Mercury and methylmercury contamination related to artisanal gold mining, Suriname
John E. Gray, Victor F. Labson, Jean N. Weaver, David P. Krabbenhoft
Abstract
[1] Elemental Hg-Au amalgamation mining practices are used widely in many developing countries resulting in significant Hg contamination of surrounding ecosystems. We have measured total Hg and methyl-Hg concentrations in sediment and water collected from artisanal Au mines and these are the first Hg speciation data from such mines in Suriname. Total Hg and methyl-Hg contents in mine-waste sediment and water are elevated over local uncontaminated baselines. Total Hg (10–930 ng/L) and methyl-Hg (0.02–3.8 ng/L) are highly elevated in mine waters. Increasing total Hg contents in discharged mine waters correlate with increasing water turbidity indicating that most Hg transport is on suspended particulates. Our Hg results are similar to those found in artisanal Au mines in the Amazon basin, where Hg contamination has led to adverse effects on tropical ecosystems.
1. Introduction
[2] Mercury is a heavy metal that is toxic to living organisms and is one of the few pollutants where ingestion of contaminated food (primarily fish) has led to human deaths [NAS, 1978]. In many areas in the USA, and worldwide, Hg contamination of fish has become severe enough to require human health warnings recommending restriction of consumption of such fish [NAS, 1978; EPA, 1997]. In humans, Hg damages the central nervous and immune systems and is especially toxic to the fetus [WHO, 1976]. Under certain conditions, less harmful inorganic Hg (e.g., elemental mercury, Hg0) may be converted to highly toxic organic forms of Hg (e.g., methyl-Hg, CH3Hg+). Mercury methylation is dependent on a number of chemical and biological factors, but is primarily a result of anaerobic microbial activity, which can be magnified in organic-rich environments [Ullrich et al., 2001]. Highly toxic organic Hg compounds are water soluble, and biomagnify with increasing trophic position in the food chain. Methyl-Hg strongly binds with lipids and proteins and is readily accumulated in aquatic organisms such as fish, which may pose a threat to humans and other fish-consuming organisms [Fitzgerald and Clarkson, 1991].
[3] Environmental problems related to the use of elemental Hg for Au and Ag recovery have been known since Roman times [Lacerda and Salomons, 1998]. Small-scale or artisanal Au mining is popular in developing countries because it involves the use of simple processes generally by a few individuals to recover particulate Au. Unfortunately, artisanal miners commonly use the elemental Hg-Au amalgamation technique. Since the 1980’s, a resurgence of artisanal Au mining in many countries has led to Hg pollution of terrestrial and aquatic ecosystems, but such mining is especially prevalent throughout South America (Brazil, Bolivia, Columbia, Venezuela, Peru, Ecuador, French Guiana, Guyana, and Suriname), China, Russia, the Philippines, Indonesia, Thailand, Tanzania, Mexico, and to a lesser extent in Australia, Canada, and the USA [Lacerda and Salomons, 1998]. Artisanal Au mining has been ongoing for over a century in Suriname and throughout the Guiana Shield of South America [Wasel et al., 1998]. An increase in artisanal Au mining in Suriname in the 1980–90’s led to increased environmental concern due to the unregulated and widespread use of elemental Hg. Estimates suggest that about 10–20 t/yr of Au are produced by as many as 25,000–35,000 legal and illegal artisanal miners in Suriname; however, Au is known to be smuggled out of the country and production figures are uncertain [de Kom et al., 1998; Mol et al., 2001]. As much as 1–3 kg of elemental Hg may be lost to surrounding environments for each kilogram of Au recovered. In and downstream from areas of artisanal Au mining, oxidation of elemental Hg to ionic Hg, and subsequent Hg methylation is potentially hazardous due to bioaccumlation of methyl-Hg in aquatic organisms. As a result of the highly toxic nature of Hg, its release into highly sensitive and diverse tropical rainforest ecosystems has been the focus of abundant research throughout South America [e.g., Malm et al., 1990; Aula et al., 1994; Maurice-Bourgoin et al., 1999; Guimarães et al., 2000; Lechler et al., 2000; Richard et al., 2000; Mol et al., 2001].
2. Study Area
[4] Surficial Au was discovered in highly weathered soils by artisanal miners in the Gross Rosebel area in 1879, and small-scale mining has continued intermittently to the present day [Wasel et al., 1998]. Recent exploration has indicated about 62 t (∼2,200,000 oz) of Au reserves in the Gross Rosebel concession [Wasel et al., 1998]. The Au deposits are located along tributaries that flow into the Suriname River and Brokopondo Reservoir (Figure 1). The Suriname River flows into the Atlantic Ocean near Paramaribo, about 80 km from Gross Rosebel. The Suriname River is a major ecosystem, and Paramaribo is the capital of Suriname with a population of about 400,000. Important fisheries are in Brokopondo Reservoir and the Suriname River, and an important shrimp fishery is located at the mouth of the Suriname River near Paramaribo. To evaluate Hg contamination in and around these artisanal Au mines, we measured the concentration of total Hg and methyl-Hg in samples of (1) mine-waste sediment and stream sediment collected near areas of artisanal Au mining, (2) stream water and mine-water discharge collected proximal to mines, and (3) stream sediment and stream water collected from uncontaminated baseline sites distant from mining activities.
3. Sample Collection and Analysis
[5] In October 2000, unfiltered water samples were collected for total Hg and methyl-Hg analysis in pre-cleaned, Teflon bottles, and within six hours of collection, these samples were acidified with ultra-pure HCl. Turbidity was measured in NTU (nephelometric turbidity units) in the field using a portable meter. Sediment was collected from several localities at each site, and composited together as a single sediment sample.
[6] Mercury was determined in the sediment and water samples by cold-vapor atomic fluorescence spectrometry (CVAFS) following EPA method 1631 [Bloom and Fitzgerald, 1988], and methyl-Hg was determined by CVAFS following EPA method 1630 [Bloom, 1989]. During methyl-Hg analysis, sediment samples were extracted into acidic/bromide methylene chloride to avoid possible methylation artifact effects [Bloom et al., 1997]. Quality control for Hg and methyl-Hg analysis was addressed with field and method blanks, blank spikes, matrix spikes, certified reference materials, and sample duplicates. Recoveries on blank and matrix spikes were 85–115%. The relative standard deviation was <12% on reference standards. Total Hg and methyl-Hg in field and method blanks were below the limits of determination. Limits of determination were 2.0 ng/g for total Hg and 0.02 ng/g for methyl-Hg in sediment samples, and 0.2 ng/L for total Hg and 0.02 ng/L for methyl-Hg in water samples.
4. Results and Discussion
[7] Mine-waste sediment and stream-sediment samples collected within 1 km of active artisanal Au mining in the Gross Rosebel area generally contain higher total Hg concentrations (5.5–200 ng/g) compared to Hg in sediments collected from local uncontaminated stream baselines (14–48 ng/g) (Table 1, Figure 2). Total Hg contents in sediments collected from the Gross Rosebel area are comparable to Hg found in sediments in other areas of artisanal Au mining in the Amazon region, which vary from 24–406 ng/g (Table 1). In addition, total Hg contents found in local uncontaminated baseline stream sediments in this study are similar to Hg in other uncontaminated baseline sites in the Amazon region and throughout the world, which vary from 5.0–93 ng/g (Table 1). Similarly, methyl-Hg contents (<0.02–1.4 ng/g) are generally higher in sediment samples collected proximal to artisanal mines at Gross Rosebel when compared to methyl-Hg in samples from local uncontaminated baseline streams (0.03–0.08 ng/g) (Table 1, Figure 2). Methyl-Hg contents in sediments in this study are also similar to those in other artisanal Au mining areas in the Amazon region (Table 1).
Table 1. Total Hg and Methyl-Hg Contents in Samples Collected Near Artisanal Au Mines in Suriname and Comparative Baselines
| Location | Sediment | Unfiltered water | ||
|---|---|---|---|---|
| Hg (ng/g) | Methyl-Hg (ng/g) | Hg (ng/L) | Methyl-Hg (ng/L) | |
| Artisanal Au mines, Surinamea | ||||
| Mine wastes | 5.5–200 | <0.02–0.83 | 11–930 | 0.05–3.8 |
| Streams below mines | 110–150 | 1.2–1.4 | – | – |
| Uncontaminated baselines | 14–48 | 0.03–0.08 | 6.4–10 | 0.08–0.28 |
| Comparative baselines | ||||
| Amazon basin | ||||
| Streams affected by miningb,c,d,e,f | 24–406 | 0.07–1.9 | 2.9–33 | 0.2–0.6 |
| Upstream from miningd,g | 67–93 | – | 2.2–2.6 | – |
| Antarctica streams and lakesh | – | – | 0.27–1.9 | 0.019–0.33 |
| Wisconsin lakesi,j | <10–80 | – | 0.72–2.1 | 0.046–0.33 |
| Lake Baikal, Russiak | 5.0–72 | – | 0.14–2.0 | 0.002–0.16 |
| Worldwide background, rivers and lakesl | – | – | 0.1–3.5 | – |
| California baseline lakesm | – | – | 0.56–2.0 | – |
- a This study.
- b Nriagu [1992].
- c Lacerda and Salomons [1998].
- d Maurice-Bourgoin et al. [1999].
- e Hylander et al. [2000].
- f Lechler et al. [2000].
- g Aula et al. [1994].
- h Lyons et al. [1999].
- i Watras et al. [1994].
- j Rada et al. [1986].
- k Leermakers et al. [1996].
- l Nriagu [1990].
- m Gill and Bruland [1990].
[8] Total Hg contents in unfiltered water samples we collected downstream from artisanal Au mines (11–930 ng/L) are elevated over those in local baseline streams (6.4–10 ng/L) (Table 1, Figure 3). Total Hg contents in these mine waters generally exceed the 12 ng/L Hg standard recommended by the U.S. Environmental Protection Agency to protect against adverse chronic effects to aquatic life [EPA, 1992]. Conversely, total Hg in all water samples collected in this study were below the 1,000 ng/L international drinking water standard for Hg recommended by the World Health Organization [WHO, 1971]. However, water flowing from these mines is not generally used for drinking water due to other adverse mining-related effects such as high water turbidity (discussed below). Methyl-Hg is also generally higher in mine water (0.05–3.8 ng/L) compared to that in the local baseline stream water (0.08–0.28 ng/L; Figure 3). Total Hg and methyl-Hg contents in unfiltered water samples collected in this study are also typically higher than those found in uncontaminated baselines worldwide, which vary from 0.1–3.5 ng/L for total Hg and 0.002–0.33 ng/L for methyl-Hg (Table 1). Total Hg contents in unfiltered water samples from our baseline sites are generally higher than uncontaminated worldwide baselines probably because of widespread use of elemental Hg and loss of Hg vapor in the region.
[9] Stream water runoff from the artisanal mines studied is highly turbid, in some instances >1,000 turbidity units (Figure 4). Highly weathered soils in the Au mining camps are oxysols (clay- and quartz-rich saprolite), which are typical of tropical environments. When oxysols are processed in sluice-boxes to recover Au during mining, resultant discharge water is highly turbid due to the abundance of suspended clay. As water turbidity increases, Hg transport also increases, and thus Hg contents are higher in more turbid water (Figure 4). Therefore, Hg fixation and transport is most significant in areas with high water turbidity indicating that Hg is attached onto suspended particulates in water. In addition to Hg contamination of local aquatic ecosystems, there is also significant deforestation and siltation of streams and rivers in artisanal Au mining areas in Suriname. Local Hg contamination of ecosystems downstream from mined areas is clearly worsened by these additional adverse environmental effects. For example, clear cutting and tropical deforestation leads to increased erosion and runoff, and thus, higher turbidity, siltation of waterways, and water quality degradation. During our fieldwork, we observed several streams that were completely choked and filled in with sediment as a result of sediment-laden mine-water discharge. In some instances, such local stream siltation resulted in cutting off a domestic water source for local inhabitants. Infilling of stream beds leads to areas of stream stagnation that are higher in organic matter and methyl-Hg (Figure 2).
[10] We estimate that over 200 t of elemental Hg have been potentially lost to the Suriname River basin during artisanal Au mining at Gross Rosebel since the 1980’s. Some portion of this elemental Hg is eventually methylated, and studies have suggested that Hg methylation is enhanced in topical environments as a result of higher temperature, higher organic matter, and more biological activity [Lacerda and Salomons, 1998]. Although there are limited methyl-Hg data for sediment or water from South America, high Hg methylation is known in parts of the Amazon basin in Brazil downstream from artisanal Au mining [Guimarães et al., 2000]. Our methyl-Hg results for mining-related water and sediment exceed those found in uncontaminated stream baselines, and are similar to those found in areas of artisanal mining in the Amazon basin (Table 1). We suggest that translocation of methyl-Hg from mine wastes and stream sediments, to stream water, and then to biota such as fish, is likely in areas of artisanal Au mining in Suriname. Methylation of Hg has led to contamination of freshwater fish in areas such as the Madeira River where fish contain Hg contents as high as 3.9 μg/g and average 0.85 μg/g [Malm et al., 1997], which exceed the 0.5 μg/g (wet weight, fish muscle) Hg standard recommended for human consumption of fish [WHO, 1990]. Elevated Hg contents in excess of 0.5 μg/g in fish collected from other areas of artisanal Au mining are also known throughout South America [e.g., Maurice-Bourgoin et al., 1999; Richard et al., 2000]. Similarly, Mol et al. [2001] reported high Hg contents (up to 4.6 μg/g, wet weight, fish muscle) in piscivorous freshwater fish collected from the Suriname River and Brokopondo Reservoir, and they related this Hg contamination of fish to upstream artisanal Au mining where elemental Hg has been used. In some instances, high Hg contents in fish have led to elevated Hg contents in local inhabitants exposed to mining related Hg in South America [Malm et al., 1997; Maurice-Bourgoin et al., 1999; Yates, 1997], often because inhabitants of the remote tropical rainforest are part of high fish consuming societies.
Acknowledgments
[11] This study was funded by the U.S. Geological Survey. We are grateful to Eric Vonkel and all of the staff of Golden Star for providing field logistics and access to sites around the Gross Rosebel concession. We also thank Mark Hines (University of Massachusetts, Lowell), Dennis Helsel (USGS, Denver), and two anonymous GRL reviewers for suggestions that improved this paper.