In many cases, especially at recalcitrant sites with complex hydrogeology, inaccurate or inadequate delineation of groundwater flow fields at appropriate resolution has resulted in poor remediation performance. Hydraulic conductivity (K) and specific storage (S_s) are the major parameters governing the fate and transport of contaminants in the subsurface. High-K zones and fractures are fast-flow conduits where transport of dissolved contaminants potentially poses significant threats to downgradient receptors. Low-K zones are potential repositories of contaminant mass that slowly release contaminants and contribute to long-term risks and liability.

The overall objective of this project was to provide the Department of Defense (DoD) and its remediation contractors with the Hydraulic Tomography (HT) technology for delineating the spatial distribution of the K and S_s parameters in high resolution. Specific technical objectives were to: 1) demonstrate that HT is superior to conventional methods for estimating the spatial distribution of hydrogeologic properties; 2) illustrate that an HT survey can be readily conducted at DoD sites using existing networks of groundwater extraction/injection and observation wells; and 3) develop guidance for HT field implementation and compare costs associated with HT and conventional methods.

The HT concept was analogous to the Computerized Tomography (CT) scanning technology, which was based on combining a series of X-ray images taken from many different angles to make detailed pictures of the physiological structures inside a human body. HT involved sequentially conducting a series of aquifer hydraulic tests (HT survey). The hydraulic stresses in the subsurface were perturbed differently in each test, and the resulting potentiometric head changes over a well network were monitored. Each test was comparable to taking a snapshot of the aquifer heterogeneity at one angle, and the whole HT survey was analogous to hydraulically scanning the subsurface from many different angles. The complete data set of observed hydraulic head responses at multiple locations were jointly analyzed through a consistent mathematical model, which provided detailed spatial distribution of hydraulic properties of the aquifer, patterns of connectivity of highly conductive zones, locations of low conductive zones, and the uncertainties associated with the spatial distribution (HT analysis).

The technical performance and cost-effectiveness of HT have been demonstrated at two field sites: (1) the University of Waterloo (UW) North Campus Research Site (NCRS), which was a local-scale site extensively instrumented at a spatial resolution critical to typical source zone remedial actions; and (2) the Air Force Plant No. 44 (AFP44) site, which was at a field-scale typical of DoD environmental sites with an existing pump-and-treat system and monitoring well network.

The results from the demonstrations at both sites confirmed that HT provides more accurate site characterization than conventional techniques. In the context of predicting hydraulic responses induced by other pumping tests, hydraulic property estimates from HT unambiguously outperform those of conventional models. The HT predictions are unbiased and have smaller errors and uncertainty.

The HT results were consistent with the current knowledge of the spatial distribution of the high-K and low-K permeable regions. The demonstration at the AFP44 site, where pump-and-treat remediation was on-going, illustrated that HT is cost-effective and can readily be applied at other sites with existing well networks and pump-and-treat systems. The only HT site characterization costs were the labor costs for conducting pumping tests and performing HT model inversion. HT is a “user-friendly” site characterization technology. The skills and equipment needed for conducting an HT survey are the same as those commonly used in conventional site characterization. The input data required for model inversion by HT are the same as the data used in groundwater model development and calibration, such as the input data for parameter estimation using the commonly used software PEST and MODFLOW. Besides, HT delineates low-K zones consistent with the available local lithologic data. HT infers the hydraulic continuity of the low-K regimes in between available lithologic information. It provides information as to whether these regimes are hydraulically functioning as competent barriers. In conjunction with available existing chemical concentration data, HT is useful for evaluating potential residual sources.

The key factors to be considered in making a decision as to whether HT is appropriate for a site include cost-effectiveness, timing and duration, knowledge of background hydraulic stresses, and chemical mobilization. The cost-effectiveness depends on the appropriate number of wells, which is dictated by the spatial resolution needed to meet the objectives and whether existing wells and treatment system are adequate. If an existing well and treatment system can be utilized, the costs associated with HT are minimal.

If additional wells are needed, and especially if they need to be installed in areas with high chemical concentrations, pertinent regulatory approval and permits might be required. This is a similar issue with conventional well installation. If the HT pumping tests involve groundwater extraction, pumping permits might be required. In addition, permits for the discharge to the onsite or off-site treatment systems need to be acquired. Depending on the application process, extraction water sampling might be necessary. Similarly, permits might have to be obtained for water injection, with a potential sampling of the injection water.

In addition, water level changes due to HT pumping tests might cause chemicals to move during the tests. The duration of the pumping tests is usually short, and the amount of the associated chemical movement is typically small. However, if the aquifer is very permeable, a large pumping rate might be required to generate a measurable hydraulic response signal. On the other hand, if the aquifer is relatively less permeable, the well yield might be small, and a longer HT pumping test duration might be needed.