Evaluasi Model Elevasi Digital dalam Pemodelan Genangan Banjir di DAS Wae Mese

Abstract

On 7 March 2019 and 31 December 2021, extreme rainfall events occurred in the Wae Mese Watershed, West Manggarai Regency, resulting in the inundation of several villages and agricultural fields. Currently, detailed records regarding the area and depth of these flood events are unavailable, hindering the formulation of comprehensive disaster mitigation plans. This study aims to reconstruct the flood inundation area and depth in the Wae Mese Watershed, while simultaneously evaluating the accuracy of various open-access Digital Elevation Models (DEM). The flood inundation simulation was performed using a two-dimensional (2D) HEC-RAS hydraulic model, driven by flood discharge values derived from a rainfall-runoff model using recorded rainfall data during the events. The flood model simulation results were calibrated against the inundation depth from the 2019 flood event in Golo Bilas, Macang Tanggar, and Gorontalo Villages. The inundation modeling incorporated four open-access DEMs: SRTM DEM, Merit DEM, NASA DEM, and DEMNAS. Simulation results indicated that DEMNAS yielded an inundation depth of 1.03–2.03 m and an inundation area closest to the actual flood event on 7 March 2019. The statistical indicators, such as an RMSE of 0.38, a BIAS of 0.15, a correlation coefficient of 0.92, and a RVE of 10.6%, show that the results are close to the actual flood.  The superior performance of DEMNAS is attributed to its higher spatial resolution compared to the other DEMs, providing a more precise representation of surface topography for modeling inundation areas and river channels. Consequently, DEMNAS is recommended as a reliable reference for developing flood hazard maps and planning flood mitigation strategies in the Wae Mese Watershed.