Citation:

Burnell, D., J. Xu, AND J. Yang. Present research results and communicate the modeling results to science community. Presented at 10th International Conference on Remediation for Chlorinated and Recalcitrant Compounds, Palm Springs, CA, May 22 - 26, 2016.

Impact/Purpose:

Present the research results in the 10th International Conference on Remediation for Chlorinated and Recalcitrant Compounds by Battelle Institute in Columbus, Ohio

Description:

Background/Objectives. As a result of subsurface heterogeneity, many field and laboratory studies indicate that the advection-dispersion equation (ADE) model fails to describe the frequently observed long tails of contaminant concentration versus time in a breakthrough curve. This inability of the ADE to represent tailing, which is caused by subscale heterogeneity, can potentially lead to a significantly underestimated cleanup timeframe, or incorrectly quantify contaminant fate and transport, from ADE-based models; the actual observed contaminant concentrations may remain for longer periods of time than the model predicts. Given the success of the continuous time random walk (CTRW) modeling framework in matching observed tailing in many field and laboratory studies, this paper presents a novel extension of CTRW to simulate sequential first-order reactions of both parent (e.g. TCE) and its degradation products (e.g. VC). This new modeling approach is expected to be useful for contaminant fate and transport modeling in general, and specifically for improving estimates of the cleanup timeframe in case of the monitored natural attenuation (MNA) requiring long-term site management. Approach/Activities. The inability of the ADE is quantified in controlled tracer test experiments that were performed using 10 ft and 20 ft heterogeneous soil columns at the USEPA Testing and Evaluation Center in Cincinnati, OH. The current ADE form yields a poor match to the observed tails of the breakthrough curve data because the ADE does not accurately represent effects of small-scale heterogeneity and the resultant broad distribution of tracer plume transport velocities. In principle, when a dissolved plume moved along preferential groundwater flow paths, the solute particles may encounter occasional low permeability zones that act to delay solute transport and cause the observed tailing. We show that the CTRW model provides a significantly improved simulation to both early and late time portions of this breakthrough curve data. We have further extended the CTRW to field applications. The new extension of the CTRW model is tested in simulation of dissolved plume mass destruction as a result of sequential first-order reactions. By using the new CTRW modeling tool, the cleanup timeframe of both parent (TCE) and degradation products (cis 1,2 DCE and VC) are estimated using field data from the Harris CERCLA site in Palm Bay, FL. Results/Lessons Learned. Consistent with many other studies, results of this study also indicate that the current ADE model fail to faithfully represent the long tails observed in controlled tracer tests. This inability indicates that ADE-based models may not be accurate for simulating the cleanup time frame particularly in highly heterogeneous aquifers. New CTRW modeling tools are developed to represent effects of subscale heterogeneity by using a single up-scaled heterogeneity parameter that is estimated by matching the model to the observed training dataset in the form of concentration versus time data in time series plots. Given that the CTRW modeling approach is relatively easy to use and is based on a theoretically consistent mathematical framework, the extension of this practical modeling tool to include mass destruction via first-order reaction may be useful to improve the contaminant fate and transport modeling in heterogeneous aquifers, and yield better estimates of the cleanup timeframe for MNA remedies as part of long term site management.

URLs/Downloads:

Record Details: