Impacts of Open Placer Gold Mining on Aquatic Communities in Rivers of the Khentii Mountains, North-East Mongolia
Daniel A. Krätz1, Ralf B. Ibisch2, Saulyegul Avylush3, Ganganmurun Enkhbayar3, Soninkhishig Nergui3 and Dietrich Borchardt2
1University of Kassel, Centre for Environmental Systems Research (CESR), Department for Integrated Water Resources Management, Kassel, Germany
2Helmholtz-Centre for Environmental Research-UFZ, Department Aquatic Ecosystem Analyses
and Management, Magdeburg, Germany
3National University of Mongolia, Faculty of Biology, Ulaanbaatar, Mongolia
Since the political change and due to high market prices for gold the mining sector has considerably grown in Mongolia and has become one of the most important economical sectors. With increasing mining activities the mining related environmental problems also increased, especially those, which are caused by open placer gold mining. Placer gold deposits are located in the alluvial sediments of river fl oodplains and the exploitation of these deposits often induces severe impacts to river ecosystem and its different components. In this paper we describe the effects of open placer gold mining on diatoms, benthic invertebrates and fi sh in four rivers in the north-east of Mongolia. Our fi ndings are based on a comparative analysis of these biocoenotic groups in pristine and mining affected river sections, taking into account also abiotic habitat characteristics. Our analyses revealed that placer gold mining causes multiple stressors acting on different trophic levels. The biocoenotic groups under investigation reacted differently against stressors, and we indentifi ed a wide range of direct and indirect effects. These fi ndings are new for Mongolia and are essential to defi ne adapted and successful strategies for an ecologically based management and monitoring of open placer gold mining pressures and ecological impacts.
Keyword: Turbidity,suspended sediments,clogging,river continuum,fi sh fauna,macroinvertebrates,diatoms
In North-East Mongolia both extremes exist: (1) large-scaled undisturbed and pristine landscapes and (2) river ecosystems, which are faced with a rapidly growing mining industry with its diverse environmental impacts (World Bank, 2006). The majority of the gold deposits of this region are located in so called ´placer deposits´ in the alluvial sediments of the river fl oodplains. The exploitation of these placer deposits causes diverse environmental problems infl uencing the river-ecosystem in multiple ways (Farrington, 2000). For Mongolia placer gold mining is seen to be responsible for the entire loss or eventually long-lasting damage of at least 29 large-scaled river eco-systems, respectively parts of it (Mongolian Ministry of Nature and Environment, 2003). Worldwide, the effects of open placer gold mining on aquatic ecosystems are well documented. Numerous authors have reported on the effects for the ecosystem itself (Newcombe & MacDonald, 1991) or specifi c components, e.g. primary production (Van Nieuwenhuyse & La- Perriere, 1986), macroinvertebrates (Wagener & LaPerriere, 1985) or fi sh (Pentz & Kostaschuk, 1999; McLeay et al., 1987). Although in Mongolia mining is seen to be a major threat to river ecosystems and its biocoenoses (Grayson, 2003; Ocock et al., 2006), little is known about the sitespecifi c impact and the scientifi c basis for targeted regional management given the specifi c situation for Mongolia. The main focus of this study was, therefore, to comparatively analyze the biocoenoses in pristine and mining affected sites and to estimate the effects of open placer gold mining on different biocoenotic groups (diatoms, macroinvertebrates and fi sh). Based on these analyses principles for an ecologically orientated monitoring and management of the open placer gold mining in Mongolia are derived. Study Area Two study areas were chosen in the border area of the strictly protected area (SPA) Khan Khentii, which is located in the north-east of the capital city, Ulaanbaatar. The study areas had both features, river stretches which can be assumed to be pristine with negligible human impact, and others which are severely affected by open placer gold mining.
The Yalbag River is located in the western border area of the SPA Khan Khentii and it belongs to the Eroo watershed (Table 1). At the river Yalbag mining activities were conducted in middle and down reaches of the watershed (see Fig. 1a). Along these stretches the natural bank structures and the riparian zone were predominantly lost due to open placer mining activities which normally start with the removal of organic top soil and the turning over of valley bottom sediments to the bedrock for subsequent processing to remove gold. In addition, mining activities caused severe changes in the matter and sediment balance downstream of the mined sites. In the study area the upstream part of the river Yalbag, called Barchuluut and another confl uences of the river Eroo, called Tsagaan Chuluut served as a reference. Our second study area was the river Terelj, which is part of the Kherlen watershed and is located south of the SPA Khan Khentii. The mining area is situated in the upper reaches of the river Terelj (see Fig. 1b) also infl uencing the middle and down reaches by high emissions of suspended sediments.
We focused our studies to this area along a longitudinal transect of 40 km in length. We also studied one site upstream of the mining area and three tributaries of the river Terelj serving as reference sites. All study areas belong to the biocoenotical region of the Epi- to Metarhithral (Jungwirth et al., 2003).
Material and Methods
The effects of open placer gold mining on the biocoenosis were studied with regard to diatoms, macroinvertebrates and fi sh. Field surveys for diatoms were conducted monthly during the icefree periods in 2004 and 2005. Within the qualitative sampling program samples from stones, fi ne substrates and macrophytes were taken and diatoms were later analyzed in the laboratory using standard methods and equipment (Krammer & Lange-Bertalot, 1986-91). The macroinvertebrate community was studied quantitatively synchronal with the sampling program for diatoms. Sampling was conducted at riffl e, pool and bank habitats using a surber-sampler equipped with 500 μm gauze. In riffl es and pools the substrate within the 32 x 42 cm frame of the sampler was kicked for about one minute to a depth of approximately 15 cm, big stones within the frame area were brushed and then set aside. In plant habitats an area equal to the frame size was washed using hands to let the currency carry the macroinvertebrates into the Surber sampler held directly adjacent to the sampling site. The material in the sampler mesh was poured through two buckets and an analysis sieve of 500 μm, and was then transferred onto white plastic trays. The samples were sorted in the fi eld, put into glasses containing 80% ethanol and then transferred to the laboratory. In the laboratory organisms were sorted from organic matter and stored in 80% ethanol until they were identifi ed to the lowest taxonomic level possible, usually to genus. The fi sh fauna was also sampled quantitatively using electro-fi shing gear (Hans Grassl GmbH, Germany; Type ELT 60). The fi sh were caught, determined and measured in length and weight. With the length and weight data the condition factor after Fulton (Ricker, 1975) was calculated and considered as fi tness indicator. Additionally to the quantitative sampling migratory patterns of fi sh were studied by installing weirs at several sites. Weirs were installed downstream and inside the mining area at the river Yalbag as well as at one by-pass channel, which spans parts of the mining area (see Figure 7). This study was performed during a two-week time period in spring 2006, whereas electro fi shing campaigns were carried out regularly from 2003 through 2006.
Laboratory analyses identifi ed 62 diatom species belonging to 35 genera with no signifi cant differences in species and genera numbers between pristine and mining affected sites. However Canonical Correspondence Analyses (CCA) based on the taxa and the relative abundances pointed out differences between two groups of sites, and cluster analyses clearly separated the two groups (Fig. 2).
Comparative analyses based on the ecological guilds revealed that the relative amount of mobile taxa (e.g. Cylindrotehca, Gyrosygma, Navicula, Nitzschia and Surirella) were signifi cantly higher at affected sites, when compared to reference sites. For macroinvertebrates the number of taxa (N) and total abundance of individuals (ind. m-2) were signifi cantly higher in pristine sites than in affected sites, especially in pool-habitats. At the study sites in the Eroo watershed this pattern was signifi cant (t-test, p < 0.05) during spring season (May and June), and visible, but not signifi cant during the rest of the sampling season (see Fig. 3). At the river Terelj the benthic invertebrate community was clearly different at sites downstream of the mining area. Here the abundance of macroinvertebrates decreased from more than 8.000 ind./m² upstream the mining area to only a few individuals at the site located directly downstream of the mining area (Fig. 4). Along the 40 km transect the benthic invertebrate abundances remained relatively low and reached maximum values of less than 2.000 ind./m². Regarding the habitat type, pool habitats seemed to be most affected, showing the lowest number of individuals when compared to riffl e and bank habitats. The mining activities also infl uenced benthic invertebrate species composition. A nonmetric, multi-dimensional scaling (NMDS) based on abundances showed that macro invertebrate communities differed signifi cantly between pristine and affected sites (ANOSIM signifi cance: < 0.001, see Figure 5).
In Figure 5 the abiotic variables conductivity, pH, oxygen concentration, temperature and suspended sediments were included in a second step by environmental fi tting. Differences between these parameters were signifi cant between reference and downstream sites (Mann-Whitney test, p < 0.05) but seemed to be ecologically relevant only for suspended sediments. For the fi sh fauna the results were somehow ambiguous. Based on the species community no differences were found between reference and disturbed sites (see Table 2). In the study area of the river Yalbag the number of species was actually higher in disturbed sites than in the reference sites (8 compared to 7 and 5, respectively). A similar pattern was found for the total abundances of fi sh at the river Yalbag, which were signifi cantly higher (Mann-Whitney-Test, p< 0.05) downstream of the mining area. At the river Terelj no such pattern could be observed.
The analyses of the fish community showed clear differences between the sites (based on a NMDS). At the mining affected sites, non-sensitive species like siberian dace and siberian stone loach were dominant, whereas in reference sites they were rare or even missing. Signifi cant differences were also found for the individual fi tness of some fi sh species (see Fig. 6). The Fulton condition factors for arctic grayling, lenok and minnow were found to be signifi cantly lower (Mann- Whitney-Test, p < 0,005) at the affected sites (Yal). At the river Terelj only minnow displayed signifi cant differences with individuals being less corpulent at the affected sites (Ter). A two-week survey in 2006 revealed that mining activities also affected the migration behavior of fish.
Lenok and arctic grayling were constantly caught downstream of the mining
area and in the by-pass channel (weir 1, 2 & 3a), whereas in the weir inside the mining area (weir 3b) these species were almost missing. Stone loach on the other hand was numerously caught in all weirs, whereas minnow was found almost exclusively in weir 3b.
Based on the comparative analyses of impacted and reference sites our study showed that the biocoenotic groups reacted differently to the disturbances caused by open placer gold mining. For diatoms we found a change in the community structure from non-mobile (affi xed) to rather mobile taxa, which could be caused by sedimentation of fi ne sediments (Passy, 2007). We assume that elevated loads of suspended sediments are likely to cause higher sedimentation rates. As a consequence benthic biofi lms get covered with sediment. Mobile species may be able to respond to this environmental change and move to the sediment surfaces in order to get access to favorable light conditions, whereas immobile taxa get covered by sediment deposits and are likely to die off due to the lack of light. This fi nally causes an increase in the relative abundance of mobile species. Comparative studies showed similar results (Dickman et al., 2005 ), but these indirect effects have not been shown for gold mining areas. In our study the macroinvertebrate communities showed direct reactions to the environmental changes downstream of the mining areas. Based on our analyses we conclude that elevated loads of suspended sediments are of major concern. Negative effects of increased suspended sediment loads are well known and have been numerously documented in literature (see review of Newcombe and MacDonald, 1991). Direct effects for example result from an increase of shear stress causing elevated drift rates of organisms (Berry et al., 2003) or clogging of respiratory organs. Indirect effects for macroinvertebrates result from lowered benthic biofi lm biomass and food quality due to light limitation (Van Nieuwenhuyse & LaPerriere, 1986, Ryan, 1991) and deposition of fi ne sediments. This results in a bottomup control of benthic invertebrate biomass and abundance (Wagener & LaPerriere, 1985; Wood & Armitage, 1997; Fossati et al., 2001) and benthic invertebrates diversity (Quinn et al., 1992). Another relevant factor is the loss of intergravel space due to clogging processes, which were found to occur at the impacted sites at river Yalbag (Ibisch et al., 2007). Clogging processes reduce the availability of refugial habitats in the intergravel space, which is especially important during catastrophic events like fl oods or temperature extremes in winter times. For fi sh we found a negative effect of gold mining on the corpulence factors of selected species. Several aspects have to be taken into account in order to explain this contradictory fi nding. For some fi sh species it is well documented, that turbidity and elevated sediment loads lower the effi ciency of food intake, and thus lowers growth rates of fi sh (Lloyd et al., 1987; Rowe & Dean, 1998). Another factor can be seen in the bottom-up control of fi sh biomass by productivity of lower trophic levels. Reduced macroinvertebrate biomass and abundances are likely to affect fi sh biomass and abundances and may at the end also reduce fi sh stocks (Harding & Boothroyd, 2004; Rowe & Dean, 1998). Our study also revealed effects of gold mining on the migratory behavior of fi sh and thus on the river continuum. Numerous authors have been described the impact of suspended sediment on the behavior of fi sh (Newcombe & MacDonald, 1991) with some studies also reporting the fact that fi sh do interrupt the migration further upstream (McLeay et al., 1987). No direct effects of gold mining on fi sh species richness or abundance were found in this study. This is possible due to relatively low suspended sediment concentrations at the studied river sections with concentrations below 200 mg/l (Ibisch et al., 2007). Following the review of Newcombe & MacDonald (1991) these values can be ranked as marginally or moderately harmful, which might be an explanation for our fi ndings.
Clear effects on the other hand were detected for fi sh community structure, with habitat alterations assumed to be a major cause for these effects. Additionally, water temperatures in Yalbag river downstream the mining sites were signifi cantly elevated (1.92 ± 4,40°C upstream, versus 4.07 ± 6.73°C downstream, yearly average values based on continuous temperature measurements, Ibisch et al., 2007). This elevation in stream temperatures are likely to cause changes in the fi sh community (Reeves et al., 1987), and may promote the dominance of siberian dace in our case. Major impacts on fi sh also result from clogging of the river bed (Ibisch et al., 2007) regarding habitat quality and habitat availability in the hyporheic zone (see detailed discussion in Krдtz, 2009). Our analyses showed that open placer gold mining causes multiple stressors in riverine ecosystems. However, these stressors act on different levels. Focusing only on the input of fi ne sediments one realizes, that the observed effects are highly interlinked (see Fig. 8). Besides the input of fi ne sediments, there are other important disturbances caused by open placer gold mining related to the hydrological regime, chemical contaminants or loss of the natural riparian zone. These disturbances on the other hand each cause another cascade of effects, which are still barely understood and where future research still needs to be conducted. In our study the biocoentic groups reacted differently to stressors and we indentifi ed a wide range of effects, both direct and indirect ones. The following table summarizes the results for diatoms, macroinvertebrates and fi sh. We classifi ed the observed effects systematically in large, moderate or minor effects. For the majority of open placer gold mines in Mongolia best management practices are not implemented (World Bank, 2006), and functional monitoring programs are not established by the offi cial administration. This is also true for the mining areas assessed in this study. This observation is in clear contrast to the results of numerous reports, which concentrated on mining and environmental issues (Farrington, 2000; Grayson, 2003; World Bank, 2006). Here the urgent need for targeted and effective management strategies and observation programs are emphasized. Our fi ndings contribute to better understand environmental effects of mining and may serve as a scientifi c basis for the formulation of adapted management and monitoring strategies for the Mongolian placer mining industry.
This project was funded by the German Federal Ministry of Science and Research (BMBF), Funding-ID 0330398.
- Berry, W., Hill, B., Melzian, B. & Rubinstein, N., 2003. The Biological Effects of Suspended and Bedded Sediments (SABS) in Aquatic Systems: a Review, Internal report prepared by the U.S. Environmental Protection Agency, Office of Research and Development.
- Dickman, M. D., Peart, M. R. & Yim, W.W.-S., 2005. Benthic diatoms as indicators of stream sediment concentration in Hong Kong. Internationale Revue der gesamten Hydrobiologie, 90(4): 412– 421.
- Farrington, J., 2000. Environmental problems of placer gold mining in the Zaamar Goldfi eld, Mongolia. World Placer Journal, 1: 107-128.
- Fossati, O., Wasson, J.G., Hery, C., Salinas, G. & Marin, R., 2001. Impact of sediment releases on water chemistry and macroinvertebrate communities in clear water Andean streams (Bolivia). Archiv fьr Hydrobiologie, 151(1): 33-50.
- Grayson, R., 2003. Impacts of placer gold mining on the red book species of Mongolia. World Placer Journal, 3.
- Harding, J. and Boothroyd, I., 2004. Impacts of mining. In: J. Harding, P. Mosley, C. Pearson and B. Sorrell (Editors), Freshwaters of New Zealand. Hydrological & Limnological Societies, Christchurch: NZ pp. 36.1-36.10.
- Ibisch, R.B., Krдtz, D. & Borchardt, D., 2007. Beeinfl usst die Kolmation des hyporheischen Interstitials den Temperaturhaushalt von Flie"gewдssern? In: Deutsche Gesellschaft fьr Limnologie. (DGL) (Editor), Tagungsbericht 2006.Dresden.
- Jungwirth M. et al., 2003. Angewandte Fischцkologie an Flie"gewдssern. 1. Aufl age, Wien, Facultas-Universitдtsverlag. Krдtz, D., 2009. Цkologie der Fischbestдnde in Flie"gewдssern des Khentii-Gebirges (Mongolei): Bestandsaufbau, Dynamik und Gefдhrdung durch den Gold-Tagebau., PhD thesis, University of Technology Dresden, 173 pp.
- Krammer, K. & Lange-Bertalot, H. 1986- 91. Sь"wasserfl ora von Mitteleuropa, Bacillariophyceae. 2/1 Naviculaceae, 876 pp.; 2/2 Bacillariaceae, Epithemiaceae, Surirellaceae, 596 pp.; 2/3 Centrales, Fragilariaceae, Eunotiaceae, 576 pp.; 2/4 Achnanthaceae, 437 pp.;
- Gustav Fischer Verlag, Stuttgart. Lloyd, D.S., Koenings, J.P. & La Perriere, J.D., 1987. Effects of turbidity in fresh waters of Alaska. North American Journal of Fisheries Management, 7: 18-33.
- McLeay, D.J., Birtwell, B.F., Hartman, G.F. & Ennis, G.L., 1987. Responses of arctic grayling (Thymallus arcticus) to acute and prolonged exposure to Yukon placer mining sediment. Canadian Journal of Fisheries and Aquatic Sciences, 44: 658-673. Mongolian Ministry of Nature and Environment, 2003. Annual Report.
- Newcombe, C.P. & MacDonald, D.D., 1991. Effects of Suspended sediments on Aquatic Ecosystems. North American Journal of Fisheries Management, 11: 72-82.
- Ocock, J. et al., 2006. Mongolian Red List of Fishes. Vol. 3, London. Passy, S.I., 2007. Diatom ecological guilds display distinct and predictible behavior along nutrient and disturbance gradients in running waters. Aquatic Botany, 86(2): 171-178.
- Pentz, S.B. & Kostaschuk, R.A., 1999. Effect of placer mining on suspended sediment in reaches of sensitive fi sh habitat. Environmental Geology, 37(1-2): 78-89.
- Quinn, J.M., Davies-Colley, R.J., Hickey, C.W., Vickers, M.L. & Ryan, P.A., 1992. Effect of clay discharges on streams - 2 Benthic invertebrates. Hydrobiologia, 248: 235–247.
- Reeves, G.H., Everest, F.H. & Hall, J.D., 1987. Interaction between redside shiner (Richarsonius balteatus) and the steelhead trout (Salmo gairdneri) in western Oregon: the infl uence of water temperature. Can. J. Fisheries and Aquatic Sciences, 44: 1603- 1613.
- Ricker, W.E., 1975. Computation and interpretation of biological statistics of fi sh populations. Bulletin of the Fisheries Research Board of Canada, 191: 1 - 382.
- Rowe, D.K. & Dean, T.L., 1998. Effects of turbidity on the feeding ability of the juvenile migrant stage of six New Zealand freshwater fi sh species. New Zealand Journal of Marine and Freshwater Research, 32: 21-30.
- Ryan, P.A., 1991. Environmental effects of sediment on New Zealand streams, a review. New Zealand Journal of Marine and Freshwater Research, 25: 207–221.
- Van Nieuwenhuyse, E.E. & LaPerriere, J.D., 1986. Effects of placer gold mining on primary production in subarctic streams of Alaska. Water Resources Bulletin, 22: 91-99.
- Wagener, S.M. & LaPerriere, J.D., 1985. Effects of placer mining on invertebrate communities of interior Alaska streams. Freshwater Invertebrate Biology, 4: 208–214.
- Wood, P.J. & Armitage, P.D., 1997. Biological effects of fi ne sediment in the lotic environment. Environmental Management, 21(2): 203-217.
- World Bank, 2006. Mongolia: A review of environmental and social impacts in the mining sector. World Bank.