Biogeochemical controls on arsenic cycling in a hydrocarbon plume

Abstract

Arsenic (As) in drinking water poses a critical threat to public health. More than 150 million people worldwide are at risk of developing diseases from unsafe concentrations of As in groundwater. Arsenic occurs naturally in rocks, soils, and sediments and generally remains associated with solid phases. However, changes in aquifer geochemistry can mobilize As into groundwater, contaminating drinking water sources. This dissertation investigates As cycling in an aquifer contaminated by petroleum hydrocarbons near Bemidji, Minnesota, where As is mobilized into groundwater due to biodegradation of hydrocarbons coupled to reduction of ferric oxides. The first project describes how aquifer sediments act as both sources and sinks for As in groundwater, depending on the prevailing redox conditions. Results show that As is released to groundwater near the hydrocarbon source but is removed near the hydrocarbon plume's leading edge. Comparison of data from 1993 to 2016 shows that As has been redistributed in aquifer sediment as the plume has expanded over time. The second project presents a mass balance for As, which shows that despite elevated As in groundwater (up to 230 μg/L), >99.7% of As mass in the aquifer is in sediments. Calculations demonstrate that As in sediment can be 22x less than the method detection limit and still cause unsafe concentrations in groundwater, suggesting that the use of standard methods limits our ability to predict where naturally occurring As poses a threat to groundwater. In the third project, a reactive transport model simulates As cycling for 400 years. Results show that sorption of As to ferrihydrite limits As transport within 300 m of the hydrocarbon source. Modeling predicts that over the plume's lifespan, more groundwater will be contaminated by As than benzene, the primary contaminant of concern in hydrocarbon plumes. Combined, these studies suggest that many aquifers are vulnerable to unsafe As concentrations due to mobilization of natural As if bioavailable organic carbon is introduced. Although aquifers can attenuate As, it may take centuries for As to be fully removed from groundwater, suggesting it is prudent to account for natural contaminants like As when developing remediation strategies at petroleum spill sites.