Bioslurping Model for Assessing Light Hydrocarbon Recovery in Contaminated Unconfined Aquifer. I: Simulation Analysis

Abstract

This study carries out a system-based assessment in search of the optimal operating strategy for light hydrocarbon recovery in a contaminated aquifer using bioslurping technology. A two-stage analytical framework involving a combined simulation/regression/optimization (S/R/O) modeling approach was designed to analyze, predict, and optimize the pumping volume with regard to the separate oil, water, and soil gas phases in the bioslurping process. The aim of the first-stage analysis is to calibrate and verify this newly developed bioslurping simulation model and to assess light nonaqueous phase liquids (LNAPLs) recovery efficiency in a heterogeneous, anisotropic unconfined aquifer. Recovery system simulations using derived linear regression submodels present high potential for exhibiting, eliciting, and summarizing the nonlinear behavior between successive batch operations in predicting residual oil levels in the surface groundwater table for model calibration and verification. The research findings indicate that the statistical paired-sample T-tests confirm the prediction accuracy of the bioslurping model during calibration and verification procedures. The simulation results presented in this paper are proved useful for finding a relatively better operational scheme to minimize the indirect operational risk throughout the on-site remediation time period. The simulation results also provide a firm database to assist in an advanced optimization study in the second stage analysis for generating risk-informed, scientifically credible, and cost-effective solutions in remediation.