12.91-10.9- l l . l-11.1-11.0-13.1

-11.7-11.5-11.7-11.3

-76.3-78.0-76.0

-73.4-73.4-77.0-R0.8-81.8-93.0

-81.3-89.0

-81.3-82.0-82. 5

-81.0

-79.1

-76.3

-93.4

-83.7

-81.7

-83.0-81. n-8(1.0-82.0

UPSUPS

UPS

UPS

DRIP WATER AT OLB MINE LEVELS

AND PLOWING WATER IN OLD MINE

•lne diacharge135 • level, drip157 " " , "310 " " , "360-360 • level , drip

II H II II

360-410 • level , dripII II II II

tII II II i*

Flooded drift at 380 •

27576/777626

8

864

72-09-0779-05-1479-05-1179-05-1177-11-1777-11-1877-09-0877-11-1779-05-1578-11-24

- 9.1-11.4», -11.5- I I . 5« , - I I . 5-11.0-10.8-10.9-10.8-10.7-11.0*-11.4

-74.6

-8«, -82-79-79.4

-79.4-79-82

* Average of »ore thjn one analyses done on different umplew collected during the stated samp-ling period.

6:3

Table 6-lb. Summary of I R 0 and 2H Data fro» borehole H 3.

STRIPA NO.KBS-LBC

DATE 6 I 8 O ° / o o S10W

H 3 77-O9-2I

77-10-19

77-11-29

78-01-24

78-02-24

-11.8

-11.8

-11.9

-11.9

- I I . 9

6 J I I " / o o SHOU ANAL.

- 8 7 . 7

UW

UU

UU

tlW

W

35 78-05-30 -12.2 -H7.2 tlW

42 78-08-15

78-09-19

7R-U-17

78-H-27

79-02-12

79-05-02

79-05-17

-12.4

-12 .3

-12 .3

-12 .4

-12 .2

-12 .2

-Af t

-90

UW

UW

UW

UW

UW

UW

UW

87 79-11-21

81-06-04

81-07-09

82-06-10

83-03-23

83-11-09

84-02-14

84-02-23

-12.3

-12.3

-12.60

-12.32

-I 2.20

-12.02

-12.1

-11.95

-90

-R7

-1D.

-m.-85.

- 8n .

-90

- 8 f t .

6

7

9

9

3

UW

UW

UPS

ursUPS

1AF.A

wIAF.A

Table 6-lc. Summary of ' 0 and H fat* from Borcholti at the 330 n Level Excavation.

NAME

R 1

R 1

It 3

11 9

HC-3

HC-4

SCU-horl. 338 •

Vant. dr i f t ccmblned

5 1

S i

S i

STRIP* NO.KBS-LBL

53

53

S3

46

79

47

48

28

65

50

51

52

DATE

78-08-2978-09-1978-11-1778-12-08

79-05-1781-06-04

78-06-1679-05-2278-06-1678-06-1677-12-0778-11-2478-09-1578-09-1478-11-2778-09-1478-11-27

«'"<> °/oo

-12.1-12.0-12.2-12.3-12.3-11.9

-12.3-12.5-12.3-12.2

-12.1-12.7-12.1-12.3-12.1-12.5-12.4

J2H °/oo

-86.0-07.0-86.0-«6.n

-B7.0-82.0-90.6-92.0-88.8-87.7-88.0

-90.0

-H7.0

-89.0

ANAL.

UW

UW

UW

UW

uwuwuwuwuwuwuwuwuwuwuwuwuw

6:4

Table 6-1d + e

Table 6-ld. nary of I R0 and 2H Data fro» Scripa Borehole E 1.

NAME

E 1

INTERVAL

3-300 a

177.5-129.5 m

0-267 •

0-267 •

open

3-300 w

DATE

81-11-11

81-11-17

82-03-21

82-06-10

82-06-10

84-02-14

84-03-06

«1B0 ° /oo SMOU

-12.7

-12.BO

-12.00

-12.66

-12 .53

-12.30

-12.34

«?H °/oo SMOU

-94.(1

-95. ' .

-Rfc.'.

-90.5

-91 .1

-92.n

-90.4

ANAL.

UU

UPS

CSF

UPS

CSF

UU

IAF.A

Table 6-le. Summary of IR0 and 2H Data from Borehole N 1.

NAME

N 1

INTERVAL

3-300 •

123-125 •

203-205 •

271.10-273.10 •

276-276 •

3-300 •

10-119 •

120-15(1 m

151-251 >

252-300 u

151-251 •

252-300 •

DATE

81-08-19

82-06-10

82-06-03

82-06-03

82-08-30

82-08-30

82-09-06

82-09-06

82-09-14

82-09-14

82-09-23

82-09-23

83-01-23

84-02-14

84-02-14

84-02-14

84-02-14

84-01-26

84-01-26

6 IB0 °/oo SMOU

-12.6-12.97

-12.8

-12.6

-12.82

-13.3

-13.59

-13.1

-13.50

-13.0

-13.11

-13.2

-13.37

-12.75

-13.5

-13.5

-12.5

-12.9

-12.43

-12.47

42II °/oo SMOU

-89.5

-93.fi

-RS.9

-89. R

-93.7

-91.4

-98.6

-92.6

-97.7

-92.4

-94.8

-91 .It

-97.4

- 9 1 . n

-96

-97

-91

-94

- o n . 5

- 9 0 . 1

ANAL.

IAEA

UPS

IAEA

CSF

IAEA

CSF

IAEA

CSF

IAEA

CSF

IAEA

CSF

UPS

IIU

in*

uwUU

I ABA

IA FA

6:5

T a b l e 6 - 1 f Suwr; of IS0 and 7H Data f ra loreholc V I.

NAHF

V 1

INTERVAL

•09 -506 •

5-506 •

10-505 •

lOfl-505

DATE

81-06-01

81-07-13

»1-08-19

81-09-08

»1-09-11

81-09-21

»1-07-09

81-08-19

»1-08-28

81-08-28

81-09-08

81-11-17

82-06-10

83-03-22

»3-03-05

83-10-03

»3-10-19

»1-11-05

81-12-07

»4-01-11

84-02-0»

84-02-J4

«'*0 °/oo SHOW

-12.8

-13.0

-13.0

-12.9

-13.0

-13.0

-13.25

-12.5

-12.92

-12.91

-12.80

-13.08

-13.10

-12.93

-12.9

-12.40

-12.73

-12.86

-12.45

-12.54

-12.69

-12.9

«?H °/oo SMOW

-93

-94

-94

-94

-94

-94

-91.5

-»9.6

-91.7

-92.7

-91.6

-91.9

-91.0

-91 .2

-95

-91 .n

-94.0

-91.3

-94.0

-90 .1

-91 .1

-93

ANAL.

UU

1IU

uvUU

uwtni

ursIAEA

CSF

CSF

CSF

ursUPS

UPS

uwIAEA

IAEA

IAEA

IAEA

IAEA

IAEA

UW

- infiltrating waters were not subject to evaporation before orduring recharge and climatic conditions determine the stableisotope contents in the recharge area. They decrease with thenumber of condensation stages of the initial vapour, i.e. main-ly with temperature. Altitude effects, seasonal effects and pa-leoclimatic effects are the result. On a delta H - delta 0diagram, sample points usually follow a straight line with aslope of about 8.

- Water can evaporate before or during the recharge event: stab-le isotope contents increase with the amount of evaporation,following a variable slope which is lower than 8 on a delta H -delta18O diagram.

A crucial parameter for the discussion of the origin of a givenwater is the deuterium excess (Dansgaard, 1964), which is defin-ed as

d (per mil) - delta2H - 8 delta18O

Values of d are generally close to 10 °/oo in present day ocea-nic precipitation (Craig, 1961; Yurtsever and Gat, 1981) and, de-pending on the slope, are obviously much lower for evaporated wa-ters. The d-value is a reflection of the origin and history of avapour mass. Since in any given area meteorological regimes arenot subject to major ultra-annual fluctuations, this value re-

6:6

Table 6-lg s«

NAME INTERVAL

V 2 8 - 3 9 a

6.3-50 a

6.3-50 a

332-359 a

401-428 •

401-428 a

356-470 a

0-356 a

356-471 a

6-822 a

6-i,22 a

406-410 a

406-410 a

406-410 a

413-416.74 a

490-493.74 a

549-552.74 a

0-822 a

490-493.74 a

549-552.74 a

382-4 23 a

424-490 a

500-561 a

562-822 a

424-499 a

382-4 23 a

500-561 a

•ary of "*0 and

STKIPA NO.KBS-LBL

(43)

(86)

(17)

(17)

(68)

(59)

(69)

2H Data on Saaplei

DATE

78-06-06

78-11-20

79-09-06

79-09-14

77-09-22/10-20

77-09-22/10-20

79-03-16

78-11-20

79-04-06

79-04-24

79-05-04

79-05-17

81-06-03

81-11-

81-11-30

82-04-21

82-04-

BZ-l 1-2*

82-11-24

82-11-24

82-12-14

82-12-14

83-01-19

83-02-07

83-02-07

81-03-

83-01-

83-02-07

84-02-14

84-02-14

84-02-14

84-02-14

83-11-29

83-11-28

84-02-28

84-02-28

froa Borehole V 2

lX*0 °/oo SHOW

-12.6

-12.4

-12.6

-12.6

-12.2

-12.0

-12.2

-13.2

-13.6

-13.6. -13.7

-13.7

-13.3

-13.6

-13.19

-12.9

-12.8

-12.84

-13.0

-13.22

-12.8

-13.30

-12.77

-12.35

-13.13

-13.03

-13.10

-13.10

-13.0

-12.8

-12.62

-12.74

-12.70

-12.70

(410 a hole).

«2H °/oo SHOW

-90.7

-90.0

-92.0

-88.8

-89.0

-95.0

-98.0

-96, -98

-97.0

-96-2

-93.1

-90.2

-91.9

-91.4

-92.4

-95.1

-92.5

-95.9

-89.3

-91.1

-88.2

-90.0

-94.2

-93.2

-94. r

-94.0

-94.0

-92.0

-92.4

-95.9

-94.8

-93.5

ANAL.

UW

uwUW

UU

uwuwuwUU

uwuwuwuwuwUPS

IAEA

IAEA

CSF

IAEA

UPS

CSF

IAEA

CSF

IAEA

IAEA

IAEA

UPS

CSF

CSF

UW

UW

uwuvIAEA

IAEA

IAEA

IAEA

mains rather constant in the average annual precipitations at agiven location. However, major modifications would be introducedif:

vapour originating in closed, evaporating basins and from eva-potransplration were added to the condensing atmospheric vap-our.

a global change in climatic conditions of oceanic vapour forma-tion and meteorological circulation patterns were to occur (pa-lftocllmatic effect).

6:7

The former case is recognized in the Mediterranean basin (Gatand Carmi, 1970) where d values reach +22 °/oo, whereas the lat-ter case is strongly suspected for some ancient groundwaters,e.g. in deep confined aquifers of the Sahara (Gonfiantini etal., 1974; Fontes, 1981) and in Saudi Arabia (Hötzl et al. 1980)where values as low as 5 °/oo are found. Direct evidence for dvalues lower than present and close to +5 °/oo were recently ob-tained on Antarctic ice core profiles from the last glaciation(Jouzel pers. comn.). The lower d values are probably due to anaverage relative humidity over the oceans higher than at pre-sent, reflecting cooler conditions over tropical and equatorialoceanic masses (Merlivat and Jouzel, 1979). A modern example forthe dependence of d-values on the origin of the vapour is seenin precipitations over Canada where Pacific/Arctic dominatedweather regimes in Western Canada produce rain with d < 5 °/oo andthe warmer Gulf of Mexico and Atlantic regimes provide precipita-tions with d-values close to +10 °/oo (Fritz et al., in prep.)

Deuterium excess values can thus be an important tool for ground-water identification because studies limited to one isotope cannot characterize the origin of the initial vapour. Thus varia-tions in delta-values might only be due to local conditions ofrecharges in general climatic conditions. However, by defini-tion, deuterium excess values accumulate both H and 0 analyti-cal errors which typically are between 2 and 3 °/oo of the d-value. Therefore, accurate and repeated measurements are requi-red in order to obtain reliable values for this parameter.

The requirement of reliable analyses for the determination of dvalues is clearly emphasized by the data given in Table 6-2 andwhich compares isotope data obtained by participating laboratori-es on two samples. Although all laboratories clearly establishthe isotopic difference which exists between the two samples,the d-values calculated from these data vary by about 4 °/oo.This is outside the possible difference which might exist betwe-en different Stripa waters and, therefore, make the considera-tion of this calculated parameter somewhat problematic.

6. 2. 2 Nojrt^oUservaMXe—St^le .isotope_ £qntents

Non-conservative stable isotope contents are found under special

hydrogeochetnical conditions such as:

18

- geothermal environments where exchange between 0 rich mine-rals and water causes an increase of the 0 contents in thewaters (see e.g. Truesdell and Hulston, 1980);

- environments with fluids rich in COj and ^ S . If their oxygenor hydrogen contents are significant with respect to the atomiccontents of oxygen and hydrogen in the water, isotopic ex-

6:8

Table 6-2.

MARCH 1984

Laboratory

Neuherberg

Waterloo

Uppsala

Paris

Krakow

IAEA

Mean

a

Results of Intercomparison of Samples MIX and VTW.

MIX

6D

-45.0

-49.0

—

-47.6

-47.6

-47.3

1.7

(excl. Paris)

61R0

-5.50

-5.48

-5.89

-5.68

-5.50

-5.56

-5.60

0.16

VTW

6D

-82.6

-81.0

—

-84.2

-83.5

-82.8

1.4

(excl. Paris)

618O

-11.5

-11.4

-11.72

-11.52

-11.66

-11.27

-11.51

n.i6

change between gas and water will be significant. Exchangewith gaseous carbon dioxide will deplete the water in 0 andH2S exchange produces an increase in the H content of thewater. These are not common phenomena.

crystallization of clay minerals during weathering of felds-pars and micas causes under closed system conditions, an en-richment in oxygen 18 and a depletion in deuterium (see Savin,1980); i.e. the residual fractions of oxygen and hydrogen inthe water may be significantly depleted in *80 and enriched inH respectively (Fritz and Frape, 1982). A similar effect canalso be observed if exchange takes place between clays ormicas and limited amounts of water as may be the case in low-permeability rocks.

6.3 Discussion

18,Histogrammes depicting the distribution of lo0 in groundwaterat different depths and environments from the Stripa area areshown in Figure 6-1 whereas Figure 6-2 presents a summary of ^ 0and 2H data. Both show that consistent and substantial differenc-es do exist.

6:9

> WåTfH T å k f WELLS4 J Mil v A U »ELLS

nrv,4 J D*m

.1

ii1

•i|•i]

IN OLD MNC HIODIIINGS

nA.

JHCXn »O«[HOLf5 ÄT550 m LCVCl. (ICåvÄTlON

•OHEHOLCS Ni B EI

• O K H O L E V I

•OWHOLE V2

8 *0 %. SMO*

Figure 6-1. The oxygen-18 contents in different types of gro-undwaters at Stripa. Note, that shallow groundwa-ters and old mine drips are the only waters whichcontain substantial amounts of tritium. All datalisted in Table 6-1 are shown.

-70

O2

I

4 0 - 9 0

-100

4- SURFACE WATERSO WELLS WT, PW, SBH-3• BOREHOLES MJ, R1,ft3,R9. S I .S5• BOREHOLES E l . NlA BOREHOLE V I , BELOW 4 0 0 m• BOREHOLE VZ, BELOW 380 m

8 2H • (8 8*0 • 10)

-13 -12IB

-II

0 •/.. SMOW

-10

Figure 6-2. 1 ft ?6 0 versus 6 H in Stripa groundwaters. Only samp-les collected from short boreholes or between limi-ted packer intervals are shown. For clarity notall data points listed in Table 6-1 are shown.

6:10

Figure 6-3. 1 8Average annual 0 content in precipitation overSweden (From Burgman, et al., 1981).

Surface waters in the Stripa area lie on or to the right of theglobal meteoric waterline. This shift indicates surface evapor-tion. Such evaporated water can also be found in some shallowwells (e.g. PW 1) although this is an exception because the majo-rity of all shallow groundwaters are close to or on the meteoricwaterline. The best "definition" of the isotope contents of shal-low groundwaters is probably provided by "drips" in the old mineworkings. These waters are rich in tritium (77 to 160 T.U.) andrepresent a rather well mixed reservoir whose average composi-tion is close to -11 °/oo (see Figure 6-1 and Table 6-1). Mostshallow groundwaters in private and watertable wells closely

1 ftagree with this value, which is also the average annual 0 con-tent in local precipitations (Figure 6-3, Burgman et al. , 1981).

Deep groundwaters are depleted in heavy isotopes as compared toshallow groundwaters and waters discharging from boreholes atthe 300 m levels. The 0 and deuterium contents place them onthe meteoric waterline (Figure 6-2). Deviations most probably re-

6:11

fleet the analytical differences discussed above. They are thusnormal groundwaters which have preserved their composition sinceinfiltration.

However, heavy isotope contents of deep groundwaters are rathervariable between different boreholes and within a given well, de-pending on the packer intervals that were sampled. Nevertheless,the data for boreholes such as M 3 or specific levels in V 2 in-dicate consistent isotopic differences ovet a 7-year period, alt-hough, minor time-variations do occur. As in most confined aqui-fers, these are small.

D and 0 contents of deep groundwater (Boreholes V 1 and V 2)are similar and in the case of borehole V 2 indicate a small dec-rease with depth (a decrease of about 6 °/oo in deuterium and0,7 °/oo in 180 was observed in 1978-79 and about 3 °/oo in deu-terium and 0,4 °/oo in 180 in 1981-83). This trend has not chang-ed much over the seven years period although some variations areobserved and are attributed to the fact that samples were measu-red in different laboratories and at different time. It shouldbe noted that the differences in question are of the order of 2-3 times the analytical error.

Isotopic differences between samples from the different levelswithin the same borehole are an indication that different frac-ture systems deliver different types of water. Mixing may occurbetween, at least, two different types of waters, in order to ex-plain observed compositions at the different levels. However,the rather regular change in chemical and isotopic compositionswith depth in borehole V2 is probably a result of mixing in theborehole rather than a reflection of mixing between fracture sys-tems.

Considering only the most saline waters in the deep portions ofboreholes V 1, V 2 and possibly N 1, a depletion in 0 by up to2.5 °/oo with respect to the shallow groundwaters is observed.Comparing this figure with common altitude gradients for isotopecontents in precipitations, this isotopic difference could implyan altitude difference of several hundred meters between areasof recharge of shallow and deep groundwaters (possibly exceeding500 m). However, this signifies that regional flow systems wouldhave to be invoked. Under those conditions the chemistry of thedeep Stripa waters would reflect an evolution in a multitude ofdifferent rock types and could not be discussed in terms of ageochemical evolution within the Stripa granitic rock mass.

Exceptional precipitation events can sometimes be called upon toexplain differences in heavy isotope contents of groundwaterssince it is known that heavy precipitations can be depleted ifcompared to average values at any given location (Yurtsever andGat, 1981). However, it is unlikely that such events can accountfor the differences seen in the Stripa groundwaters.

6:12

6n 4

2-1

SURFACE WATERS

6

n 4-

2

L_3 SHALLOW G1OUNDWATER

d ] MINE DRIPS

6-n 4 -

2-

4 -n

BOREHOLES AT 330 m LEVEL

L_3 BOREHOLE N!

C D BOREHOLE ElI t" 'A T/////Å

2-

0-

8-

6-

4 -

2-

i

C 3 BOREHOLE

1 1 BOREHOLE

1

VI

V2,

!

BELOW 360 m

15

DEUTERIUM EXCESS

Figure 6-4. Deuterium excess (in °/oo) in Stripa groundwater

samples. Only sarnies analyzed by UW are shown.

An evolution outside the granitic type rocks at Stripa may also

have occurred if the waters had a local origin. However, their

lower heavy isotope contents must then reflect paleoclimatic con-

ditions. The observed 0 difference between water recharge and

the deep waters agrees well with differences observed at locali-

ties in Germany (Eichinger, et al., 1984), the UK (Bath, et al.,

1979) and Austria (Andrews, et al., 1984) where groundwaters

which formed during interstadials (> 25 ka) and during postglaci-

al cold periods (> 12 ka) have similarly lower heavy isotope con-

tents. The deep Stripa groundwaters would then be at least 10 ka

old. Tritium and C data do not contradict this statement.

As mentioned above, a possibly important parameter for interpre-

ting the origin of a given water is the deuterium excess (T)ans-

gaard, 1964). The deuterium excess of the Stripa groundwaters is

shown in the histogrammes of Figure 6-4. The data show that the

deeper waters have essentially the same d-values as the shallow

waters. Minor differences can be due to analytical variability

and, furthermore, some of the shallow groundwaters are influenc-

ed by evaporated surface waters with d-values as law as -2.7.

6:13

This similarity of d-values between deep, old groundwaters andshallow, young groundwaters is also observed in waters with esti-mated ages between 20 and 30 ka which are found in a Triassicaquifer in the UK. (Bath, et al. 1979). The 180 contents of theold waters are also about 2 °/oo below modern recharge values.Both observations compare well with our findings at Stripa. Onthe other hand, younger, postglacial (about 12 ka) groundwatersfound in the Tertiary basin of S. Germany have d-values whichare about 6 °/oo below modern but are paralleled by a 2 °/oo de-

1 ft

crease in 0 (Eichinger, et al., 1984); similar, lower d-valuesare recognized in Pleistocene groundwaters throughout North Afri-ca (Moser, et al., 1983).

Can the similarity of d-values in shallow and deep groundwatersat Stripa be used to argue that the deep waters must be pre- orinter-glacial (> 25 ka) in nature? One may not be able to answerthis question conclusively but there is little doubt that thelower 0 contents in the deep, saline groundwater at Stripa re-flect cooler recharge conditions. Burgman, et al., (1981) estab-lished a temperature coefficient of 0.547 to a network of Swe-dish meteorological stations which would signify that the isoto-pic difference between shallow and deep groundwaters reflects achange in average annual temperature of 3-4 °C. This assumesthat groundwaters always reflect the compositions of average an-nual precipitations. However, it must be noted that some shallowgroundwater reflect the seasonal variations of precipitation andcan reach the low values of the deep groundwaters. If, in thepast, selective recharge had occurred, then the lower heavy iso-tope contents of the deep water would not necessary reflect cli-matic change. Fortunately isotope data on deeper but young gro-undwaters (as represented by mine drips) show a smoothing ofsuch variations and suggest that the observed isotopic differen-ces reflect climatic differences rather than selective recharge.

It is very unlikely that glacial meltwaters participated in theformation of the deep groundwaters. These waters are considerably

1 ftmore depleted in 0 than the lowest values we observed and thus

only a minor percentage of the deep water could have such an ori-gin. Such a mixture would require mixing before the watersenter the fracture systems at Stripa, an interpretation whichwould greatly complicate the hydrogeological regimes in thisarea.

The origin of the waters which discharge from different bore-holes at the 300-360 m levels is not easily defined, despite the factthat they display a remarkable constancy in time of their isoto-pic compositions. This is especially true for the waters in M 3which since 1977 discharged several times 10 liters of water.In terms of their 0 and H contents most waters at these le-vels fall between the compositions of the shallow and deep wa-ters. Thus, they could be considered as mixed waters althoughchemical and tritium data do not substantiate this interpreta-tion.

6:14

18,Table 6-3. Cl~ and 1H0 Contents in Samples Collected 1984-02-13.

Borehole

M 3

N 1

V 1

V 2

Interval

10 - 119120 - 150

151 - 251252 - 300

0 - 550

382 - 423424 - 490

500 - 561562 - 822

Cl ng 1 '

34.5

18514829.037.1

553

467487430526

180 °/oo

-12.4

-13.5-13.5

-12.5-12.9

-12.9

-13.1-13.1

-13.0-12.8

The problem is further compounded by analyses done on samplescollected from specific intervals in borehoe N 1 (Table 6-3).The 61Ho values range from -12.5 to -13.5 which corresponds tovalues seen in M 3, for example, and the deep waters. These datawould indicate that some of the fractures encountered in N 1 dis-charge waters which are similar to the deep saline waters. Triti-um data appear to exclude that hydraulic connection exists, alt-hough the chlorinities are highest in the isotopically lightsamples and could lend some support to this view.

Hydraulic testing, however, exclude a direct link because the1 Q

fractures carrying low 0 water in Nl do not connect hydrauli-cally with V 1 or V 2. The testing does indicate that the deeperportion of Nl connects with the BMT area and the R 1 borehole.

6.4 Regional saline groundwater survey

A regional survey of saline groundwaters was undertaken in orderto determine whether Stripa groundwater was geochemically andisotopically unique or whether it was necessary to propose moreregionally valid interpretations. The results of 0 and 2H ana-lyses are listed in Table 6-4, their location is seen in Figure6-5.

Comparison with the isotopic composition of average annual rain-fall at the different localities establishes that all but threesamples could have a local origin. However, samples 2, 3 and 4have 0 and 2H contents which are lower than would be expectedfor modern recharge. Their formation occured in a cooler clitna-

Figure 6-5

6:15

S - SWEDEN

1 - HÄSSELBY

2 - WILHELMSLUND3-KAGA4-NORRA STENE5-STORA SUNDBY6-HAMMARÖ7-SKOFTEBY8-HANGELÖSA9-SMEDTOFTA»-ROCKAGARDENII - ÅKER

f

Table 6-4 ia,O and H In groundwitcr In South Central Sweden. Collected 84-04-22.

Locality Aquifer rock 6I8O "7oo «2H °/oo Lab

HXSSLEBY

WILHELMSLUND

KACA

NORRA STENE

STORA SUNDBY

HAMMARS

SKOFTEBY

HANCELOSA

SMEDTOFTA

ROCKACXROEN

AKER

Crystalline rock

Canbrlan Sandstone

Crystalline rock

Crystalline rock

Crystalline rock

Crystalline rock

Quatern. aedlaenta

Crystalline rock

Quatern. icdlments

Quatern. icdlaents

Quatern. aedlmcnts

11.5211.27

12.3312.57

13.9*15.12

14.3214.53

11.9212.03

11.7812.44

-8.25-9.08

-9.80-9.94

-9.31-9.83

-9.03-8.99

-8.80-8.44

- 7 7

-81.1

-83.5-87.3

-98.0-109.7

-102.2- J O * . " ;

-83.7-88.1

-86.3-90.2

-60.1-63.5

-64.8-70.0

-64.0-6B.8

-60.6-63.0

-58.0-60.6

IAEACSF

IAEACSF

IAEA

CSF

IAEACSF

IAEACSF

IAEA

CSF

IAEA

CSF

IAEACSF

IAEA

CSF

IAEACSF

IAEACSF

6:16

te, whereas the Kaga (No. 3) and Norra Stene (No. A) samples aredepleted in heavy isotopes even with respect to Stripa waters.It is interesting to note that the Kaga sample is one of twosamples that has a Br/Cl ratio significantly higher than seawa-ter and an anomalously high Ca/Mg ratio. The other sample is Wil-helmslund. In this respect these resemble the deep Stripa wa-ters. For further discussions, see sections 4.6 and 4.8.3.

6.5 Conclusions

Stable isotope analyses are well suited to characterize diffe-rent groundwaters in the project area. Results show that theStripa groundwaters are normal meteoric waters which have notbeen subject to surface evaporation or secondary isotope ex-change processes with rocks and minerals. Their compositions re-flect recharge conditions.

to 2

The lower °0 and H contents of the deep waters indicate an in-

filtration during a cooler climate than exists today in this1 ft

area. In this respect, these waters are similar to low 0, lowH waters encountered in the regional survey and is suggestedthat these waters are relatively old. "Age" estimates will be at-tempted on the basis of other data presented here.However, it is important to notice that at Stripa in generallow heavy isotope contents are paralleled by increasing salini-ties, although the parallelism is far from perfect. Therefore,one can suggest that the observed compositions are not only dueto the mixing of two different water types but also to geochemi-cal reactions which modify chemistry and total dissolved solids.That mixing occurred is relevant for the discussion of water"ages" (Fritz, et al. , 1983).

Remarkable are the differences which exist for both chemistryand isotopic compositions in adjacent fracture systems. This mayreflect different flow paths. The appearance of isotopicallylight water i.i fractures in different boreholes may reflect hy-draulic connection even where hydraulic testing was not suffici-ent to substantiate this.

7:1

STABLE ISOTOPE GEOCHEMISTRY OF SULPHUR COMPOUNDS

7.1 Introduction

Sulphur geochemistry in the Stripa groundwater system is of spe-cial interest because (1) large variations in sulphate contentoccur in groundwaters from different boreholes, (2) there is arather systematic increase of sulphate concentrations with chlo-ride at depth, from less than 1 mg 1 to well over 100 mg 1 ,(3) both reduced and oxidized forms of aqueous sulphur occur and(4) the presence of solid sulphides (such as pyrite and chalcopy-rite) are found in the rock matrix and on fracture surfaces.

The discussion of available data is preceded by a basic reviewof natural isotopic variations in sulphur compounds.

7.2 Variations of stable isotope contents of sulphur compounds

Biological and chemical reactions involving oxidized and reducedsulphur compounds generally cause very significant isotopefractionations for both sulphur and oxygen isotopes. The sulphur-34 and/or oxygen-18 differences between two compounds 1 and 2 ina geochemical cycle of sulphur are expressed as a or e, stand a Bidnotations for all environmental isotopes (Friedman and O'Neil,1977):

and e (epsilon) = a-1 - In a - 6j - &2a (alpha) •

where

STD

and

R - 3AS/32S

1)

or

'R2 and

x 1000

180/160

Epsilon is also called the "enrichment factor" which directlydescribes isotopic differences between two compounds. For conve-nience epsilon is generally expressed in per mil (i.e. multi-plied by 103). Standards are

the troilite phase from the Canyon Diablo meteorite (CD) for S

and

Standard Mean Ocean Water (SMOW) for 1 R0.

7:2

Equilibrium isotope effects are generally very high for sulphurcompounds especially at low temperatures (Friedman and O'Neil,1977).

However, elements such as sulphur enter into metabolic cyclesand isotope effects may occur under the influence of biocata-lysts which increase rates of chemical and isotopic reactions.Under these conditions isotope effects are no longer controlledby equilibrium but by kinetic fractionations. Kinetic isotopefractionation effects are not only temperature dependent but de-pend also upon environmental conditions and are defined withinranges rather than as precise values. Biochemical catalysts "pre-fer" light isotopes which results in heavy isotope concentra-tions in the unmetabolized fraction.

Due to the occurrence of kinetic and reservoir effects duringthe formation of reduced species, a large range of S contentsis observed in natural sulphides. For instance, sedimentary pyri-tes may show 634S values from -50 to 70 °/oo (Krouse, 1980) butare generally depleted in S with respect to coexisting sulpha-tes. When formed at high temperatures from deep (crustal or mant-le) sulphur, volcanic H0S remains generally close to the stan-dard troilite and shows 6 S values near zero or slightly negati-ve.

Fortunately, in most aquifer systems, reducing conditions arenot frequent and because the solubility of reduced sulphur speci-es is low, the most frequent form of dissolved sulphur is sulpha-te.

Oceanic masses provide the main reservoir of sulphate.

The isotopic composition oceanic sulphate is very constant becau-se of a steady state process between input (river) and output(precipitation and reduction):

618O (SO2~) - +9.5 SMOW

634S (S0^~) - +20.0 CD

Meteoric sulphates from oceanic rains and aerosols (unpolluted byindustrial dusts and smokes) show marine values (Mizutani andRafter, 1969; Rightmire et al., 1974).

During sulphate precipitation in the form of gypsum as anhydritean isotopic fractionation occurs. Experimental and field valuesof epsllon for gypsum and anhydrite fall in the range of:

4.0 > epsilon(18O) > 3.0 % o (Lloyd, 1968; Pierre, 1982)

and

3.4 > epsilon(3AS) > 1.4 °/oo (Thode and Monster, 1965; Pierre,1982).

7:3

ouins10

»M

• 21

• H

(

/

f

t/

/

/

/

;/

/

/t

i

•71

//t

f

1

t1

1

1>

1 */ *

1 PRESENT

I

t

O>*

//

" / f

i '/ '

f // S/ t

/ /t *

f /

' \

*•*

SEA WATER SULPHATE

•

'II 6"0 SMOW

Figure 7-1. Sulphur 3A and oxygen 18 evolution of present dayaqueous oceanic sulphate during crystallization ofsulphate and sulphate reduction. Solid arrows: onestage process.

As a consequence, after precipitation, the remaining aqueoussulphate is depleted in heavy isotopes and further precipitationstages lead to solid sulphate depleted in -* S and 180 withrespect to the initial aqueous So|~ (Figure 7-1). Thus, marineMg-K sulphates show 3 S contents below average ocean values (Nielsenand Ricke, 1964).

7:4

7.2.1

Reduction hardly occurs at low temperatures except if sulphatereducing bacteria are present in the environment, e.g. Desulfovi-brio desulfuricans. Kinetic enrichments and steady state condi-tions become predominant. Whereby, products from this biosynthe-sis are depleted in heavy isotopes by 40 +10 °/oo (Claypool etal., 1980). Therefore, the remaining portion of SO^ is enriched inheavy isotopes. Under quilibrium conditions, the isotopic dif-ference (epsilon ) would be about 80 °/oo at 30°C (Friedman andO'Neil, 1977).

The behaviour of oxygen isotopes during sulphate reduction can

only be investigated indirectly since COj - the most common re-

sulting oxygen bearing compound in the reaction - is mixed with

environmental COn and reequilibrates with water. However, the re-?- 18 14

maining aqueous S0| becomes less enriched in 0 than in S byapproximately 1/4, (Rafter and Mizutani, 1967; Mizutani and Raf-ter, 1969). One assumes that typical values of epsilon (kin) for180 are close to +10 °/oo (Claypool et al., 1980). However,rates of kinetic reactions are generally specific for a given en-vironment and, therefore, values of epsllon (kin) may exhibit alarge range of variations. Mizutani and Rafter (1969, b) reportvalues of the ratio epsllon S(kin)/ epsilon 0(kin) between-7.8 and +23.5 °/oo. In reducing evaporitic ponds, Zak et al.(1980) and Pierre (1982) found values close to +1.5 for thisratio.

7.2.2 Sulphide and HjS oxidation

Reduced sulphur is quickly transformed into S , So2~ and finallyS0£ when groundwater circulation reaches an oxidizing zoneand/or mixes with superficial waters saturated with oxygen. How-ever, besides inorganic processes, oxidation may occur throughbiological (bacterial) activity (Thiobacillus thiooxldans).

Further experimental data are still needed about sulphur isotopefractionation occurring during oxidation. It is generally found(see Krouse, 1980; Pearson and Rightmire, 1980) that the purechemical process is not fractionating whereas slight depletionin S may occur through bacterial formation of So|~ (Kaplan andRittenberg, 1964).

The atoms of oxygen participating in the formation of the sulpha-te ion from HjSCfor HS~ and S^~) may have several origins. Ithas been shown by Lloyd (1967, 1968) and by Mizutani and Rafter(1969 a, b) that the oxygen involved in the process comes fromenvironmental water as well as from dissolved (molecular) oxygen.

According to Lloyd (1968), the isotopic balance requires that aproportion of 1/4 of the oxygen atoms be supplied by the waterand 3/4 by dissolved oxygen. The incorporation of water oxygen

7. 2.3

7:5

into the sulphur bond would occur without fractionation. The ad-mixture of dissolved oxygen occurs with an isotope fractionationeffect, where:

618O (aqueous S0^~) = <5 0 (dissolved oxygen) - epsilon ox.

The isotope enrichment factor epsilon ox. is 8.7 °/oo (Lloyd, 1968).If dissolved oxygen is of atmospheric origin and not modified byorganic processes one finds 5l80(ox) = 23.5 °/oo (Kroopnick andCraig, 1972). However, isotopic exchange processes are ocurringbetween water sulphite ions so that the respective participationof H2O and O2 in the oxidation are given by:

618O (S0^~) = 0.66 618O (environmental H20) + 4.9

This leads to sulphate ion with a 61 0 close to +5 °/oo of oxida-tion takes place in sea water (6 0 = 0) and to -3.7 °/oo inwater with an 0 content of -13 °/oo (Stripa deep water).

According to Cortecci (1973), however, the proportions of atmos-pheric and water oxygen participating in the formation of sulpha-te ion approach stoichiometric proportions, where:

618O(SO|~) oxidation =

{2[<5(O2) - 8.7 + 6(H2O)]2 + <5(O2) + 6(H2O)}Mi.e. +9.6 and +3.1 in waters with 618O = 0 and -13 °/oo respec-tively.

However, this picture may also be modified if dissolved oxygenis partially used for metabolic respiration. In that case an oxy-gen enrichment occurs in the remaining fraction of the initialdissolved oxygen of atmospheric origin (Fontes and Michelot,1983).

This brief discussion shows that more data are needed before iso-tope effects caused sulphate production through oxidation of redu-ced sulphur can be fully evaluated. Nevertheless, for practicaluses one can assume that oxidation processes in groundwater pre-serve S contents of the initial reduced sulphur and produce0 contents much lower than those known for marine evaporites.

Lloyd (1967) as well as Longinelli and Craig (1967) establishedthat ionic SO4 in the oceanic reservoir is far from isotopicequilibrium with sea water.

This disequilibrium is probably due to the extremely slow rateof reaction of the S-0 bond with water at neutral pH. Values forthis reaction rate have been experimentally determined by Lloyd

7:6

(1968) for different pH values in the temperature range 298 to721 K. From these values Pearson and Rightraire (1980) derivedthe following relationship between halftime of reaction tj/2>temperature T in K, and pH:

log t,/2(hours) = 2.15 x 103 x T"1 + 0.44 pH - 3.09

which theoretically could provide a geochronometer since the dis-tance to equilibrium is time dependent.

However, care should be taken in any attempt to use this formula

for time estimates in groundwater systems, because:

(a) to obtain the experimental data base, experiments were perform-ed over limited ranges of time (some months). Low temperatureruns were thus conducted at very low pH; therefore extrapola-tions to conditions f normal environments and to long periods(e.g. milenia) would be risky since no error estimate is avail-able;

(b) kinetic isotopic reactions may interfere in the SO?" - H~O sys-

tem as well as catalytic or inhibiting effects, specifically

during biochemical reduction or oxidation:

(c) equilibrium values for isotope fractionation effects betweenS0£ , HSOj ions and water at low temperature are still underdiscussion.

Furthermore, it must be remembered that the residence time ofany aqueous compound is not necessarily that of the groundwaterin which is dissolved.

7.3 Sampling and analyses

Because of the highly variable SO^ content (from less than 1 toabout 100 ppm) In the Stripa waters the amount of water collec-ted during this phase of the program varied from 2 to 60 litres.

Where necessary (well N 1) the water sample was treated with rea-gent grade HC1 in order to remove H2S and HS" whose further oxi-dation may modify the stable isotope content of aqueous S0?~.In the case of surface water, precipitation of sulphate on a 25litre sample was unsuccessful.

Samples were treated with a solution of barium chloride In orderto precipitate S0£ as barium sulphate* Barium carbonate is thenremoved by H addition and a further precipitation step is made.The final barium sulphate rinsed and dried, Is allowed to reactwith pure graphite at about 1000°C under vacuum. CO2 and carbonmonoxide are produced. The latter is converted Into COj. An all-quote of BaS produced through the reduction is dissolved and con-

7:7

verted into Ag_S and oxidized in pure and dry oxygen at torchtemperature. Oxygen and sulphur isotope contents are then deter-mined through CO- and SO2 analyses on a VG Micromass 602 D massspectrometer. Uncertainties are about +0.25 on both measure-ments.

A number of samples were collected on an experimental basis withion exchange resins. The result are encouraging and further stu-dies will be undertaken using this method.

A sample of 250 ml of water is also collected for SO?" determina-

tion and 0 measurement in the water.

7.A Results and discussion

All data are summarized in Table 7-1, and graphically shown in

Figure 7-2.

7.4.1 Aqueoujs ulphate_i_n j>ubsjjrf_a£e__waters_

Representative points of shallow groundwaters (three measure-ments) fall within a heavy isotope range which was previouslyconsidered as intermediate between marine sulphate (sea spray)and sulphates resulting from oxidation of reduced species ofsulphur (Fritz et al., 1983, Fontes and Michelot, 1983). Such va-lues were also observed in New Zealand (Mizutani and Rafter,1969c), in Italy (Cortecci and Longinelli, 1970) and attributedto a mixture of sea spray and sulphates produced through oxida-tion of fuel-sulphur.

A detailed study of •* S and 180 contents of meteoric sulphate inPoland (Trembaczowski and Halas, 1984) show a rather constant634S ^ +3.7 °/oo and a 618O strongly correlated to that ofprecipitation water: 618O(So|") - 16.0 + 0.9 + (0.38 + 0.07)618O(H2O). For Stripa this relationship yields a 618O value ofabout +12 °/oo, using the average 180 content of -11 °/oo forthe local precipitation (Burgman et al.t 1981). Thus the inter-cept value (+16 °/oo) or slope do not apply for the Stripa gro-undwaters. Another possibility is that these sulphates contain acomponent very depleted in heavy oxygen such as oxidized sulphid-es. Sample PWI is probably affected by a secondary enrichment inheavy isotopes due to a partial reduction of the SO?" bulk faci-litated through water stagnation (the same well shows a low deu-terium excess which indicates an evaporation effect from thewell).

Shallow aqueous sulphates do not seem to have any connectionwith sulphates of V 1 and V 2.

7:8

Table 7-1 °0 and S conr*nta of aquaoua aulphatea fro Stripe.

Borehole

FU 1

PU 3PU 4

WT 2SBH 3M 3

II 1E 1N 1

V 1

V 2

Hteeclbjr

K.gaNora Stene

Storl Sundby

Hanar»

Skofteby

Hangeloia

Rockagarden

Aker

Interval

89-104

3-300151-251252-300409-506409-506

5-5065-506

100-505100-505100-505100-505100-505

401-4 28401-428356-470406-410424-499562-822382-4 23424-490500-561562-822382-423424-499500-561562-822

Date

79/05/1882/06/1079/05/1679/05/1782/06/0979/05/0579/05/2878/1178/1179/0579/0579/0579/11/2179/11/2282/06/1084/02/1478/08/0984/02/1482/06/1084/02/1484/02/1481/06/0381/07/0981/11/1782/06/1083/10/0383/10/1983/12/0784/01/1184/02/0884/02/1478/11/2079/05/0481/06/0382/11/2483/11/2883/11/2984/02/1484/02/1484/02/1484/02/1484/04/2684/04/2684/04/2684/05/0383/04/2083/04/2083/04/2183/04/2283/04/2283/05/0283/05/0283/05/0383/05/0383/05/0483/05/0483/05/0483/05/0483/05/0583/05/0583/05/0583/05/05

« l 80 SMOU

•4.28

-1.7+1.02

+0.78+0.9**, +3.3+8.8•5 .7* *+5.27

+11.1**+11.7**+7.80+7.30+8.10+7.81+7.58•7.58•8.26•7.83+7.55+7.4**, +7.6+8.5

+7.95+7.54+9.15+9.72+7.7**, +8.1+ 8 . 7 " , +8.6+9.4**, +9.3• 9 . 6 " , +9.8+8.70+9.58

•10.59•10.06•H4.0•9.9-0.3•8.0•7.7+8.5+7.7

+10.8+8.7

•18.4+12.4+17.9+15.1+17.8•11.8+19.3+15.1

«>*S CD

+7.6*+5.13+8.3+2.8+3.60+5.2*+7.2

+16.5•17.6•13.2+12.6

(+14.2)+12.6*+13.0*+11.28+13.0*»,+20.7+5.8

+27.92+39.3**+27.3**+12.84+12.85+13.96+13.79+13.84+14.08+14.54+13.91+14.38.+13.9•17.4+20.1+15.04+15.35+18.94+25.50*+16.6**,+21.6**,+27.2**,+ 2 9 . 1 " ,+14.90+21.60+24.30+24.80•Hi. 9+26.9+5.1

+18.3+23.2+18.0+20.1*•14.9+17.0*+69.7+74.6+42.1+46.1+44.3•47.7+74.7+79.6

+13.4

+14.3**

+17.3+22.0+27.0*+27.8

Lab

UWUPSUWUWUPSUWUWUWUWuwuwuwuwuwUPSuwuwuwUPS

uwuwUPSUPSUPSUPSUPSUPSUPSUPSUPS

uwuwuwUPSUPSUPSUPSuwuwuwuwUPSUPSUPSUPSUPSuwUPSUPS

uwUPSuwUPS

uwUPS

uwUPSuwUPS

uwUPSuw

* Average value.** Saaple collected ualnf Ion exchange nilnii.

UPS: Unlveriltt ParU-Sud, France.UW: University of Waterloo, Canada.

7:9

CD

• 25

• 2 0

• 15

• 10

•»5

0

0

\ C0

— •

m m

I I I I-10 -5 0 +5

^ Pr*»nt day » a «ai«r

Eys-Dashwa syptum

M Pcrmo-Trlauic •«apern«s

• 10 e'V> swow

• PW

O M3

a NI

V R!

• V1

O V2

Figure 7-2. Oxygen 18 vs sulphur 34 contents in aqueous sulphat-es from Stripa.

7.4.2

The sulphates of Stripa groundwaters may have several originscorresponding to various ages:

a) recent sea water Intrusions, i.e. lateral intrusion;

b) Holocene Baltic sea intrusion, i.e. vertical and/or lateralintrusion;

c) leaching of fluid inclusions or some form of rock sulphate;

d) leaching of evaporites or contribution of associated brin-es.

In addition, geochemical reactions especially redox processes in-volving biological activities may have modified the originalsulphate concentrations and compositions. This, however, is

7:10

least likely for V I, whose waters show rather constant i°otopiccompositions and highest sulphate concentrations.

In the ö'*S - 6 0 diagram (Figure 7-2) most of the points fromV 1 fall close to the range of sea water derived sulphates asdocumented by the average values, where SO^ has 6 0 = 7.76 +0.29 and 634S « +13.R0 + 0.59.

This could reflect different sources, e.g.:

1) a succession of biogeochemical processes which could havegiven such a heavy isotope content starting from any sulphurspecies;

2) a true sedimentary origin with preservation of the originalheavy isotope content.

Hypothesis (1) would, for instance, imply the following geochemi-cal pathways:

a) a dissolution of a fossil evaporite wif.h a S content closeto +14,

b) a complete reduction of this sulphate which maintains a sulp-hur 34 content of +14 °/oo,

c) an oxidation of the reduced sulphur by dissolved oxygen con-siderably enriched in heavy isotopes through an intense res-piration process.

Condition (a) requires an evaporite source.

Condition (b) requires that reduction products remain homogeneiz-ed after completion of the reaction, i.e. a stagnant flow regi-me.

Condition (c^ implies a transfer from the reducing to an oxidi-zing reservoir.

This scheme appears somewhat complicated (despite the fact thatit represents one of the most simple pathways to reach the obser-ved heavy isotope contents) and therefore preservation of origi-nal sedimentary compositions appears probable.

7.4.3 I)iscus£ion_o;f_ ^h£ vaj_lou£ j>OBsi^Le_o>ri£ins_of_ the^ £queouj[ _sulpha-te

o Hypothesis (a): No hydraulic data are available on a possiblepenetration of present day marine waters or Baltic seawater tothe Strlpa region. However, V 1 aqueous sulphate has an aver-age S content much lower than modern open marine sulphate. A

7:11

contribution of sulphate produced through oxidation of reducedsulphur species could account for this low S content. How-ever such a process would have led to a significant decrease of

l ftthe 0 content of the sulphate which cannot be recognized.Furthermore, in the case of a supply of sulphate derived fromreduced sulphur into a bulk of sulphate of marine origin thefollowing balance should match simultaneously where m, c and sstand for marine continental and Stripa waters respectively:

Wm + Wc = Ws = 1 : water balance

Wm618m(H2O) + WC618

C(H2O) = WS618S(H2O) : 1 80 balance in water

Wm x Cm(SOA) + Wc x CC(SO4) = Ws x CS(SO4) : Sof~ balance

Wm x Cm(S04> x 6 V S 0 4 > + WcCc(S0A)6 loc(S0A) = Wg x CS(SO4)

518S(SOA) : 180 balance in So|~

Wm x Cm x fi34 + Wc x Cc x 634C = Wg x Cg x 5 3 4

S : 34S balance

Wm x Cm(Cl) + Wc x CC(C1) = Ws x CS(C1) : chloride balance

Wm x Cm ( B r ) + Wc x cc

( B r ) = Ws x C s ( B r ) : b r o m i d e balance

With the following additional constraints for the marine compo-nent:

Cm(Cl) = 7.1 Cm(SO4) 6 l 8

m ( H 2°) = ° C m ( C l ) = l 7 0 0 °

634m(SOA

2") = +20 ö l 8m

( S O 4 ~ ) = + 9 ' 5 C m ( B r ) = 6<5

This system can not obey the necessary constraints Wm, Wc, Cm,Cc >0» except for Wm £0 .03:

¥

0

0

0

a

.01

.02

.03

Wc

0.

0.

0.

99

98

97

« C1 8 (H 2 O)

-12.9

-13.1

-13.2

Cc

76

S3

29

(so4)

.8

.2

. 0

«c

• 7

• t

+3

"(so4)

.2

.2

.3

« « *

+11.4 8 .

- 2 .

(SO,)

8

1

0

< cc(ci)

43*

265

928

C

5

S

4

c(»r)

.9

. 3

.7

The discussion of such a small contribution appears meaning-l e s s . However, the continental component would be veryrich in sa l t s (So|", Cl~ and Br~) the origin of which wouldbe di f f icul t to explain.

7:12

T a b l e 7 — 2 Stable Isotope contents of waters and aqueous sulphates fro» Baltic Sea.

SamplingDepth

- 50 •

-200 •

Cl-8 I"1

4200

7800

sn;"- 8 I"1

560

870

1R0 (H,0)°/oo SMOU

-6.97

-5 . ft*

2H (HjO)°/oo SHOW

-57 .7

l8o (so^-)°/oo SHOU

+6.93

+8.83

34S (SOJ-)°/oo CD

+19.22

+19.40

It is concluded that a supply of marine sea water can not be

invoked.

Therefore hypothesis (a) which, furhermore, is not in agree-ment with other geological and chemical data (Nordstrom, 1983)will be discarded.

o Hypothesis (b): Another possible source of sulphate in theStripa system are the Quaternary precursors of the Baltic Sea,e.g. Yoldia or Littorina seas (Fritz et al.. 1983). We have nodata for these waters but measurements of S and 0 contentsof Baltic Sea sulphates show that the marine isotopic signatu-re is rather well-preserved even when the water is highly di-luted (Table 7-2). Since the heavy isotope content of openocean so|" has remained constant through the entire Quaternary(Claypool et al., 1980) a simple intrusion of Holocene BalticSea (hypothesis b) cannot provide an explanation for the ori-gin of deep groundwater S0?~ in the Stripa system. In that casea "minor admixture of isotopically light sulphate" (Fritz et_al., 1983) could hardly account for the observed differencesfor the same reasons as above.

o Hypothesis (c): Fluid inclusions of granite contain sulphate(Nordstrom, 1983, and this report). However, the concentrationis low (less than 20 ppm in the rock according to our measure-ments). Because of this scarcity, it was not yet possible todetermine the stable isotope content of this sulphate and todiscuss its origin. Measurements of Cl indicates the chlori-de is not in secular equilibrium with the granite, although itmight be leaching out of the leptite. Thus the sulphate is pro-bably not coming directly from the granite. However, dissolu-tion of secondary sulphate could be possible even if no solidsulphate was recorded from fracture filling minerals.

Fracture sulphate (gypsum) was observed in Eye-Dashwa lakespluton (Ontario). The sulphur 34 content of this gypsum raneesfrom +5.3 to +8.5 (Kamlnemi, 1983). This range, much below -**Scontents of Stripa aqueous So|~ (Figure 7-2), is attributed toa preservation of sedimentary Precambrian sulphate. Further-more, the 0 content is not available. Since a complete discus-sion of the origin of the sulphate is hardly possible withoutthe knowledge of *°0 values (Michelot et al., 1984), one can-

7:13

not draw any further conclusion on the comparison with theStripa system.

o Hypothesis (d): Leaching of evaporites would probably have toinvolve Permian deposits since Zechstein evaporites have heavyisotope contents which are very close to those observed insulphates from borehole V 1 (Claypool et al., 1980). Sinceborehole V 1 shows the highest sulphate content of the system,its heavy isotope content should have been the most preservedfrom significant modification due to partial reduction or furt-her mixing. No Permian deposits are presently known in theseregions of Sweden. However, recent paleogeographical recon-structions (Ziegler, 1982) map a Permian gulf in the Oslo re-gion. Permian salts may thus have been deposited in the Fenno-scandian peninsula and successively dissolved giving rise todense solution which could have infiltrated the Fennoscandianshield.

An alternative hypothesis takes into account the occurrence ofgypsum, anhydrite and complex sulphate salts of Permian agethroughout Northern Germany and Poland. Rivers from the catch-ment area of the Southern shore of the Baltic may have dischar-ged diluted solutions of these sulphates into a "pre-Baltic"basin isolated from open ocean. Infiltrations would havebrought this sulphate into the Stripa groundwater system. A verysimilar explanation may involve small supplies of Permian brin-es to the deep groundwater system (Michelot et al., 1984).Sulphates from evolved (post halite) brines would be slightlydepleted in 0 (and subsidiary in S) by previous gypsum, an-hydrite (and polyhalite) crystallization. It should be notedthat heavy brines (denser than Baltic seawater) have beenfound in southern most Sweden (Smellie, 1984, pers. comm.)with a Br/Cl ratio close to 0.0114 and from southern Finland(Kankainen and Hyyppä, 1984).

Regardless of the "origin" of this Baltic sulphate the ques-tion arises as to when this infiltration could have occurredand whether it took place before or after the last glacial pe-riod (assuming that no infiltration occurred during glacialtimes which still requires confirmation).

7.4.4 i11!.8^^ _reiactions_ nvolying_ jsul phur jcanpouncls_:_V_2_L fJ j_, M }_,and JR J

All these boreholes show variable but much lower SO?" contents thanV 1. Most of them contain reduced species of sulphur.

1 ftThe 0 isotope contents of V 2 appears sometimes much higherthan those of V 1. Furthermore, values lie along a line with aslope close to 4. This is indicative of a partial reduction of asmall amount of the initial sulphate bulk by bacteria. A partial

7:14

bioreductlon of So£~ could also explain the much smaller ratioSO|~/C1~ in V 2 (= 0.11) than in V 1 (= 0.16). The reduction linepasses through V 1 values confirming that V 1 contains the initi-al, not reduced, bulk of sulphate.

For samples very depleted in sulphate (N 1, M 3 and R 1) theheavy isotope content may also have been modified by a partialreduction of sulphate (e.g. in R 1) or oxidation of reduced sulp-hur (e.g. in M 3). A very pronounced partial reduction couldalso account for the very high S of Nj. For both N 1 and M 3,the reduction process might have occurred from an initial sulpha-te completely different from V 1 and which could be the shallowsulphate (see Figure 7-2). This hypothesis, which would imply anhydraulic connection between surface and the upper part of thedeep groundwater system, still requires further measurements andinvestigations.

7.4.5 eg_ional_s_urvey_ ^f_sal_in£ £roundwa_tejrs

Heavy isotope contents of aqueous sulphates show extreme varia-tions (about 70 °/oo in 3/|S and 20 °/oo in 1 80). Values are notscattered: oxygen 18 contents follow an asymptotic tvend to about+20 °/oo (Figure 7-3). Similar large variations have been repor-ted from areas in which successive cycles of reduction and oxida-tion may occur (Gilkelson et al., 1981; Basharmal in Fritz,1983). As suggested by Fritz (1983) the evolution may be due tothe combination of (i) a Rayleigh process (incomplete reductionwith removal of the reduced species through precipitation ordegassing), (ii) a reaction between S0^~ and HjO which tends to-ward isotopic equilibrium. Calculations of the various fractionsof remaining sulphate after one single stage after reduction pro-cess in closed system give reasonable values for the initialcontent in aqueous S0^ , assuming (i) a kinetic enrichment factorof 40 °/oo during reduction (Claypool et al., 1980), and (ii) onunique initial bulk of aqueous sulphate defined by the heavy iso-tope content of sample from V 1 (see Table 7-3). This enhancesthe possibility of V 1 to be representative of the regional bulkof aqueous sulphate: it would also strongly suggest that highcontents in aqueous S0£ in the regxon are:

- mainly derived from a common source;

- independent from their geological location-

Sample nr 3 in granite has a relatively low S0^~ content (35 ppm).

It could represent either the oxidation of a reduced

sulphur species previously derived from the same S0^~ bulk, or a

S0^~ supply of meteoric origin as in the Stripa shallow water

(see PU4).

At 10°C (average temperature for these groundwaters), the enrich-ment factor epsilon* (SO^ - 1^0) is approximately +35 °/oo. At equili-brium the aqueous S0^~ of sample nr 11, which is the most en-riched, should reach a ' 0 content close to +25 °/oo. None of the

7:15

• 80

6**S(SO4)CD

•60 —

+40 —

420 —

-20 -106 O(H2O,SO4)SM0W

• Water Sulphate

Figure 7-3. Oxygen 18 vs sulphur 34 contents in aqueous sulpha-tes from central Sweden wells.

aqueous sulphate appears to be in equilibrium with the groundwa-ter.

Theoretically one could then evaluate the residence time of SOf~ions within a groundwater system through the knowledge of en-vironmental temperature and pH which control the reaction. Satis-factory agreements are reported between 1 C and ^8O(So|~) age es-timations from a carbonate aquifer in Manitoba (Mkumba, inFritz, 1983).

In the case of sample nr 11, a half-reaction time of 8360a iscalculated at T - 283 K and pH • 7.64: according to Pearson andRightmire (1980) equation, the corresponding time t of contactbetween water and aqueous sulphate is thus given by:

A - epsiloneq (1 - e"(ln2/Tl/2)l:)

with:

A - 618O(SO?~) - 618O(H,O).

7:16

Table 7-3. Regional program sulphates.

Nr34S °/oo CD f (S0|~)1 ppm

11 +74.7 0.22 5457 +69.7 0.25 26410 +44.3 0.47 2008 +27.8 0.70 1161 +24.9 0.76 1134 +18.3 0.89 855 +18.0 0.90 1226 +14.9 0.97 186

Calculation of the remaining fraction of SO,2" (f) and of the SO2~Initial content

(Single Raylelgh reduction process, assuming:

- epsllon 3 4S - 40 °/oo = delta 34S(So|~) - delta 34S(S2~)

- delta 0 - delta vl - +13.8 °/oo)

The time contact Is approximately 19,000a. This surprising age(since the Fennoscandian peninsula was covered by ice at thattime) is, however, probably irrelevant.

As a working hypothesis, it is proposed that, more than time,the microenvironmental pH Is the main controlling factor in thatsystem. Through processes of oxidation-reduction the microenvl-ronment of the reaction has become highly acid during oxidationsteps:

H2S + 3/2 0 2 - SO2." + 2 H +

or

H2S + 3/2 0 2 • HSOj + H+

Before the drop in pH is buffered through release of cationsfrom the matrix, the rate of the lsotopic reaction between SO2."or HSOX and water is greatly increased. Furthermore, recent dataobtained on very old *4C groundwaters from Saharian "ContinentalIntercalalre" (the world's largest confined aquifer) suggestthat isotcpic equilibrium between SO2" and H20 within neutralrange of pH, is probably longer to reach than indicated from theextrapolation of Lloyd's experimental data (Fontes and Guendouz,

7:17

1984). As discussed before the So|~ - 1^0 "chronometer" should be

carefully reassessed before any attempt of age estimation.

In conclusion, it appears that:

- redox processes may play a major role in the control of heavyisotope contents of the regional aqueous SO* and any age esti-

1 f t

mation from the 0 content of the sulphate is probably risky.

- Stripa deep groundwaters from V 1 could be representative ofthe initial isotope and S0^~ contents of a regional supply ofsulphate of sedimentary origin.

The conclusions of the stable isotopes of sulphate are not inagreement with those of the hydrochemistry. The necessary recon-ciliation of all of the data requires further research includingefforts to find the possible source materials for sulphate eit-her locally (from rocks along the groundwater flow path) or re-gionally. This research is being performed in the ongoing PhaseII investigations.

8:1

THE CARBON AND OXYGEN ISOTOPIC COMPOSITIONS OF AQUEOUS CARBONATEAND CALCITES

8.1 Introduction

Within the Stripa project a large number of ^C, * C and ^ 0 ana-lyses were carried out on aqueous and solid carbonates. The ana-lytical work, was initially done within the LBL-KBS programme bythe University of Waterloo (UW) and the International AtomicEnergy Agency (IAEA). Dr. J.F. Barker, Dr. J. Gale and Mr. D.Reimer participated in this phase (Fritz, et al., 1979). Subse-quent analyses were done by UW, the University of Paris-Sud (UPS)and the University of Bern (UB).

The following discussion is based on all data available and ex-plores first the ^C contents of aqueous carbon (TIC). There-after, the C abundances in different water are discussed withthe aim to evaluate this tool for the determination of residencetimes. The last section deals with the isotopic composition offracture calcites and their role as indicators of hydrologic/geo-chemical processes.

8. 2 Sampling and analyses

8. 2.1 Aqueous, £a_rbonate

Samples for ^C and * C measurements were collected either byprecipitation of all aqueous carbonate (TIC) with BaClj x 2 HjOor were shipped in aqueous form to the laboratory and extractedby acidification with phosphoric acid. Sampling methods are indi-cated in Table 8-1.

Precipitation with BaCl2*2 H2O is a standard procedure whichproduces very reproducible results for both C and C determi-nation. However, the low alkalinities of the deep Stripa watersproved to be a challenge and a number of different arrangementswere used. For example, a number of 60 1 bottles were filled inseries to minimize air contamination, and precipitation had tobe done with fresh, decarbonated reagents. After a precipita-tion, samples were shipped as alkaline slurry to the laboratori-es or washed with boiled, distilled water before filtration. Thelowest C contents were measured by the latter method, which,therefore appears to be most reliable. For samples collected in

8:2

a flow-through shipping system developed by UW for the LBL/KBSproject phase as well as most other samples which relied onlarge volumes some air contamination may have occurred.

To overcome this, three samples were collected by UPS acceleratormeasurements through the UB. The results are encouraging, al-though it appears that again air contamination cannot be excluded»This is especially true for samples which were shipped as waterto UPS and extracted there.

Samples for C - T I C analyses (usually 500 ml 1^0) based on acidextraction were poisoned with HgClj before shipment. Results areexpressed as permille differences (6 °/oo values) from the PDBstandard. Analytical precision is better than +0.15 °/oo and over-all reproducibility is better than about +0.5 °/oo.

8.2.2 Fracture calcite

Fracture calcites were collected from borehole fractures andfracture surfaces exposed in the excavation. Since most sampleswere very small, no chemical or mineralogical analyses were at-tempted. Isotope analyses were done via a reaction of the calci-te with 100% H3PO4 at 25°C and results are expressed as permilledifferences from the PDB standard for 0 and C. However, forthe determination of isotope fractionation factors e t c it is ne-

1 ftcessary to use 6 l o0-SM0W (Standard Mean Ocean Water) values

which are obtained using the following conversion formula

<518O SMOW = 1.03086 6 1 8O PDB + 30.86

8. 3 Aqueous carbon

The carbon isotopic composition of the aqueous carbon is a di-rect reflection of the geochemical history of a groundwater.This evolution begins in the recharge environment and continuesin the subsurface where mineral-water interaction will dominate.

1 -i

The possible expected changes in C contents are summarized inFigure 8-1. It shows that in most recharge environments the upta-ke of isotopically depleted 80H-CO2 produces 6^C values in theaqueous carbon which are close to -20 °/oo, more positive valuesare found in high pH environments, lower values where low pH va-lues dominate. The subsequent dissolution of carbonate mineralswill usually cause an enrichment in C and values as high as 0°/oo can be reached if the incongruent dissolution of dolomitesoccurs. Biological processes can further modify the isotopic com-position of the aqueous carbon and generate 6 C values as highas +20 °/oo if methane production occurs or very negative valueswhere organic compounds are oxidized.

8:3

The results of C and C analyses on aqueous carbonate in theStripa groundwaters are summarized in Table 8-1. They show rela-tive uniformity which a first indication that the carbon geoche-mistry of these different waters has a similar history. Further-more, all 613C-TIC values are below -10 °/oo, which is a strongsuggestion (see Figure 8-1) that biogenic processes are impor-tant.

8.3.1 _Shal_l ow g romid wafers

Geochemically least evolved are watertable wells in which thecarbon geochemistry is largely determined by the PCO2 of thesoil-zone and the pH of the water. The lowest C contents arefound in watertable well WT2 which has a 613C = -23.2 °/oo PDBand a pH = 5.1. Assuming open system equilibration the soil-CO-would be close to -23.0. Chemical data agree with this since thefield measured alkalinity corresponds to 2.9 ppm HCO, and the wa-ter is saturated with atmospheric oxygen. The water in this wellrepresents most closely the recharge condition for present day(and past?) Stripa groundwaters.

Calcite dissolution probably occurs in the shallow groundwatersat Stripa yet only deeper waters are saturated with respect tocalcite. This is graphically shown in Figure 8-2; saturation in-dices for calcite were calculate with the WATEQF speciation pro-gramme by Plummer, et al. (1976). This figure is from Fritz e_£al, (1979), and a more complete picture is seen in Figure 5-7.

The uptake of calcium carbonate carbon in the shallow groundwa-ters is in agreement with the observed isotope data if, as sugges-ted above, the open system uptake of soil-CX^ does occur and ifsubsequently a closed system dissolution of calcite occurs withcalcite - 6*3C below about -5 °/oo.

The required, non-marine calcite values are encountered in thefracture calcites of the Stripa rocks. Watertable well WT3 alreadyhas 33.1 ppm HCO3, a pH • 6.6 and a 613C - -21.8, all of whichsuggest that carbonate dissolution has already taken place.

Further evolved are the waters in the private wells, wherebyespecially the deeper ones reach calcite saturation. This is ac-hieved through the dissolution of carbonate minerals. However,the 6I3C values of PW 2, 3 and 4 are amongst the highest of allsamples measured and model calculations (Fritz, et al. , 1979)show that they must have dissolved a calcite with higher 6 C va-lues than found on the fractures of the crystalline rocks. Thiscould be represented by a local marble which has a 61JC • -1.4°/oo PDB. PW 1 and 5 are sunk into granitic rocks and theirwater did not see carbonate with such high 6 C values.

8:4

T a b l e $~"1 The carbon Itotoptc composition of aqueous Inorganic carbon tn Serf pa grnundwaters.

Location

Tai l ing pond

Watertable wellsUT-2WT-3

I r i v a t e wellaPU 1

PW 2

PW 3PW 4PW 5

SBII-3

M3

Rl

1(9

N-l hor i r .

E-l

V- l

V-2

V-2-3

V-2-4

Interval (

89-104

~ 10

- 30

- 30

252-300151-251120-15010-119

openhole

10-550

6-508-40

below 50below 280below 380332-359400-4 28

400-4 28

0-4 28

424-499

562-822500-561424-490382-4 23

n) Date

77-09-12

79-05-05/1879-05-01

77-09-2779-05-1877-10-24/2777-10-24/2779-05-1679-05-17/2477-10- 6/1379-05-15

79-05-22/28

77-09-09/2177-11-15/2379-05-02/1079-11-11/1279-11-20/2284-03

78-08-09/2678-11-17/2479-05-02/0879-05-09

79-05-22

84-02-14

83-84-

84-02-14

84-02-14

77-09-09/2078-06-0878-11-16/2079-09-11/1377-09- 8/1477-09-14/2078-06-12/2479-01-31 to 03-1678-11-2078-11-22 to 12-2079-04- 6/2779-05- 7/1179-05-18/2177-10-24/26

83-11-28

84-02-1584-02-1584-02-1584-02-15

« I 3C TIC

-23 .2

-23 .2-21.8

-15 .2-14.0

-15.2-13.4-13.7-19.1-22.3

-15.6

-13.2-16.8-16.2-15.9-16.0-15.9(-16.7)d )

-17.1-16.4-16.1

-15.5

( -17.3) d )

- I 5 .9 ( - I7 .1 )(-13.0)(-14.0)

(-17.6)

(-23.4)

-16.1-16.6-16.5-15.9-18.3-18.6-18.7

-16.9

-14.0-14.4-16.1

(-19.6)(-30.8)(-28.3)(-16.9)

-15-14

-14-18

-15

-15

-16-16-15

-16-19-17

-23-18

-15-15

-16-16-17-17-18

-13-13-13

-15-18

-35

C-BaCO,

.0

.9

.1

.1

.2

.8

.3

.9

.8

.2

.0

.7

.1+2.4. 7+3 .2

.5

.7

.9

.9

.8

.5

. 2

.3

.3

.6

.9

.5

.6+7.8

I 4 C pmC

53.8h>54 . l c >

52.1')89.3b>

6:.RC>

2.5b>3.5C'6.'.c '2.S<:'3.3C'

4.5C>

6.6C)

7.0c '

3.95+3.B3*'16.06+O.33('

6 .0«2. l c '2.nc>

2 . 8 C '

I 3 . 6 c )

I 9 . 4 C >7.8C>5.5 C >

5 . l r '4 . 7 C >

1 4 . 6 7 + 0 . 7 5 f )

a The ä " c , TIC (Total Inorganic Carbon) values refer to samples collected "n.l m l<l extruded spe-c i f i c a l l y for C analysis, whereas the t C - laCOj values were obtained mi h.irlnm c.nrlinnateprecipitated fro» large volunes for C analyse». The ä C - T IC, values ari' . onsldrrufl to be•ore r e l i a b l e .

b C-B«CC>3 and C analyses done at In te rn . Aton. Energy Agency, Vienna, Aus t r l i .

c "c-BsCOj and " c analyses done at Univ. of Waterloo, Waterloo, Canada.

d Samples were shipped to UW In plast ic bott les. Al l other TIC fanples were r n i l c r t i . l In glassbott les.

e Analyses done by accelerator •easureaicnts by the University of Bern on bsrltm-cirhmiatc.

f Analyses done by accelerator acasurasent by the University of Bern on acld-cxi r n 1. I Ci),.

8:5

!«C

• 10

-10-

-20-

-30-

SOIL-ZONE

MARINE J

CARBON Art

—•"iK—^_ 1coi \ »?re M

OPENSYSTEM

INORGANIC CARBONATE

CO} from

CH.pmlX

MOONQRUENTDISSOLUTION j r -

OF DOLOMITE *."

Of'' /^k

/ > r DISSOLUTION I

i r OPQANIC/

MATTER* V

CLOSED

BKXOQIC PROCESSES .

[

OXIDATION OF CH4 1

SYSTEM

Figure 8-1 Possible range of 6 C values in the aqueous carbo-nate of groundwaters. Dark areas indicate expected6^C-TIC values whereas shaded areas represent so-urce terms.

-t

1-I

• 4

-

^ " * •

" ' ' 1 '

/-C*l£IT(r MTiauriM

- 4 - > . » .1 0

fATUMTSN WDEX -CAU3TE

Figure 8-2. Total inorganic carbonate and calcite saturationIn Stripa groundwaters. From Fritz, etal. (1979).

8.3.2 Deep groundwaters

All deep groundwaters collected at various mine levels and diffe-rent boreholes are slightly supersaturated with respect to calci-te. As indicated in Figure 8-2, this supersaturation appears tobe paralleled by a loss of inorganic, aqueous carbon (TIC) whichsuggests that calcite precipitation is occurring. The precipita-tion of calcite will preferentially remove some * C but the Iso-tope effects are small and significant shifts towards lower 61 Cvalues can only be expected if more than 70% of the TIC is remo-ved as calcite (Figure 8-3). No actual calculations can be per-

8:6

CO

,-10-

-20-

RESIDUAL FRACTION0.8 0.6 0.4 0.2

• — •

A -

12

^ 1 0

a eo

6

5*V

2

N

5 0

».

( f

\\ \\1 6 \

3o\\

6 0

.

' " 1 | X

V\

7.0pH

^ a • 1 0025

lOO^V.

\

•

83 B.0

0.8 0.6 0.4 0.2

Figure R-3. Isotope effects caused by precipitation of calciumcarbonate from solution. Since the enrichment fac-

I O

tors (a) cause preferential removal of C in thesolid phase the residual fraction of the aqueouscarbon (f) becomes depleted in this isotope.

formed because nothing is known about initial TIC contents inthe deep waters. However, the more saline waters have considera-bly less TIC than the shallow, fresh groundwaters and If thisdifference is applicable to the evolution of the deep jroundwa-

ters then 6

tatlon.

13C values may have been affected by calcite precipi-

1 "K

The similarity of 6iJC-values In deep and shallow waters is re-markable and suggests a) that either no carbon is added to thewaters following an initial evolution similar to what was seenin the shallow wells in crystalline rocks or b) that carbonadded has an isotopic composition similar to the TIC of these wa-ters.

A case could be made for both situations and, indeed one mayhave to assume that in some samples only a loss had occurred where-as others have seen the addition (followed by loss?) of car-bon. The latter is recognized in groundwaters which dischargetoday from V-2 at depth between 424 and 560 m. They display verylow 6 nC values of -35.6, -30.8 and -28.3 °/oo (Table 8-1).These values are probably not due to the loss of carbon by carbo-nate precipitation but may reflect the presence of CO2 which wasgenerated by oxidation of organic matter or methane. A characte-ristic feature of many Stripa calcites are low 61 C-values andsuch CO2 may well play an important role, both for C and 1 Cconcentrations as will be shown below.

8:7

Based on the information available to date, it appears safe toconclude that the carbon in the aqueous carbonate of the Stripagroundwaters has largely a biogenic origin. The agreement betwe-en deep and shallow waters could suggest similar evolutionarypath at least as far as the carbon geochemistry of these watersis concerned. The addition of marble or marine carbonate is notfound to be an important factor.

8.4 Carbon-14 measurements

The basic concept underlying the carbon-14 dating requires thatwaters infiltrating through vegetated soils become charged withsoil-CC>2 before they become part of a groundwater reservoir. Be-cause this soil-CC>2 has a partial pressure up to two orders ofmagnitude higher than the atmospheric CO2, it dominates the car-bon isotope content of infiltrating water. Its C activity isvery close to the C activity of the atmosphere. If no othercarbon were added, and only decay altered the C contents ofthe dissolved carbonate, this residual activity would be a func-tion of time only, where

Pleasured Ainltial ' e

with A = C activity in pmC and A - decay constant.

Unfortunately, most groundwaters get their aqueous carbonate notonly from the soil reservoir, but also from the aquifer carbonat-es and other sources. These are normally free of C and theircarbon will therefore "dilute" the **C contents of the Initialsoil carbon. The measured ages then become too old.

Many attempts have been made to quantify this geochemical dilu-

tion and to establish correction factors which would permit re-

calculation of the water ages. Three approaches have been taken:

- a statistical approach:

- chemical analyses to assess the amount of rock carbonate disso-lution and precipitation;

- C is used as an indicator for dead carbon contributions fromisotopically distinct sources.

The statistical approach does not attempt to understand the geo-chemical processes which control the carbon geochemistry of agroundwater. For crystalline terrains, Geyh (1972) proposed "cor-rection factors" which reduce the Initial activity by 0 to 20percent. The numerical value of such "q" factors thus varies bet-ween 1 and 0.8 and appears in the decay equation as

8:8

A^ = qA6e-At.

This approach is not very satisfactory primarily because the q-factors were developed only on data from water samples which con-tained tritium and, therefore, any further geochemical dilutionoccurring in waters older than 25 to 30 years could not be takeninto account.

The C contents of the aqueous carbon of a groundwater is strong-ly dependent on the carbon geochemistry of the system. The simp-lest "chemical" correction factor is derived from a comparison ofinitial versus final carbonate content. Such comparison implies(1) no carbon loss through mineral precipitation and (2) thatisotope exchange/loss with the rock matrix was not important.Both processes do, however, take place, and more sophisticatedmodels based on a total assessment of the geochemistry of a

groundwater and its evolution have been developed. These models1

combine geochemical considerations with observed C variations

(Reardon and Fritz 1978; Wigley et al., 1978; Fontes and Garni-er, 1979).

If the 6 C-TIC is used as an indication for the uptake of inor-ganic carbon, then a carbon isotope mass balance in a two compo-nent system will lead to the following relationship to obtain anapproximate correction factor q:

613C-TIC - 6l3C CARB

613C-soil - e - 613C CARB

613C-TIC is the measured 613C value of the sample, 613C-CARB isthe value for the dissolving rock-carbonate, 6 C-soil is theisotopic composition of the S0H-CO2 and e is the pH dependentisotopic difference between soil-C09 and inorganic carbon TICunder open system equilibrium conditions. This value will beclose to 0 °/oo for recharge environments with pH close to 5 -such as found at Stripa but becomes important as the pH in therecharge area rises and the system remain open to equilibrationwith soil-C02« Where recharge conditions cannot be estimated, eis usually neglected.

The final choice of how to transform C, ages obtained on aqueouscarbonate into water ages depends on the amount of informationavailable. For the Stripa project, samples were collected fromprivate wells in the area as well as from all flowing wells andboreholes in the underground working area, in an attempt to un-derstand as completely as possible the geochemistry and isotopicevolution of the Stripa groundwaters.

Finally, it is again necessary to address the concept of water"age" since it might appear that here it is used to describe the"age" of a single water mass. This is not so, since at best onecould talk about "mean residence times" of a groundwater. How-

8:9

1000

60 80

30000 26.000 20,000 15,000 laOOO 6000

UNCORBECTED ' " C A Q E '

Figure 8-4. Radiocarbon measurements on Stripa groundwaters.The data do not include the accelerator measure-ment given in Table 8-1.

ever, even this term may be misleading because as has been men-tioned in Section 6 of this report, it is quite likely that manyof the waters collected in this study represent mixtures of twoor more components. For example, a saline old component may havebeen added to less saline, younger groundwaters and the radiocar-bon measurement would reflect the average of the two carbon sour-ces. Thus, it is suggested that absolute age determinations arenot possible for the Stripa groundwaters and, instead, such ana-lyses were used to assess the relative "age" of the differentaqueous geochemical systems in the Stripa granite.

All n,C and C results obtained on aqueous carbonates are lis-ted in Table 8-1. Four of these samples come from shallow priva-te wells from farms in the vicinity of the mine, and the remain-der from flowing boreholes at the test site and from the flowingwell V-2 at the 410-m level. The ^C-ages for all or much oftheir water should be modern, i.e., have a C activity corre-sponding to at least 100 pmC. It is unlikely that in the shallowwells an old, low I4C water mixed with modern water and one canassume that the lower activity levels document a significant di-lution of initial activity as a result of geochemical reactions.

8.4.1 jShalJU>w J rotmdvratejrs

The dilution of initial activities has been evaluated on thebasis of chemical and isotope analyses, using model calculationswhere the measured carbon isotopic composition can be comparedwith a theoretical composition calculated for assumed initial

8:10

conditions and different evolutionary paths (Reardon and Fritz,1978).

The principal conclusions of these calculations are summarizedas follows (Fritz, et al., 1979).

- soil C02 with a 613C = -23 °/oo has equilibrated with the in-

filtrating water at pH = 5,

- the ^C activity of the soil C02 was between 100 and 130 pmC,

and

C-free rock carbonate with an average 6 C value below about-5 °/oo dissolved under closed system conditions. As will beshown below, most fracture calcites analyzed in this study fitthis requirement and the only exception is a sample of marblefrom the Stripa area (SBH-3) which had a 613C = -1.4 °/oo.

These assumptions are not unreasonable: The low pH implies a p 2of 10 atm which is normal for soils in these environmentswhere, under moss covers, pCOp values can be even higher. TheC activities will be above 100 pmC for very young waters and

at about 100 pmC for the somewhat older shallow systems. The sur-face environments are usually free of carbonate and most dissol-ution must occur in closed systems within the fractures. This ex-plains why a C dilution exists: if this dissolution took placein contact with the soil atmosphere then the C activities ofthe aqueous carbonate would remain close to modern even if thecarbonate were free of C. The calculation suggests that the6IJC values of this carbonate should be more negative than the-1.4 °/oo measured on the marble and that fracture calcites depo-sited much earlier are now partially redissolved by infiltratinggroundwaters. Such dissolution must have taken place whereverthe shallow waters are in contact with carbonate, since most ofthe samples of shallow waters are undersaturated with respect tocalcite.

The dilution factors (q) derived for the shallower samples varyfrom about 0.7 to 0.5. The ultimate correction factor for deeperwaters could be larger, since these shallow waters have not reac-hed calcite saturation, and will continue to dissolve calcium car-bonate. Exchange with carbonate minerals and diffusive loss intothe rock matrix will be of no importance in these shallowsystems.

8.4.2 Deep groundwaters

The total carbonate contents of the groundwaters decrease withdepth and one could assume that the "evolution" from shallow todeep groundwater involves calcite precipitation. This is possib-le because the pH and Ca concentrations of the waters increases,

and thus maintain calcite saturation. Hydrolysis and dissolutionof silicate minerals are probably responsible for this pH increa-se. If calcite dissolution were paralleled by precipitation,then the observed chemistries could also be explained by closedsystem dissolution.

The removal of calcium carbonate from solution has little effecton the ^C activities, because the C isotope effects are onlytwice those known for ^C. Thus a 2 °/oo decrease in C corres-ponds to a 4 °/oo or 0.4 percent decrease in C activity. Thischange is within analytical error. One exception could the in-congruent dissolution of dolomite. However, since this mineralhas not been found this possibility can be disregarded. Thus,in correcting measured C ages, only dilution occurring duringcarbonate uptake until calcite saturation is reached has to betaken into account. Applying this correction to the results pre-sented in Table 8-3, the waters discharging at the 330-m levelwould have an mean residence time between 23,000 and 25,000years. Samples from the shallow part of the deeper V-2 hole tendto be slightly younger apparently!

In these deep and possibly old waters the possibility of C ex-change between the fracture calcites and the aqueous carbon willalso have to be considered, because other studies have shownthat it could be important - at least in systems where fine grain-ed carbonate is exposed to migrating groundwater (Mozeto etal., 1984). There, however, the aqueous carbon approaches isoto-pic equilibrium with the carbonate minerals. We have no indica-tion that thiri is the case at Stripa.

Neretnieks (1980) showed that diffusive loss of radiocarbon intothe rock matrix can become very significant, depending on fracturespacing and width. For example, for a ? m fracture spacing and afracture opening of 0.1 mm the ratio of actual to measured waterages can become as high as 100. However, the most important frac-ture systems at Stripa appear to have larger openings and smal-ler spacing and, therefore it is felt that at least the deep, sa-line waters, are little affected by this process.

The question also arose as to whether the subsurface productionof C might be possible in an environment locally enriched inuranium bearing minerals. An evaluation by researchers at theUniversity of Arizona (7ito, et al., 1980) concluded that thismay well be possible, although the amounts involved would probab-ly not exceed the equivalent of a few pmC. Further work in ura-nium bearing fracture systems is, however, warranted because thetheoretical data must be supported by field evidence before theybecome acceptable.

The most important problem in this study was the collection ofuncontaminated samples. Because of the low carbon contents up toseveral thousand litres of water had to be processed which cau-

8:12

sed considerable sampling problems. The high variability of the14C activities listed in Table 8-1 for the V-2 borehole attestto this. However, a continuous stripping device was constructedand directly connected to the flowing borehole» CC>2 was evolvedby continuous acidification and flushing with purified N£. Thegas was absorbed in a 5N NaOH solution. Two samples from the dee-pest portion of V-2 were collected and measured 7.8 and 5.5 pmC.Assuming that both still show minor contamination with atmosphe-ric CO-, then the mean residence time of these waters would begreater than about 25,000 years. This assumes that geochemicalcorrections based on carbonate dissolution-precipitation, as des-cribed earlier, can be applied.

The new accelerator techniques should be able to overcome suchproblems. Table 8-1 list three results which show, however, thatproblems may persist even if much smaller samples are collected:N-l was sampled twice and one results is comparable to the lowerC contents measured by conventional techniques whereas the se-

cond yielded much higher radiocarbon concentrations. It is notcertain whether some contamination occurred or some young water com-ponents are present induced by a changing flow pattern. The wellshould be resampled. However, the measurement for V-2 is alsomuch higher than conventional determinations and provides astrong indication that contamination had occurred.

8.5 Fracture calcites

Fracture mineral studies using environmental isotope analyseswere, to date, limited to studies of fracture calcltes, but willbe expanded in the future. Calcites were collected from cores of anumber of boreholes but no attempts were yet made to classifythe calcites and associated minerals. All results are listed inTable 8-2. The oxygen and carbon isotopic composition of calcitewhich form in isotopic equilibrium is determined by the 0 inthe water, the C of the aqueous carbon and the temperature. Fi-gure 8-5 compares measured 0 and C data whereas Figure 8-6and 8-7 show equilibrium fractionation effects for 6 and Crespectively.

Calcium carbonate occurs on many fractures in the Stripa gra-nite, usually in association with quartz, epidote, chlorite andmica, (Carlsson and Olsson, 1983). These assemblages are thusdifferent from those investigated by Tullborg and Larson (1982)in the crystalline rocks at the Finnsjön test sites or the Gideåtest site which was studied by Tullborg and Larson (1983).There, calcite is often found in association with prehnite andlaumontite. For comparison, Tables 8-3 and 8-4 show electronmic-roprobe analyses of the dominant fracture mineral at Stripa andFinnsjön respectively, which document that, indeed, the mineralo-gical composition of the fractures fills analyzed are different

8:13

T a b l e 8 — 2 Carbon and oxygen l totoplc compositions of fracture ca l c l t e s froa Strlpa boreholes.

Fracture c a l c l t *

Borehole Depth (m) 613C PDB «180 PDB

SBH-1

SBH-2

0V2 BHH1BHH3BHH6

Back w a l l oftiiae-icaleroom 14C:below detec-tion Unit

VI

V2

SBH-3

106.43107.85124.03132.31152.90

174.67178.35189.95206.63247.11278.40

305.7306.20

9.479.4392.4399.43

103.82104.27

9.546.641.64

493495495496497

316317318

18242465

.69

.75

.61

.36

.09

.2

.8

-10.7

- 7.0

- 4.5

- 9.0

-10.1

-12.4

- 6.H

-13.2

-11.3

-11.6

- 8.3

- 4.1

- 3.9

-15.2

- 3.8

- 1.4 (marble)

-13.5

- 6.5

-11.t

+15.3

+13.0

- 9.6

- 9.7

- 6.4

-11.1

-11.5

- 9.5

-10.3

-23.7

-21.7

-12.7

- 9.8

- 9.8

-15.8

- 4 . 2

8.416.2

35.7

16.8

14.3

18.9

17.1

14.3

4.84.24.74.4

-20.9

-17.4

-13.4

-19.3

-17.4

-18.2

-17.9

-16.8

-22.5

-20.5

-21.5

-23.3

unless the Ca-mica analysed by Reimer (1980) are In reality preh-nite. Unfortunately, however, no efforts have yet been undertak-en to study the Strlpa fractures with the same techniques as ap-plied at Finnsjön. This Is a shortcoming which will be correctedin future studies*

If present groundwaters in the Stripa granite were to depositcalclte then the temperature of deposition would be between 5and 15°C. Using this range and the fractlonatlon effects shown inFigure 8-6 and 8-7, it is possible to calculate the isotoplc com-position of calcltes which coi.ild be deposited. This range is in-dicated in Figure 8-5. It is noteworthy that this range does notcomprise any samples below 300 m which could suggest that thedeep waters at present do not precipitate any significant amountof calcite.

8:14

tsi

Figure 8-5.

<D

C i•«o -a

t "c %.(PDII

The oxygen and carbon isotopic composition of cal-cites from the Stripa granite.

o aa. m0 r 30-

-10

-20

20

10

-3 39

O 100 200

TEMPERATURE CO

Figure 8-6. Equilibrium isotope effects for * 0 in the calclte18,water system. The histogram show 6 0 values for

l ft 1ftStripa calcites. The conversion 6lö0 PDB to 6IO0SMOW is by the following equation:

6Ib0 SMOW = 1.03086 618O PDB + 30.86.

In total 3 groups of calcite appear to exist:

Group I would be modern calcites in possible isotopic equili-brium with the water.

8:15

•K)BO

-30

1000 Ina I M 9

-i-— .

"20-

-25.

30

0 100 200TEMPERATURE C O

Figure R-7. Equilibrium carbon isotope effects in the CO2 -

CaCOo system. The histogram shows 613C values of

Stripa calcites with indications to which group

they might belong.

1 Group II has the same range of 6 C values as seen in Group I

but distinctly lower ' 0 values and

Group III comprises samples with relatively high 613C but very

low 6lfi0 values.

The "modern" calcites (group I, Figure 8-5) are isotopically com-

parable to many of the "open fissure" c.alcites noted by Tullborg

and Larson (1982). This is especially true if only 0 where con-

sidered which has a much narrower range than C. Figure 8-8 is

a comparison of Stripa, Finnsjön and Gideå data which documents

the similarities and differences. This figure shows that the lat-

ter prevail.

Tullborg and Larson (1982) recognized for Finnsjön at least

thrie calcite generations on the basis of fluid inclusion data,

mineralogical assemblages and oxygen isotopic compositions.

Their group I has 6^0 values above -10 °/oo (PDB) and shows

also the highest 613C values. Both open and sealed fissures

occur and their genesis .can possibly be linked to the participa-

tion of seawater. This group is only recognized at Finnsjön and

does not appear at Stripa or Gideå. Finnsjön Group II appear to

be a mixture of "modern" and hydrotherraal calcites whereas group

III at Finnsjön represents hydrothermal calcites collected below

300 m depth. These latter calcites are associated with prehnite

and laumontite.

8:16

T a b l e 8~3 Mlcroprobe analyses on Strip» fr«ctur* nlnerals.

CHLORITES

SAMPLE

Slant I

Slant 1

Slant 1

V-2

V-2

V-2

STRIPA

MICAS -

SAMPLE

SERICITE

Slant 1

Slant 1

V-2

V-2

A

B

A

B

C

D

A

B

C

AB

C

D

E

A

A

BCDEF

II

A

AB

A

A

B

DEPTH

(«)

19.7

16.9

27 .8

19.1

14.65

409.9

DEPTH

<m)

31.7

19.7

355.7

14.65

sio2

24.4

28.4

21.4

24.7

23.923.9

24.6

25.637.5

23.0

22.6

25.726.731.7

31.4

26.A27.928.329.128.930.0

46.4

44.1

42.8

46.6

37.043.2

AljO3

19.9

18.A

17.8

20.7

20.2

20.6

19.519.719.2

19.617.821.921.619.9

20.5

19.317.418.018.718.22 0 . 1

A12O3

26.4

26.624.8

26.1

24.8

25.A

COMPOSITION AS

FeO

35.828.9

35.237.136.236.8

37.935.7

26.9

37.234.9

34.2

32 .925.1

21.6

20.117.717.7

18.116 .91 7 . 1

MgO

7 . 1

5.2

4 . 7

6 . 25.0

6.1

5.86.54 . 4

4 . 4

6 . 9

5.63 . 6

8 . 1

5 . 4

17.821.321.520.221.615.7

COMPOSITION AS

FeO

4 . 9

3.311.0

6.0

18.68.2

HgO

2.8

I.I3.1

2.2

3.72.8

WT. X OXIDE

CaO

nd

t r

nd

nd

nd

nd

nd

nd

nd

ndnd

nd

nd

tr

5.1

ndnd

nd

nd

nd

nd

WT. X OXIDE

CaO

1.5

ndtr

3.2

nd5.9

Na2O

nd

nd

nd

ndnd

nd

nd

nd

3 . 3

nd

nd

nd

ndnd

nd

ndnd

trtrtrtr

N«2O

nd

nd

nd

nd

ndnd

K2O

nd1.2

t r

nd

t r

nd

t r

trt r

nd

nd

t r1.52 . 1

1.8

trnd

nd

t r

nd1.7

K2O

8.7

8.87.0

7.8

5.9

6 . 1

TOTAL

8 7 . 282 .5

79.1

88.7

85.387.4

A7.887.5

91.3

84.282.2

8 7 . '

86.386.9

85.8

84.084.385.586.185.684.6

TOTAL

90.7

83.988.7

91.9

90.092.0

C i-MICA

V-2

V-2

V-2

V-2

Fro*:

ABCD

A

A

AB

Rtlmcr,

355.

404.

14.

9.

1980

7

2

65

55

In

38.743.036.237.4

35.2

40.4

43.843.4

Fritz et

23.020.522.122.9

24.2

19.7

18.627.2

• 1 . , 1980.

11.69.2

10.9

10.5

10.2

16.0

11.67.5

5.t r1.

1.

7.

2.

6.t r

1

83

6

9

3

in.820.117.320.0

11.9

10.2

4 .717.5

ndndndnd

t r

nd

nd1.1

1.7t rl.nt r

t r

t r

1.3t r

90.992.A89.3

92.1

89.1

89.2

86.591.7

8:17

T a b l e 8~4 Mlcroprobe analyses of mineral». (Contents are given In weight-!).

Laumontlte

F15:264.4 a 0.53

0.51

Na20

0.23

0.34

C O

11.1

11.3

MgO

0.1

0.1

21.8

22.2

SlOj

52.1

50.4

Prthntte

F15:264.4 a

F17:335.I m

F17:379.5 n

Calclte

F17:41.1 m

T10, C O MgO MnO Al,0

(Prismatic)

F17:335.1

F17:494.2

F17.-519.I

(Prismatic)

2U3 FeO S1O,

--

-

o.tn.i-

BaO

_

-

n.i-

0.2--0.2-

-_

27.327.326.526.427.727.8

CaO

54.554.755.556.055.855.B55.655.955.856.156.456.256.2

0.20.10.20 .20.10.2

CO 2

4 3 . 24 4 . :

42.942.844.543.84 4 . 243.343.344.443.043.042.7

--0.1--—

MnO

0.20.5O.J0.10.10.20.3-0.10.1-0.1_

23.322.623.122-921.221.9

SrO

0.1---0.10.1--0 .10.1--0 .1

1.92.71.71.83.73.1

MgO

0 . 8

1.0

0.60.70.60 . 60 . 7

0 . 4

0 . 4

0 . 6

0 . 60 . 4

0 . 4

43-643.844.844.441.841.4

FeO

0 . 40 . 4

---

-

-

--

--

--

From: Tullborg and Larson (1982).

Group II calcltes at Finnsjön are characterized by 618O valuesbetween about -10 and -16 °/oo PDB. The Stripa 1 calcttes arealso in this range, although they tend to have distinctly lower*C values than the Finnsjön calcites. A better agreement exists

for both ]80 and ] C between Stripa 1 and Gideå II. Open fractu-re calcites occur in these groups and it is safe to assume thatmost "modern" calcites are represented here. However, some samp-les at Finnsjön and Gideå come from closed fractures indicatingthat some hydrothermal calcites are also in these groups.

Stripa 2 calcites are not found at Finnsjön but GideS Group IIIsamples are similar. At the latter locality only calcites fromsealed fractures are found and it is suggested (Tullborg and Lar-son, 1983) that some of these carbonates are Precambrian. Wehave not sufficient specific fracture mineral data to discussthis in detail but the Stripa results indicate that hydrothermalcalcites are found in these groups. Considering only 0 and as-suming for the fluid 618O - 0 to +5 °/oo SMOW (typical for mag-matlc/geothermal waters) the temperature of deposition would beclose to 200°C (see Figure 8-6).

At Finnsjön, Group III has the same 0 contents as the Stripa 2samples yet much higher 613C values. The values observed in Finn-

1 1

sjön calcites agree with a hydrothermal origin where C- valuesreflect magmatic/metamorphic carbon dioxide. For the same tempe-

8:18

2

O

-2

-4o .o -6o.* -

• OPEN FRACTURES• SEALED FRACTURES

STRIPA I »UP

X/////Å Jt • • •!.•• J

r. .:1—'! 1 -\LJ

• i

• I 61

-24 B -20FINNSJON | in

ft

-16

1l 80 %.

-12n

PDB

-8

Figure 8-8. Comparison of 6100 and 51JC values of fracture cal-cites from Stripa (shaded area with extensives),GideS (G,, G-r-r and C,,,) and Finnsjön (open andfilled circles). The Finnsjön grouping is based on0 data only and is indicated on the X-axis.

rature regime, the Stripa 2 and Gideå III samples, however, arewith two exceptions for Stripa (Table 8-2) in equilibrium with aCOj which has a 6l^C close to -15 °/oo (Figure 8-7). This wouldrequire the presence of "biogenic" CO2 during their formationwhich was possibly generated by the oxidation of organic matter.

The Stripa 3 samples, are found near the surface in SBH-3 and atclose to 300 m depth in borehole SBH-1. Similar samples are notfound at Finnsjön or Gideå and their genesis presents a specialproblem. If a hydrothermal genesis is assumed for these samples,then the temperature of deposition would have been between about250 and 35O°C - assuming that the participating fluids wereslightly enriched in *80 with respect to SMOW. This in turnwould demand, that a carbon dioxide in equilibrium with the cal-cites have a 6 C = -2 °/oo. It is thought, however, that suchhigh values could only be generated if marine limestones wereavailable to provide the carbon for the hydrothermal solutions.We would suggest that methane equilibria could also play a role.Note, that these carbon isotope values are very similar to the6 C values of the Group III calcites in Finnsjön.

8:19

We suggested earlier (Fritz et al. 197*5) that it might also bepossible to consider these calcites to be glacial meltwater pre-cipitates. Very little is known about the composition of theaqueous carbon in subglacial meltwaters, although the few 6 Ccontents of subglacial carbonate precipitates from different en-vironments known to us tend to indicate that either isotopicequilibria with atmospheric CO2 are approached or that rock car-bonate has been taken up and reprecipitated (redeposited) fromthe glacial melt waters. In both cases the 6 C values are closeto 0 °/oo, and thus quite different from those observed in thepresent day groundwaters at Stripa. However, the 6 C values ofthe Group 3 samples are close to this postulated value. Further-more, the calculated 6 0 values for water participating intheir formation are below -20 °/oo SMOW. Such low values areknown from other environments where glacial meltwater infiltra-tion has occurred and would thus support a low temperature ori-gin of these calcites.

Our present data do not suffice to discuss this problem, yet itwould be important to document with fluid inclusion data whichof the two interpretations is correct. If one could show that in-deed glacial meltwaters are responsible for the formation ofthese calcites, then this observation would contain important in-formation about the depth of actively circulating flow-systemsin these environments.

An intriguing element in the Stripa calcites are the "abnormal"values which are found. A calcite from V-I (495 m) yielded a6 C = -35.7 °/oo which is close to the values measured in theaqueous carbon from the V-2 borehole. Its 6 0 » -13.4 °/oo(PDB) could reflect deposition at about 16°C from a water with618O = -13.5 °/oo SMOW. These values are close to the actual tem-perature and present day water composition, this calcite couldbe a recent precipitate.

The question on the origin of these low C values was already ad-dressed above and it was indicated that the oxidation of methaneor organic matter might be responsible. Traces of methane were re-cognized in the deep waters and calcite compositions attest tothe past presence of methane producing bacteria in these fracturesystems. Two samples from SBH-2 gave 6^C values of +13.0 and+15.3 °/oo. These values are typical for carbonates whose carbonoriginated in a CO2 generated by methane producing bacteria.Such bacteria co-produce isotopically light methane and very heavycarbon dioxide, with typical 6 C values below -60 °/oo andabove +10 °/oo respectively. Reoxidation of such methane couldeasily produce the low 6^C values observed elsewhere in thesesystems. A first carbon-13 analysis on methane from Stripa yield-ed a 6^C - -30 °/oo. This value is non-specific although itcould indicate that partial re-oxidation had occurred. (SwedishDeep Gas Project, Progress report, 1984.)

8:20

fi.6 Conclusions

The aqueous carbon in the Stripa groundwaters is dominated bybiogenic components. The first organic carbon contribution oc-curs in the recharge environments where young groundwaters equi-librate with a soil-C02 with 6

13C = 13 °/oo PDB. This CO?-up-take is followed by dissolution of carbonate minerals. This appe-ars to be largely a closed system process although locally marb-les and other carbonates may outcrop in surface and near-surfaceenvironments. Rising pH-values in deeper groundwaters lead quick-ly to calcite saturation and no further calcite dissolutiontakes place, but calcite deposition may occur.

This geochemical evolution is also recognized in the radiocarboncontents of the shallow groundwaters where C "dilution" doesoccur. The "dilution" factor" q is estimated to lie between 0.5and 0.7. Once calcite saturation is reached, decay would dominateC abundances unless exchange with exiting carbonate minerals,

diffusive loss into the rock matrix or dilution with biogeniccarbon dioxide generated by the oxidation of methane and otherorganic compounds does occur.

Analytical difficulties in these low-carbonate waters are suchthat it is not possible to come to clear conclusions. However,assuming that the lowest measured values are correct and thatdecay dominates over other C-removing or diluting processesthan the deep, saline waters would have mean residence times inexcess of 20,000 years.

The 0 and * C data of the fracture calcites at Stripa reflectthe complex geochemical history of the fluids in these rocks. Atleast four generation of calcites are found, of which two have alow temperature origin. These are the present day precipitatesand the "methane" calcites. The former occur all at depths ofless than about 300 m. The origin of the low 180 Stripa 3 calcit-es is unclear and could either represent a high temperature hy-drothermal generation or subglacial, low temperature precipitat-es. If the latter was the case, which would have to be proventhrough fluid inclusion analyses, the statements about groundwa-ter circulation under ice cover would be possible. All other cal-cites appear to have originated in thermal environments and/orare related to the genesis and metamorphic history of the gra-nite.

9:1

THE IN-SITU PRODUCTION OF RADIOISOTOPES AND THE 3H and 36C1 CON-TENTS OF THE GROUNDWATERS

9.1 The in-situ neutron flux in the Stripa granite

The estimation of the in-situ neutron flux in geological fractur-es is very important for assessing underground production ofnuclides such as H, C, Cl and Ar, some of which have po-tential for the determination of groundwater residence time.

9.1.1 tJeii tjroji _p roduct ion_ ue _t o_ _ _ U_ sj)£nt arieous_ _f is si on _and_ (a,_ n_)

The in-situ neutron flux in rock formations which are shieldedfrom cosmic ray secondary neutrons is due to the spontaneous fis-sion of ^ 3 % an(j (at n ) reactions with light nuclei, particular-ly Si, 0, Al, Mg. The latter reactions are more significant thanthe spontaneous fission neutrons and the neutron production ratedue to them has been estimated by Feige et al., (1968). Thetotal neutron production rate may be calculated from the equa-tion

P = p(0.476A [u] + 1.57[uJ + 0.7[Th]) neutrons cnf^a"1 (9.1.1)

where [u] and [Th] are the U- and Th-contents (ppm) of the rockrespectively and p is its density (g/cnr). The first term inthis equation is the U spontaneous fission neutron productionrate whilst the last two terms are the production rates due to(a, n) reactions by o-particles from natural U- and Th-serieselements respectively.

Free neutrons decay to protons by B-emisslon with a half-life of12 minutes which corresponds to a mean lifetime of 17.3 minutes.Thermal neutrons have a velocity of 2200 m/s at 25°C and in freespace such neutrons travel very large distances (>10 m) intheir mean lifetimes. In rock media thermal neutrons undergoscattering and absorption reactions within much shorter distancesso that their ultimate fate is absorption. The weighted-mean ab-sorption cross section, o , for a rock matrix may be calculatedfrom the element abundances (Nj, mole fraction) in the rock andtheir neutron absorption cross sections (OJ) using the equation:

ia - to N moles barns/g

l i i1 l 1 (9.1.2)

i- 0.602 x p l u N atoms/cm

1 i i

9:2

Exact evaluation of this summation requires a complete elementalanalysis of the rock. However, a good estimate may be made fromdata for just 17 elements (Andrews and Kay, 1982) although theseinclude some trace elements which are not often determined. Theabsorption mean free path, X for neutrons is then given by:

Xa = l/on cm (9.1.3)

and the corresponding mean lifetime, t , and time constant, X,for the absorption of thermal neutrons are given by the equa-tions :

t_ = X /22O.OO0 sTil ct

-1(9.1.A)

This rate constant controls the removal of neutrons from therock matrix by absorption reactions and the equilibrium betweenneutron production and absorption is established according tothe equation:

when t >

(9.1.5)

10~3 s

where nt is the number of neutrons present after time t in 1 cmof rock. Equilibrium is established almost immediately since Xis about 2500 s . The thermal neutron flux, <t>, is then equal tonv cm s where v is the neutron velocity in cm/s. Estimatedneutron production rates, fluxes and absorption data for theStrips granite and leptite are in Table 9-1.

T a b l e 9 - 1 kadloclaaant eontanta and ntutren production ratal for th» grantee and leptite atStrlpa.

1

Cranite

Lcptlte

U-contentppm

44.15.4'

Th-contentPp"

53.417.9»

EoN•olea barna

0.010681

0.0O8412

n-prodaction

310.03

57.2*

neut

5.881.38

ron

X

X

flux

ID"*

10"*

1 for t pp» LI and < ppa t2 for avaraga baaalt; J pp» LI, IS ppa 13 for 70X S1O2

* for 60X 8iO25 froa WoUtnbtri at a l . . 1980 (SAC 36)

9:3

9.1.2 m£a£urement£

For the experimental determination of the in-situ neutron flux,a high sensitivity boron trifluoride (BF,) counter was used. Neu-trons are detected by their interaction with the B nucleus,which produces the ionising nuclei, He and Li , in the coun-ter. Natural boron consists of two isotopes, 18.8% of B and81.2% of B. Neutron capture by B produces the compound nuc-leus, **B*, which dissociates as shown in the decay schemebelow. The quoted dissociation energies are partitioned betweenthe a-particles and the lithium recoil nuclei as indicated.

Decay scheme and energy partition for the reaction B(n,ot) Li

XIB*

Y 0.48 MeV

7Li

2.78 MeV

Particles

a + Li recoil nucleusi

Li recoil nucleusi

a + Li recoil nucleus2 2

a2

i2 recoil nucleus

Kinetic energy, MeV

2.30

1.47

0.83

2.78

1.77

1.01

Boron may be incorporated in a proportional counter as boron tri-fluoride gas and this is usually enriched in * B. The a-partic-les and lithium recoil nuclei are detected by the proportionalcounter, and give rise to pulses whose amplitudes are proportio-nal to the particle energies if these are completely stopped inthe counter gas. Any "Y-radiation background which is present caus-es very much smaller pulses than the heavily ionising a-partic-les and lithium recoil nuclei, so it is possible to discriminateagainst the Y-radiation.

9:4

1004 6

pul» taplieude

Figure 9-1. Pulse spectrum for BF-j counter 150EB Ser.No. 8408-

290 (Inset with linear count scale)

The maximum primary ionisation occurs within the counter when anct-particle and its associated recoil nucleus dissipate all theirenergy in the filling gas. This primary ionisation is reducedwhen one of the particles strikes the counter wall before givingup all of its kinetic energy. Figure 9-1 shows a pulse amplitudespectrum for a Centronics 150EB BF-j, neutron counter (1.5 m x 50mm diameter). This counter was used to measure the neutron fluxin various locations in the Stripa mine. The counter has a sensi-tivity of 242 counts/s for a neutron flux of 1 cm s and had anegligible background. A count rate of 1 count/minute above back-ground, corresponds to a neutron flux of 7 x 10 cm s whichcould therefore be determined to +5% (2a) with a 24-hour count.

Sensitivity of the BF3 counter for different neutron energies

The cross section of B for the reaction B(n,o)Li is propor-tional to 1/v where v is the neutron velocity, in the range ofneutron energies from 0.001 eV to 10 keV. The (n,a) reactionrate is proportional to the product:

aynvvN (9.1.6)

9:5

10,000 _

1,000 _

0.01 0.1 10 100

neutron energy, eV

1,000 10,000

Figure 9-2. Neutron absorption cross section of cadmium as afunction of neutron energy

where a andy ^ are the cross section (cm ) and neutronpopulation (cm~^) for neutrons of velocity v, $ is the correspon-ding neutron flux and N is the number of B nuclei present. Ifthe 1/v law applies to the neutron cross section, then:

a v

°v *(9.1.7)

where o"t and vt are a reference neutron velocity andcorresponding cross section. These may conveniently betaken as the thermal neutron velocity at 25°C (2200 m s"1) ardthe corresponding cross section. The (n, o) reaction rate isthen proportional to:

/ - <(>N (9.1.8)

where <J> i s the integrated neutron flux in the 1/v proportionali-ty range.

The counter sens i t iv i ty of 2U2 counts s ' V u n i t flux (1 cm s"1)can therefore be applied for neutron energies up to 10 keV.Above 10 keV the 10B(n,a)7Li cross section deviates from the 1/vlaw and becomes s ignif icantly l ess than the B neutron-capturecross section. The counter has an enrichment to 90? B andabove 10 keV the cross sections for the neutron reactions withthe two isotopes in the counter become comparable.

9:6

Table 9-2 ».„,„. nus Beaaureaent».

Count rat», aln"' lout ron tlua, en''»'1

Sit» Formation UnahlaUad Shl»td*d Th.rael Epltheraal Hot*»

> 10"* i 10"*

I» lath Juraaalc l lMatoa» I5.O3+O.IO I . I I + 0 . 0 2 9.5» 0.7» Sucfac» aaaauraaent.

alavatlon 1*0 m.

Ib Rath Juraaalc llaeatona ».»5 10 .15 I . U « 0 . M 19.JO 1.32 Surfar» •eaaurrarnt,V) at abnva ground,elevation 210 at

2 Sttlpa Leptlla 31.48 +0.4? 2.I9E 20.30 LSI Surface •••auraaant on

lrpt l t» outcrop, l l m -

tlan ti m

3 Strlpa Haaaatlte ara 0.5» + 0.0* - <0.41 - 2im • dapth In o pa•ody

4 Strlpa Upttt» 1.30+0.0* 1.00 + 0.0» n.21 0.6» 3f>0 • depth near

accaaa anal t

5a Strlpa Cranlta ».IB +0.52 0.711 2.3» 0.4? Cranlttc ca»arn at556 • dapth by Z ahaItIn granit» outcrop

5b Strlpa Cranlta 4.7» < 0.11 I.15E 3.M 0.79 lorahola H4 in jranltt

M i l of enarn

ia Strlpa Cranltt 4.»I +0.11 O.tiE 2.SI 0.57 Rranltlc ca»ern atIon • d»|>th, k)> kor«-hol» VI In SCU alta

tb Strlpa CranK» e.«»+O.IO 1.17 + 0.04 J.»4 0.S1 lorahola 30», (roa

accaaa drift to SCU

alta

E »atlaatad fraaj ratio ahlaldad/iaiahlaUad counta for othar altaa.

The neutron-absorption cross section of Cd is shown in Figure 9-2. The Intensity, Ix, of neutrons transmitted through a Cd absor-ber of thickness x cm, is related to the incident intensity, IQ,by the equation:

Ix - iQ e~ö N x (9.1.9)

where o is the neutron absorption cross section and N is the num-ber of absorbing nuclei per unit volume. For "thermal neutrons"of energy 0.0253 eV, only 10 of Incident neutrons can passthrough a 1 mm thick sheet. For neutron energies above 0.3 eV (1%transmission) cadmium is much less effective as a neutron absor-ber and above 1.0 eV it is effectively transparent to neutrons.The BFj, counter response when shielded with 1 mm of Cd metal ishence due largely to neutrons in the energy range 1.0 eV to 10keV. The difference between the count rates when unshielded andwhen shielded with cadmium is attributed to neutrons of energyless than 0.3 eV and most are likely to have energies close to0.025 eV.

9:7

9.1.3 Rejs ult£ f r ° m n £ u i r £ n ^ i u ™eas iir ements

The neutron fluxes In the thermal and eplthermal (1.0 eV to 10keV) ranges which were measured in various locations are repor-ted in Table 9-2.

The total measured neutron flux (thermal + epithermal) in 76 mmdiameter boreholes in the Stripa granite is A.7 x 10~4 cm" s"1

and this is in good agreement with the calculated value of 5.88x 10~4 cm"^"1 (Table 9-1). The epithermal flux is 0.8 x 10~4

cm s and the epitherraal/thermal ratio i s 0.2. The total fluxin the lepti te is 0.9 x 10 cm"^"1, and this compares wellwith the calculated flux of 1.38 x 10"4 cm"^"1. The epither-mal/thermal neutron ratio (3.3) is much higher in the lepti tethan in the granite.

The cosmic-ray secondary thermal neutron flux is dependent uponboth altitude and latitude. At Stripa, surface measurements showthat i t is about 32 x 10 cm s and the corresponding ' ep i -thermal' neutron flux is about 7% of the thermal flux.

9.2 In-situ production of 3H and 36Cl in the Stripa granite

9.2.1 2.H_p rpduct ^on

The rate of subsurface production may be directly calculatedfrom the experimentally determined neutron flux and the Li-con-tent of the granite. The isotope production equation is:

(1 - e~Xt) atoms (9.2.1)

where N is the number of radioactive nuclei with a decay cons-tant A, which are formed in 1 cm of rock by time t after thestart of the reaction. N£ is the number of target nuclei, o isthe cross section for the production reaction and $ is the neu-tron flux. As the H-production reaction attains equilibriumafter =50 years and since 1 T.U. corresponds to one atom of Hin every 10*° atoms of H, the H-content of a water-filled frac-ture, on the assumption that all the H formed is trapped by thewater, is given by:

water •'H-content - ; — T.U. ta •, 0\6.69 x 10* f (9.2.2)

where N is equilibrium number of H atoms per cm of rock matrixand f is the fracture porosity. The H-content of fracture fluidswas calculated for the measured flux of A.7 x 10 cm~*s in the

9:8

Stripa granite and for the range of observed Li-contents. The re-sults are given in Table 9-3b. The maximum possible H-contentof the fracture fluids for total trapping of the H within themis about 1.4 T.U. However, the rock matrix must also containsome H-atoms which will trap a proportion of the H. A typicalbiotite, for example, contains about 0.5% w/w of H as OH groups.For a 10% biotite content in the granite, the maximum JH label-ling would correspond to about 2.8 T.U. for the biotite hydro-gen. Exchange with flowing fracture fluids could result in theirbecoming tritiated to the same extent if the progression of flu-ids along the fracture leads to sufficient contact with biotite.However, if the modal H-content of the rock matrix is greaterthan 0.05%, as might be the case if alteration minerals are pre-sent, the H-labelling effect would be reduced.

In conclusion, it seems probable that some H-labelling of frac-ture fluids may result from in-situ H production. The extent ofthis labelling is not likely to exceed 1 - 2 T.U. and is probab-ly less than 0.5 T.U.

Table 9-3a. Lithium content of Stripa Granite (data from KNordstrom, USGS)

Borehole

VI

V2

Depthm

107A08AA5

456471760

Average Lippm

8.A4 .65.5

1.82.8

28.0

Number ofdeterminations

242

222

Table 9-3b. In-situ production of H in the Stripa granite.

H-content (T.U.) of fracture fluid for Li-content of:

Porosity

0.010.001

1

00

. 8 ppm

.01

.09

8

00

.0 ppm**

.04

.41

11

0.0.

.00

05556

ppm* 28

0.1.

ppm

1443

* The average Li-content derived from the ^He/^He ratio ofradiogenic helium.

** Average of rock analyses.

9:9

9.2.2 36C1 production

Some muon-induced spallation of Ca, K and Ar nuclei may produ-ce Cl in the near surface of the granite but these reactionsare negligible in comparison with the neutron capture reactionwith natural chloride, 35Cl(n,Y)36Cl. The number, 36N, of 36C1atoms produced in N atoms of target nuclei, Cl, after irra-diation in a neutron flux of <|> cm~2s for time t is given by:

N . lll (1 _ e r } (9#2.3)Ar

where a is the cross section for the n-capture reaction and Xfis the decay constant of 36C1 (7.1 x lO" 1^" 1)- Equilibrium be-tween Cl formation and decay is established after about 1.5Ma (about 5 half-lives of 36C1) and the ratio of 36C1 atoms tonatural chlorine atoms is then given by:

36N/(35N + 37N) = ?*. • f 33N „ ) (9'2'4)

Ar 35N + 37N

= 0.7553 06

Since the granite is much older than 1.5 Ma, all the naturalchloride which it contains must be labelled with 36C1. Substi-tution of the cross section (44 x 10 cm ) and neutron flux(3.9 x 1O~4 cm"^"1) into equation 9.2.4 yields a value of 180atoms of Cl per 101J atoms of natural chloride for the ato-mic ratio of 36C1 in the granitic matrix. The ratio for 36C1production in the leptite (flux = 0.9 x 10~* cnT^s"1) is 45atoms 36Cl/1015 atoms of natural chloride.

9.3 Tritium contents in groundwaters at Stripa

9.3.1 I_nt rodtict ion

Tritium (3H), the radioactive isotope of hydrogen with a half-life of 12.4 years, is a common tool in hydrology to distinguishbetween recent water (recharge after 1952) and older water (re-charge before 1952) or in special cases to date groundwater upto about 80 years old.

H is produced naturally mainly in the atmosphere (in minor amountsalso in the lithosphere and hydrosphere) ard anthropogenically bythe thermonuclear tests (mainly in the period 1952 - 1962) andnuclear facilities. In the atmosphere, H is produced by spalla-tion and by interaction of nucleons with nitrogen (e.g. N (n,3H) C), oxygen and argon. After oxidation to water it reachesthe earth and the groundwater by precipitation and takes part inthe water cycle.

9:10

3H content

ITU I

5000-

3000-

1000-

500 :

330

100-

50 :

30

10-

1961 O 63 '61. 65 66 »7 66 '69 70 '71 11 '73 7* 75 76 77 7» 79 'BO I I 12

Figure 9-3. Monthly means of H contents of precipitation inthe catchment area of river Rhine (after WEISS &ROETHER 1975) completed by data since 1974 fromupper Bavaria (after MOSER & RAUERT 1983).

The main source of subsurface H production is the neutron indu-ced reaction 6Li(n, H)^He, where the neutrons originate froma decay of U and Th in the rock material and following processesof (a, n) reactions on light nuclei (comp. FEIGE etal., 1968).The H production in groundwater can be neglected because theLi concentration in water is much lower than the Li content ofrock.

The •'H concentration of naturally produced tritium in the atmosp-here (as water) is in the range of several TU*.

ROETHER (1967) reconstructed the natural tritium level in Cent-'sral Europe by H measurements of old wines and found a mean annu-al value of 5.5 + 0.7 TU for the \ content of precipitation.This value increased strongly after 1952 by the thermonucleartests and reached several thousands TU in 1963. After the test-stop the H contents in the atmosphere decreased with seasonalperiodicity to several tens TU in 1983 (compare Figure 9-3).

* The concentration of environmental tritium is expressed in TU.1 TU (tritium unit)Bq 3H/1 H20.

environmental tritium is expressed in ru.3H/H - 10"18 * 3.2 pCi 3H/1 H2O

Ä 0.12

9.3.2

9:11

.£echni£ue s

The measurements of the weak 0-emitter tritium (maximum B energyIB keV) can be done after careful sample preparation by B coun-ting in liquid scintillation counters or in gas counters (e.g.EICHINGER et al. 1981) or by the mass-spectrometric determina-tion of 3He, which is formed by radioactive decay of 3H (CLARKEet al. 1976). For very low JH concentrations an enrichment of H(e.g. by partial electrolysis) before the measurement by B coun-ting is required. Characteristic data of the applied H measu-ring techniques are summarized in Table 9-4.

T a b l e 9—4 Characteristic data of applied H measuring techniques (Maturing roan shielded by7S cm thick walls of llaenlte concrete-, the gaa counter la additionally ehlelrird byAO ca lead (above) and 20 ca Iron (all aide»)).

Liquid sclnt-countingwithout/withelectrolyticenrichment

Cas countingwithout/withelectrolyticenrichment

Volui * of water sample(cm3)

Sclnttllator/Countlng gaa

Sample container

Background (cpm)

Counting efficiency (X)

Calibration factor(TU/cpm)

Detection Unit (confidencelevel - 97.5Xmeasuring time • 1000 mln)

ID/400

12/0.6

10/400

13 cm' Insta-Cel >£>(Packard Co.)

24 cm polyethylenevial

2.42

24.5

57

2 IwrPropon

2.6 1 coppercounter

0.48

76

23

1.5/0.1

9.3.2.1 Liquid scintillation counting

Samples which are expected to contain higher tritium contentsare directly measured in commercial liquid scintillation coun-ters. 10 cm of the water sample are mixed with 13 cm Insta-Gelscintillator (Packard Co.) in polyethylene bottles ana countedfor a measuring time of about 1000 minutes. With this simple me-thod tritium contents above 12 TU can be detected (<k .ionlimit with a confidence level of 97.5%).

9.3.2.2 Gas counting

For the detection of lower 3H contents, gas counting is prefer-red in contrast to liquid scintillation counting because of its

9:12

Mom vacuum line

0 51-St * * lcylinder .Pd-Asbestos

catolyst

Fritttd''filter

Mg-Furnoc*570 »C

VMattr »ample10-20 ml

Heater

Figure 9-4. Schematic drawing of high vacuum apparatus forsynthesis of propane from water samples for H ana-lysis (after WOLF etal. 1981).

higher sensitivity. For H measurement by gas counting the watersample is converted to propane (WOLF et al. 1981) via the reac-tions (9.3.1) and (9.3.2) in a vacuum apparatus (Figure 9-4).

570°CH20 + Mg H2 + MgO (9.3.1)

CH CH- (9.3.2)

The water sample (10 cm3) is evaporated from a quartz flask intoa furnace filled with magnesium turnings. There the water vapouris completely reduced to hydrogen at 57O°C. The produced hydro-gen then reacts overnight with propadiene (97-98% of the stoi-chiometric amount in relation to hydrogen) in a 60 1 flask contain-ing palladium catalyct (10% Pd on asbestos). On the followingday the synthesized propane is transferred to a 0.5 1 steel cy-linder for storage before measurement. The propane yield relativeto propadiene is 99 + 1%.

The H concentration in the propane is counted in a 2.6 1 propor-tional counter at a pressure of 2 bars. The counter is shieldedby 40 cm lead (above), 20 cm iron (all sides) and a plastic scin-tillator anticoincidence shield. The counting system is set upin the basement in a room with 75 cm thick walls made of heavyconcrete. For a measuring time of 1000 minutes a detection limitof 1.5 TU (confidence level 97.5%) is achieved.

9:13

9.3.2.3 Electrolytic enrichment

At very low JH concentrations the H content of the water sampleis enriched by partial electrolysis. The applied system consistsof 24 batch cells (IAEA type) with stainless steel anodes andmild steel cathodes. With an initial volume of 400 cnr* and a tri-tium recovery of 85% the enrichment factor is about 20. Withthis value the detection limits for liquid scintillation coun-ting and gas counting are reduced to 0.6 and 0.1 TU, respective-ly.

9.3.3

The results of H analysis of different water samples, obtainedat the GSF-Institut fiir Radiohydrometrie, together with data ofFRITZ 1983, FL0RK0WSKI 1984 and CARLSSON & OLSSON 1982, are sum-marized in Tables 9-5 and 9-6. The sampling points for thegroundwater samples from southern Sweden are shown in Figure 9-5.

The H concentrations c (TU) are calculated from the measured

count rate n, (cpm) by means of equation 9.3.3

x (n, - n ) F

c = •—r-i 2-T = 7 ( n r n o ) (9.3.3)A (nl,st-no> A 1 °

where x = -*H concentration in 3H standard (TU)

A • enrichment factor

n, ... « H standard count rate (cpm)i ,sc

n0 " background count rate (cpm)

F = calibration factor (TU/cpm)

For the 3H results from GSF, the given analytical error o'(c) isrelated to a 95% confidence level P in the case of a two-sidedproblem or to P - 97.5% in the case of a one-sided problem andis calculated by means of equation 9.3.4.

(9.3.4)

a' is related to P • 95% or 97.5%, respectively, based on experimen-tally obtained errors. The relative error of the tritium contentin the H standard solution (o'(x)/x) was assumed to be 0.02.For the relative «rr->r o,' the enrichment factor (a'(A)/A), the ex-perimentally obtained value 0.06 was taken. For enriched sampleswith very low H concentrations (about 1 TU and lower) theground contamination during enrichment (0.O4 + 0.04 TU) was addi-tionally taken into consideration.

9:14

T a b l e 9—5 ' H contents of S t r i p * surfsce waters, shallow groundwatera «t the Str ip» test s i t eand groundwatera fron southern Sweden.

Sample descript ion Date In te rva l II content(TU).

Into t a i l i n g s pond 790506 54 + 8

Watertable wel l and aeep

WT-2

Seep at SBI1 3

790505/18

790506

57 + 8

65 + 8

IIU

uw

Private water supply wells

Private well 1

Private well 1Private well 2

Private well 3Private well 4Private well 5Private well 5Private well 5Private well 5SWI-3

SBII-3SBII-3SBII-3

770927790518771027

771026790517771006771007771011

790515790523/24

790525790526790527

89-10489-10489-10489-104

01338

1034

B6.2122

98.3534834

5548

+ 9+ 8+ 10

+ 11+ 8+ 7.0• 10+ 2.3+ 8+ 8+ 8+ 8+ 8

IIWIIU

IIUIIU

uuIAEAUU

IAEAUWIIU

IIUIIUUU

Drip water at old

• lne levels and

flowing water In old

• lne

135 a level , drip157 • leve l , drip310 • level , drip360 • leve l , drip360 - 410 • level , drip

790514

790511

790511

790518

790515

8677

15986

1 l 0

7 i o "+ 10+ 9

IIUIIUIIUIIW

59 + 8 "

Croundwaters fro»

southern Sweden

Hässelby

WtlheUilund

Kaga

N Stene

St Sundby

Hanna rit

Skofteby

Hangelttsa

Smedtofta

Kockaglrden

»ker

8304 20B3O4 2I8304 21830422830502830503B3O5O4830504830504830505830505

4.12-15.18.3

2 3 . 7

6.50.02.3

1 2 . 30,30.5

+ 0 . 6+ 0 . 6+ 0 . 8+ 0 . 8+ 1.7+ 0 . 7

+ a .6+ 0 . 6+ I . I+ 0 . 6+ 0 . 6

CSFCSFCSFC;SK

CSFCSFCSFCSFCSFCSFCSF

uw U n l v t r s l t y of Waterloo, Waterloo, Canada ( t h e a n a l y t i c a l er ror glvun corresponds to

l o ) .

IAEA • In ternat iona l Atonic Energy Agency. Vienna, Austria ( t h e ana ly t i c» ! error given cor-responds to l o ) .

CSF • CSF- Ins t l tu t fUr Radlohydrosiatrl* , Muntch-Nsuherberg, Fed. Rep. of Germany ( thea n a l y t i c a l e r ror given l a re la ted to a 95t (two aided problem) or 97 ,5 t confidenceleve l (one aided problas) and corresponds to 1.96 o ) .

* Average of »ore than one analysis done on d i f f e r e n t samples, r n l l o r t c i l during thestated sampling per iod.

9:15

9~6 ^ contents of Strlpa nine waters from different boreholes.

Borehole Interval(m)

JH content(TU)

Lab.»

M3MlH3M3M3Ml

RlRl

Rl

VIVIVIVIVIVI

VI

(N2)

(fastfilling)VI (slowfilling)VI (trl-tlum watch)VIVIVIVIVIVI

V2V2V2V2V2

V2V2V2V2V2V2V2

V2

ElElElElEl

NlNlHINlNlNlNlNlNl

(3)O>( O(1)(2)(2)(«)(4)

(2)(2)(1)(1)

77O9J6771019831109831109840223840223

781117781208790517

810828810828810908831003831019831105831105

831105

409-506409-506409-506100-505100-505100-505100-505

100-505

0.7 + 0 .30.5 + 0 .38.3 + 0 .39.1 + 0 .6

10.2 + 0 . 39 .2 + 0 .9

8 + 8-6 + 8

6 + 8

0.7 + 0.10.7 + 0.1

0.92 + 0.181 . 3 + 0 . 21.3 + 0 .31 . 6 + 0 . 21 .3 + 0 . 2

1 . 2 + 0 . 2

IAEAIAEAIAEACSFIAEACSF

IIW

UW

CSFCSFCSFCSFCSFIAEACSF

CSF

831105 100-505 1.3 + 0 . 2

831207831207840111840111840208840208

820421821124821214830119830207831128831128831129831129840228840228840228840228

811111820323820610840306840306

820603820830820906820914820923840126840126840126840126

100-505100-505100-505100-505100-505100-505

6-822406-410413-416.74490-493.74549-552.74424-4994 24-499562-R22562-822500-561500-561382-4 23382-4 23

3-300127.5-129.5

0-2673-3003-300

3-300123-125203-205

271.1-273.1274-276151-251151-251252-30025^-300

2.81.31.31.11.81.1

0.950 . 0

0.880.120.26

1.30.06

1.80.20

0.20.08

0.00.20

1742.620.618.619.4

0.190.250.160.150.23

0.40.19

0.40.12

+ 0.2**+ 0 .2+ 0.3+ 0 .2+ 0 .?*•+ 0 .2

+ 0.14+ 0.6+ 0.27+ 0.09+ 0.11+ 0.2**+ 0.10+ 0.2**+ 0.10+ 0 .2+ 0.09+ 0.2+ 0.11

+ 2+ 3.1+ 1.8+ 0.5+ 1.4

+ 0.11+ 0. 14+ 0. 10+ 0.08+ 0.09+ 0 . 2+ 0.10+ 0 .3+ 0.10

IAEACSFIAEACSFIAEAGSF

CSFCSFCSFCSFCSF

IAEACSF

IAEACSFIAEACSFIAEAGSF

CSFCSFCSFIAEACSF

CSFCSFCSFCSFCSFIAEACSFIAEACSI'

T2 831105 13.8 + 2.9 CSF

* UW • University of Waterloo, Waterloo, Canada (the analyt ical error given corresponds tolo ) .

IAEA • International Atomic Energy Agency, Vienna, Austria ( the analytical error given cor-responds to Id; i f i t s value IR lower than 0 . 2 TU It Is claimed 0 .2 TU).

CSF • CSF-Instltut fur Radlohydronetrle, Munlch-Neuherberg, Fad. Rep. of Germany (theanalytical error given i s related to a 951 (two sided problem) or 97 ,5! confidencelevel (one sided problem) and corresponds to 1.96 0 ) .

contamination cannot be excluded.

9:16

SOUTHERN ..SWEOEN U

BALTIC SEA

Figure 9-5. Location map for Stripa mine and

ints in southern Sweden.

1: Hässelby2: Wilhelmslund3: Kaga4: N Stene5: St Sundby6: Hammarö

sampling po-

7: Skofteby8: Hangelösa9: Smedtofta

10: Rockagården

11: Åker

9.3.4 Discussion

9.3.A.I Surface waters, shallow waters and groundwaters from southernSweden

Most of the samples from Stripa surface waters, shallow groundwa-ters at the Stripa test site and groundwaters from southern Swe-den (Table 9-5) show clear influence of bomb-produced 3H (seealso FRITZ, et al. 1979) and contain as a whole or at least par-tially recent water. The 3H contents from the well SBH-3 indicateactive circulation of recent waters down to a depth of at least89-104 m. From Private well 1 (date 770927) and Private well 4,it is not clear, to what extent the samples contain or do notcontain bomb-produced 3H, because the measuring technique usedis not exact enough to measure in the range below 12 TU.

9:17

'H outputconcentrationITU)

100

50-

10-

0 10

*H input Oct.-March

3H input Jan.-Dec.

50 100

Mean residence time [years]

Figure 9-6. H contents in groundwaters from Brunswick (samp-ling year 1982) with different mean residencetimes calculated by means of the exponential flowmodel•

Groundwaters t rom Skofteby, Rockagården and Åker show no detect-

able influence of bomb-produced H and are recharged before the

year 1952.

9.3.4.2 Stripa mine waters from different boreholes

The •'H concentrations in waters from different boreholes in theStripa mine show large differences (Table 9-6). The influence ofbomb-produced H is shown in the boreholes El, F2 and M3. The Hcontent in M3 increased from 0.6 TU to 9 TU during the last 6years. The highest H content from borehole El is measured inthe borehole interval 127.5 - 129.5 m. Model calculations bymeans of the exponential flow model (corn. Figure 9-6) give meanresidence times of about 3 or 70 years (3H input Jan. - Dec.) orabout 5 or 35 years ( H input Oct. - March) for this sampleunder the assumption that the H input values from southern Swe-den are comparable with the H input values from Brunswick (Fed.Rep. of Germany). The possibility can not be excluded that thiszone is in direct contact with surface water or shallow groundwa-ter. On the basis of these results the statement in FRITZ et al.1979 has to be revised, that the groundwater at the 330 m-levelis essentially free of tritium. It is not possible to decidewhether or not the samples from borehole Rl contain bomb-produ-

9:18

ced 3H because of the low sensitivity of direct liquid scintill-ation counting.

No or at most minor influences of bomb-produced H were found inthe samples from boreholes VI, V2 and Nl. The variation of the3H concentrations with borehole depths (VI, V2) and horizontalborehole distances (Nl) are shown in Figure 9-7. The 3H contentsin borehole VI (interval 100 - 505 m) show no time variation be-tween Oct. 83 and Febr. 84. Results of two contamination tests,carried out on samples from borehole VI (date 810828 and831105), show no detectable 3H contamination. The 3H content inV2 varies remarkably with the borehole depth. A sample from thewhole borehole range shows a H content comparable to that obtai-ned on the sample from the borehole interval 413 - 416.74 m.This suggests that the main flow of the borehole is fed withwater from this zone. The 3H concentration from VI (interval 409 -506 m) is comparable to the value of V2 (interval 413 - 416.74m) and perhaps these waters have the same origin. Figure 9-8shows that the depths below ground surface of these zones withhigher H concentrations are comparable in boreholes VI and V2.At this time it is not possible to decide whether the 3H concent-rations in these boreholes result from mixing with small arao-unts of recent water, or from subsurface production of H.

Remarkably low H concentrations, constant over the whole boreho-le length, have been found in borehole Nl. From this water a mini-mum 3H model age of 60 years (piston flow model) can be calcula-ted using the following assumptions:

a) the 3H input value before 1952 is 5.5 TU (after WEISS e±al. 1979, the H input values from southern Sweden arecomparable with the H input values in Central Europe),

b) diffusive loss of 3H in fractures (NERETNIEKS, 1981) isnegligible,

c) no mixing occurs between waters with different H con-tents and the H concentration decreases only due to ra-dioactive decay.

Because small contamination of the samples during sampling,transport and storage (compare WEISS et al. 1976) can not be ex-clude-!, this low mean H concentration may indicate a maximumvalue for the subsurface production of H in groundwater from bore-borehole Nl.

9.3.5 Conclusions

Surface and shallow groundwaters from the Stripa area show, as ex-pected, high 3H concentrations and are mainly recharged recently(after 1952). Active circulation of recent water is detecteddown to a depth of about 100 m.

9:19

BOREHOLEDEPTHImJO-r

100

200

300

400

500-

600-

•ORtHOLE VI

0.5 15*H |TUl

BOREHOLEDEPTH 0

500

1000

BOREHOLE V2

05 1.0

•*H [TU]

'H 1.0ITU] J

0 100 200 300

Horizontal distance (ml —

Figure 9-7. Variation of the H contents in boreholes VI andV2 with borehole depths and in Nl with horizontaldistance»

9:20

DEPTHIm]0

50O

1000

STRIPA MINE WATERS

Figure 9-8. Variation of the 3H contents in boreholes Nl, VIand V2 with depth below ground surface.

3 <j

H analyses from Stripa mine waters show clearly, that H is pre-sent partly in remarkable contents. Groundwaters from some bore-

'sholes show bomb-produced H with indicates residence times oflower than about 30 years (recent recharge). Groundwaters fromother boreholes show low but detectable H contents. At thistime it cannot be decided whether these low contents resultfrom subsurface H production and/or by admixture of smallamounts of recent water.

Additional measurements and experiments are proposed because anestimation of subsurface production and diffusion of -*H from therock into the groundwater implies several uncertainties.

A rough estimation of the subsurface H production can be done(See Section 9.2.1 and ANDREWS and RAY, 1982) on the basis of U,

9:21

Th and Li concentrations in the granite and on fracture mineralsusing assumptions concerning the effective spectrum of neutronenergy and their range.

When calculating that amount of H in groundwater, which originat-es from subsurface production, further problems are imposed bythe diffusion of ^H out of the rock into the interstitial waterwhere dilution of the ^H content begins to prevail.

The general problem can be compared with problems of subsurfaceOQ 17

Ar- and Ar production. To investigate these processes firstexperiments were started with irradiation of core material withmonochromatic neutrons and detecting the J Ar- and Ar contentin the outgassing fraction of Ar (L0OSLI & FORSTER, 1982). Theseexperiments are done as a function of outgassing-temperature and-time and are similar to the experiments of BRERETON (1970). Itis proposed to extend the irradiation experiments to the produc-tion and detection of H. This could be performed in parallel tothe Ar- and Ar experiments and the experiments should be sup-ported by geochemical, mineralogical and gas analyses of Stripagranite and fracture minerals.

9.A Chlorine-36 in the Stripa groundwaters

9.4.1 Atmosphe_ri£ j3our£e£ £f_ _

Chlorine-36 has properties which make it suitable for the studyof confined groundwaters. Its long half-life (3.01 x 105a) andthe fact that chloride is not removed from solution by mineralinteraction or secondary mineral formation would, in principle,permit the detection of very long groundwater residence times.The very high solubility of most natural chloride species ensur-es that Cl remains in solution.

Cosmic-ray production of Cl occurs both in the stratosphereand in the troposphere. In the stratosphere, the principle pro-duction reaction is proton-induced spallation of Ar, whilst inthe troposphere the neutron induced reaction, Ar(n,p) Cl, ismore significant. Although stratospheric production is latitudedependent because of the effect of the earth's magnetism on cos-mic-ray intensities, mixing is rapid and the stratospheric con-centration of Cl is uniform. The tropospheric production vari-es little with latitude. The Cl atoms become attached to atmos-pheric aerosols in the sub-micron size range and these are subse-quently removed by precipitation in the lower troposphere. Theentry of Cl from the stratosphere into the troposphere is sea-sonally dependent and occurs via the tropopause. The ^°C1 fall-out rate is consequently higher in middle latitudes than at the

9:22

equator or poles (Lai and Peters, 1967). The average falloutrate in Arizona (32°N) is 1 6 + 3 atoms m~2s~1 (Bentley et al.,1982) and at Stripa (57°N) the rate should be about 14 + 3 atomsm s , from the latitude dependence which was derived by Laiand Peters (1967).

Nuclear weapon testing at low altitudes resulted in productionof 36Cl by neutron irradiation of ocean water, "Cl(n,ir)^Cl.This caused a large increase in fallout rates (up to 70,000atoms m-2s ) during the period 1953 to 1964 (Bentley et al.,1982). As the Cl-residence time in the atmosphere is >3 years (El-more et al., 1982), this 'bomb pulse' of 36C1 was transient and

Cl fallout rates have returned to the natural level followingthe cessation of weapon testing in the marine environment. The'bomb pulse' has been used to study groundwater migrationthrough the unsaturated zone in arid regions.

9.4.2 _ Cl_in j>r£undwaj ers_diie_to cosmog£nlc_ _C^ fallout.

The J Cl content of groundwater may be calculated if it is assum-ed that the average fallout is incorporated in the total annualprecipitation and allowance is made for concentration by evapo-transpiration. The groundwater 3°C1 content, (3°Cl), is:

n 6 , F • 3.156 x 107 , 100 ,[ b C l ] g (1 0 0 _ E) atoms/litre (9.4.1)

where F = fallout rate, atoms m s

R » mean annual rainfall, mm a

E - evapotranspiration, %

At Stripa, the cosmic-ray produced Cl content of shallowgroundwater would be about 1.4 x 10 atoms/litre (R » 780 mm a , E =60%). This is at least 150 times less than the observed 36C1 con-tents of the Stripa groundwaters (Table 9-7).

As it is not possible that the 'bomb pulse' of JOC1 could be pre-sant at all depths in the granite and since the JH contents ofthe groundwaters show that there can be little modern water pre-sent, it is evident that in-situ production of Cl must be oc-curring in the rock matrix.

9.4.3 EPSsiplejui.ej0!. _£1_f£r_6£ou.nÉwa. t :£r_sl.uä : l£s

The successful use of 3°Cl for the determination of groundwaterresidence times, requires that (1) the ^"Cl input of cosraogenlc36C1 at recharge can be determined, (11) that changes In the 36Cl

9:23

total or specific activity can be attributed either to decay ofif, if.

the initial -"xl or to ingrowth of JDC1 due to in-situ neutronirradiation of chloride in the migrating groundwater, (iii) thatthe groundwater forms a closed system with respect to chloridewhich must neither be lost nor gained from the rock matrix. Inthis discussion, we will assume that the groundwater is suffi-ciently old and unmixed for there to be no complications due tothe bomb pulse of Cl. The closed system criterion for chloridein groundwaters is usually the most difficult to justify asgroundwaters generally increase in salinity during their evolution.This condition may be relaxed somewhat if in-situ production ofCl in the rock matrix is small, as may be the case for some

sandstones and limestones.OZL 'if.

The number of Cl atoms, N£, present in a groundwater at timet after recharge is given by the equation

36Nt = 36NQ e"

At + 3 6N e q (1 - e"Xt) (9.A.2)

decay of ingrowth of Cl due to in-situ

cosmic in- neutron irradiation of Cl" in

put solution (closed system)

If the in-situ neutron flux for the aquifer is small, this equa-tion reduces to the decay term only. Such a simple model hasbeen applied in the case of the confined aquifer In the Great Ar-tesian Basin of Central Australia where the 3 Cl activity decreas-es downdip over a distance of 700 km from the recharge zone

(Alrey et al., 1984). If the in-situ neutron flux is high and

the groundwater is closed to chloride, the ingrowth term con-trols the 36Cl-actlvity change with time.

In cases where the groundwater chloride content increases due tointeraction with the rock matrix, the specific activity of 3°C1in solution must attain the equilibrium specific activity ofCl In th«> rock matrix, as solution proceeds. No information con-

cerning groundwater residence time can then be deduced, but theadmixture of rock chloride with input chloride may be estimatedif the residence time is known.

"9.4.4 "Clconteiits

The total concentration of JDCl and its specific activity forchloride in the Stripa groundwaters are tabulated in Table 9-7.As shown in Section 9.4.2, the Cl contents are much higherthan can be attributed to concentration of cosmic fallout by eva-potranspiration and must be due to in-situ production. The atom-ic ratio 36N/(35N + 37N) for the Nl sample is the same, withinerror, as that estimated for the equilibrium ratio of chloridein the granite matrix. This strongly suggests that the origin ofthe groundwater chlorinity at this depth is due to water-rock in-

9:24

Table 9-7 36cl in the Strip* deep (roundiwters.

Borehole

Nl

V2U)

VI

V2(B)

IsolatedInterval•

203-205

0-822

409-506

•06-410

Depth belowsurface•

386.7-387

408-1230

766-B63

814-81»

C l "

ppa

67

300

580

620

«C1ttomt/tx 10»

2.3 + 0.

4.1 + 0.

6.2 + 0.

7.0 + 1.

4

3

7

0

36C1/C1ratio, io - ' 5

2C3 + 29

80 + 7

63 + 7

66 + 10

teractlons, either by leakage of saline fluid inclusions or by mi-neral alteration. The enhanced *°Ar/^°Ar ratios for dissolved Arin the deep groundwaters suggest that some mineral alterationdoes occur. For the samples from greater depths in the granite,the 3 Cl ratio is much lower than the granite equilibrium ratioand suggests that all of the chloride in solution at thesedepths may not have been entirely derived from the granite. Thedeep fracture system may have acquired chloride from some alter-nativ» source which has a lower -*°Cl atomic ratio than the grani-te. 'KJ possibilities have been examined, firstly, that some ofthe s /indwater salinity has been derived from interaction withthe icrounding leptite during its early recharge history and,se ; J\f, that the deep fracture system was at one time inunda-te . •>• a solution of an evaporite deposit. The neutron flux inti-e I iptite shows that the ^Cl atomic ratio of its chloridewou / be only 45 x 10"1* compared with 180 x 10"^ for the grani-te \n evaporite deposit would have a very low U-content and neg-1; tble Th-content so that the in-situ neutron flux would be ex-t -t .nely low. Evaporated sea water, for example, would containo' y 0.1 pptn U in the residual solids. The ^ Cl atomic ratio atecj ilibrium for an evaporite would consequently be negligiblysm:ll and a zero value has been assumed for the mixing modeld.i :ussed below.

Thf concentration of -'"Cl atoms in a groundwater which has de-

rived its total chlorinity, V mg/A, from two sources is given by:

36N - WJSJ + (W - Wj)S2 atoms/Ä (9.4.3)

where Wj is the chlorinity derived from source one and Si, Soare the specific concentrations (atoms/rag) of "ci in the twosources. The -*°C1 atomic ratio, R, for the mixture is given by theequation:

WJRJ f (W - Wj)R2R - y (9.4.4)

where Rj and R2 are the 36Cl atomic ratios for the two chloride

sources. The amounts of chloride which must have been derivedfrom granite/leptlte and granite/evaporite sources to yield the

9:25

T a b l e 9 - 8 Hlxlng aodele and chlorlnity aourcee In the Strip* groundwaters.

Sorehole

Nl

V2(A)

VI

V2(B)

crppa

67

300

580

620

granfte

67

79

77

99

Cl~ aourcee (ppa)

leptlteleptlt*

0

221

503

521

for alilng with:

evaporitegranite

67

114

203

229

•vaporlte

0

166

377

391

36C1/C1(calculated)x 10"15

ISO

80

63

66

Value) of ipeclflc - " d content, atcas/ag: granite 3.06 % 10*, leptlte 7.6* x 105, »vaporlte 0.

Values of 36C1 atoalc ratio: granite 110 x 10"15: leptlte 45 x 10"15.

observed Cl contents of the Stripa groundwaters have been cal-culated and are listed in Table 9-8. It may be noted that formixing with a leptite source, the amount of interaction requiredwith the granite at depth is small and most of the chloride isderived from leptite. A more significant interaction with thegranite is required by the evaporite mixing model.

It should be noted that both models imply the deep production of

^Cl if the incoming Cl is weak in the granite; this means that

the residence time in the granite is not significant.

Figure 9-9 shows the relationship between the Cl content andchlorinity for the deep groundwaters. Also shown, are lines forthe total derivation of the chlorinity from the granite and fromthe leptite. Figure 9-10 shows the Cl atomic ratio for bothmixing models compared to the field data.

The leptite and evaporite mixing models explain the observed•^Cl data equally well but the hydrological implications of themodels are different. If the chloride is derived from the grani-te/leptite model, the deepest groundwaters must undergo consider-able evolution in the leptite and <20% of the Cl" is derivedfrom the granite. There is only a small variation in the extentof interaction with the granite for the different groundwaters.Hydrologically, this implies short flow paths through the grani-te and progressively longer flow paths through the leptite forthe deeper groundwaters. As the Stripa granite is a relativelysmall intrusion, this is not implausible. The granite/evaporitemixing model implies that the fracture fluids in the system werecompletely replaced by an evaporite solution at some time in thepast and that shallow groundwaters from near the granite outcropare gradually replacing them. This is also a plausible scenarioin the hydraulic sink which has been created by the mine. It issupported by the heavy isotope composition of the dissolved sulp-hate which suggests that brines of Permian origin provided thesulphate content of the groundwaterr (Fontes and Michelot, 1983;

9:26atons/C

»Cl

2.0

1.0 _

Cl

Figure 9-9. The 36C1 content of Stripa groundwaters plottedagainst their chlorinity. The triangular symbolsindicate the granitic component for mixed Cl~ sour-ces from the granite and leptite. Mixed Cl~ sour-ces for the granite evaporite model are coincidentwith the observed data*

Michelot et al», 1984). The occurrence of evaporites in the re-

gion of the Oslo Gulf has been reported in recent restorations

of Zechstein times (Ziegler, 1982).

9.4.5

The mixing models which have been discussed above assumed thatonce the low Cl-content evaporite solution or groundwater fromthe leptite had entered the granite, there was no further increa-se in the 5 Cl-content due to irradiation of the fracture fluidsby the neutron flux generated within the granite. Since the mix-ing models may explain the 36C1 data without such correction,

9:27

D'Z 200

100

5C

\

-

b i ; : ' : ' • • - •

* '. ' ; • : '\ ' • '

v : i :\ : '

i \ ' ; • , • ; • ;

; \ : ; • ; ; !

: 1 " 1 : :

: : v! : • * » : : •

: : * » •

i i ' t

Figure 9-10.

200 400

Cl" ag/1

600 800

The 36C1/(35C1 + 37C1) atomic ratio of dissolvedchloride in the Stripa groundwaters plotted aga-inst chlorinity. The broken line is the dilutionline for admixture of low 36C1 saline water (Table8.A.2).

the further ingrowth of 36C1 in the fracture fluids could besmaller than the experimental error on 3^Cl-content measure-ments, which ranges from 7 to 14% for the low 3f*Cl content deepgroundwaters. Application of the ingrowth law (equation 9.4.2)shows that a 14% ingrowth of 3^C1 could occur in 70 ka. For theevaporite model, this suggests that the saline incursion intothe granite must have occurred less than 70 ka ago and that mostof the Cl activity has been derived by subsequent interactionof the fluid with the granite. A maximum residence time for thefluids may be calculated on the assumption that no 3^C1 has beenderived from the granite and that it is entirely due to ingrowthwithin the fracture fluids. Although this is certainly not so,as it would allow no chemical interaction with the granite, itdoes establish that, whether or not the mixing model is entirelycorrect, the deep fluids cannot be older than 170 ka.

These studies are still preliminary because there are only fourdata points and there are no measurements of 3^C1 in the rockmatrix to substantiate the calculated value. Further research isunder way to resolve these questions.

10:1

10 RADIOELEMENTS IN THE STRIPA GRANITE AND GROUNDWATERS

10.1 Radioelement content of the Stripa granite

The uranium, thorium and potassium contents of core samples fromboreholes El, Nl, VI and V2 have been determined by Y-ray spec-trometry and are reported in Table 10-1. The U-content of samp-les from VI averaged 44.6 + 10.4 wg/g and from V2 averaged 43.6 +3.7. These values may be compared with measurements by Nelson e_t<Q., (1979) for the M3 borehole, which has an average U-content

Table 10-1. Y-Spectrometric determination of

contents of the Stripa granite.

the U, Th and K

AnalystsNo.

601602

603604

605606607608609610

611612613614615

625626627628629630631632633634

63563663763863964064164 2

Borehole

NlNl

ElEl

VIVIVIVIVIVI

VIVIVIVIVIExc

V2V2V2

V2V2V2V2V2V2V2V2V2V2V2V2V2V2V2

Depthm

22.53- 22.60282.03-282.11

22.37- 22.43286.41-286.47

25 .51 - 25.5975.45- 75.52

I 28.44-1 28. 54186.34-186.41

228.75-228.82282.47-282.53

330.41-330.49385.46-385.54432.29-432.38498.16-498.25

-

Average (605-615)

00.97- 01.0655.22- 55.31

106.03-106.13147.08-147.19196.04-196.15249.68-249.78303.10-303.203*8.52-348.61400.54-400.64448.79-448.90503.84-503.95548.78-548.86599.21-599.30645.11-645.19701.54-701.66750.93-751.03800.14-800.25817.44-817.53

Average (625-642)

U-contentwgg- '

42.1620.24

26.7243.35

40.8375.1739.0938.7842.8948.7547.9735.4042.7939.6739.18

44.59

42.5443.9836.1330.7940.2342.3442.3442.5942.9543.1345.0741.9042.5344.8446.4747.8947.5153.74

43.61

±

0.250.20

0.220.25

0.250.300.240.240.250.260.260.240.250.240.24

10.38

0.240.240.220.230.230.240.240.240..»40.240.240.230.230.240.240.250.250.26

3.72

Th-contentPgg" 1

49.9729.00

48.8850.30

59.1455.2245.2447.1747.4253.3754.3852.0249.2546.2244.86

50.39

54.8055.7841.5945.9355.5156.9857.3159.3259.8656.5757.2652.9552.2757.0158.0660.8464.4670.30

56.49

±

0.890.83

0.880.89

0.910.900.8H0.890.880.900.900.890.890.880.87

4.49

0.900.900.860.870.900.900.900.910.91O.900.900.890.890.900.900.910.920.93

6.08

Th/U

1.191.43

1.831.15

1.450.731.161.221.111.091.131.471.151.171.14

1.21

1.291.271.151.181.381.351.351.391.391.311.271.261.231.271.251.271.361.31

1.2940.07

K-contentX

4.052.9R

4.344.18

3.914.223.844.053.994.064.164.163.864 . ? 7

4.05

4.05

4.274.273.673.914.254.364.334.374.234.634.344.214.133.864.304.314.364.09

4.21

±

0.040.03

0.030.04

0.040.040.030.040.040.040.040.030.040.040.04

0. 14

0.040.040.040.040.040.040.040.040.040.040.040-040.040.040.040.040.040,04

0.21

Note: The error* quoted for Individual determination» arc 20 errora bated on counting a t a t l a -t l c » . The crrora for the set averages are the standard deviation for t lic s e t .

10:2

of 44.0 + 24.6 Wg/g, som measurements by delayed neutron activa-tion analysis for the 330 m and 410 m levels ranged from 37.3 -39.9 pg/g (Andrews et al., 1982). Uranium in the granite is con-centrated as uraninite within open microfractures in feldspars(Nelson et al., 1979) and it is therefore readily accessible toaqueous leaching and can be almost totally removed from crushedsamples by leaching with dilute mineral acids.

The U-content of the granite is uniform throughout the unweather-ed part of the intrusion and is about 15 times the world aver-age for granites. The Th-content averages 50.4 + 4.5 yg/g (VI)and 56.5 + 6.1 ug/g (V2) and the Th/U ratio is 1.2 - 1.29. TheTh-content is also greater than the world average for granitesbut the Th/U ratio is less than the world average (2.8). The re-markably high radioelement contents of the granite are reflectt-din the high U and Ra found in its groundwaters; by a very highHe production rate in the granite and by a high neutron produc-tion rate due to (a,n) reactions (see section 9.1).

10.1.1 Uranium £er^i£s_e£uJ[lJLbria^ i.n_the_S.tri.pa.

The isotopes 2 3 8U, 2 3 4U and 23OTh are genetically related, beingmembers of the 4n + 2 decay series. (Table 10-3) Radioactiveequilibrium throughout this series would be established within1.25 million years as required by the half-lives in the decay se-quence. The equilibria 2 3 4U/ 2 3 8U, 23OTh/238U and 23OTh/234U maytherefore be used to determine whether or not closed system con-ditions for U and Th have persisted within the rock over thistime scale.

Samples of the Stripa granite from four locations were crushed tosub-micron size and then dissolved in HF in a Teflon pressuriseddigestion vessel. U and Th were subsequently separated by anionexchange, electrodeposited on stainless steel and determined byalpha-spectrometry. The natural U and Th contents and the vario-us activity ratios are reported in Table 10-2.

The granite from the extensometer drift is in isotopic equilibri-urn < » V 2 3 8 U B 230Th/238u . 230Th/234u . j, and has t h e r ef Ore

been undisturbed over the last 1.25 Ma. That from the VI excava-tion, however, is apparently depleted in Th. Since Th is notmobile in natural waters, this can only be explained as a conse-quence of uranium deposition*

Tie grey granite (the most typical of the granites at Stripa)from outcrop has a depleted 23*U/238U ratio. This behaviour isexpected when uranium is mobilised by groundwaters and when Uis preferentially mobilised relative to U. The pink graniteis in U/Th equilibrium and shows no evidence of uranium loss.Its radioelement contents (both U and Th) are much less thanthose for the granites at depth. The grey granite has a Th-con-

10:3

Table 10-2. Radioelement determinations on Stripa granite»

Saaple

A

B

C

D

U-contentug/g

19.118.5

24.226.1

38.638.2

36.536.0

+ 0.8+ 0.6

+ 0.7+ 0.8

+ 0.9+ 0.9

+ 0.8+ 0.6

Th-contentUli/g

24.427.2

12.015.0

25.824.8

19.216.1

+ 0.9+ 1.9

+ 0.6+ 0.7

+ 1.1+ 1.2

+ 1.3+ 0.9

0.740.75

0.981.00

1.001.02

0.990.96

++

++

++

++

0.050.04

0.040.04

0.030.03

0.030.03

23OTh/238u

activity ratloa

1.15 + 0.041.27 + 0.07

0.78 + 0.030.94 + 0.04

1.00 + 0.030.91 + 0.03

0.72 + 0.030.61 + 0.02

23OTh/2D4u

1.551.69

0.800.94

1.000.89

0.730.63

A.B.C.D.

Grey granite froa outcrop.Pink granite froa outcrop-Granite froa extentoaeter drift, 350 m depth.Granite froa VI excavation, 350 • depth.

Table 10-3. The 238U and 232Th decay series.

92T

90

23490"

Th

p

91

U92

4.5 x 10* a

24.1 d

1.18 aln

2.5 x 10s a

228R

88

22889

Ac

22890"

'Th

1.41 x 10

6.7 a

6.13 h

1.91 a

10

230,90"

Th 22488

Ra

8.0 x 10* a 3.6 d

226,88

Ra R

86

1.620 a 54 a

222,86

Rn

3.825 d

210,82

Pb

i 22 a

10:4

tent similar to that for the deeper granites. If it had an origi-nal U-content similar to that in the deep granite, the23 Th/238U ratio would be about 2 for recent uranium loss andthe observed ^ Th/ U ratio shows that the loss must have oc-curred over a period in excess of 150,000 years. Alternatively,if the uranium leaching is a recent occurrence, the original U-content of this outcrop granite cannot have exceeded 24 ug/g. How-ever, the latter is very unlikely in view of the otherwise uni-form radioelement distribution in the granite.

10. 2 Uranium and thorium solution by groundwaters

The solution of the natural radioelements in a groundwater is de-pendent upon its chemical character, that is its pH, redox poten-tial, salinity and the particular dissolved species present. Ura-nium in solution, for example, is stabilized as carbonate com-plexes of both U and U , and the importance of these complex-es is dependent upon the pH, Eh and bicarbonate content of thewater (Langmuir, 1978). Radioactive equilibrium is establishedthroughout the 238U decay series (Table 10-3) within 1.25 Ma inclosed systems containing 8U (rock matrices or solutions).Very frequently, however, the 23*U/238U activity ratio for urani-um dissolved by groundwaters is not in equilibrium, generallybeing greater than unity. Such disequilibrium may be a consequen-ce of either preferential solution of U or of a-recoil induc-ed solution of U. Preferential solution of U rather than2*1 o

U atoms may occur because the former are the decay productsof U and must, therefore, be present in lattice regions whichhave suffered recoil damage during the decay process. Such dam-age involves lattice displacement and is not readily annealed. Itis also probable that the U atoms are oxidized to the mere so-luble UV* oxidation state during the recoil process (Ros'.iolt e_tal., 1963). U atoms which decay in the rock surface close tothe rock-water interface may result in the ejection of a-recoil23^Th atoms into the solution (Kigoshi, 1971). These short lived

Th atoms either decay in solution or are deposited on therock surface wh**re, on decay, the U formed may be readily dis-solved.

The rate of 234Th recoil solution, 234Threc, due to 238U decay

at the rock-water interface, is given by:

234Threc - 0.7336[u]r • 0.235 pR min'Vcm2 (10.1)

where [u]r is the natural uranium content of the rock, lig/g, Pis the rock density, g/cm3, R is the recoil range of Th inthe rock matrix, cm, 0.235 is the fraction of recoil atoms fromwithin the recoil range of the surface that enter solution, and0.7336 is the 238U specific activity in natural uranium,min" /gg, which equals the equilibrium specific activity of234U.

10:5

The extent of rock surface, S, in contact with unit volume of agroundwater is given by:

S =-— cnr/cra of ground water (10.2)

where 4> is the fractional porosity of the aquifer and s is thespecific internal surface area giving rise to the aquifer porosi-ty. The value of S is very dependent upon the nature of the aqui-fer porosity but in many cases it is likely to be in the range 1 -20 cm In the case of crystalline rocks which are dominated byflow in tight fractures and especially where the micro-fractureporosity is significant, the value of S may be of order 10 cm" .This parameter is the primary control of the recoil Th solu-tion process and it is shown below that, together with the ura-nium content of the groundwater, it is the most significant con-trol on the 234U/238U activity ratio change with time.

The decay rate of U which is unsupported by its parent c Uand the rate of ingrowth of 3 U in a solution containing 23 Thare controlled by the decay constant, A, for U. From thestandard decay and ingrowth equations it may be shown that aground water which has acquired an initial natural uranium con-tent of [u]gug/g, will undergo change in the

234U/238U activityratio, AR, of this uranium according to the equation:

ARt = l + 0.7336[Uj

0.7336[u]r 0.235 pSR(l - e~234Xt)

s

e~234

O.7336[U]S

where 234Ug and 238Ug are the activities of these nuclides insolution, min~Vg» The activity due to 238U in solution is relat-ed to the natural uranium content of the solution by the equa-tion:

2 3 8U S - 0.7336[u]s (10 4)

so that equation (10.3) becomes:

234At "234AtARt - 1 + (ARt - l )e" + 0.235 pSR(l - e

(10.5)

where ARj is the initial activity ratio of the dissolved uraniumin the groundwater (t • 0). An initially enhanced activity ratiomay be a consequence of either preferential solution of U re-lative to 238U or of a-recoil induced solution of 234Th. How-ever, the recoil process requires time for subsequent U-in-growth to become significant and enhanced activity ratios forgroundwaters close to recharge must generally result from the pre-

10:6

ferential U-solution process. The recoil process becomes moresignificant as the residence time of the ground water increasesand chemical solution ceases, or uranium deposition occurs, asthe ground water evolves towards more reducing conditions. It isevident from equation (10.5) that groundwaters with very low ura-nium contents are most likely to undergo activity ratio increasedue to the a-recoil process. This will tend to counterbalancethe decay of any excess U which was dissolved by preferentialsolution closer to recharge. The estimation of groundwater resi-dence times from 234U/238U activity ratio changes is, in princip-le, possible but requires a knowledge of the geochemical proces-ses which occur in different aquifer zones and of its hydrologi-cal character. This is only likely to be sufficiently well knownin aquifers for which samples can be obtained from various loca-tions. It is unlikely that the uranium geochemistry can be dedu-ced for fluids from a single exploration borehole in an other-wise unknown acquifer.

In contrast with the mobility of uranium in the hydrosphere, tho-rium is very insoluble and is generally not present above detec-tion limits in natural waters. This is a consequence of itsready hydrolysis to the insoluble hydroxide at the pH values ty-pical of groundwaters. The immediate decay product of ^Th is

Ra and this may be ejected into solutions from a rock surface99ft

by the a-recoil process. The ZZÖRa (half-life 6.7 years) canthen support Th (half-life 1.91 years) production in solutionalthough this may be readily hydrolysed and absorbed on any col-loidal particles in solution.

10. 2.1 Analy_tJLcal_metho< foj: II and_ Th_isoto£e£ A

Water samples (20-25 kg) were generally filtered through 0.45 umfilters and acidified to pH<2 by addition of Analar HC1 as soonsas possible after collection. About 300 mg of Fe3+ and 20-40 dis-integrations min. of U were added to each sample. The aci-dified solution was outgassed with CC^-free air or nitrogen toremove carbonate which could complex U at high pH. U was copreci-pitated with Fe(OH).j on raising the pH to ~8.5 by addition ofNH^OH solution. The precipitate was recovered dissolved in 6 MHC1, and Fe3+ was then extracted into an equal volume of methyl-isobutyl ketone. The acid solution of U was further purified byanion exchange, first on a Cl~ and then on a NOZ column of Dowex-1-X8 100-200 mesh resin. U was finally eluted from the NO3 columnwith 0.1 M HC1 and after evaporation to dryness was dissol-ved in 10 cm3 1 M (NH^SO^ solution (acidified to pH 2.4) andtransferred to a Teflon electrolysis cell. Electrodeposition ofU on a stainless steel plachet was complete after 3 hr. at a cur-rent density of 1 A cm" . The source was counted with a Li-drif-ted surface barrier alpha-spectrometer.

10:7

0.8

0.6

o 0

-0.2

-0.4

-0.6

Figure 10-1. The stability of uranyl species in solution as afunction of pH and Eh. The minewaters are indicat-ted by full circles and the shallow groundwatersby open circles.

If Th isotopes were to be detemined, they were separated by furt-her anion exchange on the eifluent from the Cl~ anion exchangecolumn on which uranium was initially retained. Following separa-tion of Th from other cations, it was electrodeposited on a sta-inless steel planchet prior to alpha spectrometry.

10. 2. 2 U-£°Iutii>n_and_ ^ U / ^ U . ictivity_ratio_in ^he _shal_low_grpundwa_-

The 234U/238U activity ratios of the dissolved uranium in theshallow groundwaters at Stripa (Table 10-4) range from 2.18 to4.8 and their uranium contents range from less than 1 to almost100 ug/kg. These are the groundwaters closest to recharge andthey have high tritium contents, indicative of their recent ori-gin. Their Eh values show that uranium solution ic taking placeunder oxidising conditions (Figure 10-1). It Is possible thatpreferential etch solution of "3*U Is the cause of the activityratio enhancement for the dissolved uranium in these groundwa-ters. However, laboratory etch experiments on crushed samples of

10:8

W '

7 •

6 •

b •

3 •

0-

/^ —

(b)

10 10J 101

Groundwater age, years.

Figure 10-2. U/238U activity ratio change with age for urani-um in groundwater present In the micro-fracturesof a granite which has a uranium content of 5000Ug/g in fracture surfaces, calculated from equa-tion 3. The activity ratios are calculated for (a)fracture openings of 0.5 wm (S » 40000 cm2/cm3)and (b) fracture openings of 2.0 um (S • 10000cm /cm ). In each case the groundwater uranium con-tent is 10 pg/kg.

the Stripa granite resulted In an activity ratio of only 1.15for the dissolved uranium. The activity ratio for the uraniumwithin a granite sample from outcrop was found to be only 0.72 +0.3. It was only possible to obtain an activity ratio for thedissolved uranium, which was greater than that for the rock mat-rix, when etch was carried out under very slow and non-oxidisingconditions. Preferential etch solution of 23*U can probably alsooccur to some extent under natural conditions of groundwater/rock interaction but 234Th recoil due to the large concentrationof uranium in micro-fractures could also be a significant mecha-nism for generating the observed 23^U/238U activity ratios inthe shallow groundwaters. The increase of this activity ratiowith time due to continued 23^Th recoil within the micro-fractu-res is shown in Figure 10-2. The computed changes are shown for0.5 and 2.0 gm fracture separations, for a uraninlte coating onthe fracture surfaces equivalent to a surface uranium content of5,000 ug/g. For a groundwater uranium content of 10 vg/kg, theactivity ratio could increase to 4 within 10 years as a resultof such Th recoil in the micro-fractures. If most of the ura-nium were present within the 234Th-recoil range of the rock-water interface, the recoil process would eventually establish a

U/238U activity ratio of 0.75 in the uraniferous surface.This is close to the observed ratio in outcrop samples of thegranite and suggests that the recoil process is important in

10:9

4-1

>

<

u •

9 •

7 .

5 •

1 .

10

• —

-

3 10* 1

k36 10

Groundwater age, years.

Figure 10-3. 234U/238U activity ratio change with age for urani-um in groundwater present in a granite with fractu-re openings of 0.2 cm (S = 10 cm /cm ) and a uni-form uranium content of 40 yg/g calculated fromequation 3. The activity ratio/age relationship isshown for various initial activity ratios and fora groundwater uranium content of 10 pg/kg.

these groundwaters. The effect of alpha-recoil for fluids contain-ed between fracture walls which have the average U-content ofthe bulk granite is shown in Figure 10-3. In such situations the234U/238U actWity ratio cannot increase with time and enhancedratios must be due to preferential solution of 23^U, followingwhich decay of excess U occurs.

10. 2. 3 Uranium cherai try_ in_the__dee£e£ g_roundwaters

Table 10-4 reports measurements of the U-content and 234U/238Uactivity ratio for dissolved uranium in the groundwaters from1977 to 1984, and field data for recent measurements are givenin Table 10-5. Apart t rom the increased U-content for the 1981measurement, both parameters are remarkably constant in the M3borehole. The activity ratio average is 10.9 with a standard de-viation of only + 0.2. This may be attributed to the constantflow conditions which existed in this borehole over the periodof measurement, so that U-solution and ot-recoil effects havebeen unaffected by variations of the sampled interval or in theflow pattern.

The records for boreholes VI and V2 are much wore variable and

this is largely due to the variable sample intervals which were

selected and to variations in the flow history of the borehole

10:10

Table 10-4. Natural U-contents and 2 3 4U/2 3 8U act iv i ty ratiosof dissolved uranium in Stripa groundwaters.

Analyilanuaber

Sn.pl In*date

Shallow iroundwatera

PU1PW1PW2PW2PW3PW4PU5PW5W 2MineDrlpvater

111R9

Borehole M3

1452 2F1485-831520-84

Borehole El

1455 IP1535-84

Borehole Nl

1454 IP1454 3P1510-84 I1515-84 11

Borehole V]

1450 IP1450 2P1450 3P1450 5P1450 6P1450 7P1450 SP1450 9P1470-831475-831480-831500-841505-84

18.05.7921.01.8124.10.7724.10.7716.05.7917.05.794.10.77

15.05.7918.05.79

16.05.79

16.05.7916.05.79

9.09.779.09.776.06.78

16.05.793.06.819.11.83

23.02.84

11.11.816.03.84

14.07.816.10.81

26.01.8426.01.84

15.01.8111.06.8117.07.818.09.819.09.81

16.09.8121.09.816.10.SI3.IO.B3

19.10.835.11.83

11.01.848.02.84

BoreholeInterval. •

0 - 2 60 - 3 0

0 - 1 40 - 1 40 - 1 40 - 1 40 - 1 40 - 1 40 - 1 4

3 - 3 0 03 - 3 0 0

2 - 3 0 03 - 3 0 0

252 - 300152 - 251

92 - 94410 - 506409 - 506409 - 506

4 - 5064 - 5064 - 5064 - 506

100 - 505100 - 505100 - 505100 - 505100 - 505

Depthbelowground, •

60

<100<100

12040404050

157

310 - 336336 - 366

336 - 350336 - 350336 - 350336 - 350336 - 350336 - 350336 - 350

357 - 385357 - 385

357 - 401357 - 401397 - 401379 - 394

449 - 451767 - 863766 - 863766 - 863361 - 863361 - 863361 - 863361 - 863457 - 862457 - 862457 - 862457 - »62457 - 662

U~eontentU«/kg

90.290.402.842.26

19.477.60l.ftl0.810.85

14.72

11.6910.99

8.259.228.33

11.1818.0310.1210.40

9.507.40

8.461.831.271.24

34.760.420.095.350.060.350.360.281.490.210.170.170.22

+ 1.80

+ 0.49• 0.10+ 0.05

+ 0.24+ 0.06

+ 0.23+ 0.08+ 0.02+ 0.02

+ 0.46+ 0.04+ 0.13+ 0.13+ 0.04• 0.05+ 0.03+ 0.03+ 0.03+ 0.01+ 0.01+ 0.01+ 0.01

234u/238,,Activity ratio

3.223.23 + 0.072.62*2.574.802.743.11*2.182.73

2.91

11.1111.75

10.75*If).65*10.7010.6811.24 ^ 0.3210.97 + 0.1111.09 + 0.05

4.7' + 0.134.16 +0 .04

3.22 + 0.105.64 + 0.269.00 + 0.189.65 + 0.20

2.90 • 0.043.96 + 0.443.37 + 0.563.06 + 0.084.91 + 0.385.14 + 0.734.65 + 0.494.52 + 0.453.20 + 0.074.81 + 0.265.41 • 0.324.55 + 0.243.70 • 0.13

10:11

Table 10-4. Cont'd.

Analysisnuaber

Borehole V2

1453 IF1453 2F1460F1461F1462F1*95-83 I1525-84 I I1530-84 IV

Samplingdate

9.09.7710.11.778.09.77

20.09.7729.01.7816.05.7911.06.8119.11.8124.11.8213.1 8219.01.8329.;1.8328.02.8428.02.84

BoreholeInterval, •

6 - 5 00 - 470

150 - 470285 - 470376 - 470401 - 428356 - 470356 - 471406 - 510413 - 417490 - 494562 - 822500 - 561382 - 423

Depthbelow(round, •

416 - 460410 - 880560 - 880695 - 880786 - 880811 - 838766 - 880766 - 881816 - 920823 - 827»00 - 904972 -1232910 - 97i792 - 833

U-content

10.436.244.564.121.030 .50.250.140.260.480.580.080.130.25

+ 0+ 0+ 0+ 0+ 0+ 0+ 0+ n

.06

.03.38.04.19.01.01.01

234U/238O

Activity ratio

5.55»5.87*3.87*4.08*4.023.743.98 + 1.003.14 + 0.721.97 + 0.353.77 + 0.357.40 + 0.262.54 + 0.196.13 + 0.294.15 + 0.14

•Analysed at Florida State University.

(Andrews, 1983). The increase in U-content observed on 8 Septem-ber, 1981 for VI followed a shut-in period of no flow. A similarincrease was observed on 3 October 1983, when flow was restoredafter shut-in for several months. The average ZJ Vr °U activityratio for groundwater from this borehole, neglecting the high U-content measurements, is 4.5 + 0.6. The U-content for V2 shows amarked decline with depth. The activity ratio average is 3.8 +0.4 for depths greater than 560 m below surface, neglecting thehigh ratios tound in discrete intervals at 900 - 904 and 910 -97C m below surface. The lowest activity ratios occur at thegreatest depths, where the isotopic composition of the groundwa-ter is also distinctly different. The main features of uraniumchemistry in the Strip* groundwaters are summarized in Table 10-6. The U-content is highest in the shallow groundwaters and dec-lines with depth whilst the activity ratio increases to a max-imum for groundwaters from about 350 m depth and then declinesfor deeper groundwaters. The pattern of change Is consistentwith alpha-recoil enhancement of the 234U/238U activity ratiofor the shallow groundwaters and a somewhat larger recoil effectaround 350 m depth. This could result from U-deposition on frac-ture surfaces as shallow groundwater migrates downwards. Altho-ugh the field measurements of Eh/pH for the 350 m groundwatersdo not suggest that U-deposition should occur the high ^Rn con-tents in these groundwaters (10.2.3) are consistent with such de-position. This incursion of meteoric water from outcrop to about350 m level in the granite is consistent with the expected flowpattern once a hydraulic sink was created by opening the minegalleries to depths of 350 m and, locally, to >; 10 ra.

The decrease in 234U/238U activity ratio in the samples from VIand V2 may be due to any of the following possibilities:

10:12

Table 10-5. Physical data for groundwater samples from Stripafor sampling in period October 1983 - March 1984.

loreholc

M3ElNl-IN1-1IVIV2 IV2 IIV2 IIIV2 IV

Interval. •

0 - 1 *3 - 3 0 0

252 - 300152 - 251100 - 505562 - 822500 - 561424 - 499382 - »23

Flow rate.ea3/nln

75*00560555

7-18,00011

25221092

Teaperature°C

14.510.310.510.312.3A.28.69.08.0

pH

9.098.908.898.999.129.94

10.0610.049.51

Conductivity|iS/c«

235165200215

15251215109012101140

Eh•V

+100+109• 33+ 39+114+239+ 43+ 46+ 26

Rrlatlvc borehole dlapoiltion:

3 5 6 . 7 m

4 0 7 . 7

V2

IV

III

11 .,

VI

EHJ.5* dowti)

NI (S.S* down)

(i) incursion of meteoric water from shallower depths and ad-

mixture with stored fluid in the deep fracture system.

(ii) an alternative flow pattern at depth, perhaps involvingrecharge through the leptite so that the U-chemistry ispartially determined within the leptite.

iO: 13

Table 10-6. Summary of the main features of uranium solutionchemistry in the Stripa groundwaters.

Croundwater type

Approximate

depth below

g-ound, •

2 3 8U content

Mg/kg activity ratio

Shallow

Intermediate depth

Borehole El

Borehole» M , R9

M3

Borehole Nl (Inner)

Deep »roundwatera

Borehole VI

V2

V2, >816 • below•urfa-e

<150 1 - 100 3.0 + 0.7

370350350380

7.411.319.A

1.25

0 .20.10.1

-+++

-

9010

000

.5

.4

.1

.2

.4

.5

.3

4.411.410.99.3

4.53.82.U

+ 0.3+ 0.3+ 0 .2+ 0.3

+ 0.6+ 0.4+ 2.5

( i i i ) recharge of the deep groundwater occurred through theshallower levels at some time in the past, acquiring ahigh 2 3 4U/2 3 8U ratio similar to the M3 values and has sub-sequently decayed to the low levels found at the bottomof V2. This decay would require in excess of 500 ka.

This last poss ib i l i ty can be excluded on the basis of the Cldata (Section 9.4) which shows that the deep groundwater residen-ce time i s <70 ka. The incursion of a brine into the deep fractu-re system, probably at a time of lower sea l eve l , i s suggestedby the isotopic composition of dissolved sulphate (Fontes and Mi-chelot, 1983), and would also explain the 36C1 data. Such abrine would have a low U-content and the 23AU/238U ratio wouldprobably be close to equilibrium. The f i r s t explanation for thepresent U chemistry in the deep groundwaters, therefore seemsthe most probable. This i s also the most probable explanation ona hydrological basis , as recharge would always have taken placedirectly through outcrop and would not require long flow-pathsthrough low permeability strata.

10.2.4 Thorium in solution

The f irs t few members of the 232- decav ser ies are:

23 2Th 228Ra

1.4 x 1010a 6.7a

228Ac | 228Th

6.13h 1.91a

The isotope 230Th ( h a l f - l i f e 8 x 104a) belongs to the 238U decayseries and was found to be absent from a l l of the groundwaters.232Th and 228Th are present (Table 10-7) and 228Th i s generally

10:14

Table 10-7. 222Th content, 23OTh/23AU ratio and 228Th contentof Stripa groundwaters.

Analyst»

lor .hol .

1452-2U

lorehole

1454-1

lorrholc

JO

• 1

VI

1.06-111.06.11

14.07.11

bilktatar•

0 -0 -

2 -

• 1 .»•1.

It14

MO

Hi

S:

)J7 -

1. •

>50)50

M 5

« • • • '

0.11) •

0. 2)2 *

wteat

0.0)5

0.011

7.M5

I45O-2F I I . M . I I 40» - 506 767-16) 0.MI • 0.027 0.00511410 IF 14.07.11 40» - 506 767 - 16) -1450-1» 14.07.11 40» - 506 7 6 7 - 1 6 ) 0 .274*0 .06 * 0.0)1»

Inrriiol» V2

1451-IF I I . M . I I )56 - 470 766 - M0HM-lli I I .M.I I )56 - 470 766 - M0 0.071 * 0.021 0.0221

F • 0.45 urn filtered eeapleu • unflllered iMple

present in an amount exceeding equilibrium with ZJ^Th. Thoriumis generally absent from groundwaters because of the insolubili-ty of its hydroxide. At Stripa, the amounts of 232Th present inthe waters are low but significantly above detection limits anda proportion of it may be carried on particulates of colloidaldimensions. Th is produced by an a-recoil ejection of Rainto solution and this decays to Th, transient equilibriumbeing established after about 35 years. The Th content wouldbe indentical to the "°Ra content for a constant ^z°Ra recoilrate.

The alpha-recoil rate for Ra and Ra may be estimated fromthe amount of Th in solution. The 228Th activity in solutionis equal to that of about 30 ug kg"1 of 232Th (Table 10-7), thatis. to about 7.4 disintegrations min" kg . Provided that the2 Th residence time is greater than a few years, this should

equal the Ra recoil rate. As the concentrations of U and Thin the Stripa granite are about 40 ug kg and 30 vg respective-ly, the corresponding recoil rate for Ra is given by:

Ra recoil rate - U series recoil rate

„ 40 x 0.7336 , ,30 x 0.246 x '*

- 29.4 disintegrations min^kg"1

- 13.4 pCi kg"1

10:15

This estimate of the Ra recoil rate assumes that both U andTh are uniformly distributed in the rock surface. The very high

Rn contents in the groundwaters and the ready solution of2 3 8and 238U indicate that the U concentration in the fracture

surfaces may be much greater than that of Th. Ra content ofthe groundwaters exceeds that suggested by the above recoilrate, by a factor of about 10 (Section 10.3). This confirms thatU is preferentially concentrated close to fracture surfaces,whilst Th is probably incorporated in accessory minerals to alarger extent than U.

The number of ratio of 232Th fo 228^ a t o m s for radioactive equi-

librium is about 1010/l, so that the mass of 228Th in solution

is negligible even though its activity is readily determined.

The relatively high activities of 228Th do not exceed solubili-

ty limits and this suggests that other high specific activity 4-

valent nuclides amongst the actinides could have appreciable mo-

bility in the Stripa groundwaters.

10.3 Radon and radium solution in groundwaters

222The dissolution of Rn by a groundwater occurs predominantlyat the rock-water interface by an a-recoil process (Andrews andWood, 1972). The 222Rn activity in solution increases with timeuntil it reaches an equilibrium value determined by the decayrate of its parent 2 2 Ra, in the rock surface. This equilibriumis attained when the residence time of the water in the aquiferexceeds 25 days, according to the equation:

, = 222Ae(l - e-222AC) (10.6)

where 2 2 2 A t , 2 2 2A e are the 222Rn a c t i v i t i e s at time, t , and at

equilibrium, respectively, and 222X is the decay constant for222Rn.

The equilibrium 2 Rn content, [Rn], of a groundwater i s relatedto the uranium content, [u]riig/g; bulk density, pg/cm and frac-tional porosity, $, of the aquifer rock by the equation:

F x 0.7336 x p x [u] x 103

[Rn] - uCi/kg (10.7)<t> x 2.2 x 10°

where F i s a factor which accounts for the overall efficiency of2 Rn solution due to ct-recoil ejection at the rock-water inter-face and i t s diffusive movement to the flow porosity. The valueof F i s generally in the range 0.01 - 0.05.

222The diffusion of c Rn in the x-direction may be representedas:

[Rn] - [Rn] exp(-xvT7D) (10.8)

10:16

where [Rnl , [Rnl are the Rn concentrations at x = 0 and x,1 J o L J x 9?2

respectively, and D is the diffusion coefficient for Rn. Ther J 9 79 S?

diffusion coefficient for z/zRn in water is 10 cm /s (Tanner,1964) and equation (10.8) shows that the 222Rn concentration de-creases by a factor of 0.37 for diffusion over a distance equal to/D/X (the diffusion length) which is 2.18 cm in water. Over adistance equal to 10 diffusion lengths the Rn concentrationis reduced by a factor of 4.5 x 10" . In more practical tertmthis means that Rn atoms decay before they can move signifi-cant distances by diffusion. In a still fluid the Rn concentration decreases to IX over a distance of 10 cm.

909Transport of Rn by groundwater movement is, therefore, muchmore significant than its diffusive movement. For movement by

999flow in fractures or conduits the Rn-contenc of the water is

99A

unsupported by its parent " D R a in the rock surface and it maydecay significantly if transported over long distances. Transportof Rn by interstititial flow can maintain the 222Rn-contentof the water at its equilibrium value as it is then always sup-ported by Rn recoil from the interstitial surfaces.The following processes are possible for the solution ofby groundwater:

(1) chemical solution by rock-etch processes,

(2) decay of 230Th in solution,

(3) ot-recoil of 226Ra on decay of 23OTh

(a) which was deposited from solution onto the rock sur-face,

(b) which was formed by U decay in the rock surface.

O 9£

The chemical solution of Ra should increase in importance asthe salinity of the groundwater increases, provided that suffi-cient barium or calcium ions are present in solution. The resi-

?26dence time of Ra in solution depends upon the congruent - in-congruent solution of barium and calcium as well as upon its half-life (1620 years). The formation of 226Ra due to decay of 23OTh insolution is negligible since the Th is never present in solu-tion to any significant extent. The activity of Ra dissolv-ed by the recoil mechanisms increases with time according tothe equation:

where 226A,. and 226A_ are the 226Ra activities at time, t, andat equilibrium, respectively, and \ is the decay constant for226Ra. The 226Ra activity in solution should, if o-recoil isthe dominant mechanism, become constant after about 8000 years.

10:17

However, the incongruent solution of calcium carbonate resultsin the exchange of calcium between solution and rock-carbonateand co-precipitation of Ra with calcium may prevent its acti-vity ever attaining the equilibrium value in solution. Neverthe-less, relatively high Ra contents are generally found in themost evolved groundwaters and as geothermal waters and in oil-field brines.

10.3.1 An£ly_tj.c£l_method fojr f^Rn .and

10.3.1.1 222Rn

Rn in 1-litre water samples was determined by outg^ssing witha stream of radon-free nitrogen and trapping the radon on an ac-tivated charcoal trap at -80°C. More than 99 per cent of theequilibrium radon was recovered from the solution by outgassingwith ten times its volume of nitrogen at a flow rate of approxi-mately 1.4 1/min. The radon so released was held on the charcoaltrap even after prolonged passage of nitrogen, provided that thetrap was maintained at -80°C. After radon collection, the charco-al trap was connected to a 50-cm conical flask, the conicalwalls of which were coated with zinc sulphide scintillator. Withthe trap maintained at -80°C, the apparatus was evacuated to apressure less than 1 mbar and the charcoal trap heated to 200°Cto desorb the radon. The desorbed radon was transferred to thescintillator flask by admitting a stream of dry air through thetrap and into the flask until the pressure equalled atmosphericpressure. After closing the stopcock on the flask, it was placedon the photocathode window of a 3-in diameter photomultiplier(E.M.I, type 9708 KA). The o-particle emission from 222Rn andits daughters was recorded with a pulse amplifier/sealer system.

The counting rate was determined after a delay of 3

h to allow ingrowth of Po in the decay sequence:

2**Rn a 2J8Po a 2:itPb P

3.825 days 3.05 min 26.8 min

21*Bi B ?i"Po a

19.9 min 1.6 x iO"*1 sec

The counting rate so obtained was corrected for counter back-ground and for radon decay during the time elapsed since the watersample was collected. Each scintillation flask was periodicallycalibrated by outgassing radon from a standardized radium solu-tion in which radon-radium equilibrium had been established. Thebackground counting rate for a freshly prepared flask was less

10:18

than 0.5 counts/min and increased gradually with use to about2.5 counts/min. This increase in background is due to the accumu-lation of the radon decay product ^10Pb (half-life 22 ye*rs)which has an o-particle emitting daughter, Po. The countingefficiency of a scintillator flask was generally about A counts/min per pCi ot radium.

10.3.1.2 226Ra

926 222Ra was determined by estimating the equilibrium c Rn activi-

22?ty. The Rn contents of groundwaters are frequently high, andvery much greater than the Ra content of the water. This ex-cess radon was removed from the sample by outgassitig with nitro-gen before the radium determination. The sample was then allowed

99 0 99 9

to stand for > 20 days to permit ingrowth of "^Rn. The Zl"Rnthen present vas determined as described above and the activitywas corrected to the equilibrium value by use of the knowngrowth time. The removal of excess radon must be at least 99.9per cent efficient to avoid errors in the radium determination,and this was achieved by outgassing the water sample with 20times its volume of nitrogen. After outgassing 99.9 per cent ofthe radon present in a sample with an excess radon content of800 pCi/1, 0.8 pCi/1 would remain, which after a 20-day decay pe-riod, is reduced to 0.02 pCi/1 of 222Rn or 226Ra.

22210.3.2 Rii contentsof

The 222Rn contents of the deeper Stripa groundwaters are excep-tionally high (Table 10-8). The waters from shallow boreholes(up to 60 m depth) have 222Rn contents of 5 - 2 0 nCi/kg which istypical of other granitic provinces (Edmunds et al., 1984). Forthe deep groundwaters, values range from 200 - 2,000 nCi/kg.There are marked temporal variations in the Rn content forsome of the deep groundwaters. For example, for determinationsbetween October 1983 and February 1984 for VI, there is a largeincrease in Rn content. This suggests that the flow regime inthe borehole must have changed during this period. The Rn con-tent in this borehole decreased during an earlier shut-in periodand was attributed to decay within the borehole and its associa-ted fractures (Andrews, 1983).

222The Rn content of a groundwater is related to the U-contentof the rock, [u]r ppn, and its porosity, $, by equation (10.7).For the shallow groundwaters, this equation shows that less than1% of the Rn generated within the rock matrix is dissolved inthe fracture fluids for a porosity of 1%. The fraction of 2 Rndissolved in the minewaters is much greater and up to 50% releasehas been observed for M3. These high efficiencies of Rn relea-se indicate that the U within the rock matrix must be in close

10:19

Table 10-8. Rn and Ra contents of Stripa groundwaters.

DepthAnalyala Saapllng Borehole belownuabcr date Interval, a ground, a 22hn nCl kg"1 226Ra pCl kg"1

Shallow troundwatera

PW1

PW3

PU4

pus

WT2

Borehole H3

1452

1452

1485

1520

- 1

- 2

- 83

- 84

18.05.79

16.05.79

17.05.79

15.05.79

18.05.79

14.02.78

16.05.79

2.01.81

10.06.81

9.11.83

23.02.84

60120

40

40

SO

6 . 2

S . I

10.3

8 . 1

19.3

0.570.A4

0.83

0.41

0.72

0 -

0 -

0 -

0 -

0 -

0 -

14

14

14

14

14

14

336 - 350

336 - 350

336 - 350

336 - 350

336 - 350

336 - 350

1300.02010.0

1111.8

847.21037.7

1168.31230.4

942.5813.9

+ 1.5

+ 3.4• 4.0

+ 6.0+ 6.1

+ 4.7+ 3.9

34.04.596.88 + 0.045.89 + 0.045.24 + 0.055.79 + 0.054.23 + 0.114.68 + 0.04

4.90 + 0.12

Borehole El

145S -

1455 -

1535 -

1

2

84

11

24

6

.11

.03

.03

.81

.82

.84

3 -

128 -

3 -

300

130

300

357

367

357

- 385

- 369

- 385

402.5 + 1.13S7.O + 0.8

81.8 + 0.884.1 + 0.8

383.7 + 4.8351.4 + 4.5

5.14 + 0.045.21 + 0.04

5.41 + 0.11

Borehole Nl

1454 - 1 14.07.81

14S4 - 2 19.08.81

1454 - 3 Ö.10.81

1 5 1 0 - 8 4 26.01.84

ISIS - 84 26.01.84

2-300

3-300

252 - 300

152 - 251

357 -

356 -

397 -

379 -

401

401

401

394

1.71.9

171.7173.7

642.0639.0

591.4480.3

+ 0+ 0

+ 1+ 0

+ 5+ 4

+ 4+ 3

.1

.8

.1

.5

.3

.7

2 . » +2.30 +2.90 +2.14 +2.47 +

3.26 +3.67 +

12.56 +11.74 +

12.01 +10.25 •

0.040.020.030.020.03

0.030.03

0.200.14

0.190.12

10:20

Table 10-8. Cont'd.

Analytle Sampling

number date

Depth

Borehole below

Interval, • ground» • Rn nCi kg" 1 22nRa pCl k g ' 1

Borehole VI

1450 - 2

1450 - 3

1450 - 4

1450 - 5

1450 - 6

1450 - 7

1450 - 8

1450 - 9

1470 - 83

1475 - 83

1480 - 83

1500 - 84

1505 - 84

11.06.81

14.07.81

19.08.81

8.09.81

9.09.81

16.09.81

21.09.81

6.10.81

3.10.83

19.10.83

5.11.83

11.01.84

8.02.84

409 - 506

409 - 506

409 - 506

409 - 506

4 - 506

4 - 506

4 - 506

4 - 506

100 - 505

100 - 505

100 - 505

100 - iOS

100 - 505

767 - 863

766 - 863

766 - 863

766 - 863

361 - 863

361 - 863

361 - 863

3M - 863

457 - 862

457 - 862

457 - 862

457 - 862

457 - 862

172.1 + 1.4123.4 7 1.3

218.4 + 0.9222.7 + 0.9

11.0 + 0.19.2 + 0.1

178.2 + 0.8198.3 + 0.2

20».1 + 2.8212-2 + 1.0

1 8 2 . 2 + 1 . 5183.7 + 0.5

170.6 + 0.7173.2 + 0.8

99.0 + 0.388.2 + 0.3

125.3 + 0.6

1 9 5 . 2 + 1 . 1200.8 + 1.2

2 5 3 . 1 + 1 . 32 4 2 . 1 + 1 . 5

763.6 + 5.7609.1 + 4 . 5

1)2.2 + 0.22128.4 + 0.53

134.1 + 0.33146.8 + 1.60

101.9 + 0.15I0S. 7 + 0.24

102.1 + 0.6895.5 + 0.85

125.6 + 1.02103.4 + 0.71

111.2 + 1.31104.3 + 0.92122.5 + 0.96120.9 + 0.36

120.1 + 0.63109.6 + 0.91

131.3 + 0.8141.1 + 0.8

119.6 + 0.9119.3 + 1.3

103.9 + 0.8103.2 + 1 - 2

155.3 + 2.5151.3 + 0 . 7

116.5 + 0.41103.5 + 1.78

Borehole V2

1453

1453

1453

1460

1461

1462

1461

1490

1495

1525

1530

- 1

- 2

- 3

- 83

- 83

- 84

- 84

14.02.78

14.02.78

14.02.78

16.05.79

11.06.81

19.11.81

22.04.82

24.11.82

13.12.82

19.01.83

25.02.81

28.11.83

29.11.81

28.02.84

28.02.84

0 - 2 2

132 - 359

176 - 471

401 - 428

356 - 470

156 - 471

6 - 822

406 - 410

411 - 417

490 - 494

549 - 553

423 - 499

562 - 822

300 - 461

182 - 421

416 - 412

742 - 769

786 - 881

811 - 838

766 - 880

766 - 881

416 -1212

816 - 920

821 - 827

900 - 904

959 - 961

811 - 909

872 -1232

910 - 971

792 - 833

360.0

560.0

810.0

224.8 + 2.0500.7 + 0.6

471.9 + 0.5453.8 + 0.5440.0 + 0.3501.0 + 0.4

388.0 + 1.0434.0 + 1.0

574.0 + 1.0418.0 + 1.0

713.0 + 1.0731.0 + 0.6

497.4 + 4.5618.5 + 4.9

185.0 + 2.7115.3 + 1.8

492.1 + 2.6527.9 • 2.9

281.7 + 2.1356.7 + 3.5

40.0

56.0

56.6 +

56.9 +

71.7 +

85.8 +

44.1 +

45.9 •

11.2 +

13.5 +

42.1 +

43.4 +

45.7 +

48.2 +

56.4 +

64.1 +

75.6 +

71.6 +

96.8 +

79.6 +

0.51

0.66

0.41

0.49

0.34

0.15

0.27

0.29

0.13

0.34

0.30.41

0.80.4

0.350.5,

0.46

0.56

10:21

contact with the fracture fluids. The highest 222Rn contentswere found In the M3 borehole and this would be consistent withthe deposition of uranium in the fracture porosity in the shallow-er part of the system as sugges.ed In Section 10.2. As onlyuranium is mobilised by groundwater and 23OTh (half-life 80,000years) requires a long time, such uranium deposition must have oc-curred naturally before the hydraulic sink was formed by miningoperations. Alternatively, as U solution by the shallow groundwa-ters must also dissolve Th, this nuclide may have been trans-ported to depth after adsorption on colloidal matter. The deposi-tion of U on fracture walls is unlikely to be revealed by U-ana-lyses on bulk rock samples.

10.3.3 _ _Ra coilt_ents_o_f_ j[t£i£a_gjroundwat er£

The 2°Ra contents of the Stripa groundwaters (Table 10-8) gene-rally increase with depth to a maximum value between 50-100 pCikg"1 for V2 and up to 150 pCi kg"1 for VI. The recent shallowgroundwaters all have 22°Ra contents <1 pCi kg"1 and contentsfrom 2 - 1 2 pCi kg"1 are present in M3, El and Nl.

226As discussed in Section 10.3, Ra solution may be strongly in-fluenced by a-recoil following Th decay at the rock-water in-terface. The ^ Rn content of the minewaters shows that much99£

Ra must also be present at this interface and that it must(for M3) be at least at 50% of equilibrium with the U content of

99fi

the matrix. It is not possible for Ra solution to proceed to722 226

722 226

the same extent as Rn solution since Ra would become deple-ted in the rock surface and it would then be impossible to main-

222tain the Rn content of a flowing groundwater. The ratio226Ra/222Rn increases in the sequence: M3 < Nl < V2 < VI, andthe values range from 3 x 10"6 (M3) to 2 x 10"4 (VI). These verylow ratios suggest that diffusion out of uraninite is a much more

999

significant solution mechanism for ^"Rn than direct a-recoil,and this is consistent with the known porosity of uraninite inmicrofractures to the aqueous phase. In contrast, diffusion of

Ra is a very slow process and a-recoil must be much more sig-nificant. The amount of Ra present, however, is greater thanthat which the 23OTh recoil rate (Section 10.2.4) would indicate

77 f

77 ftand it is likely that geochemical factors also influence Rasolution.

Barium contents of the Stripa minewaters average about 0.03 mg/1and slight variations from this do not correlate with 22°Ra con-

9 9ft9 9fttent variations. There is no general correlation between " Racontent and the Ca-content of the minewaters. Within borehole V2,however, both the Ca"1"1" and Ra4"1" contents for water from the 406-410 m interval are approximately 50% higher than those for the356-470 m interval. In borehole VI, the groundwaters have simi-

10:22

lar Ca++ contents to the 406-410 m interval in V2 but the

contents are significantly greater than in the latter zone.

The Ra content of the waters appears to be specific for parti-cular fracture zones but does not generally correlate with eit-her Ca++ or Ba-H- contents.

11:1

11 ATMOSPHERIC AND RADIOGENIC GASES IN SOLUTION

11.1 Atmosphere derived gases

A groundwater dissolved atmospheric gases in the unsaturated zoneaccording to Henry's law from which it may be shown that:

VSTP = K p

where Vg-p is the volume of gas dissolved (measured at STP), Pis the partial pressure of the gas and K is the Bunsen coeffici-ent for the dissolved gas. The value of the constant, K, varieswith temperature, so that the volume of any gas in solution isdetermined by its partial pressure in the atmosphere and the tem-perature of equilibration of the groundwater with air. Sincethe annual average of atmospheric pressure is almost constant,the average temperature during the recharge season is the onlyother factor which can influence gas solubility. The solubilityrelationship with temperature for atmospheric gases has been well-established (Morrison and Johnstone, 1954; Benson and Krause,1976) and it is possible to determine the temperature during re-charge from measurements of the amounts of dissolved gases in agroundwater. For example, the recharge temperatures for some pa-laeo-groundwaters have been related to the palaeo-climatic recordby determination of their inert gas contents (Andrews and Lee,1979).

11.2 Radiogenic helium

The atmospheric gases which were dissolved at groundwater rechar-ge are supplemented by solution of radiogenic gases in the confin-ed zone of the aquifer. The most significant radiogenic gas isHe, produced by a-decay of tiie radioelements uranium and thori-um and their daughter nuclides. The amount of dissolved radioge-nic He has been shown to increase with groundwater age (Andrewsand Lee, 1979; Marine, 1979; Heaton and Vogel, 1981). For anaquifer of porosity, $, in which all the radiogenic He dissolv-es in the water, the He-content of the water after t years isgiven by the equation:

[He] - p$~lt(\.\9 x 10"13[u] + 2.88 x 10"1A[lh])cm3STP/cm3 H20(11.2)

where p is the bulk density, g/cm ; [u]pg/g is the natural urani-um content and [Th] yg/g is the natural thorium content of therock.

11:2

t o "i

The neutron-induced reaction Li(n,a) H followed by decay of Hproduces He at very low rates in geological formations. It maybe shown that the He/ He ratio for radiogenic helium is primari-ly dependent upon the lithium content of the rock and is charac-teristic of the formation (Andrews, 1983). It is, therefore, pos-sible to use measurements of the He/ He ratio to determine whet-her the radiogenic helium in a groundwater is characteristic ofthe aquifer or has diffused into it from other formations. Chang-es in the He/ He ratio as atmospherically derived helium mixeswith radiogenic helium may also enable groundwater ages up toabout 105 years to be determined for aquifers in which there isno ingress of helium from adjacent formations.

11.3 Radiogenic argon

The *°Ar/36Ar isotopic ratio for atmospheric Ar is 295.5. Decayof K by K-electron capture in potassic minerals of a rock mat-rix can provide an additional source of Ar which on mixingwith dissolved atmospheric Ar can increase its Ar/ Ar ratio.Alternatetion are:Alternative mechanisms which enable radiogenic Ar to enter solu-

a) direct K decay along rock surfaces in contact with thegroundwater. This is unlikely to be a significant cause of Kgeneration in times less than 10 million years since consider-able *°Ar must be formed by decay to influence the *°Ar/36Arratio of the relatively large amount of dissolved atmosphe-ric argon.

b) Leakage of radiogenic Ar stored in the minerals of therock (feldspars and micas). Diffusive leakage of Ar fromthese minerals would be aided by elevated temperatures atdepth and long groundwater residence times.

It is difficult to estimate the time required for such diffusiveargon loss from the rock minerals. The argon diffusion coeffici-ents for Ar very over some orders of magnitude for feldsparsand are very dependent upon the extent of crystal imperfectionsand impurities. Argon diffusion is also very temperature-depen-dent, the rate increasing by a factor of 10 for a 100°C tempera-ture rise. If the maximum temperature encountered by a thermalwater in a feldspar dominated system was 200°C, the time requi-red for a 1% argon loss is 10' years using the data of Dalrympleand Lanphere (1969).

11:3

11.4 Biogenic gases

Carbon dioxide is first dissolved by a groundwater in the soilzone by equilibration with the soil air in which the partialpressure of CO- is enhanced by biological activity. With subsequ-ent groundwater evolution, dissolved CO- is controlled by pH andthe Co|~, HCO3, C02 equilibria.

Reduction and thermo-degradation of organic matter can producesignificant amounts of dissolved hydrocarbons, particularly CH^,in deeply circulating, reducing or thermal groundwaters. The amo-unts of CO2 and CH^ present can have a significant influence onthe gas solubility in groundwaters and on de-gassing processes.

11.5 Analytical methods

11.5.1 amplLiiig_f or_i_ne_r t_gas_analy£e£

For inert gas analyses, it is important to adopt procedureswhich retain the gases in the same state as in the formation.The amounts of gases in solution at formation pressures general-ly exceed those resulting from air-equilibration at recharge dueto (i) solution of entrained air at increased hydrostatic pressu-res (excess air), (ii) radiogenic gas solution in the formation(He, Ar), and (iii) solution of biogenic gases (CH^, CO2) or ni-trogen formed by biodegradation. An artersian well which is allow-ed to flow naturally, is depressurised to atmospheric pressureat the surface. At this pressure the water is supersaturated andgases must be evolved. Degassing occurs in the well to a depthsuch that the hydrostatic pressure is sufficient to prevent it.This depth depends upon the extent of excess air and additionalgases dissolved in the formation and upon the temperature. Ifthe water is to be sampled without gas loss, it must be keptunder a pressure corresponding to this minimum depth which pre-vents degassing. This requires the adoption of either down holesampling or the pressurisation of the well head.

Down-hole sampling must be carried out at a depth greater thanthat required to prevent degassing. In this procedure the sampleis isolated between automatic shut-off valves once the samplingdepth is reached. It is Important to ensure that there was ade-quate flow through the sample tube before these valves are operat-ed. For a pressurised well-head, which is the method adopted inthis work, the procedure needs to be carefully controlled toavoid de-pressurisation of the well and to ensure that it is ade-quately flushed following pressurisation. This requires a comple-te flushing to a depth at least equal to the degassing depth. Fi-gure 11-1 shows the necessary well head arrangements. Valve A is

11:4

Valv* for pruiurt I flow •ajuttarnt

Fraiturt (augr

pinch-offclaapi

\

Wtll-b*ad »alv*. Do not allow th*well to b* coaplttclydc-prcisuriicd.

Figure 11-1. Arrangement for sampling groundwater under artesi-an pressure for dissolved gas analysis.

placed directly on the well head and the pressure of gauge GA ad-justed to a minimum pressure equal to the degassing pressure.The well should then be allowed to flow at this pressure to remo-ve water from the degassed zone before sampling. The sampling ap-paratus consisting of a copper tube with pinch-off clamps, pres-sure guage GB and valve B is then attached and adjusted to ob-tain the desired sampling pressure. After adequate flushing atthis pressure, the sample is isolated between the pinch-offclamps.

11:5

CalibrationAir »pike

S ml

Bulb for N^ undAr tub-sample 5 ail water sample

glass/metal seal

to diff..um c.tr c

and rotary _8Q. [gr . , „ . fp u mP 8 Kr, Xe Ar

'spectrometer

metering valvesNupro SS4H - T1 • 3welded to 1/V'O.D. S.S. tube

Figure 11-2. Gas extraction line for dissolved inert gas analy-

sis.

11.5.2 diluti^n

Samples were initially collected in 1 ctsr glass sample tubes,isolated between vacuum stopcocks. Later in the programme, samp-les were collected in 5 cm annealed copper tubes, being isolat-ed between pinch-off clamps which effected a vacuum-tight coldweld of the copper tube. In either case, the fluid was caused toflow through the sample tube to flush it completely free of air.

For analysis, the tube containing the groundwater sample was at-tached to the gas extraction line (Figure 11-2) with an O-ringseal and the system was then evacuated to better than 10 mb.A metered volume of isotopically enriched tracers for the inertgases was then admitted to the vacuum system, followed by thewater sample. The water was vaporised by warming trap A and thewater vapor excluded from the remainder of the system by coolingTrap B with solid C02/ethanol. A sub-sample for N2 and Ar analys-is was trapped in bulb E. Nitrogen (and any oxygen) were thenremoved from the main sample on a titanium getter at 800-900°C.The heavy inert gases, Kr and Xe were then adsorbed on a charco-al trap (D) at -78°C and after this Ar was adsorbed on a secondcharcoal trap (C) at -196°C. The remaining gases, He and Ne, werethen admitted to a 5 cm radium, 180° magnetic deflection massspectrometer (Kratos, MSIOS) for isotope ratio analysis. Aftermeasurement, residual He and Ne in the extraction line was pump-

11:6

ed away and the Ar/Np sub-sample was admitted for isotope ratioanalysis. Residual gases were again pumped from the extractionline and Kr + Xe were then desorbed from trap D by heating it to200°C. The isotope ratios of these gases were then determined inthe mass spectrometer.

The volumes of the inert gases released from the water samplewere calculated from the volume of tracer admitted and thechange in its isotopic ratios after mixing with the releasedgases, using the standard relationships for isotopic dilutionanalysis. The Ne content of water equilibrated with air at 0°Cis 2.3 • 10"' cm3 STP/cm 1^0 and this is the maximum dissolvedNe content which a groundwater may contain since 0°C is the mini-mum possible recharge temperature. Samples which contained Ne inexcess of this value were considered to contain entrained air inthe form of microscopic bubbles. In such cases the inert-gas con-tents were corrected by subtracting the Ne, Ar, Kr and Xe con-tents of small volumes of air until as close as possible agree-ment was obtained between the air equilibration temperatures indi-cated by the corrected contents of these gases. Since the solubi-lity temperature coefficients for these inert gases are diffe-rent, this procedure leads to a unique solution for the dissolv-ed inert gases in the sample. The equilibration temperature de-rived in this way is termed the 'recharge temperature' of thegroundwater.

11.6 Radiogenic helium

All of the Stripa minewaters contain large amounts of radiogenicHe (Table 11-1). The groundwater He content generally increas-es with depth in the granite (Figure 11-3) and the shape ofthis concentration/depth profile is very similar to those calcu-lated for diffusive loss He from He generating rock with onesurface open to the atmosphere (Figure 11-4). This suggests thatthe He content of the water is due to diffusive equilibrationbetween the rock and water phases. The amount of He generatedwithin a rock of age t years is given by:

[He] r o c k - pt {l.l9xlO~13[u]r+2.88xlO"

14[Th]r} cm3 STP/cm3 rock

(11.3)

The time required to generate the 4He concentrations observed inLhe VI/V2 groundwaters for the radioelement contents typical ofthe Stripa granite is about 200 Ma which may be compared withthe pre-Cambrian age of the granite. It is apparent that eitherthe groundwater has not totally equilibrated with the granite orthat He has been lost from the rock matrix because of prolongedgroundwater flow or by diffusion processes.

The diffusion distance from rock matrix to the water phase Inthe fracture system is so small that the fracture residence

11:7

Table 11-1. Inert gas

STP/cm3)

contents of Stripa Groundwaters (cm

Analyaie Samplingmaaber date

DepthInterval . • *K* x 10* Ne x

Eat l u t e dRecharge

in '

Borehole M3

KeenargeAr x 10* Kr x 10* Xr x 10* Temperature

1452

1452

14R5

1520

- 1

- 2

- 83

- 84

Borehole El

U55

1455

1535

- 1

- 2

- 84

Borehole Nl

1454

1454

1545

1540

- 1

- 2

- 84 .

- 84

Borehole VI

1450

1450

1450

1450

1450

1450

1450

- 1

- 2

- 3

- 4

- J

- 6

- 7

12.

4.

9.

23.

1 1 .

24.

6.

14.

19.

7.

13.

16.

4.

14.

19.

8

9.

1»

01.81

06.81

11.83

02.84

11.81

03.82

03.84

07.81

08.HI

03.84

03.84

01.81

06.81

07.81

08.81

09.81

09.81

09.81

3 -

127 -

3 -

2 -

3 -

252 -

152 -

409 -

409 -

409 -

4 -

4 -

300

130

300

300

300

300

251

506

506

506

306

506

2018723194

»1967620043265924479

6930167113

37727381R038890

1020810759

5488

528143746780

22451152971473014236

9361425267

60917286791S923

332233222132414

2910428361

2198721107

387403335867103122

16*08813335271911

371585160718145270

2994 28297042238288

8903716«?4ft15414393301

88849107998108318

2.77

2.69

6.304.853.24ft. 28

3.843.56

3.463.783.58

8.878.656.20

2.852.802.85

6.308.483.013.14

4.364.13

3.164.423.70

4.234.09A. 35

3.873.75

11.0711.03

5.842.723.23

3.302.883.50

4.003.408.19

1.80

6.07

J.383.22

2.713.102.66

3.712.01

7.908.596.087.46

6.085.72

6.326.636.28

9.699.417.83

5.295.365.33

7.108.275.485.40

6.796.27

5.896.896.09

7.227.387.37

6.856.71

9.789.20

8.918.608.62

7.866.966.00

8.337.587.85

8.928.167.68

7.747.077.206.72

7.177.637.04

1.03

15.1517.9613.2717.54

12.8812.14

12.5213.3113.04

Average

-16.6215.54

11.7312.0111.43

13.9215.2318.4411.80

Average

13.0611.89

11.4713.4520.71

14.3514.1512.31

12.11412.78

Average

1.331.98

15.3914.5614.37

13.8111.6911.35

30.9412.7237.52

27.0015-6414.78

14.3813.3314.9711.79

13.8214.2714.57

-

.52

.91

.48

.87

.83

.67

.74

.79

.64(M3)

.89

.92.96

.24

.36

.32

.64

.73

.60

.59(F.I)

.fli

.31

.841.54

.73

.74

.74

.75

.711.73

( N I )

3.71

2.701.583.70

1.701.271.70

1.811.66

2.061.891.94

1.871.771.461.66

1.791.791.63

02.1

1.93.5

2.20.71.22.6

5.34.62.2

3.42.54.2

3.33.45.53.74.0

2.3

1.91.53.9

4.24.04.1

2.22.03.6

4 . 3

1.6

1.6

3.9

2.32.74.8

- 2.5- 4.8

- 3.6- 3.1- 5.3+ 1.5

- 5.8+ 1.1

- 2.9

- 4.7- 7.9

- 4.4

- 4.4- 4.0+ 1.5

- 5.7

- 2.5

- 3.5

- 4.7

11:8

rabl e 1

Anal yt lalumber

1450

1470

1475

1480

1500 •

1505 -

- 8

- 83

• 83

- «3

- 84

- 84

L - l .

Staplingdate

21.09.01

3.10.8319.10.83

5.11.83

11.01.84

8.02.84

Cont d.

DepthInterval,

4 - 506

100 - 505

100 - 505

100 - 505

100 - 505

100 - 505

• *He x 10*

96958111139107574

174350

386026180658165920

33237i355070

188887186824

436887215006

Ne x 107

3.142.824.46

5.40

5.204.404.93

4.023.97

4.633.89

4.903.99

Ar x 10*

7.407.128.18

7.80

7.327.818.21

7.437.56

8.298.25

8.987.98

Kr

17-

15

13

121313

1313

1414

1313

x 10*

.98

.43

.61

.43

.13

.80

.18

.29

.55

.52

.99

.34Average

Xe x 10*

1.81-

1.80

1.87

1.911.971.79

.81

.87

.85

.89

.85

. 81(VI)

F.stlaitedRechargeTeaperature

2.3

3.4

2.6 -

2.3 -0 . 9 •1 .3 •

1.4 -1.1 -2.71.7

0.8 -1.0 -2.7 H

• 3.0

• 5 . 6- 2 .2- 3 .9

2.92 .0

2.92.9

h 1.3

Borehole V2

1453

1453

1453

1460

1461

1462

1463

1490

1525

1530

- 1

- 2

- 3

- 83

- 84

- 84

11.06.81

19.11.81

27.04.82

24.11.82

13.12.82

19.01.83

25.02.83

29.11.83

28.02.84

28.02.84

356 - 471

356 - 471

6 - 822

406 - 410

413 - 417

490 - 494

549 - 553

424 - 499

500 - 561

382 - 423

288818257024171585154769145127

148241138565

1154J011245492278

261812301004

89740104448

223224322474

59603

24 2680273270315015367993

1844501876791.4819

159697

--

4.315.763.83

10.3010.57

3.463.543.04

-

-

-

-

-

4.254.253.867.18

4.123.853.89

4.17

--

8.109.136.65

10.009.86

6.867.066.61

8.159.14

7.257.0

7.877.97

3.68

7.257.297.438.56

7.957.937.49

8.58

--

13.2113.77

-

15.8515.54

13.3212.6312.35

13.7714.70

13.5812.78

13.7913.41

8.01

11.1910.8213.4814.37

12.6712.9113.89

13.90

--

1.881.711.32

0.951.04

1.371.441.44

1.801.73

1.871.75

1.791.76

1.25

1.901.911.801.86

1.771.781.64

1.89Averag; (V2)

1.8 - 22.8 - 5

0 . 32 .21.9

1.81.70.4 - 23.6 - 4

3.0 - 31.8 - 35.6

1.92.7 + 1

.1

.9

.9

.5

.6

.3

.5

Hotel

1. The analytical data without any correction! are tabulated. The recharge tenperaturci were ee-tlMted fro» Ne/kr and Ne/Xe teapvratura aatchee where poaalble.

2. from aample 1460 (Noveabcr, 1982) onward», al l eeaplca were collectedtubei. Earlier eeaplei ware collected In glaai tubca.

In crimped copper

times needed to establish concentration equilibrium between the

water and rock, are short. The He flux from a uniform rock mat-

rix which loses He at one surface Is given by (Andrews, 1984):

1/2(11.A)

11:9

c

100

MC

»Ok

•00

f l

c

-

1 1 1 11

\

III

\

M3.RV\ftl

Ni N

1 1 1 1 1

\

I l f"

1

' V I

1 1 t 1 1

ic,-s le*4

*M« cent

10

. ' / c . 3 ,

Figure 11-3. The variation of the He content of Strlpa ground-waters with depth.

Figure 11-4. The dependence of the radiogenic **He content ondepth and age of rock, (for U - 1.5 ppm, Th = 4.5ppm and diffusion coefficient " 3 . 1 6 x 10 m a .

11:10

Table 11-2. 3He/4He Ratios of dissolved helium in Stripa

groundwaters.

Borehole Sampling date He/ He

VI 15.07.81 5.58 + 0.50 x 10"9

VI " 5.29 ± 0.53 x 10~9

VI " 6.28 + 0.62 x 10"9

average 5.72 + 0.42 x 10 9

where G is the generation rate of He by radioelement decay withinthe rock, D is the diffusion coefficient for He in the matrixand t is the age of the rock. If this flux of He is dissolved bygroundwater within fractures, then the time for the fracturefluid to attain the observed He concentrations may be estimated.This time is dependent upon the values adopted for the diffusioncoefficient and for the fracture width but for groundwater move-ment in 0.1 mm fractures in the Stripa granite, application ofequation 11.4 suggests that the maximum observed He-contents inVI and V2 would require rock-equilibration times of about 150years. The low He contents found in El could diffuse from therock matrix within 10 years. The El borehole has an anomalouslylow He content in its groundwater and this suggests that thereis a rapid flow of water from shallower depths in this locality.The high H content of this water also suggests that it is of re-cent origin.

The He/ He ratio of radiogenic He is principally determined bythe Li-content of the rock matrix (Andrews, 1984). The observed3He/*He ratio of dissolved He (Table 11-2) requires a Li-contentof 11 ppm in the Stripa granite, on the assumption that the Heis entirely radiogenic. This value may be compared with the aver-age analytical Li-content of 8 ppm (Table 2-3).

11.7 40Ar/36Ar ratios

Argon of atmospheric origin has an ^uAr/J0Ar ratio of 295.5. Thewaters from borehole El clearly contain atmospheric areon andthe M3 and Nl waters have only slightly enhanced ^Ar/^Ar ra-tios (Table 11-3). The added radiogenic 40Ar is about 3% oftheir atmospheric Ar content* The Ar/ Ar ratios for dissolv-ed argon in the VI and V2 groundwaters are greater than the at-

11:11

Table 11-3. 40Ar/36Ar ratios of dissolved argon in Stripagroundwaters.

Analysis

number

Samplingdate

Depth

interval, m *°Ar/36Ar

Borehole M3

1485 •1520 -

- 83- 84

Borehole El

9.23.

11.02.

8384

0 -0 -

99

301316

.6

.3+ 4

± °.4.5

1535 - 84 6.03.84 3 - 300 298.3 + 3.3294.3 + 0.5

Borehole Nl

1545 - 84 I1540 - 84 II

Borehole VI

1480 - 831500 - 84

1505 - 84

Borehole V2

1460 a1462 c1525 - 84

7.03.8413.03.84

5.11.8311.01.84

8.02.84

24.11.82

19.01.8328.02.84

252 - 300151 - 251

100 - 505100 - 505

100 - 505

500 - 561

305.2 +304.1 +

342.6 +

371.5 +353.1 +

368.1 +

346.8 +

341.0 +336.2 +

3.73.9

1.73.43.1

2.8

1.61.81.8

mospheric ratio and correspond to the addition of up to 25% ofradiogenic argon. However, diffusive loss of ^ A r from feldsparsand micas is generally very slow and at the prevailing groundtemperatures would require several million years to proceed to asignificant extent. Therefore, mineral alteration would releaseAr and such Ar-release might be proportional to groundwater sa-

linity. The Cl" contents of VI and V2 are approximately 10 timesthose of M3 and Nl, and suggest that the 40Ar release is due tomineral alteration by the migrating fracture fluids. As Ar is

11:12

produced by K decay in the micas and feldspars of the matrix,the release of Ar into the groundwaters suggests active alter-ation of feldspars at depths of at least 1 km. The VI and V2groundwaters have enhanced Ar contents and higher salinitiesthan those from M3 and Nl, suggesting that the former groundwa-ters have been produced by a longer period of chemical altera-tion of the granite.

11.8 Recharge temperatures

The inert gas contents of the Stripa groundwaters are reportedin Table 11-1. The agreement between replicates is normally bet-ter than +5% for such analyses but much larger variations areobserved in the Stripa groundwaters. These variations are attri-buted to the difficulty of obtaining samples for which the bore-holes had been adequately pressurised beforehand. All of thesamples have Ne-contents which indicate the presence of conside-rable amounts of excess air. This may be due to the entrainmentof air with the groundwater during recharge or could be causedby sample outgassing in the borehole and consequent incorpora-tion of released gases in the sample.

For the estimation of recharge temperatures, the measured Ar-con-tents were first corrected for the presence of radiogenic Aras indicated by the Ar/^ Ar ratio determined in other samplesfrom the particular borehole. Excess air corrections were thenmade by subtracting aliquots of air until the best match was ob-tained between the recharge temperatures indicated by Ne, Kr andXe. Normally, tl. > match between Ne and Ar temperatures is thebest recharge temperature indicator, but in all cases the Ar con-tents, even after correction for radiogenic Ar, indicated nega-tive recharge temperatures. This suggests that the excess gasespresent are exsolved gases rather than excess air; which wouldcause over-correction of Ne contents and undercorrection of Arcontents for recharge temperature estimation. The large He con-tents of the waters also made Ne analysis more difficult, whichalso effects the precision with which recharge temperatures canbe estimated.

The average recharge temperatures for the water from boreholesVI, V2 and M3 are indistinguishable at about 2.6 + 1.5°C (Table11-1). For the samples from boreholes Nl and El, average rechar-ge temperatures are 3.6 + 1.5°C and 4.0 + 1.1°C respectively.This trend reflects the changes in the stable Isotope composi-tion of the groundwaters.

11:13

11.9 Np/Ar ratios

The ^/Ar ratio for dissolved gases in a groundwater which hasbeen equilibrated with air is 37.70 for 10°C equilibration and37.33 for 5°C equilibration. The atmospheric N2/Ar ratio is83.54. The Nj/Ar ratios for all of the minewaters are greaterthan the expected air-equilibration value and indicate that ex-cess air has been incorporated in all of them. The ratio of thetotal dissolved Ne to the volume of Ne due to air-equilibrationhas been used as a contamination index. This index was evaluatedfor samples which were analysed for inert gas contents. Thevalue of the contamination index required to produce the observ-ed Nj/Ar ratio, is compared with the observed values from inertgas analyses in Table 11-4. The high No/Ar ratios confirm that

Table 11-4. Nj/Ar ratios of dissolved gases in Stripa groundwa-

ters.

Anal yela

number

Sampling

date

Depth

Interval, n

N2/Arratio

Contamination IndexA B

Borehole M3

1485 - »3 9.11.83

1520 - 84 23.02.84

55.8

49.051.5

3.40

2.202.59

1.63 - 1.72

1.56 - 1.681.56 - 1.68

Borehole El

1515 - 84D 6.03.84

1535 - 84E 6.03.84

3 - 3 0 0

3 - 300

48.3

55. R53.7

2.10

3.402.97

1.40 - 3.93

1.40 - 3.931.40 - 3.93

Borehole Nl

1540 - 84 II

1545 - 84 I

Borehole VI

1480 - 83

1500 - 84D

13

7

5

11

.03

.03

.11

.01

.84

.84

.63

.84

151 -

252 -

100 -

100 -

251

300

505

505

61.6

57.5

5» !

56.9

55.6

47.5

1500 - 84E

1505 - 84

11.01.84

8.02.84

100 - 505

100 - $05

44.7

50.652.8

44.946.4

5.013.80

4.223.65

3.36

1.991.66

2.442.81

1.691.86

1.69 - 1.751.69 - 1.75

1.87 - 2.001.87 - 2.00

1.76 - 1.79

.73 - 2.08

.73 - 2.08

.73 - 2.08

.73 - 2.08

.80 - 2.18

.80 - 2.18

Borehole V2

1525 - »4 28.02.M JO0 - 561 41.7

43.01.35J.4*

1.73 - 1.861.73 - 1.86

A. To correct Nj/Ar ratio to 37.7 (10°C a ir equil ibration).

I . Obaerved range fo r Inert gaa analjraee.

11 s 1A

either there is generally excess air entrainment at recharge orthat the samples generally contain exsolved gases. There are no^ / A r ratios greater than the atmospheric ratio, which would indi-cate the presence of nitrogen due to degradation of organic mat-ter.

11.10 1 5N/ 1 4N ratios

Seven groundwater samples were collected from boreholes Nl, V2,and M3 for isotopic determination of 1 5 N / 1 4 N ratio at the U.S.Geological Survey. The 6 N values (relative to air) are givenin Table 11-5. These values are identical to atmospheric valuesfor the expected uncertainty in the analyses. Hence, the dissolv-ed nitrogen in the Stripa groundwaters appears to be atmosphe-ric in origin, and there is no clear indication of nitrogenousorganic matter being decomposed from these determinations.

Table 11-5. Stable nitrogen isotopes from nitrogen gas extrac-ted from Stripa groundwaters.

Borehole

N1820830N18209O7N182D914V2820928V2820928M383O4O7M383O4O7

I n t e r v a l (m)

123-125203-205

217.1-273.14-8224-8223-103-10

615N AIR

-0.15 °/oo-0.05 °/oo-0.35 °/oo40.10 °/oo

0.00 °/oo-0.30 °/oo-0.25 °/oo

12:1

12 CONCLUSIONS

Several important conclusions can be summarized from the inves-

tigations carried out during Phase I:

A. Apparent hydraulic connections have been demonstrated betwe-en VI and V2 boreholes and between Nl borehole and the innerpart of the Buffer Mass Test area. There is no apparent hy-draulic connection between VI or V2 and Nl; neither is thereany apparent connection between El and Nl. The groundwaterchemistry between VI and V2 is very similar, supporting thehydraulic connection. However, the ion ratios of Nl are alsovery similar to VI and V2. This suggests that the same geo-chemical processes are occurring in different parts of thegroundwater system regardless of the connectivity of flowpaths. In other words, water-rock interactions are generalpatterns that are not affected by flow path except to con-trol the total amount of dissolved solids. This concept isconsistent with the suggestion that geochemical processesare largely governed by processes internal to the bedrockrather than imposed by external sources of solutes.

B. The groundwater chemistry and the total dissolved solidscannot be accounted for by unmodified seawater intrusion.The source of the salt components might be accounted for byresidual salts associated with the crystalline bedrock,such as fluid inclusions. Other hypotheses have been consi-dered, such as Holocene seawater intrusion and Permian eva-porates transported to this region. If a marine or sedimenta-ry source is postulated for the origin of the dissolvedsalts, then extreme chemical changes must be invoked to ex-plain the enormous differences in ion ratios between Stripagroundwaters and marine waters. These extreme changes aregenerally consistent with the high-temperature processes as-sociated with metamorphic reactions whose signature couldremain in residual fluid inclusions.

C. The groundwater chemistry and total dissolved solids re-flect an irregular increase with depth with distinct hetero-geneity in chloride concentrations between water-bearingfracture zones. Although a distinct increase in chlorideconcentrations occurs below 700 m, the distribution of chlo-ride values is neither gradual nor sharp, but rather irregu-lar.

12:2

D. Several water-rock interactions are occurring under present-day conditions, including the dissolution and precipitationof calcite, fluorite, barlte, and Fe(OH)3 (ferrihydrite),and the dissolution of feldspars. Carbonate geochemistry,coupled with feldspar dissolution, appears to dominatemuch of the water chemistry, especially changes in the pHand alkalinity with depth. These processes are likely to belinked to changes in the hydraulic conductivity with depth,and should be generally applicable to other granitic envi-ronments in the absence of other influences, such as the in-trusion of Holocene (or older) seawater, geothermal activi-ty, and large differences in rock composition and/or hydrau-lic conductivity profiles.

E. Chemical trends, as expressed by ion ratios, are not uniqueto Stripa. Other deep groundwaters in Sweden and Finlandhave very similar ion ratios; some are within Holocene-sea-water-intrusion limits, and some are outside. However,whether these groundwaters occur on a large regional scaleor not is unknown.

F. Redox processes involving sulfur, iron and carbon speciesmay be important in deep granitic groundwaters. The redoxcapacity of granitic groundwaters is very low, and may beoverwhelmed by Introduced radionuclides; nevertheless,evidence does suggest that disequilibrium processes, inclu-ding sulfate reduction and iron oxidation, are active in cer-tain localized parts of the groundwater system. Iron and sul-fide redox species appear to dominate redox processes. Stab-le isotopes of sulfate and SO^/Cl ratios suggest that sulfa-te reduction is more pronounced at intermediate depths andlow sulfate concentrations. The stable isotope content ofaqueous sulphate is similar to those recorded from sedimen-tary brines. Further sampling and analysis of iron, sulfur,and carbon species during Phase II investigations will bedone to clarify these findings.

G. All groundwaters have a meteoric origin and are unaffectedby evaporative, exchange or geothermal processes based onibo/lbQ a n d ZH/1H data.Deep groundwaters are depleted Tnthe heavy Isotopes of 1^0, suggesting that older waters in-filtrated under cooler environmental conditions. The lackof correlation between chloride and 1 0 stable isotope ra-tios suggests that the dissolved salts have a different ori-gin than the water Itself.

H. The mean residence time of the groundwater based on C mea-surements suggests that the deeper waters are in excess of20,000 years using the conventional approach; however, seve-ral factors could affect these values, such as the presence

12:3

of dissolved organic carbon, matrix diffusion, subsurfaceproduction, and (least likely) calcite precipitation.Aqueous carbon has a biogenic signature (based on C) whichcould suggest the deep groundwaters infiltrated during aninterstadial period or that biological processes are activein the deep groundwaters. The exceptionally low C in V2 in-dicates redox processes. Isotopic signatures of fracture-fill calcite indicates 3 or 4 types of which only those thatare modern are similar to those from other Swedish sites,such as Finnsjön and Gideå.

I. High concentrations of tritium at intermediate depths indi-cate rapid infiltration of meteoric waters. Values of 10-4 2T.U. at depths of 330-355 m (in M3, El, and F2 boreholes)must be due to rapid inflow of surface waters since their re-sidence time is calculated to be less than 60 years. This ob-servation is consistent with the chloride concentrationtrends in M3 and El boreholes; i.e., chloride concentrationdecreases as tritium concentration increases. The change inthese parameters for M3 have occurred in only the last 7years. The Nl borehole has not yet been affected by the in-trusion of young meteoric water because consistently low va-lues of 0.1-0.2 T.U. have been measured, and these valuesare most likely due to subsurface production or contamina-tion.

J. Values of about 1 T.U. in the deep groundwaters from the VIand V2 boreholes suggest subsurface production since conta-mination has been checked and found negligible while youngmeteoric input seems to be inconsistent with the other isoto-pic and chemical data. The suggestion of subsurface produc-tion is entirely consistent with theoretical calculationsand direct measurements of neutron flux in the Strlpa grani-te.

K. The Stripa site is excellent for studying the effect of sub-surface production of jH, 14C, 3tlCl and other radiolsotopesbecause the production rate is quite high (300 neutrons/cmJ/yr) or about ten times the average granite. Hence, theStripa studies can provide some idea of the upper limits ofnatural radionuclide production for granites.

L. Preliminary -^Cl measurements indicate that it is nearlyall produced in the subsurface by neutron flux. Deep ground-waters appeartohavesi^nificantlyless^^l than that re-quired for secular equilibrium. It is suggested that theselower values may reflect equilibrium values for the lepti-te, or alternatively, that chloride with a zero activity in-filtrated the bedrock and, during a residence time of lessthan 200,000 years, acquired the observed values. Further

12:4

detailed studies during Phase II should clarify these preli-minary findings. Certainly Cl promises to be a valuabletool in the elucidation of the origin and evolution of thegroundwaters.

M. Uranium concentration and activity ratio measurements inthe Stripa groundwaters have shown a general pattern withdepth that indicates: oxidizing conditions in the shallowgroundwaters (high concentrations and lowactivlty ratios),some deposition and effects of a-recoil in the intermediategroundwaters of about 350 m depth (moderately low concentra-tions and high-activity ratios), and large amounts of urani-um deposition (lowest concentrations) with variable actlvi-ty ratios indicating different degrees of evolution foreach flow path in the deep groundwaters. These data again de-monstrate that some generalizations can be made assumingevolutionary histories can be related t< depth profiles.The data also show little or no mixing between closest neigh-bor fracture zones, consistent with the chemical data.

N. Although Ra has similar chemical properties to Ba, these twoelements do not correlate, and Ra appears to increase withthe degree of evolution 01 the groundwater. A Ra recoil-based model suggests residence times of about 8,000 yearsfor VI groundwater, and about 3,000 years for V2 groundwa-ter. These calculations are not necessarily Inconsistentwith other radioisotope data, and may simply reflect diffe-rences in residence time for each element, as well as uncer-tain assumptions in the model.

0. Chemical and isotopic composition of the gases have provedvaluable in documenting the high production of radiogenic Heand Ar, demonstrating rock-water equilibration at depth, in-dicating diffusive loss of He, indicating an atmosphericorigin for non-radiogenic gases, indicating cooler environ-mental conditions during recharge of deep groundwaters con-sistent with the lii0 and H data, and Indicating local lntru-slon of young meteoric water (El borehole) consistent withthe tritium and chemical data.

P. Finally, It cannot be overemphasized that groundwater agedetermination based on single-element radioisotope determi-nations may be entirely misleading for deep groundwaters incrystalline bedrock. In fact, the concept of "groundwaterage" may not be meaningful. The results from Stripa indicatethat different elements can have different "mean residencetimes" because they may have different origins and diffe-rent processes which affect their concentrations during theevolution of the groundwater. The investigations at Stripa

12:5

have made considerable progress towards defining what thoseorigins and evolutionary processes are, and how such con-cepts may be employed towards other investigations in crys-talline bedrock.

13:1

13 ACKNOWLEDGEMENTS

A program of this scope could not have been possible without theassistance of many individuals who deserve recognition. Firstand foremost, we are deeply indebted to Hans Carlsson (SKBF/KBS)whose managerial skills and patience allowed a somewhat feistygroup of scientists to collaborate on a most exciting and chal-lenging task. We are very appreciative of Henrik Norlander whoassisted on the field sampling and who helped to coordinate thescheduling of the various research programs at the Stripa mine.

J.-L. Michelot contributed substantially to the results and in-terpretations presented in Chapter 7; M. Wolf and M. Forster con-tributed Sections 9.3 and 9.3.5, respectively. In addition, thefollowing list of individuals are acknowledged for their vari-ous contributions to the HAG members and the research at Stripa:

J.W. Ball - analysis of major and trace elements

J.M. Burchard - analysis of anions

R.J. Donahoe - analysis of major and trace elements,preparation and analysis of fluidinclusions

J. Ek - tabulation of water analyses

A. Frayne - analysis of uranium-series elements

B.F. Jones - review of Chapters 4 and 5

J.D. Hem - review of Chapters 4 and 5

L. Ranhagen - analysis of major and trace elements

W. Rauert - isotope analysis and helpful

discussions

E. Roedder - review of Chapters 2, 4 and 5

K. Skagius - preparation and leaching of granite

W. Stichler - isotope analysis and helpfuldiscussions

D.O. Whittemore - analysis of iodide

M.J. Youngman - analysis of dissolved gases

14:1

14 REFERENCES

CHAPTER 1

CARLSSON, L. and OLSSON, T., 1981:

Hydrogeological investigations in boreholes. Summary of defined

programs. Stripa Project 81-01, 1-26.

CHAPTER 2

BARKER, F., 1981:

Introduction to special issue on granites and rhyolitesr a com-mentary for the nonspecialist, J. Geophys. Res. j}6_, 10131—10135-

CARLSSON, L., OLSSON, T. and STEJSKAL, V., 1981: Core-logs of

borehole VI down to 505 m. Stripa Project 81-05, 28 pp.

CARLSSON, L., OLSSON, T. and STEJSKAL, V., 1982: Core-logs ofthe subhorizontal boreholes Nl and El. Stripa Project 82-04, 44pp.

CARLSSON, L., EGERTH, T., OLSSON, T. and WESTLUND, B., 1982:

Core-logs of the vertical borehole V2. Stripa Project 82-05, 40

pp.

FRITZ, P., BARKER, J.F. and GALE, J.E., 1979: Geochemistry andisotope hydrology of groundwaters in the Stripa granite. Resultsand preliminary interpretation. LBL-8285, SAC-12, 105 pp.

FRITZ, P., BARKER, J.F., GALE, J.E., WITHERSPOON, P.A., ANDREWS,J.N., Kay, R.L.F., LEE, D.J., COWART, J.B., OSMOND, J.K. andPAYNE, B.R., 1980: Geocheraical and Isotopic investigations atthe Stripa test site (Sweden). In: Underground Disposal of Radio-active Wastes, Vol. II, IAEA, Vienna, 341-366.

GALE, J.E., 1982: Hydrogeologic characteristics of the Stripasite. Univ. Waterloo. Res. Inst. Rep. 003C, 60 pp.

GAMBELL, A.W. and FISHER, D.W., 1966: Chemical composition ofrainfall in eastern North Carolina and southeastern Virginia,U.S. Geol. Survey Water-Supply Paper 1535-K, 41 pp.

14:2

GARRELS, R.M., 1967: Genesis of some groundwaters from igneousrocks, in Researches in Geochemistry, P.H. Abelson, ed., Vol. 2,405-420.

GEIJER, P., 1938: Stripa odalsfälts geologi. SGU, Ser Ca 28.

HASKIN, L.A., FREY, F.A., SMITT, R.A. and SMITH, R.H., 1966: Me-

teoritic, solar and terrestrial rare-earth distributions, Phys.

Chem. Earth 7, 169-321.

HASKIN, L.A. and HASKIN, M.A., 1968: Rare-earth elements in theSkaergaard intrusion, Geochim. Cosmochim. Acta 2.» 433-450.

HEIJER, K.F. and ROGERS, J.J.W., 1963: Radiometric determina-tions of thorium, uranium and potassium in basalts and in twomagmatic differentiations series. Geochim. Cocmochim. Acta, Vol.27, pp 137-154.

HENDERSON, P., ed. (1984) Rare Earth Element Geochemistry, Else-vier, 510 pp.

JACKS, G., 1978: Groundwater chemistry at depth in granites and

gneisses. KBS Tech. Rept. 88, 28 pp.

KOARK, H.J. and LUNDSTRÖM, I., 1979: Berggrundskartan LindesbergSV. SGU Ser Af 126.

LANDSTRÖM, 0., LARSSON, S-Å, LIND, G. and MALMQVIST, O., 1980:Geothermal investigations in the Bohusgranite area in Sweden.Technophysics, (in press).

LINDBLOM, S., 1984: Hydrogeological and hydrogeochemical investi-gations in boreholes - Fluid inclusion studies in the Stripa gra-nite. Stockholm University, Sweden, Stripa Project. Internal re-port 84-07, Oct. 1984.

NORDSTROM, D.K., 1983: Preliminary data on the geocheraical cha-

racteristics of groundwater at Stripa, in Geological Disposal of

Radioactive Waste: In Situ Experiments in Granite, OECD/NEA

Workshop, 143-153.

OLKIEWICZ, A., HANSSON, K., ALMN, K.-E. and GIDLUND, G., 1978:Geologisk och hydrologisk dokumentation av Stripa försöksstation.SKBF/KBS Tech. Rep. 63, 37 pp.

OLKIEWICZ, A., GALE, .I.E., THORPE, R. and PAULSSON, B., 1979:

The geology and fracture system at Stripa. LBL-8907, SAC-21.

PAULSRYD, B., 1941: Ore mapping in the Stripa Mine. StällbergsGrufve AB.

14:3

ROEDDER, E. 1958: Technique for the extraction and partial che-

mical analysis of fluid-filled inclusions from minerals, Econ.

Geol. 53, 235-269.

ROEDDER, E., INGRAM, B. and HALL, W.E. , 1963: Studies of fluidinclusions III: Extraction and quantitative analysis of inclu-sions in the milligram range, Econ. Geol. 58_, 353-374.

ROEDDER, E., 1972: Composition of fluid inclusions, U.S. Geol.Survey Prof. Paper 4AOJJ, 164 pp.

ROEDDER, E., 1984: Fluid inclusions, Revs. Mineral. j_2, Mineral.

Soc. Am., 680 pp.

STRECKEISEN, A.L., 1973: Plutonic rocks: Classification and no-menclature recommended by the IUGS Subcommission on the Systemat-ics of Igneous Rocks, Geotimes, October, 26-30.

STRECKEISEN, A.L., 1976: To each plutonic rock its proper name,

Earth Sci. Rev. 2_2> 1-33.

WELIN, E., 1964: Pitchblende-bearing vein fillings in the Stripaand Blanka iron ores. Transactions Geological Society of Sweden,Vol. 86, pp. 215-270.

WOLLENBERG, H., FLEXSER, S. and ANDERSSON, L., 1982: Petrology

and radiogeology of the Stripa pluton, LBL-11654.

CHAPTER 3

CARLSSON, A., 1979: Characteristic features of a superficialrock mass in southern central Sweden. Horizontal and subhorizon-tf>l fractures and filling material - Striae. Vol. 11, 79 pp.

CARLSSON, L., and OLSSON, T., 1985a: Hydraulic investigations inboreholes. Hydraulic testing Part 1 - Shut-in tests. Stripa Pro-ject, Technical report, In press.

CARLSSON, L., and OLSSON, T., 1985b: Hydraulic investigations inboreholes. Hydraulic testing Part 2 - Injection-recovery testsand interference tests. Stripa Project, Technical report, Inpress.

CARLSSON, L., and OLSSON, T., 1985c: Hydraulic investigations inboreholes. Final report. Stripa Project, Technical report, Inpress.

GALE, J.E., 1982: Hydrogeological characteristics of the Stripasite. University of Waterloo, Rep. 003C.

14:4

NORTON, D. and KNAPP, R., 1977: Transport phenomena In hydrother-mal systems: The nature of porosity. - American Journal of Scien-ce. Vol. 277, p. 913-936. New Haven.

OLKIEWICZ, A., GALE, J.E., THORPE, R., PAULSSON, B., 1979: Geolo-

gy and fracture system at Stripa. - SAC Report 2!.

PARSON, M.L., 1972: Determination of hydrogeological propertiesof fissured rocks. - 24th International Geological Congress.Vol. 11, p. 89-99. Montreal.

SNOW, D.T., 1968: Rock fracture spacings, openings and porositi-

es. - Journal of Soil Mechanics and Foundation Division. Vol.

94, No. 1, p. 73-91.

STRELTSOVA-ADAMS, T.D., 1978: Fluid in naturally fractured reser-voirs. - Proceedings of the Second Invitational Well-Testing Sym-posium, LBL Oct. 1978, pp 71-77.

STRELTSOVA, T.D., McKINLEY, R.M., 1984: Effect of Flow Time Dura-tion on Buildup Pattern for Reservoirs With Heterogeneous Proper-ties. - Soc. Pet. Eng. Jour., June 1984, pp 294-306.

WITHERSPOON, P.A., COOK, N.G., GALE, J.E., 1980: Progress withfield investigations at Stripa. - SAC Report 27.

CHAPTER 4

AGERSTRAND, T., HANSSON, G. and JACKS, G. , 1981: Effect ongroundwater composition of sequential flooding of aquifers withfresh and saline water. I_n Intruded and relict groundwater of ma-rine origin. Seventh Salt Water Intrusion Mtg., 59-64.

AVOTINS, P. and JENNE, E.A., 1975: The time stability of dissolv-ed mercury in water samples—II. Chemical stabilization. J. En-viron. Qual. £, 515-519.

BARKER, C , and ROBINSON, S.J., 1984: Thermal release of waterfrom natural quartz, Am. Mineral. j>9_, 1078-1081.

BRINKMANN, R. (1969) Geologic Evolution of Europe, J.E. Sanders,transl. Hafner Publ., 161 pp.

BROTZEN, F. and ASSARSSON, G., 1951: Brines in mesozoic strata,

Scania, Sweden. Int. Assoc. Hydrol. Sci., 222-223.

ENGQVIST, P., 1981: Some wells with high content of chloride incentral Sweden. Ijn Intruded and relict groundwater of marine ori-gin. Seventh Salt Water Intrusion Mtg., 33-44.

14:5

EUGSTER, H.P., 1980: Geochemistry of evaporltic lacustrine depo-sits. Ann. Rev. Earth Planet. Sci. 8, 35-63.

FRITZ, P., BARKER, J.F. and GALE, J.E., 1979: Geochemistry andisotope hydrology of groundwaters in the Stripa Granite: Resultsand preliminary interpretation. LBL-8285, SAC-12, 105 pp.

FRITZ, P., BARKER, J.F. and GALE, J.E., 1980: Summary of geoche-mical activities at the Stripa test site during FY 1979/80. Wa-terloo Res. Inst. Rept. 803-12.

FRITZ, P., BARKER, J.F. and GALE, J., 1983: Isotope hydrology at

the Stripa test site. _IH Geological disposal of radioactive

waste. In situ experiments in granite. OECD/NEA, 133-142.

GIBBS, C.R., 1976: Characterization and application of ferrozineiron reagent as a ferrous iron indicator. Anal. Chera. 8_, 1197—1200.

HYYPPX, J., 1984: Pohjaveden Kemiallinen Koostumus Suomei Kallio-

perässä. Nuclear Waste Commission of Finnish Power Companies,

Rept. YJT-84-10, 69 pp.

IVES, D.J.G. and JANZ, G.J., eds., 1961: Reference Electrodes

Theory and Practice. Academic Press, New York, Publ., 651 pp.

JACKS, G., 1973: Chemistry of some groundwaters in igneous

rocks. Nordic Hydrol. 4 207-236.

JACKS, G., 1978: Groundwater chemistry at depth in granites andgneisses. KBS Tech. Rept. 88, 28 pp.

KREMLING, K. , 1969: Untersuchungen iiber die chemische Zusammen-setzung des Meereswassers aus der Ostsee vom Friihjahr 1966. Kie-ler Meeresforschungen. _25_, 104 pp.

KREMLING, K., 1970: Untersuchungen liber die chemische Zusammen-setzung des Meerwassers aus der Ostsee. II. Friihjahr 1967 -Friihjahr 1968. Kieler Meeresforschungen. _26_, 20 pp.

KREMLING, K., 1972: Untersuchungen iiber die chemische Zusamraen-setzung des Meerwassers aus der Ostsee. III. Frlihjahr 1969 -Herbst 1970. Kieler Meeresforschungen. J8_, 99-118.

LINDEWALD, H., 1981: Saline groundwater in Sweden, ^n Intruded

and relict groundwater of marine origin. Seventh Salt Water In-

trusion Mtg., 24-32.

MEANS, J.L., 1981: The organic geochemistry of deep groundwa-ters, ONWI/Battelle Rept., 51 pp.

14:6

NORDBERG, L., 1981: Problems in Sweden with intruded and fossil

groundwater of marine origin. In Intruded and relict groundwater

of marine origin. Seventh Salt Water Intrusion Mtg., 20-23.

NORDSTROM, D.K., 1977: Thermochemical redox equilibria of Zo-Bell's solution. Geochim. Cosmochim. Acta. 4J_, 1835-1841.

NORDSTROM, D.K., 1982: Recent investigations of the major ele-ment, trace element, and isotopic geochemistry of deep graniticgroundwaters at the Stripa test site, Sweden. Ann. Mtg., Geol.Soc. Am., New Orleans, LA, p. 577.

NORDSTROM, K., 1983a: Chemical data. In Geochemical and isotopecharacterization of the Stripa groundwaters - Progress report.Stripa Project Rept. 83-01, 106-115.

NORDSTROM, D.K., 1983b: Preliminary data on the geochemical charac-teristics of groundwater at Stripa. OECD/NEA, 143-153.

SNELLMAN, M., 1982: Groundwater chemistry in bedrock. NuclearWaste Commission of Finnish Power Companies, Rept. YJT-82-42, 65pp.

SUND, B. and BERGMAN, G., 1981: Seawater intrusion in drilled

wells, ^n Intruded and relict groundwater of marine origin. Se-

venth Salt Water Intrusion Mtg., 45-58.

WHITE, D.E., HEM, J.D. and WARING, G.A., 1963: Chemical composi-tion of subsurface waters. U.S. Geol. Survey Prof. Paper 440-F,67 pp.

CHAPTER 5

BALL, J.W., JENNE, E.A. and NORDSTROM, D.K., 1979: WATEQ - acomputerized chemical model for trace and major element specia-tion and mineral equilibria of natural waters, _Tn Jenne, E.A.,ed. , Chemical modelling in aqueous systems. Am. Chem. Soc. Symp.Series 93_, 815-836.

BALL, J.W., NORDSTROM, D.K. and JENNE, E.A., 1980: Additionaland revised thermochemical data and computer code for WATEQ2—acomputerized chemical model for trace and major element specia-tion and mineral equilibria of natural waters. U.S. Geol. SurveyWater Res. Invest. 78-116, 109 pp.

BALL, J.W., JENNE, E.A. and CANTRELL, M.W., 1981: WATEQ3--A geo-chemical model with uranium added. U.S. Geol. Survey Open-FileRept. 81-1183, 81 pp.

BEHNE, W., 1953: Untersuchungen zur Geochemie des Chlor undBrora. Geochim. Cosmochim. Acta 2» 186-214.

14:7

BIRD, D.K. and HELGESON, H.C., 1981: Chemical interaction ofaqueous solutions with epidote-feldspar mineral assemblages ingeologic systems. II. Equilibrium constraints in metamorphic/geo-thermal processes. Am. J. Sci. 281, 576-614.

BIRD, D.K., SCHIFFMAN.P., ELDERS, W.A., WILLIAMS, A.E. and McDo-

well, S.D., 1984: Calc-silicate mineralization in active geother-

mal systems. Econ. Geol. 22.» 671-695.

BISCHOFF, J.L. and DICKSON, F.W., 1975: Seawater-basalt interac-tion at 200°C and 500 bars: implications for origin of sea-floorhevy-metal deposits and regulation of seawater chemistry. EarthPlanet. Sci. Letters 25, 385-397.

BOLES, J.R. and FRANKS, S.G. , 1979: Clay diagenesis in Wilcoxsandstones of southwest Texas: Implications of smectite diagene-sis on sandstones cementation. J. Sediment. Petrol. 9_, 55-70.

BRACE, W.F., SILVER, E., HADLEY, K. and GOETZE, C., 1972: Cracks

and pores: a closer look. Science 178, 162-164.

DALE, T.N., 1923: The commercial granites of New England. U.S.

Geol. Survey Bull. /38, 14-26.

DICKSON, F.W., 1977: The reaction of granite with seawater at300°C and 1000 bars. Am. Geophys. Union Trans. 8_, p. 1251.

DREVER, J.I. (1982) The Geochemistry of Natural Waters, Prentice-Hall, N.J., 388 pp.

DUCE, R.A. and HOFFMAN, EVA J., 1976: Chemical fractionation at

the air/sea interface. Ann. Rev. Earth Planet. Sci. ji, 187-228.

EDMUNDS, W.M., KAY, R.L.F. and McCARTNEY, R, 1983: Origin of sa-line groundwaters in the Carnmenellis granite: nature processand reaction during hot dry rock reservoir circulation. Proc. Fo-urth Water-Rock Interaction Symp., Japan, 127-131.

EDMUNDS, W.M., ANDREWS, J.N., BURGESS, W.G., KAY, R.L.F. andLEE, D.J., 1984: The evolution of saline and thermalgroundwaters in the Carnmenellis granite. Mineral. Mag. 8_, 402-424.

ELLIS, A.J. and MAHON, W.A.J., 1967: Natural hydrothermal sys-tems and experimental hot water/rock interactions (Part II). Geo-chim. Cosmochim. Acta 3}_, 519-538.

ELLIS, A.J. and MAHON, W.A.J., 1977: Chemistry and geothennalsystems. Academic Press, N.Y., 392 pp.

EUGSTER, H.P., 1980: Geochemistry of evaporitic lacustrine depo-sits. Ann. Rev. Earth Planet. Sci. J5, 35-63.

14:8

FABER, H., 1941: On the salt-solutions In microscopic cavitiesin granites. Dan. Geol. Unders., 44 pp.

FETH, J.H., 1981: Chloride in natural continental water—A re-view. U.S. Geol. Survey Water-Supply Paper 2176, 30 pp.

FOLK, R.L., 1955: Note on the significance of "turbid" feldspars.

Am. Mineral. 40_, 356-357.

FOURNIER, R.O. and POTTER, R.W., 1979: Magnesium correction tothe Na-K-Ca chemical geothermometer. Geochim. Cosmochim. Acta42, 1543-1550.

FREEZE, R.A. and CHERRY, J.A., 1979: Groundwater. Prentice-Hall,

Inc., Englewood Cliffs, N.J., 604 pp.

FRITZ, P., BARKER, J.F. and GALE, J.E., 1979: Geochemistry andisotope hydrology of groundwaters in the Stripa Granite: Resultsand preliminary interpretation. LBL-P285, SAC-12, 105 pp.

FRITZ, P., BARKER, J.F. and GALE, J.E., 1980: Summary of geoche-mical activities at the Stripa test site during FY 1979/80. Water-loo Res. Inst. Rept. 803-12.

FUGE, R., JOHNSON, C.C. and PHILLIPS, W.J., 1978: Iodine in gra-nitic and associated rocks. Chem. Geol. 22_, 347-352.

FUGE, R., JOHNSON, C.C. and PHILLIPS, W.J., 1978: Iodine in gra-nitic and associated rocks. Chem. Geol. _22, 347-352.

FUGE, R, 1979: Water-soluble chlorine in granitic rocks. Chem.

Geol. _25, 169-174.

GRAUSTEIN, W., 1981: The effects of forest vegetation on soluteacquisition and chemical weathering, Ph.D. Thesis, Yale, 695 pp.

HALL, W.E. and FRIEDMAN, I., 1963: Composition of fluid inclu-

sions, Cave-in-Rock fluorite district, Illinois, and Upper Mis-

sissippi Valley zinc-lead district. Econ. Geol. 8_, 886-911.

HAWKINS, D.B. and ROY, R., 1963: Experimental hydrothermal studi-es on rock alteration and clay mineral formation. Geochim. Cosmo-chim. Acta _27_, 1047-1054.

HITCHON, B., BILLINGS, G.K. and KLOVAN, J.E., 1971: Geochemistryand origin of formation waters in the western Canada sedimentarybasin. III. Factors controlling chemical composition. Geochim.Cosmochim. Acta jJ5_, 567-598.

JACKS, G., 1978: Groundwater chemistry at depth in granites andgneisses. KBS Tech. Rept. 88, 28 pp.

14:9

KAY, R.L.F., 1984: Generation of groundwater salinity by low-tem-perature granite-water interaction: a comparison of results fromtwo British studies. Proc. Intern. Groundwater Symp. GroundwaterResour. Utiliz. Contain. Hydrol., 21, Canada, 432-444.

KELLER, W.D. , BALGORD, W.D. and REESMAN, A.L., 1963: Dissolvedproducts of artificially pulverized silicate minerals and rocks:Part I. J. Sediment. Petrol. J3_, 191-204.

KERRICH, R., 1984: Chronology and ambient temperature/pressureconditions of fluid flow through the Eye-dashwa lakes plutonbased on the 18O/16O ratio and fluid inclusions. AECL Rept. TR-267, 32 pp.

KHARAKA, Y.K., SPECHT, D.J. and CAROTHERS, W.W., 1985: Low to in-termediate subsurface temperatures calculated by chemical geo-thermometers (abs.). Am. Assoc. Pet. Geol., Ann. Mtg.

K0ZL0WSKI, A. and KARWOWSKI, L., 1974: Chlorine/bromine ratio in

fluid inclusions. Econ. Geol. S9_, 268-271.

LANGMUIR, D., 1971: The geochemistry of some carbonate groundwa-ters in central Pennsylvania. Geochim. Cosmochim. Acta 35_, 1023-1045.

LINDBLOM, S., 1984: Fluid inclusion studies of the Stripa grani-

te. Stripa Project. Internal Report 84-07, Oct. 1984.

McDOWELL, S.D. and ELDERS, W.A., 1983: Allogenic layer silicateminerals in borehole Elmore #1, Salton Sea Geothermal Field, Cali-fornia. Am. Mineral. 68_, 1146-1159.

MONTGOMERY, C.W. and BRACE, W.F., 1975: Micropores in plagio-

clase. Contrib. Mineral. Petrol. _52.» 17-28.

MOORE, D.E., MORROW, C.A. and BYERLEE, J.D., 1983: Chemical reac-tions accompanying fluid flow through granite held in a tempera-ture gradient. Geochim. Cosmochim. Acta <47_, 445-454.

MOTTL, M.J. and HOLLAND, H.D., 1978: Chemical exchange during hyd-rothermal alteration of basalt by seawater. I. Experimental re-sults for major and minor components of seawater. Geochim. Cosmo-chim. Acta 4J2, 1103-1115.

NORDSTROM, D.K., JENNE, E.A. and BALL, J.W., 1979: Redox equili-bria of iron in acid mine waters. Jta Chemical Modelling inAqueous Systems: Speciation, Sorptlon, Solubility and Kinetics,Jenne, E.A., ed. Am. Chem. Soc. Symp. Series J33_, 51-79.

NORDSTROM, D.K., 1983a: Chemical data, ^n Geochemical and isoto-pe characterization of the Stripa groundwaters - Progress re-port. Stripa Project Rept. 83-01, 106-115.

14:10

NORDSTROM, D.K., 1983b: Preliminary data on the geochemicalcharacteristics of groundwater at Stripa. OECD/NEA, 143-153.

NUR, A. and SIMMONS, G., 1970: The origin of small cracks in ig-neous rocks. Int. J. Rock Mech. Min. Sci. 7_, 307-314.

PLUMMER, L.N. and BUSENBERG, E. , 1982: The solubilities of calci-

te, aragonite and vaterite in CO^-H^O solutions between 0 and

90°C, and an evaluation of the aqueous model for the system CaCO^-

CO2-H2O. Geochim. Cosmochim. Acta 1011-1040.

ROEDDER, E., INGRAM, B. and HALL, W.E., 1963: Studies of fluidinclusions III: Extraction and quantitative analysis of inclu-sions in the milligram range. Econ. Geol. 8_, 353-374.

ROEDDER, E. and SKINNER, B.J., 1968: Experimental evidence that

fluid inclusions do not leak. Econ. Geol. 63_, 715-730.

ROEDDER, E., 1972: Chapter JJ. Composition of fluid inclusions._In Data of Geochemistry, M. Fleischer Tech., ed. U.S. Geol. Sur-vey Prof. Paper 440-JJ, 164 pp.

ROEDDER, E., 1984: Fluid inclusions. Revs. Mineral. J_2, Min.

Soc. Am., 680 pp.

RYZHENKO, B.N., MEL'NIKOVA, G.L. and SHVAROV, YU. V., 1981: Com-puter modelling of formation of the chemical composition of natu-ral solutions during interaction in the water-rock system. Geo-chem. Intern. _18_, 94-108.

SCHIFFMAN, P., ELDERS, W.A., WILLIAMS, A.E., McDOWELL, S.D. and

BIRD, D.K., 1984: Active metasomatism in the Cerro Prieto geo-

thermal system, Baja California, Mexico: a telescoped low-pressu-

re low-temperature metamorphic facies series. Geology ^2, 12-15.

SCHWERTMANN, U. and TAYLOR, R.M., 1977: Iron oxides. I_n Mineralsin soil environments, J.B. Dixon and S. B. Weed, eds. Soil Scl.Am., 145-180.

SCHWERTMANN, U., 1979: Is there amorphous iron oxide in soils?

(abs.). Soil Sci. Soc. Am., Ann. Mtg.

SELLA, C. and DEICHA, G., 1962a: Etude au microscope electroniquedes pores intergranularies des gangues et des roches. Compt.Rend. Acad. Sci. Paris _25 , 2796-2798.

SELLA, C. and DEICHA, G. , 1962b: Lacunes de cristallisation etpores intergranulaires du quartz (abs.). ^n Electron Microscopy,S. Breese, Jr., ed. Academic Press.

14:11

SELLA, C. and DEICHA, G., 1963: Importance des cavites Inträ etintercristallines dans l'architecture des mlneraux et des roc-hes. J. Microscopie 2> 283-296.

SEYFRIED, W.E., Jr. and BISCHOFF, J.L., 1979: Low temperture ba-salt alteration by seawater: an experimental study at 70°C and150°C. Geochim. Cosmochim. Acta 43_, 1937-1947.

SEYFRIED, W.E., Jr. and MOTTL, M.J., 1982: Hydrothermal altera-tion of basalt by seawater under seawater-dominated conditions.Geochim. Cosmochim. Acta 46_, 985-1002.

SPRUNT, E.S. and BRACE, W.F., 1974: Direct observation of microca-

vities in crystalline rocks. Int. J. Rock Mech. Min. Sci. Geo-

mech. Abstr. J_l_, 139-150.

TAPPONNIER, P. and BRACE, W.F., 1976: Development of stress-induc-ed microcracks in Westerly granite. Int. J. Rock Mech. Min.Sci. Geomech. Abstr. 22> 103-112.

TRUESDELL, A.H., THOMPSON, J.M., COPLEN, T.B., NEHRING, N.L. andJANIK, C.J., 1981: The origin of the Cerro Prieto geothermalbrine. Geotherraics _10_, 225-238.

TULLBORG, E.-L. and LARSON, S.A., 1982: Fissure fillings fromFinnsjön and Studsvik, Sweden. SKBF/KBS Tech. Rept. 82-20, 76pp.

VINOGRADOV, A.P., 1944: Geochemistry of dissolved elements in Mos-cow waters, Uspekhi Khimii J^, 3-34.

WHITE, D.E., 1970: Geochemistry applied to the discovery, evalua-tion, and exploitation of geothermal energy resources. Geother-mics, special issue, 2_> 58-80.

WISE, D.U., 1964: Microjointing in basement, middle rocky moun-

tains of Montana and Wyoming. Geol. Soc. Am. Bull. T5_t 287-306.

ZEN, E-An and THOMPSON, A.B., 1974: Low grade regional metamorph-

ism: mineral equilibrium relations. Ann. Rev. Earth Planet.

Sci. 2, 179-212.

CHAPTER 6

ANDREWS, J.N., BALDERER, W., BATH, A.H., CLAUSEN, H.B., EVANS,G.V., PLORKOWSKI, T, GOLDBRUNNER, J. , ZOJER, H., IVANOVICH, M.and LOOSLI, H.H., 1984: Environmental isotope studies in twoaquifer systems: A comparison of groundwater dating methods.IAEA (Vienna), Symp. 270. In press.

14:12

BATH, A.H., EDMUNDS, W.M. and ANDREWS, J.N., 1979: Paleoclimatictrends deduced from hydrochemistry of a trlassic sandstone aqui-fer, U.K. In: Isotope Hydrology 1978. IAEA (Vienna) Symp. Proc.545-568.

BURGMAN, J.O., ERIKSSON, E., KOSTROV, L. and WESTMAN, F., 1981:Oxygen-18 variation in monthly precipitation over Sweden. Avdel-ning för Hydrologi, Univ. of Uppsala, Sweden. Dec. 1981, 39 pp.

CRAIG, H., 1961: Isotopic variations in meteoric waters. Scien-

ce, _[33_, 1702-1703.

DANSGAARD, W., 1964: Stable isotopes in precipitation. Tellus,

J£, 436-468.

EICHINGER, L, RAUERT, W., STICHLER, W. , BERTLEFF, B., EGGER, R.,1984: Comparative study of different aquifer types in central Eu-rope using environmental isotopes. IAEA, (Vienna), Symp. 270,Sept. 1983. In press.

FONTES, J.-C, 1981: Paleowaters. In: Stable Isotope Hydrology.

IAEA Tech. Rep. No 210, 273-302.

FRITZ, P. and FRAPE, S.K., 1982: Saline groundwaters in the Cana-dian Shield, a first overview. Chem. Geol. J6_, 179-190.

FRITZ, P, BARKER, J.F. and GALE, J., 1983: Isotope Hydrology atthe Stripa Test Site. IN: Geological Disposal of RadioactiveWaste, in situ Experiments in Granite. OECD and KBS, Paris 1983,133-142.

FRITZ, P., BARKER, J.F. and GALE, J.E., 1979: "Geochemistry andIsotope Hydrology of Groundwaters in the Stripa Granite," Univ.of California, Lawrence Berkeley Laboratories, Berkeley, CA,Rep. LBL-8285, pp. 107.

FRITZ, P., BARKER, J.F., GALE, J.E., ANDREWS, J.N., KAY, R.L.F.,LEE, D.J., COWART, J.B., OSMOND, J.K., PAYNE, B.R. and WITHERSPO-ON, P.K., 1980: "Geochemical and Isotopic Investigation at theStripa Test Site (Sweden), "Proc. Symp. Underground Disposal ofRadioactive Wastes, IAEA, Otaniemi, Finland, July 1979, Symp.243/6, 341-365.

FRITZ, P., DRIMMIE, R.J. and O'SHEA, K. Deuterium, tritium and

0 in precipitations over Canada. In preparation.

GAT, J.R. and CARMI, I., 1970. Evolution of the isotopic composi-tion of atmospheric waters in the Mediterranean Sen. J. Geophys.Res., 75, 3039-3048.

14:13

CONFIANTINI, R., CONRAD, G., FONTES, J.C., SANZAY, G. and PAYNE,B.R., 1974: Etude isotopique de la nappe du Continental Interca-lalre et de ses relations avec les autres nappes du Sahara sep-tentrional. Isotope Techniques in Hydrology 1974. Proc Symp.Vienna, 1974, Vol. I, 227-241.

HÖTZL, H., JOB, C , MOSER, H., RAUERT, W., STICHLER, W. and

ZÖTL, J.G.: 1980. Isotope methods as a tool for Quaternary studi-

es in Saudi Arabia. In: Arid Zone Hydrology. IAEA (Vienna),

Panel Proc. 1980, 215-235.

IAEA. 1981: Stable Isotope Hydrology. Deuterium and oxygen-18 inthe water cycle. IAEA, (Vienna). Tech. Rep. No 210.

IAEA. 1983a: Guidebook on Nuclear Techniques in Hydrology. IAEA

(Vienna), Techn. Rep. 91.

IAEA. 1983b: Isotope techniques in the Hydrogeological Assess-ment of Potential Sites for *.he Disposal of High-Level Radioacti-ve Wastes. IAEA (Vienna), Techn. Rep. 228.

MERLIVAT, L. and JOUZEL, J., 1979: Global climatic interpreta-tion of the deuterium-oxygen-18 in the water cycle. J Geophys.Res., _34_, 18, 5029-5033.

MOSER, H., STICHLER, W. and TRIMBORN, P., 1983: Stable isotopestudies on paleowaters. In: Paleoclimatic and paleowaters. IAEA,Vienna, 1983, 201-204.

SAVIN, S.M., 1980: Oxygen and hydrogen isotope effects in low—temperature mineral water interactions. In: Handbook of Environ-mental Isotope Geochemistry, P. Fritz and I.C. Fontes. Edts. El-sevier, Publ. Co., Amsterdam. Vol. 1, 283-328.

TRUESDELL, A.H. and HULST0N, J.R.: 1980: Isotopic evidence on en-vironments of geothennal systems. In: Handbook of EnvironmentalIsotope Geochemistry, P. Fritz and J.C. Fontes Eds. ElsevierPubl. Co., Amsterdam, Vol. 1, 179-226.

YURTSEVER, Y. and GAT, J.R., 1981: Atmospheric waters. In StableIsotope Hydrology. IAEA, Vienna, Technical Rep. No. 210, 103-142.

CHAPTER 7

CLAYPOOL, G.E., HOLSER, W.T., KAPLAN, I.R., SAKAI, H. and ZAK,I., 1980: The age curve of sulfur and oxygen isotope in marinesulfate and their mutual interpretations. Chem. Geology, 28:199-260.

14:14

CORTECCI, G. and LONGINELLI, A., 1970: Isotopic composition ofsulphate in rainwater. Pisa, Italy. Earth Planet. Sci. Lett.,8:36-40.

CORTECCI, G., 1973: Oxygen-isotope variation in sulfate ions inthe water of some Italian lakes. Geochim. Cosmochim. Acta,37:1531-1542.

FONTES, J.CH. and MICHELOT, J.L., 1983: Stable isotopes geoche-mistry in groundwater systems from Stripa site. Report 83-01(Chap. 4) SKBF-KBS. Stockholm.

FONTES, J.CH. and GUENDOUZ, A., 1984: Deuterium, oxygéne 18,carbone 13, carbone 14, soufre 34 dans les eaux, le carbonate etle sulfate dissous des nappes profondes du Sahara. Proc. Symp.Etude des ressources en eaux du Sahara. IAEA. Vienna (to be pub-lished).

FRIEDMAN, I. and O'NEIL, J.R., 1977: Compilation of stable isoto-pe fractionation factors of geochemical interest. In: M. Flei-scher (Ed). Data of Geochemistry. U.S. Geol. Sur., Prof. Paper,440-KK:l-12 (6th ed).

FRITZ, P., BARKER, J.F., GALE, J.E., WITHERSPOON, P.A., ANDREWS,J.N. , KAY, R.L.F., LEE, D.J., COWART, J.B., OSMOND, J.K. andPAYNE, B.R., 1980: Geochemical and isotopic investigations atthe Stripa test site (Sweden). In: Underground Disposal of Radio-active Wastes, vol. II, IAEA, Vienna: 341-366.

FRITZ, P., BARKER, J.F. and GALE, J.E., 1980b: Summary of Geoche-mical Activities at the Stripa Test Site during FY 1979/80. Sub-mitted to LBL.

FRITZ, P. and FONTES, J.CH., 1980: Handbook of Environmental Iso-

tope Geochemistry. Vol. I, Elsevier, Amsterdam.

FRITZ, P., BARKER, J.F. and GALE, J.E., 1983: Isotope hydrologyat the Stripa test site. Proc. Symp. Geological Disposal of Ra-dioactive Waste. In Situ Experiments in Granite. OECD/NEA.Paris. 133-142.

FRITZ, P., 1983: The isotopic composition of sulphur compounds

in the hydrosphere (in prep.).

GIKELSON, R.H., PERRY, E.C. and CARTWRIGHT, K., 1981: Isotopicand geologic studies to identify the sources of sulfate ingroundwater containing high barium concentrations. Report 81-01165.University of Illinois, Water Res. Cent.

KAMINENI, D.C., 1983: Sulphur-isotope geochemistry of fracture-filling gypsum in an archean granite near Atikokan, Ontario, Ca-nada. Chem. Geol., 39: 263-272.

14:15

KANKAINEN, T. and HYYPPÄ, J., 1984: Isotoplc analyses of ground-water from the Rapakivi granite, Southern Finland, for the ma-nagement of radioactive waste» Proc Int. Symp. Isotope Hydrologyin Water Resources Development. IAEA, Vienna, 801-803.

KAPLAN, I.R. and RITTENBERG, S.C., 1964: Microbiological fractio-

nation of sulfur isotopes. J. Gen. Microbiol., 34:195-212.

KROOPNICK, P.M. and CRAIG, H., 1972: Atmospheric oxygen: Isoto-

pic composition and solubility fractionation. Science, 175:54-

55.

KROOPNICK, P.M. and CRAIG, H., 1976: Oxygen isotope fractiona-tion in dissolved oxygen in the deep sea. Earth Planet. Sci.Lett., 32:375-385.

KROUSE, H.R., 1980: Sulphur isotopes in our environment. In: P.

Fritz and J.Ch. Fontes (Eds). Handbook of Environmental Isotope

Geochemistry. Vol. I, Elsevier, Amsterdam: 435-471.

LLOYD, R.M., 1967: Oxygen 18 composition of oceanic sulfate.Science, 156:1228-1231.

LLOYD, R.M., 1968: Oxygen isotope behaviour in the sulphate-

water system. Jour. Geoph. Res., 73:6099-6110.

LONGINELLI, A. and CRAIG, H., 1967: Oxygen 18 variations in sul-fate ions in seawater and saline lakes. Science, 156:56-59.

MICHELOT, J.L., BENTLEY, H.W., BRISSAUD, I., ELMORE, D. and FON-TES, J.CH., 1984: Progress in environmental isotope studies(36Cl, ^ S , 180) at the Stripa site. Proc. Int. Symp. IsotopeHydrology in Water Resource Development. IAEA. Vienna, 207-229.

MIZUTANI, Y. and RAFTER, R.A., 1969a: Oxygen isotopic composi-tion of sulfates. 3. Oxygen isotopic fractionation in the bisul-fate-water system. N.Z.J. Sci., 12:54-59.

MIZUTANI, Y. and RAFTER, R.A., 1969b: Oxygen isotopic compositionof sulphates. 4. Bacterial sulphate and the oxidation of sul-phur. N.Z.J. Sci., 12:60-68.

MIZUTANI, Y. and RAFTER, R.A., 1969c: Oxygen isotopic composi-tion of sulphates. 5. Isotopic composition of sulfate in rainwa-ter, Gracefield, New Zealand. N.Z.J. Sci., 12:69-80.

NIELSEN, H. and RICKE, W., 1964: Schwefel-IsotopenverhMltnissevon Evaporiten aus Deutschland; ein Beitwag zur Kenntnis von63^S in Meerwasser-Sulfat. Geochim. Cosmochim. Acta, 28:577-591.

NORDSTROM, K., 1983: Conceptual framework for the chemical pro-cesses in the Stripa groundwaters. Rep. 83-01 (Chap. 5) SKBF-KBS. Stockholm.

14:16

PEARSON, F.J., Jr., RIGHTMIRE, C.T., 1980: Sulphur and oxygen iso-topes In aqueous sulphur compounds. In: P. Fritz and J.Ch. Fon-tes (Eds). Handbook of Environmental Isotope Geochemistry. Vol.I, Elsevier, Amsterdam: 227-258.

PIERRE, C , 1982: Teneurs en isotopes stables (180, 2H, 13C,

S) et conditions de genése des évaporites marines: applica-

tions å quelques milieux actuels et au Messinien de la Méditer-

ranée. These Doct. Sc. Nat. Université Paris-Sud, Orsay, 266

P*

RAFTER, R.A. and MIZUTANI, Y., 1967: Oxygen isotope compositionof sulfates. 2. Preliminary results on oxygen isotopic variationin sulfates and relationship of their environment and to their<534S values. N.Z.J. Sci., 10:816.

RIGHTMIRE, C.T., PEARSON, F.J., Jr., BACK, W., RYE, R.O. and HAND-SHAW, B.B., 197A: Distribution of sulphur isotopes of sulphatesin groundwaters from the principal artersian aquifer of Floridaand the Edwards aquifer of Texas. U.S.A. In: Isotope Techniquesin Groundwater Hydrology. Vol. 2, IAEA, Vienna: 191-207.

THODE, H.G. and MONSTER, J., 1965: Sulfur-isotope geochemistryof petroleum evaporites and ancient seas. In: A. Young and J.E.Galley (Eds). Fluids in Subsurface Environments. Am. Assoc.Petr. Geol. Mem., 4:367-377.

TREMBACZOWSKI, A. and HALAS, S., 1984: Oxygen and sulfur isotoperatio in sulfate from atmospheric precipitation, Lublin, Poland.Proc. Int. symp. Isotope Hydrology in Water Resources Develop-ment. IAEA, Vienna, 819-820.

ZAK, I., SAKAI, H. and KAPLAN, I.R., 1980: Factors controllingthe 1*W*"O and 3*S/3^S isotopes ratios of ocean sulfates, evapo-rites and interstitial sulfates from modern deep sea sedimentIn: Isotope Marine Chemistry. Uchida Rokakuho, tokyo: 339-373.

ZIEGLER, P.A., 1982: Paleogeography of western and central Euro-pe. Elsevier, Amsterdam.

CHAPTER 8

CARLSSON, L. and OLSSON, T., 1983: Geological and hydrogeologi-cal characterization of the Stripa granite. SKBF/KBS. InternalReport 83-01.

FONTES, J.C. and GARNIER, J.M., 1979: Determination of the initi-al C-14 activity of the total dissolved carbon, a review of theexisting models and a new approach. Water Resour. Res. 15, pp399-413.

14:17

FRITZ, P., BARKER, J.F. and GALE, J.E., 1979: Geochemistry andisotope hydrology of groundwaters in the Stripa granite. LBL-8285, SAC-REPORT 12, 105 pp.

FRITZ, P., BARKER, J.F. and GALE, J.E., 1980: Summary of Geoche-mical Activities at the Stripa Test Site during F.Y. 1979/1980.Waterloo Research Inst., Rep. 803-12, 1980.

GEYH, M.A., 1972: On the determination of the dilution factor ofgroundwater. Proc 8th Int. Conf. Radiocarbon Dating, New Zea-land, pp 369-380.

M0ZET0, A.A., FRITZ, P. and PEARSON, E.J., 1984: Experimental ob-servations on carbon isotope exchange in carbonate-water sys-tems. Geochim. Cosmochim. Acta, 48, pp 495-504.

NERETNIEKS, I., 1980: Diffusion in the rock matrix, an importantfactor in radionuclide retardation? J. Geophys. Res. 85, B8,4379-81.

PLUMMER, L.N., JONES, B.F. and TRUESDELL, A.H., 1976: WATEQF, a

FORTRAN IV version of WATEQ, a computer programme for calcula-

ting chemical equilibrium of natural waters. U.S. Geol. Survey,

Water Resour. Invest. 76-13.

REIMER, J., 1980: In Fritz et. al., 1980.

REARDON, E.J. and FRITZ, P., 1978: Computer modelling of groundwater C-13 and C-14 isotope compositions. Jour. Hydrol. JI6_, pp201-224.

TULLBORG, E.L. and LARSON, S.A., 1982: Fissure fillings fromFinnsjön and Studsvik, Sweden. Identification, chemistry and da-ting. SKBF/KBS Techn. Report 82-20.

TULLBORG, E.L. and LARSON, S.A., 1983: Fissure filling from

Gideå, Central Sweden. SKBF/KBS Techn. Report 83-74.

WIGLEY, T.M.C., PLUMMER, L.N. and PEARSON, F.J. Jr., 1978: Masstransfer and carbon isotope evolution in natural water systems.Geochim. Cosmochim. Acta, , pp 1117-1139.

ZITO, R, DONAHUE, D.J., DAVIS, S., BENTLY, H. and FRITZ, P.,1980: Possible subsurface production of C-14. Geophys. Res.Lett. 7U_, pp 235-238.

14:18

CHAPTER 9

AIREY, P.L., BENTLEY, H., CALF, G.E., DAVIS, S.N., HABERMEHL,M.A., PHILLIPS, F., SMITH, J., TORGERSEN, T. , ELMORE, D. andGROVE, H., 1983: Isotope hydrology of the great artesian basin.Proc. Int. Conf. Groundwater and Man. December 1983. Sydney (tobe published).

ANDREWS, J.N., 1983: Radioelements and inert gases in the Strlpa

groundwaters. Report 83-01 (Chap. 3). SKBF-KBS. Stockholm.

ANDREWS, J.N.L. and KAY, R.L.F., 198?: Natural production of tri-tium in permeable rocks. Nature 298, 361-363.

BENTLEY, H.W., 1978: Some comments on the use of chlorine-36 fordating very old groundwater: Proc. Symp. Dating of Very OldGroundwater. Tucson, Arizona. 102.

BENTLEY, H.W., PHILLIPS, F.M., DAVIS, S.N., GIFFORD, S., ELMORE,D., TUBBS, L. and GOVE, H.E., 1982: The fallout of thermonuclear36C1. Nature, 300, 737.

BENTLEY, H.W., PHILLIPS, F.M., DAVIS, S.N., 1984: Chlorine-36 inthe terrestrial environment. In Handbook of Environmental Isoto-pe Geochemistry (FRITZ, P. and FONTES, J.Ch., eds). Elsevier,Amsterdam (in press).

BRERETON, N.R., 1970: Corrections for interfering isotopes inthe 4OAr/39Ar dating method, Earth Planet. Sci. Letters J5, 427-433.

CARLSSON, L., OLSSON, T., 1982: Stripa hydrogeochemistry - compi-lation of analysing results, Internal Report Stripa Project,Stockholm, pp. 10.

CLARKE, W.B., JENKINS, W.J., TOP, Z., 1976: Determination of tri-

tium by mass spectrometric measurement of He, Int. J. Appl.

Rad. Isot. _27, 515-522.

EICHINGER, L. , FORSTER, M., RAST, H. , RAUERT, W., WOLF, M.,1981: Experience gathered in low-level measurement of tritium inwater. - In: Low-Level Tritium Measurement, IAEA-TECDOC-246,Vienna, 43-64.

ELMORE, D., TUBBS, L.E., NEWMAN, D., MA, X.Z., FINKEL, R., NISHI-

IZUMI, K., BEER, J., OESCHGER, H. and ANDREE, M., 1982: 36C1

bomb pulse measured in a shallow ice core from Dye 3, Greenland.Nature, J300, 735-737.

FEIGTI, Y., 0LTMANN, B.G., KASTNER, J., 1968: Production rates ofneutrons in soils due to natural radioactivity, J. Geophys. Res.73, 3135-3142.

14:19

FLORKOWSKT, T., 1984: personal communication.

FONTES, J-Ch., BRISSAUD, I. and MICHELOT, J.L., 1984: Hydrogeolo-gical implications of deep production of chlorine-36. Proc. 3rdInt. Symp. Accelerator Mass Spectrometry. Nuclear Inst. Met.,B5, 303-307.

FONTES, J-Ch. and MICHELOT, J.L., 1983: Stable isotope geochemi-stry of groundwater systems from Stripa. Report 83-01 (Chap. 4).SKBF-KBS. Stockholm.

FRITZ, P., 1983: personal communication.

FRITZ, P., BARKER, J.F., GALE, J.E., 1979: Geochemistry and iso-tope hydrology of groundwaters in the Stripa granite, Univ. ofCalifornia, Lawrence Berkeley Laboratories, Berkeley, CA. Rep.LBL-8285, pp 107.

LAL, D. and PETERS, S., 1967: Cosmic ray produced radioactivityon the earth. In: Handbuch der Physik, 46/2 (SITTE, K., ed.).Springer-Verlag, Berlin. 551.

LOOSLI, H., FORSTER, M., 1982: Ein Beitrag zura Vergleich vonAr- und C-Konzentrationen im Grundwasser. - In: Beiträge

iiber hydrologische Tracermethoden und ihre Anwendungen, GSF-Be-richt R 290, Munich/Neuherberg, 133-138.

MICHELOT, J.L., BENTLEY, H.W., BRISSAUD, I., ELMORE, D. and FON-

TES, J-Ch., 1984: Progress in environmental isotope studies

(36C1, 3*S, l80) at the Stripa site. Proc. Int. Symp. Isotope Hy-

drology in Water Resources Development. IAEA, Vienna, 207-229.

MOSER, H., RAUERT, W., 1983: Determination of groundwater move-ment by means of environmental isotopes - State of the art. IAHS-Symposium "Relation of Groundwater Quantity and Quality", Ham-burg, IAHS Publ. (in print).

NERETNIEKS, I., 1981: Age dating of groundwater in fissured

rock: Influence of water volume in micropores, Wat. Resourc.

Res. J7, 421-422.

ROETHER, W., 1967: Tritium im Wasserkreislauf, Thesis Univ. Hei-delberg.

WEISS, W., BULLACHER, J., ROETHER, W., 1979: Evidence of pulseddischarges of tritium from nuclear energy installations in Cent-ral European precipitation. - In: Behaviour of Tritium in the En-vironment, IAEA, Vienna, 17-30.

WEISS, W., ROETHER, W., 1975: Der Tritiumabfluss des Rheins 1961 -1973, Deutsche Gewässerkundliche Mitteilungen 21» 1~5.

14:20

WEISS, W., ROETHER, W., BADER, G., 1976: Determination of blanksin low-level tritium measurement, Int. J. Appl. Rad. Isot. 27,217-225.

WOLF, M., RAUERT, W., WE1GEL, F., 1981: Low-level measurement oftritium by hydrogenation of propadiene and gas counting of propa-ne, Int. J. Appl. Rad. Isot. 2£» 919-928.

WOLLENBERG, H., FLEXSER, S. and ANDERSON, L., 1980: Petrology andradiogeology of the Stripa pluton. Laurence Berkeley Laboratoryrep. 11654 Berkeley, California.

ZIEGLER, P.A., 1982: Paleogeography of Western and Central Euro-pe. Elsevier, Amsterdam.

CHAPTER 10

ANDREWS, J.N., 1983: "Radioeleraents and inert gase<= «" rhe Stri-pa groundwaters". SKBF/KBS report 83-01, Stockjlm.

ANDREWS, J.N., GILES, I.S., KAY, R.L.F., LEE, D.J., OSMOND,J.K., COWART, J.B., FRITZ, P., BARKER, J.F. and GALE, J., 1982:"Radioelements, radiogenic helium and age relationships forgroundwaters from the granites at Stripa, Sweden". Geochim. Cosmo-chim. Acta, 6, 1533-1543.

ANDREWS, J.N. and WOOD, D.F., 1972: "Mechanism of radon release

in rock matrices and entry into groundwaters". Inst. Min. Me-

tall. Trans., Sect. B81_, 198.

EDMUNDS, W.M., ANDREWS, J.N., BURGESS, W.G., KAY, R.L.F. andLEE, D.J., 1984: "The evolution of saline and thermal groundwa-ters in the Carnmenellis granite". Min. Kag., 48 407-4 274.

FONTES, J.C. and MICHEL0T, J.L., 1983: "Stable isotope geoche-mistry in groundwater systems from Stripa". SKBF/KBS report 83-01,Stockholm.

KIGOSHI, K., 1971: "Alpha-recoil 234Th: dissolution into waterand the 234U/238U disequilibrium in nature". Science _173 47-48.

LANGMUIR, D., 1978: "Uranium solution - mineral equilibria atlow temperatures with applications to sedimentary ore deposits.Geochim. Cosmochim. Acta 4_2 547-569.

NELSON, P., PAULSSON, B., RACHIELE, R. , ANDERSSON, L., SCHRAUF,T., HUSTRULID, W., DURAN, D. and MAGNUSSON, K.A., 1979: "Prelimi-nary report on the geophysical and mechanical borehole measure-ments at Stripa". Report LBL-8280, Lawrence Berkeley Lab., Univ.California.

14:21

ROSHOLT, J.N., SHIELDS, W.R. and GARNER, E.L., 1963: "Isotopic

fractionation of uranium in sandstone". Science 139 224-226.

TANNER, A.B., 1964: "Radon migration in the ground: a review".In The Natural Radiation Environment, (eds. J.A.S. Adams andW.M. Lowder), Chicago University Press, Chicago.

CHAPTER 11

ANDREWS, J.N., 1984: "The isotopic composition of radiogenic he-lium and its use to study groundwater movement in confined aqui-fers". Chem. Geol. in press.

ANDREWS, J.N., 1983: "The isotopic composition of helium and itsuse to study groundwater movement". Proc. 4th Int. Symp. Water-Rock Interaction, Misasa, Japan, 17-21.

ANDREWS, J.N. and LEE, D.J., 1979: "Inert gases in groundwaterfrom the Bunter Sandstone of England as indicators of age and pa-laeoclimatic trends". J. Hydrol. M_ 233-252.

BENSON, B.B. and KRAUSE, D., 1976: "Empirical laws for dilute

aqueous solutions of non-polar gases". J. Chem. Phys., 4_ 639.

DALRYMPLE, G.B. and LANPHERE, M.A., 1969: "Potassium-Argon Da-ting". W.H. Freeman and Company, San Fransisco.

HEATON, T.H.E., and VOGEL, J.C., 1981: "Excess air' in groundwa-

ter". J. Hydrol. _5£ 201-216.

MARINE, I.W., 1979: "The use of naturally occurring helium to es-timate groundwater velocities for studies of geologic scorage ofradioactive waste." Water Resour. Res. Jj 1130.

MORRISON, T.J. and JOHNSTONE, N.B., 1954: "Solubilities of inertgases in water". J. Chem. Soc, p. 3441.

Strips Project - Previously Published Reports

1980TR 81-01"Summary of defined programs"L Carlsson and T OlssonGeological Survey of Sweden, UppsalaI NeretnieksRoyal Institute of Technology. StockholmR PuschUniversity of LuleåSweden November 1980

1981TR 81-02"Annual Report 1980"Swedish Nuclear Fuel Supply Co/Division KBSStockholm, Sweden 1981

IR 81-03"Migration in a single fracturePreliminary experiments in Stripa"Harald Abelin. Ivars NeretnieksRoyal Institute of TechnologyStockholm, Sweden April 1981

IR 82-02"Buffer Mass Test - Data Acquisition andData Processing Systems"B HagvallUniversity of Luleå, Sweden August 1982

IR 82-03"Buffer Mass Test - Software for the DataAcquisition System"B HagvallUniversity of Luleå, Sweden August 1962

IR 82-04"Core-logs of the SubhorizontalBoreholes N1 and E1"L Carlsson, V StejskalGeological Survey of Sweden, UppsalaT OlssonK-Konsult, Engineers and Architects, StockholmSweden August 1982

IR 81-04"Equipment for hydraulic testing"Lars Jacobsson, Henrik NorlanderStallbergs Grufve ABStripa. Sweden July 1981

IR 81-05Part I "Core-logs of borehole VIdown to 505 m"L Carlsson, V StejskalGeological Survey of Sweden, UppsalaT OlssonK-Konsult, Stockholm

°art II "Measurement of Triaxial rockstresses in borehole VI"L Strindell, M Anr'erssonSwedish State Po.ve* Boa'd, StockholmSweden July 1981

1982TR 82-01"Annual Report 1981"Swedish Nuclear Fuel Supply Co/Division KBSStockholm, Sweden February 1982

IR 82-05"Core-logs of the Vertical Borehole V2"L Carlsson. T Eggen, B WestlundGeological Survey of Sweden, UppsalaT OlssonK-Konsutt. Engineers and Architects. StockholmSweden August 1982

IR 82-06"Buffer Mass Test - Buffer Materials"R Pusch. L BörgessonUniversity of LuleåJ NilssonAB Jacobson & Widmark, LuleåSweden August 1982

IR 82-07"Buffer Mass Test - Rock Drilling andCivil Engineering"R PuschUniversity of LuleåJ NilssonAB Jacobson & Widmark, LuleåSweden September 1982

IR 82-08"Buffer Mass Test - Predicitions of thebehaviour of the bentonite-based buffermaterials"L BorgessonUniversity of LuleåSweden August 1982

1983IR 83-01"Geochemical and isotope characteriza-tion of the Stripa groundwaters -Progress report"Leif Carlsson.Swedish Geological, GöteborgTommy Olsson,Geological Survey of Sweden, UppsalaJohn Andrews.University of Bath, UKJean-Charles Fontes,Université, Paris-Sud, Paris, FranceJean L Michelot,Université, Paris-Sud, Paris, FranceKirk Nordstrom,United states Geological Survey. Menlo ParkCalifornia, USAFebruary 1983

TR 83-02"Annual Report 1982"Swedish Nuclear Fuel Supply Co/ Division KBSStockholm. Sweden April 1983

IR 83-03"Buffer Mass Test - Thermal calculationsfor the high temperature test"Sven KnutssonUniversity of LuleåSweden May 1983

IR 83-04"Buffer Mass Test - Site Documentation"

Roland PuschUniveristy of Luleå and Swedish State Power BoardJan NilssonAB Jacobson & Widmark. Luleå,Sweden October 1983

IR 83-05"Buffer Mass Test - Improved Models forWater Uptake and Redistribution in theHeater Holes and Tunnel Backfill"R PuschSwedish State Power BoardL Börgesson, S KnutssonUniversity of LuleåSweden, October 1983

IR 83-06"Crosshole Investigations — The Use ofBorehole Radar for the Detection of Frac-ture Zones in Crystalline Rock"Olle OlssonErik SandbergSwedish GeologicalBruno NilssonBoliden Mineral AB, SwedenOctober 1983

1984TR 84-01"Annual Report 1983"Swedish Nuclear Fuel Supply Co/Division KBSStockholm, Sweden, May 1984.

IR 84-02"Buffer Mass Test — Heater Designand Operation"Jan NilssonSwedish Geological CoGunnar RamqvistEI-teknoABRoland PuschSwedish State Power BoardJune 1984

IR 84-03"Hydrogeological and HydrogeochemicalInvestigations—Geophysical BoreholeMeasurements"Olle OlssonAnte JämtlidSwedish Geological Co.August 1984

IR 84-04"Crosshole Investigations—PreliminaryDesign of a New Borehole Radar System"O. OlssonE. SandbergSwedish Geological Co.August 1984

IR 84-05"Crosshole Investigations—EquipmentDesign Considerations for SinusoidalPressure Tests"David C. HolmesBritish Geological SurveySeptember 1984

IR 84-06"Buffer Mass Test — Instrumentation"Roland Pusch, Thomas ForsbergUniversity of Luleå, SwedenJan NilssonSwedish Geological, LuleåGunnar Ramqvist, Sven-Erik TegelmarkStripa Mine Service, StoraSeptember 1984

IR 84-07"Hydrogeological and Hydrogeochemical"Investigations in Boreholes — FluidInclusion Studies in the Stripa GraniteSten LindblomStockholm University, SwedenOctober 1984

IR 84-08"Crosshole investigations — Tomographyand its Application to Crosshole SeismicMeasurements"Sven IvanssonNational Defence Research Institute,SwedenNovember 1984

1985IR 85-01Borehole and Shaft Sealing — SitedocumentationRoland PuschJan NilssonSwedish Geological CoGunnar RamqvistElteknoABSwedenFebruary 1985

IR 85-02Migration in a Singl9 Fracture-Instrumentation and site descriptionHarald AbelinJardGidlundRoyal Institute of TechnologyStockholm, SwedenFebruary 1985

IR 85-04Hydrogeological and HydrogeochemicalInvestigations in Boreholes—Compilation of geological dataSeje CarlstenSwedish Geological CoUppsala, SwedenJune 1985

IR85-05Crosshole Investigations—Description of the small scale siteSeje CarlstenKurt-Åke MagnussonOlle OlssonSwedish Geological CoUppsala, SwedenJune 1985

IR 85-03Final Report of the Migration in a SingleFracture — Experimental results andevaluationH. AbelinI. NeretnieksS. TunbrantL. MorenoRoyal Institute of TechnologyStockholm, SwedenMay 1985