Temporal and spatial influences of flooding on dissolved mercury in boreal reservoirs.

Dissolved total mercury and methylmercury concentrations ([HgT]D and [MeHg]D, respectively) in natural lakes and reservoirs of varying age (2 months to 69 years) were monitored between 1993 and 1997, during different times of the year and at several sites within the same reservoir. Both HgT-D and MeHgD in flooded environments were significantly greater (Student's t-test, P < 0.001) than those determined for neighbouring lakes, with average concentrations of 2.35+/-0.85 ng(-1) (n = 79) vs. 1.51+/-0.43 ng l(-1) (n = 70) for HgT-D and 0.28+/-0.10 ng Hg l(-1) (n = 87) vs. 0.05+/-0.025 ng Hg l(-1) (n = 30) for MeHgD. Furthermore, the proportion of HgT-D in the methylated form was, on average, nearly four times greater in the reservoir sites as compared with those of the lakes (12% vs. 3.6%). Simple linear regressions used to compare [MeHg]D with nine water quality parameters suggest that factors controlling the presence of MeHgD are not the same in all flooded environments. Overall, our results demonstrate that [MeHg]D are higher in reservoirs as compared with natural lakes and that up to 18 years after reservoir construction [MeHg]D remain elevated as compared with background levels.

Methylmercury in water, seston, and epiphyton of an Amazonian river and its floodplain, Tapajós River, Brazil.

Levels and accumulation of MeHg were characterized in filtered water, suspended organo-mineral matter, phytoplankton, zooplankton and epiphyton during the dry season and the wet season. In open water of the lentic and lotic ecosystems, the MeHg in filtered water (< 0.2 microm) was near or below the detection limit (< 0.02-0.03 ng/l). These concentrations represent < 5% of the Hgtot. content in filtered water. Inundated forests (Igapó) and macrophyte floating mats were the only sites where MeHg was significantly detected (0.07-0.24 ng/l), representing 3-22% of the Hgtot. in filtered water. MeHg concentrations in organo-mineral suspended matter (2-26 ng/g dry wt. representing 0.6-7.3% of Hgtot.) were correlated with the N content but not with the C content. Data suggest that MeHg enrichment of suspended matter is strongly influenced by the presence of degraded planktonic remains relatively rich in N and MeHg. In zooplankton, MeHg concentrations (20-140 ng/g dry wt.) increased from the dry season to the end of the wet season. This increase was followed by higher proportions of MeHg during the wet season in comparison to the dry season (15-40 to 50-70% of the Hgtot.). The epiphytic material collected from the roots of macrophyte floating mats contained 2-8.5 ng/g dry wt. of MeHg. The proportion of MeHg to Hgtot. in epiphyton (1.5-8.3%) correlated with its C and N contents. The data suggest a greater bioavailability of MeHg in the Tapaj6s River ecosystems due to the seasonal increase in water level and the consequent inundation of the floodplain. Inundation favours the development of large macrophyte floating mats which increases the bioavailability of epiphyton to herbivorous/detritivorous fish. The root zone of floating macrophytes and the flooded organic horizon of the Igapó forest are the only sites along the Tapajós River where significant MeHg can be detected in the water column and sediment. This new study supports the hypothesis that MeHg production and transfer to the first link of the food chain in Amazonian river systems is closely related to organic matter biogeochemistry in the floodplain environment.

The geochemistry of mercury in central Amazonian soils developed on the Alter-do-Chão formation of the lower Tapajós River Valley, Pará state, Brazil.

In an oxisol-spodosol system developed on the terrestrial surface of the lower Tapajós Valley, the determination of total mercury (Hg), organic carbon (C), iron and aluminum oxy-hydroxide (Fe(cdb) and Al(cdb)) concentrations in the surface soil horizons are used to characterise the geochemical processes controlling the accumulation of Hg in soils under natural vegetation cover and in deforested and cultivated sites. Oxisols from the plateau have homogeneous and relatively high background Hg contents and burdens constituting an important natural reservoir of Hg for the region (90-210 ng/g dry wt. and 19-33 mg/m2 for the first 20 cm). The Fe(cdb) and Al(cdb) contents associated with the fine fraction (< 63 microns) of the soil suggest that oxy-hydroxides and, particularly Al-substituted Fe oxy-hydroxides, control the Hg concentrations observed in all of the soils of the study region. Consequently, the geochemistry of these minerals along the slopes governs the accumulation or the release of the Hg according to the natural evolution of the soil cover and/or following the degradation of soils by erosion after deforestation and cultivation. These observations have important implications for the interpretation of Hg contamination patterns observed in Amazonian aquatic systems that could be linked to different drainage sources of Hg from the terrestrial surface. The sandification and podzolisation that is characteristic of the evolution of numerous pedological systems in the equatorial Amazon could be responsible for exportation of the naturally accumulated Hg, as for other metals, by acidic complexation and migration to the black waters of the Amazon. In the central Amazon region, as a result of the fragility of the soil cover, deforestation and cultivation, affecting principally the superficial soil, promote the selective erosion of fine particles enriched in oxides and Hg. The erosion of soil could be responsible for an important release of Hg, transported in particulate form by drainage waters.

Evidence of early nervous system dysfunction in Amazonian populations exposed to low-levels of methylmercury.

There is increasing concern about the potential neurotoxic effects of exposure to methylmercury in Amazonian populations due to mercury (Hg) release from gold-mining activities. A preliminary study was undertaken in two villages on the Tapajos River, an effluent of the Amazon, situated over 200 km downstream from the extraction areas. The study population included 29 young adults (< or = 35 years), 14 women and 15 men, randomly chosen from a previous survey. Hair analyses were conducted with cold vapor atomic fluorescence spectrophotometry. Total hair Hg (THg) varied between 5.6 micrograms/g and 38.4 micrograms/gl, with MeHg levels from 72.2% to 93.3% of the THg. A quantitative behavioural neurophysiological test battery, designed for use under standard conditions, in an area without electricity and for persons with minimal education was administered to all participants. The results of visual testing showed that although all participants had good near and far visual acuity, color discrimination capacity (Lanthony D-15 desaturated panel) decreased with increasing THg (F = 4.1; p = 0.05); near visual contrast sensitivity profiles (Vistech 6000) and peripheral visual field profiles (Goldman Perimetry with Targets I and V) were reduced for those with the highest levels of THg. For the women, manual dexterity (Santa Ana, Helsinki version) decreased with increasing THg (F = 16.7; p < 0.01); this was not the case for the men. Although the women showed a tendency towards reduced grip strength, muscular fatigue did not vary with THg for either sex. The findings of this study demonstrate that it is possible, using a sensitive test battery, to detect alterations in nervous system functions, consistent with knowledge on Hg toxicity, at levels below the currently recognized threshold of 50 micrograms/g THg.