Journal of Environmental Radioactivity 138 (2014) 1e10
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad
Radioactive and chemical contamination of the water resources in the former uranium mining and milling sites of Mailuu Suu (Kyrgyzstan) € llin, A. Jakob, M. Burger J.A. Corcho Alvarado*, B. Balsiger, S. Ro Federal Office for Civil Protection, Spiez Laboratory, Physics Division, Labor Spiez, CH-3700 Spiez, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 April 2014 Received in revised form 10 July 2014 Accepted 18 July 2014 Available online
An assessment of the radioactive and chemical contamination of the water resources at the former uranium mines and processing sites of Mailuu-Suu, in Kyrgyzstan, was carried out. A large number of water samples were collected from the drinking water distribution system (DWDS), rivers, shallow aquifers and drainage water from the mine tailings. Radionuclides and trace metal contents in water from the DWDS were low in general, but were extremely high for Fe, Al and Mn. These elements were associated with the particle fractions in the water and strongly correlated with high turbidity levels. Overall, these results suggest that water from the DWDS does not represent a serious radiological hazard to the Mailuu Suu population. However, due to the high turbidities and contents of some elements, this water is not good quality drinking water. Water from artesian and dug wells were characterized by elevated levels of U (up to 10 mg/L) and some 2 trace elements (e.g. As, Se, Cr, V and F) and anions (e.g. Cl, NO 3 , SO4 ). In two artesian wells, the WHO guideline value of 10 mg/L for As in water was exceeded. As the artesian wells are used as a source of drinking water by a large number of households, special care should be taken in order to stay within the WHO recommended guidelines. Drainage water from the mine tailings was as expected highly contaminated with many chemicals (e.g. As) and radioactive contaminants (e.g. U). The concentrations of U were more than 200 times the WHO guideline value of 30 mg/L for U in drinking water. A large variation in 234U/238U isotopic ratios in water was observed, with values near equilibrium at the mine tailings and far from equilibrium outside this area (reaching ratios of 2.3 in the artesian well). This result highlights the potential use of this ratio as an indicator of the origin of U contamination in Mailuu Suu. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Uranium contamination Water resources Former uranium mines Mailuu Suu Kyrgyzstan
1. Introduction Uranium ore mining activities in Mailuu Suu started in 1946 and lasted until 1968. Scarce information about the uranium ore characteristics, mining methods and processing volumes is available (Kunze et al., 2007). Large volumes of residues from the ore mining and processing were produced at this mine site. Most of the residues were disposed in near-surface impoundments in the vicinity of the mines and mills, disregarding their potential environmental and human impacts. In total about 3 million m3 of residues were deposited near the town of Mailuu Suu in 23 mine tailings and 13 mine waste dumps (Vandenhove et al., 2006; Kunze et al., 2007).
* Corresponding author. Tel.: þ41 58 468 1788; fax: þ41 58 468 1402. E-mail address:
[email protected] (J.A. Corcho Alvarado). http://dx.doi.org/10.1016/j.jenvrad.2014.07.018 0265-931X/© 2014 Elsevier Ltd. All rights reserved.
During the processing, little or no care was taken to assure the isolation of the tailings from their environment. A typical problem arising from these tailings is the leaching of contaminants (e.g. radionuclides, arsenic and heavy metals) into surface waters and groundwater (BGR, 2008). The environmental and health risks are enhanced by their location near or in the river banks of the Mailuu Suu River and its tributaries, which are the major sources of irrigation water in the region and of drinking water in some areas. Due to the mountainous topography (steep slopes) and the geology, the occurrence of seasonal floods and landslides during rainy and snow melt periods is very common. These natural processes have affected or even destroyed some of the tailings piles (Vandenhove et al., 2006; BGR, 2008). The region of Mailuu Suu is also affected by seismic activity, which has a negative impact on the stability of the waste tailings. The intensive uranium exploitation and the inappropriate management of the residues have had a harmful impact on the
2
J.A. Corcho Alvarado et al. / Journal of Environmental Radioactivity 138 (2014) 1e10
environment in the Mailuu Suu region, especially on the quality of the drinking water resources. According to Stone (2005), from the legacies of uranium mining in Kyrgyzstan, Mailuu Suu posed the greatest risk of an environmental impact. Over the past few years, several projects have been carried out to understand the environmental impact and to develop remediation solutions (Kunze et al., 2007). The most comprehensive studies were carried out by the Belgian Nuclear Research Centre (Vandenhove et al., 2006) and the German Federal Institute for Geosciences and Natural Resources (BGR, 2008). These studies provided valuable data about the environmental situation in Mailuu Suu and the impact of the tailings and waste dumps. It was concluded that the main processes that could produce a severe impact of the tailings were due to: a) a strong erosion of tailings and waste rocks in the river banks by the rivers themselves and strong rainfall events and snowmelt, b) seismic instability, and c) landslides and mudflows (Kunze et al., 2007). Between 2006 and 2008, BGR (2008) conducted a water monitoring program to gain information on the impact of the mining and milling residues on the water resources (surface waters, seepage water and groundwater). This study confirmed that contamination of the water resources was the major hazard for the local population and livestock. It was revealed within this project that most waters were influenced by uranium mineralization or by the remnants of the mining activity, with some waters exceeding international guidelines for water quality. BGR (2008) concluded that waters from the northernmost part of Mailuu Suu were not affected by contamination of any kind. The water intake point of the drinking water distribution system (DWDS) of Mailuu Suu is located in this northernmost area. The water intake plant was designed to be fed by river water after several treatment processes (e.g. filtration, flocculation and sedimentation, and chlorination). BGR (2008) recommended a) the consumption of only processed water, and b) a continuous monitoring of this water and other sources of drinking water (e.g. groundwater). The risks related to the aging of the DWDS infrastructure were nonetheless not considered in any of the previous studies. The DWDS of Mailuu Suu date from the eighties, which means that the system is very old and high corrosion may likely be a significant problem. Aged water pipelines have higher vulnerabilities for example to external contamination. Natural processes such as landslides and flooding, which are common in the region, further increase this vulnerability. In several parts, the old pipelines have already been damaged by these natural processes. Another problem that has not been taken into account in previous projects is the potential accumulation of particles and contaminants within the DWDS. Due to an incomplete removal of suspended solids at the treatment plant, particles and contaminants may enter and accumulate in the distribution network (Vreeburg et al., 2008; Lytle et al., 2013). An associated problem is their potential release back into the water which increases the exposure of the consumers. Lytle et al. (2013) demonstrated, for example, that measurable levels of radium, thorium and uranium were accumulated in the water distribution systems of four States in the USA. The main objective of our study was to assess the drinking water quality in the DWDS of Mailuu Suu (Kyrgyzstan). In order to locate potential sources of contaminants within the distribution system, samples were collected from the water intake plant and at different points in the distribution network. The samples were analyzed for chemical and radiological contaminants. Other sources of drinking water such as surface waters and groundwater were also investigated. Seepage and drainage waters, and soil and sediments from the tailings were sampled in order to verify the success of some previously undertaken remediation measures (e.g. relocation of mine tailings).
2. Study site The town of Mailuu Suu is located in the southwestern part of Kyrgyzstan, in the Fergana Valley (Fig. 1). The town is divided by the Mailuu Suu River that later feeds the Syr Darya River. Approximately 25,000 inhabitants live today in Mailuu Suu. A large number of the households are connected to the centralized water distribution network, which is fed with river water. Before entering the DWDS, the river water undergoes several processing steps at the water intake plant (filtration, coagulation and sedimentation, chlorination). It is not certain that these processing steps are working adequately. The water intake plant is located a few kilometers north of the mining and milling areas, away from any known source of uranium contamination. BGR (2008) concluded that water provided by this plant was of good quality, however the number of complaints about the water quality of the DWDS in Mailuu Suu has increased considerably in recent years (Voitsekhovych, 2012; World Bank, personal communication, 2013). Geologically, the town is settled in consolidated and unconsolidated sedimentary rocks of Mesozoic and Neozoic age. In the northern part, the valley is formed of sedimentary strata of Jurassic age (BRG, 2008). The river valleys locally consist of Quaternary unconsolidated sediments of alluvial origin. The rocks consist mainly of fine grained limestone sequences with intercalated marl layers. In the town area, a dominant sandy component is observed. The hydrogeology of the region is dominated by two main aquifer systems. A shallow aquifer located in the unconsolidated Quaternary alluvial sediments (sands, clays and gravels) along the center of the river valleys and in the city area. The second aquifer is located at deeper depths in solid and fractured rocks, with the formation of karstic systems as in the area of Kara-Agach. A conceptual hydrogeological model of the region indicates that contamination from the deeper strata (e.g. uranium) could be transported upward to the shallow systems of Mailuu Suu (BGR Report, 2008). The significance of this contamination pathway is not known. Moreover, surface waters and groundwater are supposed to communicate with each other, which supports an exchange of solutes. Contaminated groundwater has been observed all along the valley (BGR Report, 2008). Uranium was shown to be one of the main contaminants in groundwater in Mailuu Suu Valley, but also other contaminants (e.g. arsenic) gave rise to health concern (BGR Report, 2008). 3. Materials and methods A joint field mission to the Mailuu Suu mining site was carried out in August 2013 by a joint collaboration between Switzerland (SPIEZ LABORATORY and the Swiss Embassy in Kyrgyzstan), the World Bank (Country Office in Kyrgyzstan) and Kyrgyzstan (experts from ARIS and from the Mailuu Suu Epidemiological Laboratory). The DWDS was investigated at a total of fourteen sites. Additionally, five artesian wells, one dug well, three control points in the river €k-Tash) and three mine tailings (Mailuu Suu, Kara-Agach and Ko (former TP3 and TP5/TP7) were sampled. A total of one hundred water samples, one river sediment and three surface soils were taken at 27 different locations. The locations of the sampling points are shown in Fig. 1. Taking into account the large area of the tailings and the fact that sampling was conducted in a dry month, the water sampled at the mine tailings may not be representative of the totality of water draining the sites. The river sediment was taken at the interception point of the Mailuu Suu River and the main drainage channel of the former mine tailing TP3. On-site the water parameters pH, conductivity, temperature, redox potential, dissolved oxygen and turbidity were measured using HACH LANGE GmbH field equipments and their
J.A. Corcho Alvarado et al. / Journal of Environmental Radioactivity 138 (2014) 1e10
3
Fig. 1. Map of Mailuu Suu with locations of the sampling points.
recommended methods. Indicative tests for alkalinity, hardness, ammonia, chlorine, nitrate, nitrite and sulfate were performed. In situ measurements of the external gamma dose rates at 1 m above ground surface were conducted at each sampling location using Dose Rate Meters 6150AD6 (Automess, Germany). 222Rn concentrations in air were screened using an Alphaguard (Saphymo GmbH, Germany). The mean value of three measurements of 10 min was reported for each site. Sets of water samples were filtered through 0.45 mm filters and taken in appropriate bottles (PP, Nalgene®), partially stabilized onsite (HNO3 for cations; or NaOH for anions) for selected analyses. In order to identify the fraction of radionuclides and metals associated with particles, additional water samples were taken without filtration (stabilized by acidification with HNO3). Complete analysis of the water samples for the major and trace elements, heavy metals, selected anions and cations, and radionuclides was performed at SPIEZ LABORATORY. Ion Chromatography (Metrohm IC-881, Metrohm AG) of the filtered water samples was used for the determination of: F, Cl, NO2, Br, 3 2 þ þ þ þ 2þ 2þ NO (estimated 2s3 , PO4 , SO4 , Li , Na , NH4 , K , Ca , Mg uncertainty of 5%). An aliquot of 20 ml of the non-filtrated water samples was mineralized using microwave-assisted digestion (Milestone, MLS ETHOS plus) as described in the EPA 3015 method (USEPA, 2007). After digestion, the samples were filtered through 0.45 mm membrane filters and diluted to 100 mL with double distilled water. Semi Quantitative Inductively Coupled Plasma e Mass Spectrometry (ICP-MS, PerkinElmer, DRCII) was conducted in all samples for the determination of Be, Al, V, Cr, Mn, Co, Ni, Cu, Zn, As, Se, Mo, Ag, Cd, Sn, Sb, Ba, W, Hg, Tl, Pb, Th and U (estimated 2suncertainty of 20%). For the determination of Fe, Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES, Varian, VISTA-PRO) was utilized. For the elements U, Ra and Th, and their
isotope ratios, double focusing sector field Inductively Coupled Plasma Mass Spectrometry (sf-ICP-MS, Element2, Finnigan) was used. For this analysis, the samples were acidified to 2% with nitric acid and the tracer Indium was added as an internal standard. The 238 U, 226Ra and 232Th concentrations were measured quantitatively with the external calibration. The external calibration is based on the measurement of a calibration solution that contains 100 ppt of 238U, 100 ppt of 232Th, 2 ppt of 226Ra and 500 ppt of Indium. The isotopic composition was measured without any mass discrimination corrections. The mass discrimination was less than 1%. The river sediment and two surface soil samples (0e5 cm) were collected near the output point of the drainage water flowing out of the tailings. One surface soil sample was taken at the entrance gallery of a “closed” mine. The samples were dried at 60 C until mass constancy, homogenized, ground and sieved through 2 mm mesh. The dry weight was determined for each sample. The samples were then weighed in plastic containers of known geometry (plastic beakers of 0.25 or 0.5 l) for gamma spectrometry measurement. Gamma emitters were analyzed by high-resolution gamma spectrometry using high purity germanium detectors (HPGe, resolution of 1.8 keV at 1332 keV, Canberra GmbH). 137Cs was determined through its gamma line at 661.6 keV, 210Pb through its gamma line at 46.5 keV, 214Bi through its gamma line at 609.3 keV, 214Pb through its gamma line at 351.9 keV, 235U through its gamma line at 143.8, 226Ra through its gamma line at 186.2 keV (after correcting for the contribution of the 235U gamma line at 185.7 keV), 234mPa through its gamma line at 1001 keV and 234Th through its gamma line at 63.3 keV. Efficiency calibration was performed using the mathematical efficiency calibration software LabSOCS from Canberra GmbH (Venkataraman et al., 2005). The typical counting time was two days.
4
J.A. Corcho Alvarado et al. / Journal of Environmental Radioactivity 138 (2014) 1e10
4. Results and discussion 4.1. Physical parameters and chemical composition of the water The location of the sampling points and the results of some of the field measurements are listed in Table 1. The chemical composition of the filtered (dissolved fraction) and non-filtrated (total) water samples is presented in Table 2. The water parameters pH (8.3e8.6), conductivity (150e440 mS/cm), temperature (13e24 C), redox potential (98e240 mV) and the dissolved oxygen (6.8e8.3 mg/L) in the DWDS were in acceptable ranges according to the WHO recommendations (WHO, 2011). The indicative tests for alkalinity (measured range: 100e240 mg/L), hardness (measured range: 120e250 mg/L), ammonia (not detected), chlorine (not detected), nitrate (not detected), nitrite (not detected) and sulfate (measured range < 800 mg/L) in the DWDS also fulfilled the guidelines and recommended values for drinking water (WHO, 2011) . Water from the river and the DWDS were however characterized by extremely high turbidities (46e307 NTU, Table 1) that were caused by a high particle load. The highest turbidity of 307 NTU was measured in a household located in the Pioneerska street. The water from the river and the DWDS exceeded the WHO recommended value for turbidity (1 NTU) and are not good quality drinking water. Turbidities for these samples were well correlated with their contents of Al, Fe, Mn and Cr (Fig. 2). Significant proportion of Al (>99%), Fe (>99%), Mn (81e99%) and Cr (>99%) were found to be associated to the particle fractions (Table 2). Elevated turbidity of water in the DWDS is likely caused by an inadequate treatment of the river water which is delivered into the system. Minor amounts of Fe and Al may also be present in drinking-water as a result of their use as coagulants or, in the case of Fe, the corrosion of steel and cast iron pipes during water distribution. Fe, Al and Mn are among the most abundant metallic elements in the Earth's crust, and they do not present a hazard to health (WHO, 2011). However, health-based values have been established as a precaution against storage in the body of excessive amount of these elements. Most of the water samples from the DWDS and the river do not fulfill the WHO health-based limits for Al (0.9 mg/L) and/or Fe (2 mg/L). The health-based value for Mn of 0.4 mg/L was well above the concentrations of Mn found in the river and the DWDS. In Fig. 3, the concentrations of U and the dissolved fractions of 2 Cl, NO 3 and SO4 are plotted against the Na concentrations in water. This last element was chosen as a reference because its concentration varies over a large range depending on the origin of the water (e.g. tap water, groundwater or river water). Water from the DWDS is indistinguishable in terms of chemical composition (e.g. U, Cl, SO2 4 and NO3 ; Fig. 3) and physical characteristics (e.g. high turbidities, Fig. 2) from river water upstream of the contamination sources (sample 32). This result shows that the DWDS is fed by non-treated river water and has not been affected by other water sources such as contaminated water entering broken pipelines. On the other hand, water from the DWDS differs considerably from ground and drainage waters (Fig. 3). Groundwater, which is also used for drinking purposes, has as expected, a higher content of most elements due to longer period of watererock interaction processes in the subsurface. The concentration levels of some trace elements (As, Se, Cr, V, U, F) anions (Cl, þ 2 NO 3 , SO4 ) and cations (Na ) were significantly higher than in the river and the DWDS (Tables 2 and 3). This type of water was nonetheless characterized by acceptable levels of turbidity and low contents of Al, Fe and Mn (Tables 1 and 2, Fig. 2). Relatively high levels of As, exceeding in two sites the WHO guideline value of 10 mg/L (sites 13 and 15), were observed in groundwater (Table 2).
The presence of arsenic in water samples at levels above 10 mg/L is of high concern as this element has acute toxicity in humans (WHO, 2011). These high arsenic concentrations might be caused by a naturally elevated arsenic content of the rocks in contact with groundwater. The reducing redox conditions observed in two artesian wells (sites 13 and 15) favor an enrichment of dissolved arsenic in groundwater. The concentrations of Se and Cl in four artesian wells exceeded the guideline values of the European Union Directive on the quality of water intended for human consumption (EU, 1998). One of these wells is located near a school and the water is used by a large number of children for drinking purposes. Special care should be taken in order to stay within the WHO recommended guidelines. Elevated concentrations of many chemical (e.g. As) and radioactive contaminants (e.g. U) were detected in the few water samples collected near the three waste and mine tailings sites investigated (TP5/TP7 and former TP3). The concentrations of U in the water sample taken near the TP5/TP7 sites was more than 200 times the WHO guideline value of 30 mg/L for U in drinking water (Table 3). High levels of As exceeding the WHO guideline value of 10 mg/L were detected (Table 2). Particularly important is the high level of As of 1600 mg/L detected near the TP5/TP7 sites, because at these mine tailings the water flow out of the protective fence and may be consumed by animals. 4.2. Radionuclides in water A summary of the radionuclides detected in the water samples from the Mailuu Suu area is presented in Table 3. The naturally occurring radionuclides of the uranium decay series (e.g. 238U, 234U, 235 U and 226Ra) are the main radioactive contaminants in water. The highest contamination was observed in the streams that were draining the investigated mine tailings. Uranium concentrations of 159 and 6820 mg/L (Table 3) were measured in water from the tailings TP5/7 and former TP3, respectively. These concentrations are up to more than 200 times above the WHO guideline value of 30 mg/L for uranium in drinking water. An enhanced protective fence would assure that these waters could not be consumed by animals. Elevated concentrations of uranium of 3e10 mg/L were also observed in four groundwater samples taken in the town of Mailuu €k-Tash (samples 41, 42, 51 and 55, Table 3). Nonetheless, Suu and Ko these samples fulfilled the WHO guideline value for uranium in drinking water. The uranium content in the samples taken from the DWDS varied in a narrow range from 0.27 to 0.34 mg/L (Table 3), indicating their similar origin. As expected, this range of variation does not differ much from the value observed in the Mailuu Suu River upstream of the former mining areas and the tailings (sample 21, near the hot water plant). However, uranium concentrations of an order of magnitude higher were observed in the Kara-Agach €k-Tash, 4.2 mg/ River (3.3 mg/L) and in the Mailuu Suu River (in Ko L), downstream of the tailings. Although these concentrations are low, this result confirms that river water is being contaminated with uranium from the tailings and potentially from uraniumbearing rocks. A similar result was obtained by Vandenhove et al. (2006) in a previous study in this region. In drainage water from the tailings, uranium isotopes were in their natural radioactive equilibrium (activity ratio of 234U/238U ¼ 1, and 235U/238U ¼ 0.047, Fig. 4). In the rest of the samples, except in € k-Tash, the uranium isotopes were in the river water from Ko disequilibrium. All these samples contained an excess of 234U with respect to the radioactive secular equilibrium with 238U (234U/238U > 1, Fig. 4). Elevated ratios are normally expected in surface waters and groundwater due to a preferential removal of 234 U from the rock minerals (Suksi et al., 2006; Porcelli, 2008;
J.A. Corcho Alvarado et al. / Journal of Environmental Radioactivity 138 (2014) 1e10
5
Table 1 Sampling locations for water, soil and sediment samples and results of some field measurements. Site code
GPS coordinates N
Location
pH
E
Drinking water distribution system 11 41 19.539 072 31.978 Sedimentation pounds, water intake plant 12 41 19.543 072 32.032 Reservoir, water intake plant 14 41 18.679 072 30.086 Household, Pioneerska Street 23 41 15.331 072 28.459 Epidemilogical laboratory (SESS) 31 41 17.254 072 28.888 Household, Kara Agach Village 34 41 16.000 072 28.586 Household, Mailuu Suu Town 35 41 15.443 072 28.689 Household, east side river 36 41 16.066 072 27.847 Bazarne street (pipeline) 37 41 15.532 072 28.435 High school, Mailuu Suu 43 41 14.873 072 27.816 School no. 1, outside 44 41 14.896 072 27.178 Household, Bedresay 45 41 14.146 072 26.571 Technical school 46 41 13.584 072 26.352 Household, east side of river 47 41 14.657 072 27.763 Household, Mailuu Suu Town River water 21 41 17.932 072 29.564 Close to thermal water plant 32 41 18.107 072 27.761 Kara Agach, north €k Tash 52 41 10.415 072 24.178 Ko Artesian wells 13 41 18.704 072 30.090 Pioneerska street 15 41 18.168 072 29.698 Thermal water plant 41 41 14.962 072 28.046 Distrcit 17 (Mosque) 42 41 14.833 072 27.818 Near school 1 €k Tash 51 41 11.289 072 24.413 Ko 55 41 14.720 072 27.634 Center of Mailuu Suu Drainage water from tailings 22 41 17.615 072 29.078 Tailings TP5/TP7 54 41 10.824 072 28.971 Former tailing TP3 Soils and sediment 22 41 17.615 072 29.078 Soil, tailings TP5/TP7 33 41 16.487 072 28.650 Soil, mine entrance, at 200 m from the ISOLIT plant 53 41 10.824 072 28.971 Soil, former tailing TP3 54 41 10.824 072 28.971 River sediment, former tailing TP3
Conductivity [mS/cm]
Water temperature [ C]
Redox potential [mV]
Dissolved O2 [mg/L]
Turbidity NTS
Radon in air [Bq/m3]
Mean gamma dose rate [mSv/h]
7±5
0.04
8.3
351
16.4
240
7.8
87
8.4 8.4 8.4 8.5 8.5 8.5 8.4 8.5 8.5 8.5 8.5 8.5 8.4
210 210 211 210 210 211 221 210 217 220 153 213 215
18.0 16.0 15.2 15.3 16.8 18.0 23.7 18.2 21.0 19.9 19.6 22.3 22.4
180 180 160 185 130 130 120 122 110 98 125 119 110
6.8 7.5 8.1 7.9 8.0 8.0 6.8 8.0 7.4 8.1 7.9 7.3 8.0
46 307 145 131 156 114 137 116 68 111 92 90 110
7±5 10 ± 5 60 ± 15 22 ± 9 56 ± 30 18 ± 8 e 22 ± 9 42 ± 10