Formatted Title
Real-Time Radon Measurement in the Headspace of Monitoring Wells for the Assessment of Residual LNAPL
Background/Objectives
In recent decades, there has been growing interest in utilizing naturally-occurring radon (Rn) as a method for detecting and quantifying light non-aqueous phase liquids (LNAPLs) in the subsurface. Because of its preferential partitioning into oily phases, the natural concentration of Rn in soil gas at the subsurface can be influenced by LNAPL contamination. Specifically, Rn concentrations in soil gas are expected to decrease in areas impacted by LNAPL compared to background values in uncontaminated zones, creating a measurable Rn deficit that can be used to identify and quantify LNAPL in the subsurface. In this study, we first introduce an analytical model to examine the feasibility of using Rn monitoring data in soil gas to identify and quantify LNAPL in the subsurface. Next, we present a protocol to use real-time Rn analyzers in the headspace of monitoring wells to assess the contamination of LNAPL in the smear zone.
Approach/Activities
An analytical model was developed to describe the expected Rn behavior in the presence of separate phases in the subsurface. Using the developed solution, the expected vertical Rn concentration profiles in the soil gas above the source zone and in the area not affected by LNAPL were evaluated under different conditions. Based on the simulations performed, it was observed that the parameter that most influences the effectiveness of using Rn as a tracer for LNAPL is the vertical distance of the soil gas probe from the source zone. In particular, at distances greater than 2 m, the method appears to be no longer sensitive to the presence of LNAPL. Therefore, the use of soil gas probes to apply the Rn deficit technique to estimate LNAPL would only be applicable if the probes are installed in close proximity to the source zone. This study presents an alternative method for applying the technique in the headspace of groundwater monitoring wells. The protocol developed, designed for groundwater monitoring wells with a portion of their screen in the vadose zone, is based on the use of portable equipment that allows rapid measurement of the Rn soil gas activity in the vadose zone close to the water table (i.e., smear zone) where LNAPL is typically expected. A preliminary evaluation of the potential of the method was conducted at two Italian sites characterized by accidental subsurface releases of gasoline and diesel from underground storage tanks.
Results/Lessons Learned
In all tests performed, the Rn concentrations in the hydrocarbon-impacted zones were lower than in the background areas, indicating that the proposed protocol is applicable to determine the presence of residual phase in the unsaturated zone, which is not possible with the sole monitoring of the free product in monitoring wells or with water sampling. The method was also effective in estimating the expected concentrations in the investigated area, with results consistent with those available from the characterization phase. In the application of the method, some uncertainties were found related to the presence of heterogeneities in the subsurface (which may affect the estimation of the Rn deficit) and to the temporal variability of Rn emissions from the subsurface, suggesting that in such cases these estimates should be considered semi-quantitative. The results obtained suggest that, from a qualitative point of view, Rn monitoring in the headspace of monitoring wells is a promising, rapid, and minimally invasive screening method that could also potentially reduce the costs associated with field data acquisition. This method proves to be suitable for detecting the presence of LNAPL in both the mobile and residual phases with results consistent with the other lines of evidence available at the sites, such as groundwater and soil gas monitoring. Future efforts should be directed toward evaluating the accuracy of this method for a quantitative assessment of residual LNAPL saturations.