REACTIVE TRANSPORT MODELING OF ARSENIC CYCLING IN A PETROLEUM-CONTAMINATED AQUIFER

A crude-oil spill in 1979 released 10,700 barrels of petroleum hydrocarbons to a surficial aquifer near Bemidji, MN, creating a dissolved hydrocarbon plume. In the plume, biodegradation coupled with reductive dissolution of ferric (Fe(III)) hydroxides (HFO’s)) in aquifer sediments has caused the mobilization of naturally occurring arsenic (As) from sediment into groundwater via desorption from Fe(III). Thirty-five years after the spill, dissolved As concentrations were as high as 23x the drinking water standard of 10 μg/L, above which As exposure has been linked to skin, lung, and bladder cancers.

In addition to acting as a source for mobilized As, sediments can also attenuate As in groundwater via resorption under oxic and suboxic conditions. Low dissolved oxygen concentrations at the plume's leading edge can oxidize dissolved ferrous iron to form fresh Fe(III) precipitates, which can sorb As. However, downgradient expansion of the anoxic plume can cause re-mobilization of attenuated As. Given the complex cycling of Fe and As in the aquifer, we are developing a reactive transport model to simulate As cycling in the Bemidji aquifer using observations of As and Fe in groundwater and sediment in the modern plume in combination with historical Fe data documented over the plume’s history. This model builds upon a previous model of Fe cycling, which includes reductive dissolution of Fe(III), Fe(II) sorption, and Fe(II) oxidation, by adding As(III/V) surface complexation on HFO’s, which controls As mobilization and attenuation by resorption to aquifer minerals. Modeling objectives include: 1) testing mechanisms important for the attenuation of dissolved As; 2) estimating when in the future the dissolved As plume will be completely attenuated; 3) quantifying the transport of As in groundwater over the plume’s lifespan. Insights from these questions will be integral in developing remediation strategies for aquifers where As can be mobilized into groundwater via biodegradation of organic carbon. Results from this study will provide insight into how important it is to conduct long-term monitoring of naturally occurring and potentially toxic inorganic species at fuel spill sites; once mobilized, inorganics can pose an additional, and potentially greater, risk to groundwater and human health than the original components of the spill.