Considering Monitoring Well Bias in the Delineation of Benzene Plume Lengths

Underground storage tanks (USTs) primarily store and dispense fuel products (petroleum hydrocarbons) such as gasoline and diesel as well as some hazardous substances. When a release occurs from an UST, it must be cleaned up and remediated as necessary. As part of the site characterization process, monitoring wells are installed to measure petroleum hydrocarbon concentrations, such as benzene, to help site managers make decisions on remediation strategies and effectiveness, determine site status, and whether sites pose a risk to local receptors—with implications to human health and the environment. Previous studies have used various approaches to estimate benzene plume lengths using these monitoring well data. Examining one state’s large database of monitoring well data from leaking underground storage tank (LUST) cleanup sites, and through statistical and spatial analysis, we find that studies that used existing monitoring well data may have underestimated plume lengths. Our study identified elevated benzene concentrations at the boundaries of monitoring well extents, including at the leading edge of plumes. For example, in 3,495 LUST sites, 83% exceeded a selected threshold concentration of $5\mathrm{ug}/\mathrm{L}$ in a monitoring well at the site boundary, and 76% of sites exceeded $5\mathrm{ug}/\mathrm{L}$ at the monitoring well furthest from a defined source, the point used to delineate length. Of the total 17,816 monitoring wells that delineated UST site boundaries, 47% had concentrations exceeding $5\mathrm{ug}/\mathrm{L}$. These data analyses and results are done over the lifetime of a monitored UST site, not targeted over time periods when decisions are being rendered. However, results suggest that studies to date using existing monitoring well data may have underestimated plume lengths and extents. Furthermore, results suggest monitoring wells in combination with other site information, such as modeling, should be used to guide site managers in their decisions regarding leaked UST products, the area and extent of contamination, risks to receptors, and remedial approaches.

USTs, which store fuel at locations such as gas stations, are known to leak fuel products into soil and groundwater. Fuel products contain petroleum hydrocarbons like benzene, a cancer-causing chemical. Previous studies that have attempted to identify the maximum distance benzene plumes travel have been hampered by a bias in available data—specifically, that monitoring wells are often not placed at far enough distances to be able to map a plume to concentrations below concerning levels. This paper identifies this bias and provides insight into how it may be considered in the future to more effectively map plumes and identify potential biases that can lead to inconclusive results related to plume behavior. While traditional physics-based models remain useful, they can be augmented with data-driven models to more accurately estimate plume extents, especially when real-world data are lacking.