Urban flooding has intensified due to climate change, particularly in rapidly expanding metropolitan regions. This study presents an integrated remote sensing (RS) approach to monitor flood-prone areas in the Porto Alegre Metropolitan Region (Brazil). The methodology combines predictive topographic modeling using the HAND model with post-flood analysis based on water spectral indices. Flood susceptibility was assessed using HAND derived from SRTM data, while flood extent was estimated with NDWI and WNDWI applied to Landsat imagery and validated against HEC-RAS hydrodynamic simulations for the 2015 and 2024 events. HAND effectively identified low-lying, high-risk areas near watercourses but showed limitations in densely urbanized zones. WNDWI mapped up to 2423 km2 flooded in 2024 (22 % of the area), versus 1206 km2 by NDWI. Thematic accuracies for WNDWI reached 98-99 %, with Kappa values of 0.75 (2015) and 0.82 (2024), and AUC of 0.55 and 0.74, indicating strong agreement with HEC-RAS. WNDWI outperformed NDWI in detecting floodwaters in highly turbid and heterogeneous areas, though both indices were less sensitive in transition zones. This study's main contribution is the novel integration of modeling, spectral indices, and hydrodynamic validation, enhancing the reliability of RS-based flood mapping and supporting urban planning, risk management, and implementation of resilient strategies in climate-vulnerable metropolitan areas.
Keywords: Flood susceptibility; HAND model; Landsat; NDWI; Remote sensing; Spectral indices; Urban flooding; WNDWI.
Copyright © 2025 The Authors. Published by Elsevier B.V. All rights reserved.