Team: Jack Parker, UT; and Ungtae Kim, Cleveland State University
The objective of this project is to develop and test a methodology to periodically assess and optimize groundwater remediation systems at DoD sites contaminated with dense nonaqueous phase liquids (DNAPL). A computer program designated the Stochastic Cost Optimization Toolkit (SCOToolkit) that includes modules for (1) contaminant fate and transport in groundwater with multiple DNAPL sources, (2) cost-performance models for commonly used remediation technologies coupled with the transport model, (3) an inverse solution to calibrated model parameters from available site data, (4) a Monte Carlo model to compute the probability-weighted cost of a given remediation strategy, (5) an optimization algorithm to determine design variables that will minimize life cycle expected net present value cost to reach remediation objectives, and (6) a graphical user interface to simplify field applications.
The transport module employs a newly developed efficient 3-D semi-analytical solution for resident and flux concentrations that accounts for natural and engineered DNAPL source reduction and diffusion-limited mass transfer between high and low permeability zones. Remediation technologies that have been implemented include thermal source reduction, continuous and pulsed in situ chemical oxidation, electron donor enhanced reductive dechlorination, and plume containment. All functional modules have been tested on hypothetical problems and three demonstration sites have been selected for field testing. Efforts are in progress to develop a web-based user interface with on-line training modules.
An interesting finding from SCOToolkit optimization studies, that is contrary to current practice, is that decreasing monitoring frequency and/or number of samples per sampling event does not necessarily reduce expected life cycle. While less intensive monitoring reduces direct monitoring costs, fewer samples yield wider confidence limits which can increase the duration of long term monitoring to achive cleanup objectives with the same level of confidence.
Results from test problems and limited field site testing indicates that SCOToolkit can significantly increase the probability of meeting remediation objectives within a specified period and decrease expected cost based for one-time calibration and optimization analyses. The current program has been modified to allow periodic recalibration and optimization to take advantage of additional data collected since the prior calibration and optimization analysis. Preliminary results indicate this will substantially improve long-term site management and further decreases life-cycle costs. We anticipate average cost savings across all DoD sites of 10% to 30% or more, which can translate to hundreds of millions of dollars in cumulative savings to DoD, other public agencies, and private entities. Project completion is anticipated in mid-2017.
This project, initiated in October 2013, is being conducted for the US Department of Defense to help clean up groundwater at US military installations contaminated by persistent chlorinated solvents. The objective is to develop a practical tool for optimizing the design and operation of groundwater remediation systems that explicitly considers uncertainty in site and remediation system characteristics, performance and cost model limitations, and measurement uncertainties that affect predictions of remediation performance and cost. The approach involves periodic reassessment of current remedial actions and determination of the cost-optimal forward strategy to meet remediation goals considering additional data collected since the previous assessment.
The method is based on a semi-analytical mathematical model to simulate chlorinated solvent source depletion and dissolved phase transport in response to natural and engineered conditions. The performance model is coupled with cost functions for thermal source zone treatment and enhanced bioremediation by electron donor injection. Compliance criteria are defined by statistical rules. The performance model is also coupled with an inverse solution to estimate model parameters, parameter covariances, and residual prediction error. A stochastic cost optimization (SCO) algorithm is used to determine values for design and operation variables, including monitoring, that minimize expected net present value cost over Monte Carlo realizations. The method is implemented in the SCOToolkit software, available in MATLAB or executable code. The method was applied to two well-characterized sites where different remedial technologies were used, to evaluate its ability to reduce costs and improve remedial designs.
SCOToolkit has been applied at various field sites. At the Fort Lewis East Gate Disposal Yard (EGDY) site, optimization of thermal source treatment indicated a need for a much larger treatment area than was actually employed, to avoid a high failure probability associated with source delineation uncertainty based on available source characterization data. The method was also used to optimize source and plume bioremediation at the site, using whey injection without additional source reduction. The results indicated that this strategy hould achieve Maximum Concentration Limits by 2100, with a 94% probability of success using relatively low whey injection rates. The approach was applied to Dover AFB Area 5 to optimize dissolved plume bioremediation. Optimization to minimize long term operating costs indicated compliance criteria could be met using only five of the current ten emulsified vegetable oil injection galleries, with operating costs approximately half of current costs. Recalibration and optimization after an initial period of operation, using additional data from this period, was projected to further reduce operating costs.
The results indicate that SCOToolkit can reduce expected costs by 50% or more relative to conventional design methods, while substantially increasing the probability of meeting compliance targets. The method can also identify critical data gaps and uncertainties that will affect costs and performance projections.