Flash flood susceptibility assessment using an integrated geomorphologic framework in data-scarce arid and semi-arid regions

Abstract Flash floods are among the most frequent natural hazards, posing serious environmental challenges, particularly in arid and semi-arid regions where steep topography and compact basin morphometry accelerate surface runoff. The ungauged hill torrents of southwestern Punjab, Pakistan, are particularly vulnerable to such events, experiencing recurrent, intense flash floods. This study develops an integrated GIS-based morphometric framework to assess flash flood susceptibility across 13 major hill torrents in DG Khan and Rajanpur, Punjab. Fifteen morphometric parameters were analyzed using three complementary approaches: Hazard Degree (HD), Principal Component Analysis (PCA), and Analytical Hierarchy Process (AHP). The HD method captured the cumulative influence of all parameters. At the same time, PCA reduced redundancy and identified four dominant controlling factors: length of overland flow (Lo), form factor (Ff), circulatory ratio (Rc), and ruggedness number (Rn). The AHP further incorporated expert judgment, identifying drainage density (Dd) and ruggedness number as the most influential parameters, followed by basin relief (Bh) and drainage frequency (Fs). The convergence of results across the three independent methods significantly improves classification reliability compared to single-method analyses. Three hill torrents (Vidore, Mithawan, and Kaha) were consistently classified as highly susceptible, one as moderately susceptible, and four as low susceptibility. Hence, the proposed integrated HD-PCA-AHP model reduces methodological uncertainty and provides a transferable, data-efficient tool for assessing flash flood susceptibility and managing watersheds in data-scarce arid and semi-arid regions.