Sustainability appraisal of desired contaminated groundwater remediation strategies: an information-entropy-based stochastic multi-criteria preference model

This study proposes an information-entropy-based stochastic multi-criteria preference model (SMCPM) to assess the sustainability of groundwater remediation strategies. The model was applied to a real-world site, which includes the following steps: (1) selecting fifty remediation alternatives with four criteria, including total pumping rate, remediation cost, average contaminant concentration and general cancer risk in light of uncertainty parameter (e.g., slope factor); (2) evaluating the health risks associated with the strategies under three confidence levels (i.e., 68.3%, 95.4% and 99.7%); (3) computing the weight by information entropy; and (4) determining the ranking of remedial alternatives by establishing a random evaluation matrix. Results from the case study indicate that: (1) the most desirable actions are action A39 (i.e., the pumping rates at six wells are 3.499, 2.722, 2.624, 1.458, 7.776 and 2.527 m(3)/h, respectively) for the 5-year; (2) action A25 (i.e., the six pumping rates at injecting and extracting wells are 2.819, 2.722, 0, 3.791, 0.972 and 1.166 m(3)/h, respectively) for the 10-year; (3) action A24 (i.e., the pumping rates of wells P1-P6 are 1.944, 1.555, 1.166, 3.110, 7.193 and 0.583 m(3)/h, respectively) for the 15-year; (4) action A37 (i.e., the optimal pumping rates of P1 to P6 become 2.138, 2.722, 3.694, 2.722, 3.694 and 7.582 m(3)/h, respectively) for the 20-year remediation, respectively. Compared to the traditional MCPM, the proposed SMCPM would be more acceptable and reasonable since various uncertainties be addressed.

Automated summary

This study introduces an information-entropy-based stochastic multi-criteria preference model (SMCPM) for assessing the sustainability of groundwater remediation strategies. By selecting, evaluating, computing weights, and establishing a random evaluation matrix, the model identifies the most desirable remedial actions for different time frames, providing a more acceptable and reasonable approach compared to traditional models by addressing various uncertainties.