Rapid modeling of catastrophic floods: A case study of the Padang flood on July 14, 2023

PT Reasuransi MAIPARK Indonesia, Multivision Tower 8 th Floor, DKI Jakarta, 12960, Indonesia

^* Corresponding author: yrramadhan@maipark.com

Abstract

This study presents a rapid flood modeling approach for catastrophic flood events in Padang City on July 14, 2023. Floods pose significant risks to urban areas, exacerbated by rapid urbanization and climate change. This research provides a solution to the urgent need for quickly and accurately flood mapping, crucial for effective disaster response and mitigation. Various data such as Digital Elevation Models (DEMs), rainfall, evapotranspiration, soil types, and land use were employed in this study. The modeling process involved DEM conditioning, bias correction of rainfall, defining the model domain using the Topographic Wetness Index (TWI), and applying hydrological and hydrodynamic models. Model verification was conducted using flood location and depth data collected from mass media and social media, showing that the model achieved 77.8% accuracy in mapping inundated areas, with Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) of flood depth at 38.093 and 27.584, before the model was calibrated. The calibration process significantly improved the model’s accuracy.