Application of Four Measurement Techniques to Understand Natural Source Zone Depletion Processes at anLNAPLSite

There are several key data gaps in our understanding of Natural Source Zone Depletion (NSZD) processes at sites impacted by light nonaqueous phase liquid (LNAPL), and quantifying NSZD rates can be challenging due to the inherent differences in measurement methods. In this study, four different NSZD measurement methods (oxygen influx measured by the Gradient Method, long-term carbon dioxide efflux measured with Carbon Traps, instantaneous carbon dioxide efflux measured with Dynamic Closed Chambers (DCC LI-COR), and the long-term heat flux from biodegradation measured by Thermal NSZD monitoring), as well as LNAPL composition and dissolved gas sampling, were applied at a site in Southern California. These techniques were used to evaluate key questions such as: (1) how do different NSZD rate measurement methods compare, and what causes variability in NSZD results?; (2) to what extent NSZD processes are occurring in LNAPL within the saturated zone?; and (3) how is NSZD related to LNAPL composition change over time? Carbon Traps and Thermal NSZD monitoring measurement methods provided the most consistent NSZD data at this geologically heterogeneous site, with two location average NSZD rates of 540 and 480 gal/acre/year, respectively. Overall, comparisons of NSZD rates between methods were challenging due to different measurement timeframes, significant temporal and spatial heterogeneity, and operational challenges with two of the NSZD methods. Finally, samples of subsurface LNAPL were collected for analysis in 2007 and 2016; results indicated that diesel-range constituents were already very degraded and anaerobic degradation of gasoline-range constituents was ongoing. A LNAPL depletion model (Douglas et al. 1996) applied to the measured LNAPL composition change appeared to greatly overestimate the amount of LNAPL depletion compared to the measured NSZD rate, but did provide an independent semiqualitative line of evidence that LNAPL was being depleted by active NSZD processes at the site.