Land subsidence hazard assessment based on novel hybrid approach: BWM, weighted overlay index (WOI), and support vector machine (SVM)

Land subsidence is a morphological phenomenon, which causes negative environmental and economic consequences for human societies. Therefore, identifying the areas prone to can be one of necessary measures reducing potential risks. This study aims evaluate efficiency support vector machine (SVM) algorithm weighted overlay index (WOI) models in zoning rate land hazard Hashtgerd plain, Iran. First, 19 criteria include groundwater depletion, extraction, aquifer thickness, alluvium recharge, well density, drainage depth, lithology, bedrock average annual precipitation, temperature, climate type, agricultural use, urban industrial distance from rivers streams, roads, faults were considered. Then, layers weighed based on Best–Worst Method (BWM). The results BWM indicated that factors extraction (0.219), lithology (0.157), depletion (0.079) have greater effect hazard. Moreover, validation by performing ROC curve showed accuracy RBF-SVM, LN-SVM, SIG-SVM, PL-SVM, WOI 95.7, 94.3, 94.9, 93.2, 90%, respectively. Based results, all preparing map plain exhibit excellent accuracy. used here predict vulnerable properly. In this study, five maps as new input integrated using fuzzy gamma-ensemble methods make better maps. ensemble model 19.3% zone high very sensitivity. help planners managing possible hazards subsidence.