A SYSTEMATIC DATA-DRIVEN APPROACH TO EVALUATING HYDRAULIC CAPTURE AT SUPERFUND SITES IN USEPA REGION 5

We are developing systematic methods for evaluating monitoring data as part of the CERCLA five-year review process. These methods take advantage of standardized data formats in USEPA's Region 5 guidance for electronic data deliverables (EDDs) adopted in part by 6 of the 10 U.S. EPA regional offices, many state environmental agencies, and other users of environmental data. This presentation focuses on the use of mapping-based techniques to illustrate the likely extent of hydraulic capture by pump and treat systems. The extent of groundwater contaminated above cleanup levels for any contaminant of concern (COC) is mapped as a Composite Target Zone (CTZ). The CTZ is based upon the upper 95% confidence limit (UCL) of recent measurements for each COC at each monitoring point as calculated using PAM intrawell statistical analysis software developed by Subterranean Research.

These methods require data for water levels, contaminant concentrations, and pumping rates that can be rapidly manipulated, evaluated for quality control, and summarized into discrete data sets. The mapping uses universal kriging, embedding equations that describe the response of the potentiometric surface to extraction or injection wells and discharge to/from surface water bodies. Particle tracking is used with the mapped surface(s) to approximate capture zones. If a groundwater flow model is available, related mapping techniques can be used to incorporate simulation results as a trend or ‘drift'. Automation of database queries, kriging, and particle-tracking routines enables relatively rapid evaluation of hydraulic capture at multiple time periods, and stochastic (frequentist) approaches. Dozens of discrete time period maps may be evaluated simultaneously to develop capture probability maps.

The efficiency and cost-effectiveness of these analyses is dependent upon data standardization and quality review. While automation of mapping can be effective, it does not mitigate the need for thoughtful assessment of the site. For example, analyses can reveal significant concerns unrelated to hydraulic capture, such as conflicts between proposed treatment technologies and the groundwater redox conditions, or capture-related concerns such as inadequate pumping rate data or infelicitous monitoring well locations.