Characterization and prediction of long-term arsenic mobility, dissolution, and kinetic behavior in arsenic contaminated floodplain deposits of Whitewood Creek and the Belle Fourche River, South Dakota
Date
2021
Authors
Ji, Mu, author
Ridley, John, advisor
Stednick, John, committee member
Borch, Thomas, committee member
Gallen, Sean, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
From 1877 to 1977, the Homestake Mine discharged over 100 million tons of arsenic-rich mine-wastes from Lead, South Dakota into Whitewood Creek (WWC), which joins the Belle Fourche River (BFR). Arsenopyrite and other arsenic-bearing minerals were deposited in tailings (containing between 0.12% to 0.35% arsenic) and mixed with uncontaminated alluvium along the floodplains of WWC and the BFR as overbank deposits and filling abandoned meanders. Since it is not feasible to remove millions of tons of contaminated sediments from the area, an understanding of arsenic mobility on long timescales is vital. Many studies have laid the framework for factors controlling arsenic mobility appropriate to fluvial sedimentary systems; investigating mechanisms of arsenic mobilization, adsorption/desorption kinetics, and the effects of pH, changing redox conditions, etc., however, these studies were conducted on relatively short time scales and did not quantify arsenic mass-budget on field-scales. This study focuses on the long-term retention, dissolution, and kinetic behavior of arsenic from mine tailings. The uniqueness of this site enables arsenopyrite dissolution behavior to be constrained over a 135-year timespan (1877-2012). This allows for the investigation of changes in arsenic's residence sites, its rate of release into the environment, calculation of its transport mass-budget, and elucidation of how natural processes have or have not remediated arsenic contamination over a span of 35 years since the deposition of tailings have ceased (1977-2012). For this investigation, sediment, surface water, and seep water samples were collected along reaches of WWC and the BFR for analysis of arsenic and other geochemical constituents. Sequential extractions of the sediments were performed to determine the mineralogical setting of the arsenic as well as the proportion of arsenic available at different rates of release into the environment. Additionally, various historical data (discharge levels, geochemical analyses of water and sediment samples) were compiled from the United States Geological Survey database. Regressions were applied to historical data to estimate the rate of physical and chemical arsenic removal from the WWC watershed. Sediments collected along the floodplains of WWC and the BFR exhibited arsenic concentrations ranging from approximately 100 to 4,000 mg/kg. The results from the sequential extractions applied to the sediments suggest arsenic is predominantly located in residence sites that are not easily accessible, and arsenic is not readily mobilized or released into solution in large quantities under normal environmental conditions seen in WWC and the BFR. An average of 16% of the arsenic is weakly bound to readily exchangeable surface sites, water-soluble secondary minerals and available for rapid release, or is adsorbed to exchange sites that easily exchange PO43- ions for adsorbed arsenic oxyanions, is weakly bound in amorphous to poorly crystalline fine-grained metal oxides/hydroxides, reducible phases, and easily soluble carbonates. An average of 24% of the arsenic is moderately strongly bound in weakly soluble secondary minerals like clays or crystalline fine-grained metal oxides/hydroxides and will be released relatively slowly with time. The remaining 60% of arsenic is interpreted to be relatively immobile and locked in arsenopyrite in part due to the formation of metal oxyhydroxide coating, which slows down the degradation of the mineral. These interpretations are supported by the elevated but still relatively low total arsenic concentrations (EPA MCL for arsenic is 0.01 mg/L) of in-stream water in WWC (averaging 0.037 mg/L) and in the BFR (averaging 0.021 mg/L), considering that in-stream sediments carried by WWC and the BFR have high arsenic concentrations (264 to 694 mg/kg). Based on regressions applied to 30 years of historical sediment transport and arsenic concentration in solution and in sediment load (1982-2012), the average annual total arsenic load transported out of WWC during these 30 years was estimated to be between 34 to 71 megagrams (Mg) per year. At this rate, based on the 17,400 to 50,800 Mg of arsenic that remain in storage along the floodplains of WWC, complete arsenic transport out of the floodplains of WWC would range between 250 to 1,500 years. The actual rate of arsenic removal is expected to be longer because the model is based on a uniform movement of uniformly distributed sediment, and historical patterns may not be reflective of future trends, as evidenced by the decline in suspended arsenic transport rate starting in the early- to mid-1980s. The constant shifting of the stream creates abandoned meanders along WWC that can store contaminated sediment where the stream no longer has access. Conversely, as the meanders shift over time, the once-abandoned meanders could be again accessed by WWC. The majority of suspended sediment transport occurs during flood events; approximately 88% of the total arsenic load moved during the years between 1983 to 2012 occurred in only 3 of the years (1983, 1984, and 1995). Thus, the rate of arsenic transport for the next 30-year period is uncertain and could be lower if the number of flood events remains low. Although the WWC area once experienced heavy environmental degradation during the period of active mining, natural processes have allowed for relatively stable current environmental conditions. However, the physical transport of arsenic-contaminated sediment and the slow release of arsenic to the environment endures downstream to the BFR into the Cheyenne River and Lake Oahe and will continue for many generations.