Flood mitigation role of traditional tank cascades: quantifying land use change and siltation effects using integrated hydrologic-hydraulic modelling

ABSTRACT

Graphical abstract illustrating the workflow and major findings of a study investigating the effects of land use change and tank siltation on flo od risk in the Kaluveli watershed, Tamil Nadu, India. The figure begins with a map of the study area showing the watershed boundary, stream network, tan ks, and Kaluveli Lake. The methodology panel presents an integrated modelling framework using rainfall, digital elevation model (DEM), land use/land c over (LULC), and tank data as inputs to HEC-HMS hydrologic modelling, followed by HEC-RAS 2D hydraulic modelling and Google Earth Engine (GEE) analysis with Sentinel-1 SAR imagery for flood mapping and validation. Scenario analysis compares 2015 and 2025 LULC conditions and functional versu s silted tank systems, demonstrating increased urbanization and reduced flood storage due to tank siltation. Flood inundation maps show larger and deeper flooded areas under the 2025 LULC and silted tank scenario. Key results indicate that peak discharge increases by approximately 10–12%, flood inundatio n extent increases by up to 19%, and flood depths increase in low-lying areas when tank storage is reduced. Validation against GEE-derived flood maps sh ows good spatial agreement, with simulated and observed inundation differing by approximately 10–15%. The graphical abstract concludes that maintainin g and restoring traditional tank cascade systems can reduce flood peaks and inundation, while integrating HEC-HMS, HEC-RAS, and remote sensing prov ides an effective framework for flood risk assessment and management.

This study quantifies the impact of LULC change and tank siltation on flood peak discharge and floodplain extent using an integrated hydrologic–hydraulic modeling framework (HEC-HMS and HEC-RAS 2D). LULC maps for 2015 and 2025 were generated using supervised classification, and floodplain maps were developed for a 100-year return period under both years. Frequency analysis using the Gumbel distribution indicated that the 100-year design storm corresponds to a rainfall depth of 150 mm, which was used as input for hydrological and hydraulic simulations. Results show that excluding tanks increased peak discharge from 700 to 780 m3/s (11.4%) in 2015 and from 750 to 820 m3/s (9.3%) in 2025, while flood inundation extent increased by up to 18.6% under silted conditions. Enhanced flood depths were observed primarily in low-lying and intensively modified land use zones. The Analytical Hierarchy Process (AHP) was used to figure out what causes floods. Model validation using Sentinel-1 SAR imagery processed in Google Earth Engine showed good spatial agreement, with deviations within 10–15% and performance metrics indicating strong model reliability (CSI = 0.74; POD = 0.89). These findings demonstrate the critical role of functional tank cascade systems in attenuating flood peaks and reducing inundation.