A novel combination approach for karst collapse susceptibility assessment using the analytic hierarchy process, catastrophe, and entropy model

The evolution of cover collapse is a severe hazard in karst regions. The main objective of the present work was to develop a novel approach that combined both subjective and objective methodologies to evaluate sinkhole susceptibility. Based on the comprehensive analysis of the mechanisms for sinkholes, a typical subjective method was first built using the analytic hierarchy process (AHP) with a hierarchical structure that included nine factors. Considering the apparent disadvantage of AHP, the catastrophe theory was integrated to determine the weight of the criterion factors. To further improve and avoid the bias of the assignment of weights, the entropy method was then integrated into the model to objectively and reasonably determine the order of the index factors and weights of the sub-factors in the index layer during the calculation of the catastrophe model. The verification results showed that the combination of the subjective and objective approaches was indeed suitable to indicate collapse susceptibility. The sensitivity analysis results indicated that the thickness of the overlying layer and karst development were the most sensitive parameters, as indicated by the high rate value using the subjective method. The karst collapse area was then classified into very high-, high-, medium-, and low-susceptibility areas, which accounted for 20.09%, 19.82%, 38.58%, and 21.51% of the total area in the study region. The extraction of groundwater, especially mine draining, was the most important factor, causing more severe hazards, especially in the very high- and high-susceptibility areas.