New DRASTIC framework for groundwater vulnerability assessment: bivariate and multi-criteria decision-making approach coupled with metaheuristic algorithm

Environ Sci Pollut Res Int. 2022 Jan;29(3):4474-4496. doi: 10.1007/s11356-021-15966-0. Epub 2021 Aug 18.

Abstract

Unplanned anthropogenic activities and erratic climate events pose serious threats to groundwater contamination. Therefore, the vulnerability assessment model becomes an essential tool for proper planning and protection of this precious resource. DRASTIC is an extensively adopted groundwater vulnerability assessment model that suffers from several shortcomings in its assessment due to the subjectivity of its rates and weights. In this paper, a new framework was developed to address the subjectivity of DRASTIC model using a bivariate, multi-criteria decision-making approach coupled with a metaheuristic algorithm. Shannon entropy (SE) and stepwise weight assessment ratio analysis (SWARA) methods were coupled with biogeography-based optimization (BBO) to modify rates and weights. The performance of developed models was assessed using area under the receiver operating characteristic (AU-ROC) curve and weighted F1 score. The Shannon-MH model yields better results with an AUC value of 0.8249, whereas other models resulted in an AUC value of 0.8186, 0.7714, 0.7672, and 0.7378 for SWARA-MH, SWARA, SE, and original DRASTIC models, respectively. It is also evident from weighted F1 score that Shannon-MH model produced maximum accuracy with a value of 0.452 followed by 0.437, 0.419, 0.370, and 0.234 for SWARA-MH, SWARA, SE, and original DRASTIC models, respectively. The results indicated that Shannon model coupled with metaheuristic algorithm outperforms other developed models in groundwater vulnerability assessment.

Keywords: Biogeography-based optimization (BBO); DRASTIC model; Groundwater vulnerability; Shannon entropy; Stepwise weight assessment ratio analysis (SWARA); Weighted F1.

MeSH terms

  • Algorithms
  • Anthropogenic Effects*
  • Environmental Monitoring
  • Groundwater*
  • Models, Theoretical