Multi-year CO2 efflux measurements for assessing natural source zone depletion at a large hydrocarbon-impacted site

The changing landscape of fuel consumption related, in part, to increased engine efficiency and the inexpensive supply of natural gas, has led to the closure of multiple refineries. As the operational lifetime of many refineries exceeds 100 years, historical releases of oil and refined products is common. To evaluate remediation and rehabilitation options, there is a need to understand the rate and distribution of natural hydrocarbon mass losses across these large properties. Here, surficial CO2 flux measurements were used to evaluate naturally occurring hydrocarbon mass losses at a large-scale former refinery that has been closed since 1982. Natural source zone depletion (NSZD) rates over a five-year period (2012–2016) were derived from surficial CO2 efflux measurements on a high-resolution grid (N > 80). Results demonstrate substantial variations of mass loss rates across the site. Average site-wide mass loss rates ranged from 1.1–5.4 g TPH m−2 d−1 as C10H22 with a multi-year average of 4.0 g TPH m−2 d−1 as decane (C10H22), consistent with observations at other sites. Statistical analysis demonstrated that the same average mass loss rates would have been obtained with fewer measurement locations (N = 20–30). Comparing NSZD rates to site metadata show CO2 fluxes to be a reasonably good proxy for zones of subsurface hydrocarbon contamination – particularly with respect to vadose zone impacts. It is hypothesized that the observed decline of NSZD rates over the study period is related to rise of groundwater levels, leading to increased submergence of the smear zone. Overall, mass loss rates calculated from CO2 fluxes show NSZD can result in substantial contaminant removal, which may rival that obtained from engineered remediation, under some conditions. •CO2 effluxes were measured at a large-scale former refinery site.•Sampling density required for calculating representative NSZD rates was evaluated.•CO2 effluxes were correlated with subsurface source zone characteristics, indicating they can be used for contaminant delineation.•Site-wide NSZD rates calculated using background correction ranged from 1.1 to 5.4 g TPH m−2 day−1.•Site-wide NSZD mass losses were comparable to those achieved by engineered systems in localized areas.