Loading...
Surface spills at unconventional oil and gas sites: a contaminant transport modeling study for the South Platte alluvial aquifer
Kanno, Cynthia M.
Kanno, Cynthia M.
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2016
Date Submitted
Keywords
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
2017-09-22
Abstract
Hydraulic fracturing has revolutionized the U.S.’s energy portfolio by making shale reservoirs productive and commercially viable. However, the public is concerned that the chemical constituents in hydraulic fracturing fluid, produced water, or natural gas itself could potentially impact groundwater. Here, I present fate and transport simulations of aqueous fluid surface spills, the most likely pathway for groundwater contamination during oil and gas production operations, with four primary goals: (1) to evaluate whether or not these spills pose risks to groundwater quality in the South Platte aquifer system, (2) to identify the key hydrologic and transport factors that determine these risks, (3) to develop a screening level methodology that could be applied at other sites and for various pollutants, and (4) to demonstrate the potential importance of co-contaminant interactions using selected chemicals. I considered two types of fluid that can be accidentally released at oil and gas sites: produced water and hydraulic fracturing fluid. Benzene was taken to be a representative contaminant of interest for produced water because it is mobile, a known carcinogen, federally regulated, and a common constituent of produced water. Glutaraldehyde (biocide), polyethylene glycol (surfactant), and polyacrylamide (friction reducer) were the chemical additives considered for spills of hydraulic fracturing fluid because they represent different classes of common compounds used in hydraulic fracturing and experimental data were available. I focused on the South Platte Alluvial Aquifer, which is located in the greater Denver metro area and overlaps a zone of high-density oil and gas development. Risk of groundwater pollution was based on predicted concentration at the groundwater table. For this screening level assessment, I considered a range of representative hydrologic conditions and chemical behavior for transport of benzene in a sandy loam. Results of the study showed that the risk of benzene contamination of groundwater from a single produced water spill, based on spill data, was relatively low in the South Platte aquifer. Spill size was the dominant factor influencing whether a contaminant reached the water table. Only statistically larger spills with spill volume per area of at least 12.0 cm posed a clear risk. Storm events following a spill were generally required to transport more typical sized spills (0.38 cm spill volume per area) to the water table; typical median-sized spills only posed risk to shallow groundwater if a 500-year storm or if a 100-year storm (followed by little degradation or sorption) occurred right after the spill. However, for more typical storm intensities (1 year and 10 year), only the larger spills (with spill volumes per area ≥6 cm) posed risk to groundwater quality. The purpose of the hydraulic fracturing fluid spill simulations was to assess the importance of accounting for co-contaminant interactions when modeling these spills. Co-contaminant interactions were largely considered in terms of biocidal and salt impacts on degradation of the three chemical additives glutaraldehyde, polyethylene glycol, and polyacrylamide. Results from the study showed that co-contaminant interactions are an important consideration when simulating spills of hydraulic fracturing fluid. Glutaraldehyde removal was impacted by the presence of salt, polyacrylamide, and its own concentration. Polyethylene glycol was affected by the presence of glutaraldehyde and salt. Polyacrylamide remained recalcitrant regardless of chemical composition of the hydraulic fracturing fluid. Because no EPA concentration standards exist for the three compounds, risk was evaluated by using a hazard quotient (HQ) risk evaluation approach based on potential exposure to a substance and the level at which no adverse effects are expected. Considering the highest predicted concentration of glutaraldehyde, the HQ was below 1, indicating relatively low risk associated with a spill of hydraulic fracturing fluid under these conditions. Collectively, results demonstrated groundwater contamination potential due to produced water and hydraulic fracturing fluid spills is low in most areas of the South Platte system for the contaminants and spill conditions investigated. Substantial risk may exist in certain areas where the groundwater table is shallow (less than 10 ft below ground surface), soils contain high sand contents, and when large spills and large storms occur. By helping to identify locations in the Front Range of Colorado that are at low or high risk for groundwater contamination due to a surface spill, this research will aid in improving prevention and mitigation practices so that decision-makers can be better prepared to address accidental releases in Colorado. In addition, this research provides a methodology that could be used to evaluate risk of groundwater contamination due to spills for other aquifers. Future work could apply this methodology to study spills in other areas and to consider other compounds present in produced water and hydraulic fracturing fluid.
Associated Publications
Rights
Copyright of the original work is retained by the author.