Flood modeling using GIS-based analytical hierarchy process in Gandak river basin of Indian territory

Flood events in recent years have increased as a major catastrophic event and India is listed among top nations in the globe. The Gandak river basin in Indian Territory puts heavy flood situations every year and the present study was done to classify it into flood susceptible zones using analytical hierarchical process (AHP). The AHP was implemented through a series of steps, including the multicollinearity test, creation of a pairwise comparison matrix, the assignment of weights, consistency check, sensitivity analysis and weighted overlay analysis using ArcGIS V.10.5 platform. The Consistency Ratio (CR) values showed perfect consistency when calculated for each thematic parameter as the values were below 0.1 or 10% showing high acceptance for AHP model to run. To attain dependable precision, the study incorporated a substantial number of thematic layers (N = 12 for FSZ) and conducted multicollinearity analysis of these variables to address issues related to high correlation among them using SPSS. For the multicollinearity check, the tolerance value should be greater than 0.1, and the percentage of variance inflation factor (VIF) should be less than 10 for the accuracy evaluation of this test. The AHP was successfully employed in the present area and flood susceptibility zone (FSZ) map generated revealed elevation and slope were dominant factors affecting FSZ map. The map removal sensitivity analysis and single parameter sensitivity analysis described the importance of elevation and slope in the final FSZ map. The high flood susceptibility region mostly lies in the southern part of the study area and percent cover under high and very high is around 44.19% reflecting regions suffering extreme conditions of flood during monsoon season. Very low and low flood zones are situated in northern part of the basin which clearly indicates that slope and elevation are the predominant factors controlling the flood zones. The AUC/ ROC curve value was 83.60%. The current study finds its greater utilization for environmentalists, farmers, agricultural engineers and researchers to carry out sustainable development, and most importantly would help government bodies to regulate remedial measures.