Search USGSSearch

USGS · 70030836

Mineralogy and arsenic mobility in arsenic-rich Brazilian soils and sediments

Abstract

Background. Soils and sediments in certain mining regions of Brazil contain an unusually large amount of arsenic (As), which raises concerns that mining could promote increased As mobility, and thereby increase the risks of contaminating water supplies. Objectives. The purpose of t his study was to identify the most important factors governing As mobility in sediments and soils near three gold-mining sites in the State of Minas Gerais, Brazil. Methods. Surface and sub-surface soil samples were collected at those sites and characterized by chemical and mineralogical analyses. Oxalate (Feo) and citrate-bicarbonate-dithionite (Fed) iron contents were determined by atomic absorption spectroscopy (AAS). Arsenic mobilization was measured after incubating the samples in a 2.5 mM CaCl2 solution under anaerobic conditions for 1, 28, 56, 84, or 112 days. The solution concentrations of As, Fe, and Mn were then measured by inductively coupled plasma-mass spectrometry (ICP-MS) and AAS, respectively. Results and Discussion. Results indicated that As mobilization is largely independent of both the total As and the Feo/Fed ratio of the solid phase. Soluble As is roughly controlled by the Fe (hydr)oxide content of the soil, but a closer examination of the data revealed the importance of other highly weathered clay minerals and organic matter. Large amounts of organic matter and a low iron oxide content should favor As leaching from soils and sediments. Under reducing conditions, As is mobilized by the reductive dissolution of Fe and/or Mn oxides. However, released As may be readsorbed depending on the sorptive properties of the soil. Gibbsite is particularly effective in adsorbing or readsorbing As, as is the remaining unreduced fraction of the iron (hydr)oxides. Conclusion and Outlook. In general, low soluble As is rel ated to the presence of gibbsite, a large amount of iron oxides, and a lack of organic matter in the solid phase. This has environmental significance because gibbsite is thermodynamically more stable than Fe oxides under anaerobic conditions, such as those found in waterlogged soils and lake sediments. ?? 2006 ecomed publishers (Verlagsgruppe Hu??thig Jehle Rehm GmbH), D-86899 Landsberg and Tokyo.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

de Mello, William R. Roy, J.L. Talbott, J.W. Stucki. 2005-09-16. Mineralogy and arsenic mobility in arsenic-rich Brazilian soils and sediments. https://doi.org/10.1065/jss2005.09.144

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

Suspended-sediment and phosphorus response in a fire-affected central Montana headwater catchment

Purpose Eutrophication and nuisance filamentous algal blooms (i.e. Cladophora ) are increasingly common occurrences throughout much of the western United States. Wildfire may be contributing to the frequency and magnitude of algal blooms through excess sediment and nutrient loading to streams and rivers. Our objective was to evaluate the effects the 2021 Woods Creek Fire had on sediment yields and phosphorus (total and bioavailable) partitioning in Camas Creek, a major tributary to the Smith River in Montana where Cladophora are now consistently reaching nuisance levels. Methods We collected water quality samples during snowmelt pulsing events as well as fixed interval sampling using an established U.S. Geological Survey stream gage instrumented with a continuous water quality sonde and an automatic peristaltic pump sampler. Water samples were processed for total phosphorus (TP), sediment-bound bioavailable phosphorus (S-BioP), soluble reactive phosphorus (SRP), and suspended sediment concentrations and were evaluated using linear regression and other nonparametric statistical tests. Continuous turbidity and streamflow were evaluated using hysteresis analysis to determine sediment sourcing and connectivity. Results We found that the Woods Creek Fire did not significantly influence TP and S-BioP in Camas Creek. However, there was a significant increase in SRP and turbidity in both postfire years (2022 and 2023). Hysteresis analysis of 91 delineated events indicated positive (clockwise) hysteresis was the dominant event pattern during the snowmelt period. This may indicate a lower hillslope to channel connectivity, with the major sediment supply originating from the channel and/or riparian areas. Conclusion Results from this study demonstrate the benefits of combining discrete water quality samples with high-frequency turbidity sensors to characterize postfire sediment and phosphorus dynamics. While a lack of postfire response in TP and S-BioP is contrary to many other studies, our findings highlight the role climate and catchment morphology play in attenuating a disturbance effect.

Montana

Evaluating uncertainties with sample-collection method and source selection in sediment fingerprinting: an example from a Great Lakes tributary

Purpose Identifying suspended-sediment (SS) sources, seasonal variability, and phosphorus (P) transported with SS is critical information for basin managers, although there may be concerns about comparability between flow-integrated SS samples used for sediment fingerprinting and discrete samples used for concentrations and loads in basins where SS is mostly silt + clay and(or) one land-use predominates. Objectives were to determine if (1) sample-collection method and (2) source consideration influenced apportionment of the largest SS source. Methods Concurrent-replicate, SS samples were collected during 2022 from the East River, Wisconsin using an automated sampler, commonly used for water-quality sampling, and passive samplers, frequently used for SS fingerprinting. Samples were evaluated for differences in physical and chemical characteristics that may affect source apportionment. Considered sources included three upland land-use (cropland, forest, and roads), two in-channel (streambank and streambed sediment), and one that connects uplands to the stream channel (gullies). Source apportionment used established methods in the SedSAT tool. Source scenarios included land-use + streambank (4src), 4src + gully, 4src + streambed, and 4src + gully + streambed (6src). Results There were no statistically significant differences in median grain size, organic carbon, or sediment-bound P as a function of collection method. In-channel sources were the largest proportional SS source, regardless of season, hydrologic condition, collection method, or source scenario. Source verification highlighted which source fingerprints were most accurately defined and implications for SS target apportionment. Conclusion Varying the source scenarios for sediment fingerprinting indicated that improved management of hydrologic connectivity between upland land use and the stream channel has the potential to mitigate SS loads.

Wisconsin

Sediment budget of a Maumee River headwater tributary: How streambank erosion, streambed-sediment storage, and streambed-sediment source inform our understanding of legacy phosphorus

Objective We described source and phosphorus (P) retention potential of soft, fine-grained, streambed sediment and associated phosphorus (sed-P) during summer low-flow conditions. Combining in-channel, sed-P storage with relative age provided context on relevance to western Lake Erie Basin management goals. Methods In 2019, rapid geomorphic assessment (30 reaches) compared streambed-sediment storage (S) to streambank erosion (E), providing annual sediment budgets (S:E). Streambed sediment (13 reaches) was fingerprinted and analyzed for sed-P. The P saturation ratio (PSR; four reaches) quantified potential sorption/desorption of dissolved P (DP) between the water column and streambed sediment. Analyses were supplemented with data from 2017 and 2021. The ratio of two fallout radionuclides, beryllium-7 (54-day half-life) and excess lead-210 (22.3 years), apportioned “new” sediment based on time since rainfall contact. Results Streambed sediment was mostly streambank (54–96%) for contributing areas > 2.7 km 2 ; for upstream reaches, a larger percentage was apportioned as upland (cropland, pasture, forest, and road), with < 30% streambank. Streambank erosion correlated with contributing area; however, soil type (ecoregion), stream characteristics, and land use combined to drive streambed-sediment storage. Individual-reach S:E (accumulation of 0.01–35 years of streambank erosion) differentiated erosional and depositional in-channel environments. Most reaches indicated that 17–57% of sediment had recent contact with rainfall. Streambed-sediment PSR indicated a low potential for further sorption of DP from the water column; one reach was a P source when sampled. Conclusion Sed-P was higher in streambed sediment than in source samples, which varied by land use and ecoregion. This indicates homogenization resulting from in-stream sorption of DP during sediment transport that occurs over multiple events.

Indiana, Ohio