International Journal of Coal Geology 86 (2011) 12–19

Contents lists available at ScienceDirect

International Journal of Coal Geology

j ourna l homepage: www.e lsev ie r.com/ locate / i j coa lgeo

Evaluation of pollution levels at an abandoned coal mine site in Turkey with the aidof GIS

Firdes Yenilmez a, Nazan Kuter b, Mustafa Kemal Emil c, Aysegul Aksoy a,⁎a Middle East Technical University, Department of Environmental Engineering, 06531, Ankara, Turkeyb Cankiri Karatekin University, Department of Landscape Architecture, 18200, Cankiri, Turkeyc Middle East Technical University, Department of Mining Engineering, 06531, Ankara, Turkey

⁎ Corresponding author. Tel.: +90 312 210 5874; faxE-mail addresses: firdes@metu.edu.tr (F. Yenilmez),

(N. Kuter), emil@metu.edu.tr (M.K. Emil), aaksoy@metu

0166-5162/$ – see front matter © 2010 Elsevier B.V. Adoi:10.1016/j.coal.2010.11.012

a b s t r a c t

a r t i c l e i n f o

Article history:Received 30 May 2010Received in revised form 2 November 2010Accepted 21 November 2010Available online 28 November 2010

Keywords:Geographical information systemCoal miningPollution transportSurface runoffCase study

Determination of the level of pollution at contaminated sites can be costly due to the number of samplesrequired to identify the areas with higher contaminant concentrations. Yet, if pollution transport pathwayscan be found, then contamination level can be detected with fewer samples. In this case study, pollution levelsat an abandoned coal mine site at Ovacik–Yaprakli (Cankiri, Turkey) are evaluated with respect to topographyand surface runoff pathways derived using Geographical Information System (GIS) tools. First, surface runoffpathways are identified using ArcGIS 9.3. Then, the concentrations of trace elements (Mn, Cr, V, Cu, Ni, Zn, Ba,and Sr), pH, Fe and S contents in soil samples taken at random locations around the mine area are determinedin a laboratory. In addition, pH and electrical conductivity (EC) are measured in flowing waters in-situ atdifferent locations using pH and EC probes. The spatial distribution of pollutant concentrations is evaluatedwith respect to the surface runoff pathways and locations of potential contamination sources (i.e. open pit,coal storage and dump sites). Finally, the contamination level in the study area is assessed based on the limitvalues stated in the Soil Pollution Control Regulation of Turkey (SPCR). Results indicate that the site iscontaminated with Cr, Ni, and Cu. In general, pollutant concentrations are higher close to the contaminationsources and along the surface runoff pathways determined by the ArcGIS. Results indicate that GIS can aid inlocating the areas that are more likely to have high pollutant concentrations. This would in return preventoverlooking highly contaminated spots which may be located far away from contamination sources.Moreover, these areas can be determined using a smaller number of samples which would decrease thesampling costs.

: +90 312 210 2646.nkuter@karatekin.edu.tr.edu.tr (A. Aksoy).

ll rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Coal mining can be an economical asset for a nation. However, itmay lead to substantial environmental damage if precautions are nottaken. Mining activities can lead to land degradation, changes intopography, soil and water pollution, and ecological deterioration(Bell et al., 2001; Bhattacharya et al., 2006; Ribeiro, 2010; Si et al.,2010). Acid mine drainage is another environmental problem relatedto coal mining that can result in acidification of soil and receivingwater bodies (Bell et al., 2001). Acidification would increase thedissolution of toxic metals from water-permeable tailings, waste rockpiles and open pits. In return, soil and water ecosystems will beimpacted (Bell and Donnelly, 2006; Sams and Beer, 2000). Theseproblems can be associated with abandoned mining sites as well.These sites may continue to pollute soil and water for years (Bell et al.,

2001; Navarro et al., 2004). For example, Navarro et al. (2008) showthat an abandonedmine site in Cabezo Rajao (Spain) still continues topollute the environment with metals dispersed downstream anddownslope from tailings by surface runoff.

Contaminants originating from coal mining activities can betransported to long distances. Surface runoff and mechanicaldispersion due to wind action are the main transport mechanisms(Navarro et al., 2004). Enhanced dissolution of toxic metals viaacidification further improves the transport. In limited number of casestudies, the importance of surface runoff pathways and streams on thetransport of contaminants from coal mining sites has been shown. Forinstance, water draining freely from the Handlova–Cigel brown coaldistrict transported large amounts of toxic elements into surfacestreams, which in turn contaminated a broader area (Klukanova andRapant, 1999). The Allegheny and Monongahela Rivers transported1.2 and 1.35 million tons of sulfate, respectively, to the Ohio River inPittsburgh in 1980 (Sams and Beer, 2000). Sams and Beer (2000)showed that the trends observed in contaminant concentrationswithin years are in linewith the trends in coal production rates in sub-basins contributing to pollution transport.

Fig. 1. Location of the Ovacik–Yaprakli abandoned coal mine site.

13F. Yenilmez et al. / International Journal of Coal Geology 86 (2011) 12–19

Identification of the spatial extent and levels of pollution at miningsites are critical in determining the best approaches and techniques inpollution prevention and environmental reclamation. For this pur-pose, contaminant transport pathways should be identified as well as

Fig. 2. Geological map

the locations of contaminant sources (i.e. excavation area, dump site,storage area, etc.). This can be achieved by rigorous sampling (i.e.systematic and regular grid sampling, simple random sampling,stratified sampling, etc). The number of sampling locations isdependent on the geometric features of the area and the character-istics of contamination (Cuccu, 2002; EPA, 2002). Nevertheless, it is afact that sampling can be highly costly. This is mostly due to thenumber of samples that should be taken to assure data quality and theneed to locate the areas that have high contaminant concentrations(hot spots) (EPA, 2002). However, if locations that are prone tocontamination can be predicted, then the sampling program can bedesigned accordingly. Several studies, as mentioned before, point outthat it is more likely to observe contamination closer to the source.Moreover, contaminants can be transported by surface runoff.Therefore, if surface runoff flow pathways can be derived, then itmay be possible to identify the locations of higher contaminantconcentrations (other than the ones in the vicinity of contaminationsources), the direction of transport and to which distance contami-nants may have been transported using a smaller number of samplescompared to sample sizes required for traditional sampling methods.In this manner, GIS can be an effective tool for determination of theareas that are contaminated as a result of contaminant transport viasurface runoff.

GIS helps to integrate various types of information spatially to aidin characterization and decision making. Yet, its application in theanalysis of pollution transport, prevention and reclamation at coalmining sites is limited. Ayad (2006) used GIS in the prioritization ofstream remediation in the Clarion River basin in northwestern

of the study area.

Fig. 3. Raster DEM of the study area.

Fig. 4. Streams and levels of flow accumulation.

14 F. Yenilmez et al. / International Journal of Coal Geology 86 (2011) 12–19

Pennsylvania, which was impacted by coal mining activities. Mapswere produced showing the stream conditions in different segments.The segments that require immediate intervention were identifiedbased on the spatial distributions of acidity and Fe concentrations inthe river. deLemos et al. (2009) used GIS to identify the high riskexposure areas in the Navajo Churchrock mining district. Komnitsasand Modis (2006) used geostatistics and GIS to examine the spatialuncertainty in contaminant concentrations and hazard assessment ata site impacted by coal mine wastes.

In this case study, the contaminant distributions at an abandonedcoal mine site in Cankiri are evaluated based on the topographicalinformation and surface runoff pathways derived via GIS. Thepotential of GIS in determination of locations that are likely to havehigher contaminant concentrations is presented based on the casestudy. ArcGIS is used to assess the ground elevations and surfacerunoff pathways at the abandoned coal mine site. Then, contaminantconcentrations in water and soil samples taken at random locations inand around the mining site are compared with respect to topographyand surface runoff pathways. Finally, the contamination level in thearea is assessed.

2. Study Area and Methodology

The abandoned coal mine site that is considered in this study islocated 1.5 km south of Ovacik village within the borders of Yapraklitown in Cankiri province. The location of the site is depicted in Fig. 1.The map of the geological formations in the region is provided inFig. 2. This map is obtained from the General Directorate of Mineral

Research and Exploration of Turkey (MTA). The thickness of the coalseam at the mining site is 0.3–0.8 m with a width of about 800 maccording to a report submitted to MTA in 1987 (Koseoglu MiningCompany, 1987). The slope of the seam is 75°–80°. The calorific valueof the coal at the site is 4500–5000 kcal/kg. Both open pit andunderground coal mining operations had been conducted in the area.The open pit coal mine was operated in 1987 to 2005. Based on initialsurveys, the coal reserve at the site was predicted as 144,000 tons.However, additional coal seams were discovered during the miningoperation. Approximately 224,440 tons of coal was quarried until1999 using the open pit and undergroundmining techniques. In 2005,the mining activity was halted since the company operating the sitedecided that it was no longer economical to mine the coal at the site.In 2008mining permit was abolished. Currently, the abandoned site isunder the jurisdiction of provincial administration.

In order to determine the impact of surface runoff on the transportof contaminants in the area, flow directions, flow accumulation areasand sub-basins contributing to flows in the watershed are derived. Forthis purpose, 1:25,000 scaled digital topographical maps (prepared in1992) are supplied from the General Command of Mapping. Inaddition, the Digital Elevation Model (DEM) of the area is createdusing an ASTER satellite image obtained in 2010. Surface runoffpathways are derived using both the digital topographical maps andthe DEM generated based on the ASTER image. No significantdifferences are observed in major surface runoff routes and flowaccumulation information. As a result, 1:25,000 scaled topographicalmaps are used in the analysis of contaminant distributions for theirhigher spatial accuracy.

The ArcGIS software package (version 9.3) is used for processingthe topographical data for DEM generation. The data stored in the GISdatabase is processed by the 3D-Analyst and Spatial Analyst tools ofArcGIS 9.3. Elevation contours are extracted from the topographicalmaps in vector data format. The elevation maps are then merged andclipped to focus on the study area (a square area of 5 km by 5 km). TheDEM of the region is first constructed in the form of a triangularirregular network (TIN) using the elevation contours. Then,

Fig. 5. Soil and water sampling locations.

Table 2

15F. Yenilmez et al. / International Journal of Coal Geology 86 (2011) 12–19

conversion to raster DEM is employed in which elevation informationis stored in each pixel of 10 m pixel size (Fig. 3).

Following the generation of the raster DEM, the flow directions inthe study area are generated. This is accomplished by directing theflow in each cell to its steepest downslope neighbour. Then, flowaccumulation is retrieved using the flow direction information andaccumulating the flow weights of all cells that flow into a givendownslope cell. The flow accumulation areas derived by this processare depicted in Fig. 4 together with the locations of dump sites, openpit, and coal storage area. In this figure, higher flow quantities(accumulation) are demonstrated by deeper color. The raster DEMand the flow accumulation information are used in the evaluation ofpollution levels in the area. Concentrations of various soil and waterquality parameters at random soil and water sampling points arecompared with respect to their locations relative to the surface runoffpathways and the distances to source areas (excavation area, dumpsite, storage area). Locations and elevations of the sampling points areevaluated based on coordinate information, DEM and surface flowpathways.

In order to determine the level of contamination, soil samples aretaken at 36 locations within the sampling area of about 1 km2 (on 4thof July, 2009). Soil sampling locations are depicted in Fig. 5. It must beemphasized that the number of sampling locations is limited bybudgetary concerns which is a typical drawback for contaminated sitecharacterization. Soil sampling locations are selected randomly in and

Table 1Heavy metal limits stated in the Soil Pollution Control Regulation of Turkey (SPCR,2005).

Element Soil pH=5–6(mg/kg oven dry soil)

Soil pHN6(mg/kg oven dry soil)

Pb 50 300Cd 1 3Cr 100 100Cu 50 140Ni 30 75Zn 150 300Hg 1 1.5

around the mining area. Yet, care is given to take the samples atlocations with different altitudes (i.e. hill tops, rills, gullies, etc.) andvarying distances from the mine waste dump sites, open pit and coalstorage area. At the sampling points, samples (about 1.5 kg each) aretaken from the first 20 cm of soil depth following the removal of thetop 5 cm of soil layer. All soil samples are stored in a cooler andtransported to the laboratory for analysis. The samples are kept at 4 °Cin a refrigerator until analyses are performed.

Soil quality parameters of concern are selected based on acharacterization report on coal reserves in different parts of Turkey(MTA, 2002). Since relevant elemental characterization informationfor the Ovacik coals is not public, the characteristics of coal seamsclose-by are taken into account. In this regard, the Ilgaz-Ilisilik site inCankiri is assumed to represent the coal characteristics at the Ovacikcoal mine site. The Ilgaz-Ilisilik coals have elevated Fe, Mn, Cr, V, Cu,Ni, Zn, Ba, Sr, and S contents. The quantities of these elements in coalare also above the world average values (MTA, 2002). Therefore, Mn,Cr, V, Cu, Ni, Zn, Ba, and Sr are selected as the trace elements for theevaluation of contamination levels in soil. In addition, Fe and Scontents and pH of soil samples are analyzed. As well as the number ofsampling locations, the number of examined soil quality parameters islimited due to budgetary constraints.

Summary of the analyses results for soil samples.

Element/parameter

Minimumvalue

Averagevalue

Maximumvalue

Standarddeviation

pH 3.27 6.87 8.02 1.46Fe (%) 3.58 5.03 6.71 0.67Mn (mg/kg) 281.64 776.35 1293.50 233.44Cr (mg/kg) 195.45 348.06 652.65 103.96V (mg/kg) 83.24 131.48 191.98 23.47Cu (mg/kg) 30.51 58.05 109.58 17.22Ni (mg/kg) 167.67 296.80 504.20 69.46Zn (mg/kg) 65.11 101.09 130.13 16.33Ba (mg/kg) 48.42 129.32 260.42 44.15Sr (mg/kg) 35.46 81.36 137.70 22.44S (%) 0.30 0.99 1.61 0.49

Fig. 6. Acid mine drainage at the site.

16 F. Yenilmez et al. / International Journal of Coal Geology 86 (2011) 12–19

Soil quality analyses are conducted at the accredited laboratories ofthe Middle East Technical University. Soil samples are dried at roomtemperature and sieved to obtain a suitable particle size (b2 mm) forpH, Fe, S and trace element analyses. For soil pHmeasurement, 100 g ofsoil is saturated with distilled water and pH is read using an OAKTONpH300 pH meter. In order to determine the concentrations of traceelements, Fe, and S in soil samples,first soil extraction is employedusingthe EPA 3052 method (Mason, 1992). Following extraction, measure-ments are realized by a Perkin Elmer DRC II model ICP-MS (inductivelycoupled plasma mass spectrometer). Soil pollution levels are evaluatedbased on the heavymetal limit values set in Appendix I-A of SPCR. Theselimit values are stated in Table 1. For Cr, Cu, Ni and Zn, contaminationstatus is evaluated based on the limit values given in Table 1. For allelements including the ones which do not have limit values in SPCR,background concentrations are used to determine enrichments due tocontamination. For this purpose, the minimum concentrations (mgelement in kgof soil) among36 samples are considered as “background”

Fig. 7. Soil pH d

concentrations for the respective trace elements. This approach deemedfeasible since some of the sampling points are at high elevations awayfrom the influence of transport via surface runoff. Normalization isemployed in order to find the enrichment factors due to potentialcontamination. In this application relevant concentrations are dividedby the corresponding background concentrations. Enrichment factorsare defined for the average and the maximum concentrations of eachsoil quality parameter.

Soil characteristics and ground cover are important for thesorption of elements to soil. This would impact the concentrationsobserved in soil. However, in this study these impacts are assumednegligible and focus is given to the relative distributions ofcontaminant concentrations in soil with respect to elevations andwater flow pathways in the sampled area. Analyses results are placedon the maps obtained using ArcGIS. Pollution levels are evaluatedbased on topography and surface runoff flow directions.

In addition to soil quality, water quality in the flowing streams andcreeks is determined. For this purpose, in-situ pH and ECmeasurementsare conducted at a total of 14 points. These sampling points are depictedin Fig. 5. The pH and EC quantities are determined using an OAKTONpH300 pHmeter and a Hach Sension 156 probe, respectively. Thewaterquality status with respect to pH and EC is evaluated based on the limitvalues in the Water Pollution Control Regulation of Turkey.

3. Results and Discussion

In this section, the results of soil and water analyses will beprovided to present the contamination status at and around theabandoned coal mine with respect to the limit values and backgroundconcentrations. The analyses results will also be evaluated based ontopography and surface runoff pathways.

In-situ water pH measurements indicate that pH values varybetween 6.64 and 8.13. pH is lower in the downstream direction ofsurface runoff pathways away from the open pit. Similarly, pH valuesin the surface waters away from the coal storage area are lower in thedownstream flow direction. According to the Water Pollution ControlRegulation of Turkey, water pH ranging from 6.5 to 8.5 indicatesacceptable quality. EC quantities in water samples exhibit a wide

istribution.

0.00

1.00

2.00

3.00

4.00

5.00

6.00

Fe (%

)

Mn (mg/kg

)

Cr (m

g/kg)

V (mg/k

g)

Cu (mg/kg

)

Ni (m

g/kg)

Zn (mg/kg

)

Ba (m

g/kg)

Sr (m

g/kg)

S (%)

enri

chm

ent

fact

ors

Ca/CbCm/Cb

Fig. 8. Enrichment factors for different elements in the sampled area (Ca = averageconcentration, Cm = maximum concentration, Cb = background concentration).

17F. Yenilmez et al. / International Journal of Coal Geology 86 (2011) 12–19

variation ranging from 865 μs/sec to 6470 μs/sec. The average EC is2690±1500 μs/sec. EC is higher around the storage area and thereforestation area which had been operated as a mining site before.However, it must be pointed out that water quality measurementswere realized following a storm. Therefore, heavy rain may haveimpacted the pH and EC values.

The summary of soil analyses is provided in Table 2. When Cr, Cu,Ni, and Zn contents in soil samples are compared to the respectivelimit values in SPCR, it is seen that concentrations exceed the limits in100% of the samples for Cr and Ni. Cu contents surpass the acceptablelimit in 20% of the samples. On the other hand, Zn concentrations arebelow the limit. The Cr concentration in the sample taken from a farmis 340.22 mg/kg, which is well above (3.3 times) the limit value givenin Table 1. As a result, the area is contaminated above the limits for Cr,Ni, and Cu. However, since Cr and Ni contents are high for all samples,additional sampling should be employed to determine the back-ground concentrations and the contribution of mining activities on Crand Ni contamination.

In Table 2, it is seen that soil pH values range from 3.27 to 8.02. LowpH values (3.29–4.15) are observed around the coal storage area anddump sites. These results show that although the coalminewas closedmany years ago, environmental effects are still observable. In fact, acid

Fig. 9. Distribution of so

mine drainage is apparent even now at the abandoned coal mine siteas depicted in Fig. 6. The spatial variation in soil pH can be seen inFig. 7. The evaluation of Fig. 7 with respect to elevations and surfacerunoff pathways indicates that the pH values within potential surfacerunoff directions are relatively lower (acidic) even if the samplinglocations may be away from the coal mine and dump sites. At highelevations along the direction of surface runoff, pH values are higher.These locations are less likely to be in contact with contaminatedsurface runoff.

Variations in the concentrations of Fe, S, and trace elements areexamined based on the respective background concentrations. Asdiscussed in the methodology section, enrichment factors are deter-mined for theaverage and themaximumconcentrationsof the elementswith respect to the corresponding minimum concentrations (back-ground concentrations). These enrichment factors are shown in Fig. 8.Within the sampled locations, the maximum enrichments are observedfor S and Ba. Mn, Sr, and Cu are the following ones in decreasing order.When Cr concentrations in the sampled area are examined, 3.3 timesdifference is observed between the minimum and the maximumconcentrations.Measured Cu concentrations range from30.51 mg/kg to109.58 mg/kg (Table 2). This reflects up to 3.6 times difference in theconcentrations at different locations. The enrichment factors based onthemaximum concentrations are 3.0, 1.9, 4.6, 2.3, 5.4, 3.9 and 5.4 for Ni,Fe,Mn, V, Ba, Sr andS, respectively. Although theZn concentrations in allsoil samples are below the limit value stated in SPCR (Table 1), observedconcentrations at different locations vary by up to 3 times comparedwith the minimum concentration.

When the spatial distributions of element concentrations areexamined, it is observed that the concentrations of trace elements, Fe,and S are generally higher close to the coal storage area, open pit, anddump sites. This is the case although the mine has been abandonedyears ago. The regions in which Cu concentrations exceed the limitvalue in SPCR are situatedwithin the coal storage area and dump sites.In general, Cr concentrations are higher around the open pit, dumpsites and coal storage area (Fig. 9). The major surface runoff directionsare shown in Fig. 9 with arrows. In the figure, it can be observed thathigher Cr concentrations are in line with the flow directions. Areasthat are close to contamination sources (i.e. storage area and dumpsites) but have high elevations are relatively less impacted from the

il Cr concentrations.

Fig. 10. Distribution of soil Fe concentrations.

18 F. Yenilmez et al. / International Journal of Coal Geology 86 (2011) 12–19

pollution transported via surface runoff. In Fig. 9, the high elevationarea at north of the storage area is marked with dashed circle (at theintersection point of the arrows that show the runoff directions). Inthis location lower contaminant concentrations are observed eventhough high contaminant concentrations are observed downslope.This situation is valid for the majority of trace elements. The spatialdistributions of Fe, V, and Zn concentrations are depicted in Figs. 10,11, and 12, respectively. In general, similar spatial concentrationdistributions are observed. As can be expected, concentrations arehigher closer to contaminant sources such as the coal storage area. Athigh locations where sampling points are more likely to be unaffectedby surface runoff, element concentrations are lower. Downstream

Fig. 11. Distribution of so

locations with respect to the direction of surface runoff contain hightrace element and Fe concentrations. Therefore, GIS can be helpful inlocating the areas that are more likely to be impacted by contami-nation via transport routes.

Locations with high elevations on the surface runoff directions canbe protected against contamination, providing that topographyhinders the contact of contaminants advected by surface runoff.Examination of the locations of contamination sources, contaminanttransport pathways and topography of the area may help to selectrepresentative sampling locations. In this regard, it may be possible toassess the severity of contamination and identify hot spots using asmaller number of samples. Similarly, sampling locations for better

il V concentrations.

Fig. 12. Distribution of soil Zn concentrations.

19F. Yenilmez et al. / International Journal of Coal Geology 86 (2011) 12–19

background concentration determination can be identified. GIS toolsare helpful for these purposes. Moreover, visualization using GIS helpsto better evaluate the state of contamination.

4. Conclusions

The results of this case study indicate that the Ovacik–Yaprakli coalmine site is contaminated with Cr, Ni and Cu according to SPCR. TheMn, Ba, Sr, and S concentrations in soil samples exhibit largervariations compared with other elements based on the enrichmentfactors. In general, soil and water quality is worse around the sources(open pit, storage area and dump sites). Moreover, higher elementconcentrations are measured along the surface flow pathways.Exceptions are the locations with high elevations that are less likelyto be impacted by the contaminated surface runoff.

In this case study, it is shown that surface runoff routes andtopography of an area are important in the transport of contaminantsfrom the mining area. In this sense GIS can be useful in locating theareas that have higher potential of exhibiting elevated contaminantconcentrations. Similarly, it may be possible to select the bestlocations for background concentration determination with the aidof GIS. In fact, if these locations can be derived using GIS, then asmaller number of samples can be sufficient to define the level andextent of soil and water contamination. This in turn would reduce thesampling and analyses costs.

Acknowledgements

This study is financially supported by the Middle East TechnicalUniversity Research Fund (project number: BAP-07.02.2009.07). Weare grateful to Assoc. Prof. Dr. Sebnem Duzgun for her support inrealizing this work.

References

Ayad, Y., 2006. Prioritizingacidminedrainage streamremediation.ESRI, GISBest Practices:Mining, pp. 21–24http://www.esri.com/library/bestpractices/mining.pdf.

Bell, F.G., Bullock, S.E.T., Halbich, T.F.J., Lindsay, P., 2001. Environmental impactsassociated with an abandoned mine in the Witbank Coalfield, South Africa.International Journal of Coal Geology 45, 195–216.

Bell, F.G., Donnelly, L.J., 2006. Mining and its Impact on the Environment, first ed.Taylor&Francis Group, Oxon, England.

Bhattacharya, A., Routh, J., Jacks, G., Bhattacharya, P., Morth, M., 2006. Environmentalassessment of abandoned mine tailings in Adak, Vasterbotten district (northernSweden). Applied Geochemistry 21, 1760–1780.

Cuccu, W., 2002. Site assessment before reclamation. Proceedings of the 7thInternational Symposium on Environmental Issues and Waste Management inEnergy and Mineral Production, pp. 113–120. 7-10 October 2002. Italy.

deLemos, J.L., Brugge, D., Cajero, M., Downs, M., et al., 2009. Development of risk mapsto minimize uranium exposures in the Navajo Churchrock mining district.Environmental Health 8 article 29.

Mason, B.J., 1992. Preparation of Soil Sampling Protocols: Sampling Techniques andStrategies. EPA Office of Research and Development, EPA/600/R-92/128, Washing-ton DC.

EPA, 2002. Guidance on choosing a sampling design for environmental data collectionfor use in developing a quality assurance project plan. Office of EnvironmentalInformation, EPA/240/R-02/005, Washington DC.

Klukanova, A., Rapant, S., 1999. Impact of mining activities upon the environment of theSlovak Republic: two case studies. Journal of Geochemical Exploration 66, 299–306.

Komnitsas, K., Modis, K., 2006. Soil risk assessment of As and Zn contamination in a coalmining region using geostatisretics. The Science of the Total Environment 371,190–196.

Koseoglu Mining Company, 1987. Ovacik–Yaprakli Coal Mine Project Report. Archive ofthe General Directorate of Mining Affairs, Ankara (in Turkish).

MTA, 2002. Chemical and technological properties of Turkish tertiary coals. GeneralDirectorate of Mineral Research and Exploration, Ankara, pp. 353–354.

Navarro, A., Collado, D., Carbonell, M., Sanchez, J.A., 2004. Impact of mining activities onsoils in a semi-arid environment: Sierra Almagrera district, SE Spain. Environmen-tal Geochemistry and Health 26, 383–393.

Navarro, M.C., Perez-Sirvent, C., Martinez-Sanchez, M.J., Vidal, J., Tovar, P.J., Bech, J.,2008. Abandoned mine sites as a source of contamination by heavy metals: a casestudy in a semi-arid zone. Journal of Geochemical Exploration 96, 183–193.

Riberio, J., Ferreira da Silva, E., Li, Z., Ward, C., Flores, D., 2010. Petrographic,mineralogical and geochemical characterization of the Serinnha coal waste pile(Douro Coalfield, Portugal) and the potential environmental impacts on soil,sediments and surface waters. International Journal of Coal Geology 83, 456–466.

Sams, J.I., Beer, K.M., 2000. Effects of Coal-Mine Drainage on Stream Water Quality inthe Allegheny and Monongahela River Basins-Sulfate Transport and Trends. WaterResources Investigations Report 99-4208, Lemoyne, Pennsylvania.

Si, H., Bi, H., Li, X., Yang, C., 2010. Environmental evaluation for sustainabledevelopment of coal mining in Qijiang, Western China. International Journal ofCoal Geology 81, 163–168.

SPCR (Turkish Soil Pollution Control Regulation), 2005. Official Gazette, issue date:31.05.2005, issue number: 25831.