Environmental assessment of mercury dispersion, transformation and bioavailability in the Lake Victoria Goldfields, Tanzania.

Environmental dispersion and transformation of mercury discharged from gold mining operations has been investigated in field and laboratory studies in order to provide better understanding of the degree of mercury (Hg) pollution and bioavailability in the Lake Victoria goldfields (LVGF) ecosystems. This paper reviews results already published elsewhere and presents additional data on Hg dynamics in the LVGF. Studies conducted at the Mugusu and Rwamagaza artisanal mines indicated different degrees of Hg contamination and dispersion in environmental matrices. Mercury concentration in contaminated river sediments near the Mugusu mine varied from 6.0 to 0.5 mg/kg on a dry weight basis. The highest Hg contamination levels (165-232 mg/kg) were associated with mine tailings at the Rwamagaza mine. Mercury concentrations in fish representing different dietary habits on the southwestern shore of Lake Victoria at the Nungwe Bay were very low (2-35 microg/kg) and thought to represent background levels. These and other results suggested that the use of Hg in gold extraction in the LVGF has not caused high Hg levels in lake fish. The study of Hg in lichens showed Parmelia lichen to be an effective bioindicator for atmospheric Hg contamination due to Hg emissions from gold-amalgam firing and purification operations. The Hg levels in the lichens around the Mugusu mine ranged from 3.1 to 0.1 microg/g; the highest levels were recorded in the lichens sampled close to gold-amalgam processing sites. The regional background level in the Parmelia lichen was 0.05-0.10 microg/g, with a mean level of 0.07 microg/g. Studies of Hg transformation in the mine tailings revealed unexpectedly high methylmercury (MeHg) levels in the tailings (629-710 ng/g), which indicated that oxidation and methylation of metallic Hg in the tailings occurred at significant levels under tropical conditions. Re-equilibration of the tailings with freshwater (FW) indicated the MeHg was firmly bound in the tailings and therefore very little MeHg was released to the water column (0.2-1.5 ng/L). The methylation of Hg in tropical loamy clay soil contaminated with HgCl(2) (5 mg Hg/kg) yielded MeHg concentrations of 11 and 14 ng/g when inundated with seawater and FW, respectively, for 4 weeks. Little MeHg was transferred from the soil to the equilibrated water (< or = 0.4 ng/L). Atmospheric exposure of the soil pre-inundated with FW resulted in net degradation of MeHg during the 1st week of exposure, followed by net production and accumulation of MeHg in the soil (up to 15.5 ng/g) during atmospheric desiccation. Mercury uptake by fish from the Hg(0)-contaminated aquatic sediment-tailings system in the aquarium experiment was found to be low, suggesting the low availability of MeHg for bioaccumulation in the system. These and other results provide useful insights into Hg transformation, mobility and bioavailability in tropical aquatic systems affected by Hg pollution from gold mining operations.

Total mercury and methylmercury levels in fish from hydroelectric reservoirs in Tanzania.

Total mercury (THg) and methylmercury (MeHg) levels have been determined in fish species representing various tropic levels in four major hydroelectric reservoirs (Mtera, Kidatu, Hale-Pangani, Nyumba ya Mungu) located in two distinct geographical areas in Tanzania. The Mtera and Kidatu reservoirs are located along the Great Ruaha River drainage basin in the southern central part of the country while the other reservoirs are located within the Pangani River basin in the north eastern part of Tanzania. Fish mercury levels ranged from 5 to 143 microg/kg (mean 40 microg/kg wet weight) in the Mtera Reservoir, and from 7 to 119 microg/kg (mean 21 microg/kg) in the Kidatu Reservoir downstream of the Great Ruaha River. The lowest THg levels, in the range 1-10 microg/kg (mean 5 microg/kg), were found in fish from the Nyumba ya Mungu (NyM) Reservoir, which is one of the oldest reservoirs in the country. Fish mercury levels in the Pangani and Hale mini-reservoirs, downstream of the NyM Reservoir, were in the order of 3-263 microg/kg, with an average level of 21 microg/kg. These THg levels are among the lowest to be reported in freshwater fish from hydroelectric reservoirs. Approximately 56-100% of the total mercury in the fish was methylmercury. Herbivorous fish species contained lower THg levels than the piscivorous species; this was consistent with similar findings in other fish studies. In general the fish from the Tanzanian reservoirs contained very low mercury concentrations, and differed markedly from fish in hydroelectric reservoirs of similar age in temperate and other regions, which are reported to contain elevated mercury concentrations. The low levels of mercury in the fish correlated with low background concentrations of THg in sediment and flooded soil (mean 2-8 microg/kg dry weight) in the reservoir surroundings. This suggested a relatively clean reservoir environment that has not been significantly impacted by mercury contamination from natural or anthropogenic sources.

Methylmercury production and distribution in aquatic systems.

Mercury methylation and partitioning between sediment, water and fish were studied in river sediment spiked with mercuric chloride at levels of 1, 5 and 10 ppm as Hg. Maximum methylmercury production in the sediment occurred during the first week of incubation, with concentrations reaching 45.5 ng/g (dry wt.). Equilibrium conditions were established 3-4 weeks after the contamination of the sediment with mercury. Methylmercury partitioning between sediment and water yielded methylmercury levels of 25-154 ng/l in the fish aquariums and 0.26 ng/l in the fish-free control after 1 week of incubation. Equilibrium levels were 1.5-5.5 ng/l and 0.53 ng/l, respectively, during the 7th week. Over 50% of the methylmercury in whole water were in soluble form or associated with colloidal particulate < 1 micron. Mercury concentration in fish increased almost exponentially from 30 ppb to an average of 345 ppb within 3 weeks. Mercury uptake rates by fish were in the range of 10-18 ng/g per day during the 2nd and 3rd weeks, high rates occurring in water with methylmercury to total mercury ratios > 0.45. The partition coefficients for total mercury and methylmercury between fish and water (Kf-w) were 5000-7000 and 10,000-22,000, respectively. Large differences were observed in methylmercury production in sediment-water incubation with fish and the fish-free control. Equilibrium methylmercury concentrations in sediment were in the range 15-32 ng/g in the aquariums containing fish and 3-4.5 ng/g in the fish-free control. The significance of fish in mercury methylation in the aquariums still remains to be clarified as fish itself cannot methylate mercury in vivo.

Monitoring of fish and human exposure to mercury due to gold mining in the Lake Victoria goldfields, Tanzania.

Monitoring of environmental and human exposure to mercury in the Nungwe Bay area of the Lake Victoria goldfields, Tanzania, has revealed low mercury concentrations in fish (range: 1.8-16.9 ppb, mean: 7.0 ppb) and human hair (range: 156-5433 ppb, mean: 947 ppb) that represent background levels. Gold mining has not produced a significant increase in environmental methylmercury that is available for bioaccummulation in aquatic food chains. Urinary mercury levels in gold mine workers frequently exposed to Hg vapour in amalgamation and burning of amalgam were significantly higher (mean: 241 ng/ml) than those in the general mine population not occupational exposed to Hg (mean: 2.6 ng/ml). Rotation of mine duties reduced Hg exposure levels and hence the risk of intoxication in the gold mine workers. The lowest urinary and hair mercury levels were found in the Nungwe Bay fishing village population. This was consistent with the low mercury content of fish consumed by the inhabitants of the Nungwe Bay.