Mapping flood resilience: a comprehensive geospatial insight into regional vulnerabilities

Flood is the most frequent and destructive natural disaster, causing significant negative impacts on humans and built and natural ecosystems. While it is extremely challenging to prevent floods, their associated hazards can be mitigated through well-planned and appropriate measures. The present study combined the analytical hierarchy process (AHP) analysis and an ArcGIS-based multi-criteria decision-making (MCDM) approach to assess, categorize, quantify, and map the flood-prone areas in Khyber Pakhtunkhwa, Pakistan, a region particularly vulnerable to recurrent flooding. Eight key factors including precipitation, rivers/streams, slope, elevation, soil, normalized difference vegetation index, and land use were used for flood susceptibility modeling. The weighted sum overlay tool in global positioning system ArcGIS was utilized to give weightage to each raster layer, based on the AHP ranking to produce a flood susceptibility map for the study area. According to the AHP analysis, the most impactful factors defining the flood susceptibility in our study area were streams (0.29%), precipitation (0.23%), slope of the area (14%), and LST (10%). Our flood model achieved excellent accuracy, with Area Under the Curve (AUC) value of 0.911. The model predicted that 9% of the total area is classified as very high risk, while 14% is identified as high risk, covering approximately 923,257 hectares and 1,419,480 hectares, respectively. These high-risk zones are predominantly concentrated in the central and lower northern, densely populated districts of the province. Our flood susceptibility results would assist policymakers, concerned departments, and local communities in assessing flood risk in a timely manner and designing effective mitigation and response strategies.