The influence of the abandoned Kalecik Hg mine on water and stream sediments (Karaburun, İzmir, Turkey)

The Science of the Total Environment 312 (2003) 155–166

The influence of the abandoned Kalecik Hg mine on water and stream sediments (Karaburun, ˙Izmir, Turkey) ¨ Unsal Gemici*, Tolga Oyman ¨ University, 35100 Bornova, Izmir, Turkey Department of Geological Engineering, Dokuz Eylul Received 15 August 2002; received in revised form 19 December 2002; accepted 23 December 2002

Abstract This study covers the geochemical investigations on water and stream sediments to evaluate the influence from the abandoned Kalecik Hg mine. The groundwater samples (S5, S8, S9, WW10) are neutral, slightly alkaline waters which have pH values varying between 7.3 and 7.5. Electrical conductivity (EC) values of groundwaters for spring samples are low (250–300 mSycm). However, groundwater obtained from a deep well has a higher EC value of 950 mSycm. Hg concentrations of groundwater samples vary between 0.01 and 0.13 mgyl. Hg concentrations of other water samples taken from mining area from surface waters and adits are between 0.10 and 0.99 mgyl. Adit water (A4) collected at the mine has the highest Hg content of 0.99 mgyl and a pH of 4.4. Trace element concentrations of mine water samples show variable values. As is observed only in MW1 (310 mgyl). A4 was enriched in Cd, Co ¨ Standartları Enstitusu, ¨ ¨ 1997). Cu concentrations vary and Cr and exceed the Turkish drinking water standards (Turk between 6.0 and 150 mgyl and are below the Turkish water standards. Mn concentrations in mine waters are between 0.02 and 4.9 mgyl. Only for sample A4 Mn value (4.9 mgyl) exceeds the standard level. Ni was enriched for all of the mine water samples and exceeds the safe standard level (20 mgyl) for drinking water. Of the major ions SO4 shows a notable increase in this group reaching 650 mgyl that exceeds the drinking water standards. Stream sediment samples have abnormally high values for especially Hg and As, Sb, Ni, Cr metals. With the exception of sample Ss6 of which Hg concentration is 92 mgykg, all the other samples have Hg contents of higher than 100 mgykg. Pollution index values are significantly high and vary between 69 and 82 for stream sediment samples. 䊚 2003 Elsevier Science B.V. All rights reserved. Keywords: Mercury mining; Groundwater; Stream sediment; Heavy metal contamination

1. Introduction Aegean Region of Turkey consists of numerous ¨ ¸ Hg deposits in Karaburun, Bayındir, Tire, Odemis, ¸ Alasehir, Ulubey, and Banaz districts from which approximately 5500 t of mercury have been pro*Corresponding author. Tel.: q90-232-3882919; fax: q90232-3887865. ¨ Gemici). E-mail address: [email protected] (U.

˘ duced (Bircan and Aydoganlı, 1969; MTA, 1977) (Fig. 1). Due to increasing environmental concerns of mercury, the mines were abandoned gradually until the early 1990s. Apart from the Kalecik mine that is located 4 km SE of Karaburun on the eastern part of the Karaburun Peninsula, Karareis and Kucukbahce ¨¸¨ ¸ mercury mines are located in the western part of the peninsula (Fig. 1). Between 1906 and until the early 1970s approximately 700

0048-9697/03/$ - see front matter 䊚 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0048-9697(03)00008-1

156

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

Fig. 1. The distribution of some abandoned Hg mines in western Anatolia and location of the study area.

t of mercury and more than 200 000 t of calcines were produced during mining from the Kalecik mine (Lehnert-Thiel, 1969; Nar, 1974). Mercury contaminated stream sediments and surface from the Kalecik mine are potentially hazardous to resident wildlife when mine runoff enters local surface water and eventually flows into in Karaburun Bay approximately 4 km downstream. Karaburun is a tourist town with a summer population of 30 000 residents. Water resources are springs, wells, surface waters in and around the Kalecik mine and Karaburun. The purpose of this study was to evaluate potential mining-related Hg contamination in groundwaters, surface waters, and stream sediments in and around Karaburun. 1.1. Description of the study area Kalecik mercury mine is hosted in sandstone, mudstone, fine-layered black chert, and limestone ˘ et al., 1990) of the ˙Izmir–Ankara Zone (Erdogan that overlie massive limestone. Mafic submarine volcanite and associated tuffs cover the sedimentary rocks. Mineralized epithermal veins and

replacements at the Kalecik mine are found in silicified sandstone, clay and carbonate rocks. The main metallic minerals are cinnabar (HgS), pyrite (FeS2), marcasite (FeS2), lesser arsenopyrite (AsFeS), stibnite (Sb2S3 ), and nickel silicates. Fig. 3 represents some stream sediment samples that were investigated under the reflected-light microscope to determine the physical partition of mercury and associated ore minerals. The presence of cinnabar and Fe–oxide–hydroxide assemblage confirms that the relative resistance of cinnabar to chemical breakdown in weathering environment (Jonasson and Boyle, 1972; Rogers, 1979; Rose et al., 1979; Harsh and Doner, 1981). Study area has a Mediterranean type climate which is characterized by arid, sunny summers and wet, cool winters. The mean precipitation value for the years 1971–1998 was 490 mm. Maximum precipitation was observed in December and January. The main annual temperature is 17 8C. Yearly precipitation is not regular and shows differences from one year to another year. The minimum precipitation was 211 mm in 1989 and the maximum was 761 mm in 1996. The mine

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

157

Fig. 2. Locations of the water and stream sediment samples.

area is rugged and highly sloped. The elevation increases from sea level towards the west and reaches 500 m. The mine area is drained by an ephemeral stream and its small tributaries with a catchment area of 5 km2 (Fig. 2).

the time of collection. Cl and HCO3 were determined volumetrically and SO4 by gravimetry. The compositions of water samples are reported in Table 1. Solmineq-88 (Kharaka et al., 1988), and Hydrowin (Calmbach, 1995) computer programs were used to evaluate their geochemical properties.

1.2. Water sampling 1.3. Stream sediment sampling Chemical data for surface and groundwaters have been obtained from samples that were collected from springs (S5, S8, S9), well (WW10), adits (A2, A3, A4, A6, A7) and mine waters draining calcines (MW1) (Fig. 2). Two samples were collected from each sampling location and stored in polyethylene bottles. One of the bottles was acidified with HNO3 to decrease its pH value to 2 and the other was kept unacidified for anion analyses. Electrical conductivity (EC), pH, Eh and temperature values were measured in the field at

Approximately 3 kg of stream sediment were collected with a plastic scoop at seven sampling sites (Fig. 2). Stream sediment samples were composited by collecting material from channelbed alluvium. During collection special care was taken to avoid oxidized materials and organic substances. Following Herr and Gray (1995), stream sediment was sieved to y10 mesh (2 mm) in the field. In the laboratory, the sediment samples were air-dried and sieved to y80 mesh (0.18

158

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

Fig. 3. Photomicrographs of some metals in stream sediments samples under the reflected-light microscope.

mm). One gram sample was leached with 6 ml 2– 2–2 HCl–HNO3 –H2O 95 8C for 1 h and diluted to 20 ml. Stream sediment samples were analyzed for Hg, As, Sb, Fe, Cu, Pb, Zn, Co, Ni, Cr, Cd by ICP-mass spectrometry by ACME Analytical Laboratories Ltd, Vancouver, Canada. Lower limits of determinations and analytical data are given in Table 2. 2. Results and discussions 2.1. Water chemistry Groundwater samples (S5, S8, S9) are near neutral (pH of 7.3–7.5) and electrical conductivi-

ties are 250–300 mSycm. However, another groundwater sample (WW10) that was obtained from a deep well has an EC of 950 mSycm. Mine waters have variable pH, Eh and EC values based on their circulating conditions and the lithological properties of the basement rocks. The pH values of mine waters are between 4.4 and 8.4. Samples A4, A6 and A7 have lower pH values (Table 1). Although the residence time of mine waters significantly is less than that of groundwaters, their EC values that reach to 1300 mSycm are higher due to the acidic character by low pH causing the dissolution of the alteration minerals. The Eh

Table 1 Chemical properties of the water samples MW 1

A2

A3

A4

S5

A6

A7

S8

S9

WW10

Turkish potable water standards (TSE 1997) *

Ag As B Ba Be Bi Cd Ce Co Cr Cu Fe Li Mn Mo Ni P Pb Sb Se Si Te Ti Tl U V W Zn Hg *

mV mSycm mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl

8.4 y68 1120 0.1 123 42 33 3.5 47 385 251 4.8

7.6 y26 770 0.5 106 53 20 1.4 36 183 312 –

7.4 y14 740 0.2 76 47 26 1.4 38 145 295 –

4.4 143 1300 33 130 72 17 1.5 32 650 17 –

7.3 y10 300 0.2 21 11 25 1.2 41 26 83 –

6.7 23 170 3.1 8.7 8.4 18 3.4 26 59 37 –

5.8 103 500 2.9 21 26 18 3 34 145 20 –

7.5 y20 250 0.2 13 5 16 0.5 22 13 64 –

7.4 y19 290 0.8 15 6.9 25 1.2 33 22 79 –

7.3 y11 950 -0.1 113 14 68 1.7 170 39 234 –

mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl mgyl

-5 310 50 86 -2 -20 -2 -30 -5 -20 6 0.02 -0.05 0.06 -5 35 0.1 -10 -10 -20 8.7 -10 -10 -10 -0.05 -10 -10 982 0.04

-5 -30 -20 189 -2 -20 -2 -30 -5 -20 55 0.6 -0.05 0.05 -5 50 0.03 -10 -10 -20 6.1 -10 -10 -10 -0.05 -10 -10 45 0.01

-5 -30 -20 30 -2 -20 -2 -30 -5 -20 8 0.1 -0.05 0.02 -5 40 0.02 -10 -10 -20 8.5 -10 -10 -10 -0.05 -10 -10 746 -0.01

-5 -30 -20 -20 5 -20 13 34 532 35 153 0.4 -0.05 4.9 -5 4800 -0.02 -10 -10 -20 18 -10 -10 -10 -0.05 -10 -10 278 0.99

-5 -30 21 39 -2 -20 -2 -30 -5 -20 6 0.1 -0.05 -0.01 -5 -5 0.16 -10 -10 -20 31 -10 -10 -10 -0.05 -10 -10 30 0.01

-5 -30 229 24 -2 -20 -2 -30 -5 -20 17 1.2 0.14 0.02 -5 8 1.55 -10 -10 -20 10 -10 58 -10 -0.05 -10 -10 52 0.02

-5 -30 22 -20 -2 -20 4 -30 21 -20 33 0.1 -0.05 0.4 -5 160 -0.02 -10 -10 -20 6.2 -10 -10 -10 -0.05 -10 -10 67 0.02

-5 -30 -20 -20 -2 -20 -2 -30 -5 -20 -2 0.1 -0.05 -0.01 -5 -5 0.15 -10 -10 -20 23 -10 -10 -10 -0.05 -10 -10 34 0.08

-5 -30 -20 22 -2 -20 -2 -30 -5 -20 3 0.5 -0.05 0.01 -5 -5 0.15 -10 -10 -20 30 -10 29 -10 -0.05 -10 -10 40 0.13

-5 -30 23 57 -2 -20 -2 -30 -5 -20 4 -0.01 -0.05 -0.01 -5 -5 0.02 -10 -10 -20 5.3 -10 -10 -10 -0.05 -10 -10 38 0.04

6.5–8.5 400 0.05 100 30 20

**

MAC

6.5–9.2 2000 0.2 200 50 175

25 25

600 250

0 1000 100

50 2000 300

0

5

0 100 0.05

50 3000 0.2

0.02

0.5

0

50

0 0 0

50 10 10

100

5000 1

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

pH Eh EC Al Ca Mg Na K Cl SO4 HCO3 CO3

GL

Goal level. Maximum admisible concentration.

**

159

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

160

Table 2 Chemical properties of the stream sediment samples (in mgykg)

Ss1 Ss2 Ss3 Ss4 Ss5 Ss6 Ss7 a

As

Cd

Cu

Hg

Pb

Sb

Zn

Ni

Cr

Fe (%)

Co

PIa

277 267 286 279 185 9660 367

0.7 0.6 0.6 0.6 0.6 6.0 0.7

33 32 38 32 38 40 40

100 100 100 100 100 92 100

45 25 51 32 36 40 31

31 23 28 25 14 692 37

99 89 108 92 88 511 113

190 188 192 189 234 539 214

99 98 108 110 67 106 116

3 3 4 4 3 6 4

22 23 24 23 32 35 24

76 74 76 75 69 683 82

Pollution index.

values show significant difference between groundwaters and waters from adits. Eh values of water samples increase from surface water (y26 to y68 mV) to groundwater samples (y10 to y20 mV) and to mine waters from adits (23–143 mV). The relationship between Eh and EC against pH for the water samples is presented in Fig. 4. The pH is inversely correlated with Eh with a high correlation coefficient (rs0.98). However, relation between pH and EC is heterogeneous and show inversely poor correlation. This is probably due to the different circulating paths of water samples in the various rocks.

All of the ground water samples have low metal concentrations, generally below detection limits (Table 1). Hg concentrations of groundwater samples vary between 0.01 and 0.13 mgyl, whereas Hg mine runoff waters and adits vary from -0.1 to 0.99 mgyl. Sample A4 with pH of 4.4 has the highest Hg content of 0.99 mgyl and elevated concentrations of Cd, Co, Cu, Mn, Ni and Zn (Table 1). Arsenic is observed only in sample MW1 with a value of 310 mgyl. Fig. 5 shows the relationship between total metal contents and pH for water samples. Sample MW1 is near neutral with high metal content water. Data for A4 show-

Fig. 4. The relationship between pH, Eh and EC.

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

161

Fig. 5. Scatter diagram of the concentrations of the metals vs. pH (modified from Gray et al., 2000).

ing low pH and high contents suggest acid water leaching of sulfide minerals. Correlation analysis was carried out for pH, EC, major ions and selected trace elements (Table 3). The pH is negatively correlated Fe, Ni, Al and Hg and a positive correlation with SiO2. High positive correlation of Ca and Mg with SO4 show that the dissolution of carbonate and sulfide minerals. SO4 also has high positive correlations with the heavy metals (Mn, Zn, Cu, Ni, Al and Hg) with correlation coefficient of higher than 0.8. Positive correlations for Hg are seen between Mn, Cu, Ni and Al (r)0.9) (Table 3). Trace element concentrations of water samples from the mining area (samples MW1, A2, A3, A4, A6, A7) show variable values. Cd, Co and Cr exceed the Turkish drinking water standards (TSE, 1997) in sample A4. Cu concentrations vary between 6 and 150 mgyl, which are below the potable water standards. Mn concentrations for this group of waters are between 0.02 and 4.9 mgyl. Only for sample A4 Mn value exceeds the standard level. Ni was enriched for all mine waters (samples MW1, A2, A3, A4, A6, A7) and exceeds the safe level (20 mgyl) for drinking water. The Ni content of sample A4 reaches to 4800 mgyl. In addition, SO4 in mine water reaches 650 mgyl and also exceeds the drinking water standard (250 mgyl). The source of sulfate and heavy metals is sulfide mineral oxidation in the mining area.

The concentrations of Al, As, Ba, Cd, Ce, Co, Cr, Cu, Fe, Mn, Ni, Hg presented in Table 1 are below the Turkish drinking water standards (TSE, 1997) with the exception of Al and Fe for sample S9 for the groundwater samples S5, S8, S9 and WW10 that are used in Karaburun for mainly drinking. Al (0.8 mgyl) and Fe (0.51 mgyl) concentrations for sample S9 slightly exceed the standards for potable waters (Table 1). Sample S5 was collected from a spring where there are no known upstream Hg deposits, and thus, the 0.10 mgyl Hg concentration in this sample represents an uncontaminated baseline in this region. Hg concentrations of other groundwater samples (S8, S9 and WW10) are higher than that of S5 (0.01 mgyl) and vary between 0.04 and 0.13 mgyl. Hg contents in all collected in this study are less than the 1.0 mgyl Turkish drinking water standards and that recommended by the WHO (1993). Chemical analyses indicate that surface and groundwaters are more or less enriched in a number of elements. However, heavy metal concentrations of groundwater samples are very low. Although the leachate from the mining area seems to recharge groundwaters, sample WW10, that is one of the main sources for drinking water demand for Karaburun, was not contaminated. The low Hg concentrations in groundwater samples is due to the high resistance nature of cinnabar to chemical and physical weathering and its solubility in water

162

pH Cond. K Mg Ca Na Cl SO4 HCO3 SiO2 Mn Fe Zn Cu Ni Al Hg

pH

Cond.

K

Mg

Ca

Na

Cl

SO4

HCO3

SiO2

Mn

Fe

Zn

Cu

Ni

Al

Hg

1.0

y0.23 1.0

y0.03 0.14 1.0

y0.39 0.81 0.06 1.0

y0.06 0.96 0.07 0.77 1.0

0.31 0.36 0.03 y0.18 0.43 1.0

0.168 0.36 y0.01 y0.15 0.42 0.98 1.0

y0.54 0.81 0.22 0.86 0.71 y0.18 y0.17 1.0

0.63 0.42 y0.05 0.33 0.58 0.45 0.36 y0.03 1.0

0.003 y0.45 y0.55 y0.39 y0.49 y0.29 y0.35 y0.19 y0.49 1.0

y0.85 0.57 y0.09 0.66 0.45 y0.23 y0.14 0.84 y0.38 0.08 1.0

y0.17 y0.43 0.21 y0.13 y0.34 y0.40 y0.37 y0.09 y0.30 0.13 0.05 1.0

0.29 0.55 0.36 0.49 0.51 0.05 y0.09 0.48 0.49 y0.31 0.04 y0.39 1.0

y0.82 0.57 y0.05 0.76 0.49 y0.31 y0.19 0.83 y0.26 y0.07 0.94 0.17 y0.04 1.0

y0.81 0.57 y0.25 0.67 0.43 y0.27 y0.19 0.85 y0.44 0.27 0.99 0.06 y0.01 0.94 1.0

y0.85 0.53 y0.06 0.63 0.40 y0.25 y0.16 0.82 y0.42 0.08 0.99 0.13 0.01 0.94 0.99 1.0

y0.77 0.56 y0.19 0.66 0.44 y0.21 y0.15 0.81 y0.37 0.16 0.98 0.03 0.12 0.90 0.99 0.98 1.0

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

Table 3 Correlations of major ions and some selected metals for water samples from the study area (in mgyl)

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

163

Fig. 6. Schematic model of the groundwater flow of groundwaters around abandoned Kalecik Hg mining.

(Gray et al., 2000). Low concentrations of metals in groundwater samples are possible due to the adsorption by ironhydroxide or precipitation and combination with sulfide (Pierce and Moore, 1980; Lee et al., 2001). When the pH increases the eventual precipitation of Al hydroxides will contribute to enhance metal adsorption (Cidu et al., 1997). Carbonate minerals in limestone in the aquifer for WW10 neutralize the acid water from mining area resulting in the coprecipitation of Al and Fe compounds. This causes the heavy metals to decrease in the groundwater. In addition, this process of dilution with uncontaminated groundwater gives rise also to decrease of contaminants through to the groundwater flow direction (Fig. 6). 2.2. Stream sediments A series of stream sediment samples (Ss1, Ss2, Ss3, Ss4, Ss5, Ss7) and a sample of mine-waste calcine (Ss6) were taken from the study area (Fig. 2) to evaluate the distribution of the metals through to the environment from the mining area. These stream sediments are eroded downstream from

rocks, soils, dumps, and calcine piles in the catchment area. Metal concentrations of the stream sediments are presented in Table 2. The data show that stream sediment samples have abnormally high values especially for Hg, As, Sb, Ni and Cr. With the exception of sample Ss6 of which Hg concentration is 92 mgykg, all the other samples have Hg contents of higher than 100 mgykg. Concentrations of As are remarkably high in the range of 185–9660 mgykg. The calcine piles sample (Ss6) has relatively higher metal values especially As content of 9660 ppm and it is one of the main sources of the contaminants. As is derived from arsenopyrite, which associated with cinnabar in ore paragenesis. Water sample MW1 taken from the ephemeral stream that passes through to the calcine piles has As concentration of 310 mgyl. The other metals have contents of 32–40 mgykg for Cu, 25–51 mgykg for Pb; 88– 511 mgykg for Zn; 188–539 mgykg for Ni; 22– 35 mgykg for Co; 0.5–6 mgykg for Cd; 14–692 mgykg for Sb and 67–110 mgykg for Cr. Due to different rates of recovery as a result of retorting process, large interval of Hg concentrations in calcine material was determined in mer-

164

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

Fig. 7. The variation of some selected metals from mining area to downstream.

cury mines worldwide (Gray, 2001; Gray et al., 2002b). In abandoned mercury mines in Nevada the concentration of Hg in calcine material is significantly higher than that in the stream sediments (Gray et al., 2002a). In contrast, the calcine material from the Kalecik mine has much more lower Hg concentrations than those of downstream sediments. Similarly to Nevada mercury mines in which Hg concentrations are less than 0.012 mgy l, Hg concentrations in water samples are also low in Kalecik due to the low dissolubility of cinnabar and high dilution during runoff. Gosar et al. (1997) reported the pollution of the Idrijca River (Slovenia) caused by the activities of Idrijca Hg mine which is the second largest mercury mine in the world. Downstream sediments of the Idrijca River are highly contaminated with Hg concentrations between 300 and 1000 mgykg which is higher than those of the samples from Kalecik mine () 100 mgykg). The most intensive Hg concentrations in literature were documented in the river sediments around Monte Amiana mercury mine in Italy (228 mgykg) and in Mount Avala near Belgrad (up to 6000 mgykg) (Gosar et al., 1997).

Fig. 7 shows the variation of some selected metals from mining area to downstream. Very high values of the metal concentrations found in the samples close to the calcine piles and in the lower parts of the stream concentrations do not show a significant decrease. Distribution of the metals in the stream sediments show that metals from the mining area move downstream in the rainy season giving rise to contamination in the areas adjacent to the mining area. Although the metals in stream sediments are higher than that of the tolerable levels, groundwater samples are very slightly contaminated as in WW10 which is located downstream in karstic aquifer. This is probably due to the adsorption of metals, coprecipitation of Al and Fe compounds, increasing pH and high dilution rates with uncontaminated water in the aquifer. Since soil contamination involves a number of elements, a pollution index (PI) value was proposed by Nishida et al. (1982) to evaluate the contamination in soils and stream sediments. Many researches have used PI to estimate the degree of multiple metal contaminations (Nishida et al., 1982; Chon et al., 1996; Kim et al., 2001). The

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

PI is computed by averaging the ratios of the total concentrations of the heavy metals to the tolerable levels (Kim et al., 2001). In this study the PI value was calculated for nine elements. PIs(Asy20qCdy3qCuy100qHgy2qPby100 qSby5qZny300qNiy50qCry100)y9 The tolerable levels are the appropriate concentrations of the elements in soils above which crops produced are considered as unsafe for human or animal health (Chon et al., 1996). Tolerable level values were taken from Chon et al. (1996) for As, Cd, Cu, Hg, Pb, Sb, Zn and from Sponza and Karaoglu (2002) for Ni and Cr. PI values above 1.0 indicate the metal concentrations are greater than the permissible levels (Kim et al., 2001) and any element enrichment can be from anthropogenic inputs or natural geological resources (Nimick and Moore, 1991) and monitoring and remediation are required. The PI values of stream sediments from the study area are presented in Table 2. PI values are significantly high and vary between 69 and 82 for stream sediment samples (Ss1, Ss2, Ss3, Ss4, Ss5, and Ss7). Sample from calcine piles (Ss6) has PI value of 683. Sample Ss5 with PI of 69 was taken from the upper parts of the mining area (Fig. 2). PI values of the other stream sediment samples increase close to the calcine piles that are located in the upper part of the stream (Fig. 2). It can be concluded that calcine is the main source of contaminations in the stream sediments. Consequently, PI values for all stream sediments were found much higher than permissible pollution limit of 1.0 (Kim et al., 2001). The stream sediments derived from mining area are potentially hazardous to the environment adjacent to the abandoned Hg mine and in need of remediation. 3. Conclusions Based on the chemical properties of waters with major constituents dissolution of carbonate and sulfide minerals form the hydrogeochemical characteristics of waters from the study area. pH directly effects the metal content of water samples. High pH values are the source of the heavy metals

165

by leaching of the mineralized rocks. Water pH is negatively correlated with Fe, Ni, Al and Hg and a positive correlation with SiO2. High positive correlation of Ca and Mg with SO4 show the dissolution of carbonate and sulfide minerals in the aquifer. Positive good correlations for Hg are seen between Mn, Cu, Ni and Al (r)0.9). Studies on water samples showed that elevated metal concentrations are observed around the abandoned mining area, however, the heavy metal concentrations in groundwater sample from springs and well are under the drinking water standards. Hg contents of the groundwater samples that are currently used for drinking in Karaburun do not exceed the levels of Turkish drinking water standards. Therefore, the groundwaters around the mining area in Karaburun are very slightly contaminated but not seriously polluted by Hg and other toxic heavy metals. The low Hg concentrations in groundwater samples is due to the high resistance nature of cinnabar to chemical and physical weathering and its solubility in water and the adsorption by ironhydroxide or precipitation and combination with sulfide and dilution with uncontaminated water in the aquifer. Although the water resources for Karaburun are not highly contaminated by the abandoned Kalecik Hg mine, stream sediments that move to the plain adjacent to Karaburun have metal contents that are highly above the tolerable levels derived from the mining area. PI values for all of the stream sediments were found much higher than permissible pollution limit of 1.0. The stream sediments derived from mining area are a danger to the environment adjacent to the abandoned Hg mine and are in most urgent need of remediation. Distribution of the metals in the stream sediments show that metals from the mining area move downstream in the rainy season giving rise to contamination in the areas adjacent to the mining area. Acknowledgments The authors would like to thank anonymous reviewer for critical reviews of the manuscript.

166

¨ Gemici, T. Oyman / The Science of the Total Environment 312 (2003) 155–166 U.

References ˘ ¨ Bircan A, Aydoganlı E. Turkiye civa envanteri. Maden Tetkik ¨ ¨ Yayınları 1969;143:52. ve arama Enstitusu Calmbach L. Hydrowin Computer Programme—Version 3.0, by Institut de Mineralogie BFSH2 1015 Lausanne 1995. Chon HT, Cho CH, Kim KW, Moon HS. The occurrence and dispersion of potentially toxic elements in areas covered with black shales and slates in Korea. Appl Geochem 1996;11:69 –76. Cidu R, Caboi R, Fanani L, Frau F. Acid drainage from sulfides hosting gold mineralization (Furtei, Sardinia). Environ Geol 1997;30:231 –237. ¨ ˘ ¨ ¨ T, Ozer Erdogan B, Altıner D, Gungor T. Karaburun Yarımadasınin Stratigrafisi. MTA Dergisi 1990;111:1 –23. Gosar M, Pirc S, Bidovec M. Mercury in the Idrijca River sediments as a reflection of mining and smelting activities of the Idrija mercury mine. J Geochem Explor 1997;58:125 – 131. Gray JE, Theodorakos PM, Bailey EA, Turner RA. Distribution, speciation and transport of mercury in stream-sediment, stream-water and fish collected near abandoned mercury mines in southwestern Alaska, USA. Sci Total Environ 2000;260:21 –33. Gray JE. An overview of mercury transport, cycling, and environmental effects of mercury mining. Mater Geoenviron 2001;48:2 –7. Gray JE, Crock JG, Fey DL. Environmental geochemistry of abandoned mercury mines in West-Central Nevada, USA. Appl Geochem 2002a;17(8):1069 –1079. Gray JE, Greaves IA, Bustos DM, Krabbenhoft DP. Mercury and methylmercury contents in mine-waste calcine, water, and sediment collected from the Palawan Quicksilver Mine, Philippines. Environ Geol 2002b warticle online, DOI. 10.1007yS00254-002-0626-8x. Harsh JB, Doner HE. Characterization of mercury in a river wash soil. J Environ Qual 1981;10:333 –337. Herr C, Gray NF. Environmental impact of acid mine drainage on the Avoca River: Metal Fluxes in Water and Sediment. Part II. Metal contamination of riverine sediments. Dublin: Technical Report 14, Water Technology Research, Trinity College, University of Dublin; 1995. p. 36. Jonasson IR, Boyle RW. Geochemistry of mercury and origins of natural contamination of the environment. Can Min Metall Bull 1972;65:32 –39.

Kharaka YK, Gunter WD, Aggarwall PK, Perkins EH, De Braal JD. Solmineq 88: a computer program code for geochemical modelling of water–rock interactions. US Geological Survey Water Investigations Report 1988;88–4227. Kim KK, Kim KW, Kim JY, Kim IS, Cheong YW, Min JS. Charecteristics of tailings from the closed metal mines as potential contamination sources in South Korea. Environ Geol 2001;41:358 –364. Lee CH, Lee HK, Lee JC. Hydrogeochemistry of mine, surface and groundwaters from the Sanggok mine creek in the upper Chungju Lake, Republic of Korea. Environ Geol 2001;40(4–5):482 –494. Lehnert-Thiel K. Kalecik sianbr zuhurlari ve Karaburun Yari˘ kesiminde jeolojik ve maden yatakları madasının kuzeydogu ¨ ¨ ¨ Dergisi 1969;72:82 – uzerine incelemeler. MTA Enstitusu 109. ¨ ¨ MTA Raporu. Dunyada ve Turkiye’de metal ve mineral kay¸ naklarının potansiyeli, ticareti, beklenen gelismeleri: Civa. ¨ ¨ yayınları, 1977;166. Maden Tetkik Arama Enstitusu ¨ Civa Muracaat ¨ Nar T. ˙Izmir–Karaburun Yarımadası’nda (Turk ¸ Sahalarında) yapılan civa arastırmasina ait rapor. MTA Raporu 1974;5443:23. Nimick DA, Moore JM. Prediction of water soluble metal concentrations in fluvially deposited tailings sediments, upper Clark ork Valley, Montana, USA. Appl Geochem 1991;6635 –6646. Nishida H, Miyai M, Tada F, Suzuki S. Computation of the index of pollution caused by heavy metals in river sediment. Environ Pollut (Ser B) 1982;4:241 –248. Pierce ML, Moore CB. Adsorption of arsenite on amorphous iron hydroxide from dilute aqueous solution. Environ Sci Tech 1980;14:214 –216. Rogers RDV. Volatility of mercury from soils amended with various mercury compounds. Soil Sci Soc Am J 1979;43:289 –291. Rose AW, Hawkes HE, Webb JS. Geochemistry in mineral exploration. New York: Academic Press, 1979. Sponza D, Karaoglu N. Environmental geochemistry and ˘ metal industry district. Environ pollution studies of Aliaga Int 2002;27:541 –553. ˙¸ TSE (Turkish drinking water standards) Sular-Icme ve kullan¨ Standartları Enstitusu, ¨ ¨ Ankara, 1997. ma suları. Turk WHO (World Health Organization). WHO guidelines for drinking water quality, vol 2. Health criteria and other supporting information. WHO, Geneva, 1993.