Performance Quantification of Flood Hydrograph Models Based on Catchment Morphometry Derived from 30 m SRTM DEM

Synthetic hydrograph unit (SUH) is one type of hydrological models for predicting floods at a watershed outlet. The accuracy of this model, apart from being dependent on the distribution and intensity of the rainfall, is also largely determined by the characteristics of the watershed, especially the area of the watershed and the configuration of the drainage network system. The characteristics of the watershed that can be described by morphometric and fractal properties, where the determination of which is very dependent on the resolution of the digital elevation model (DEM) data as the basis for derivation. This paper aims to evaluate the performance of SUH models based on catchment morphometry derived from Shuttle Radar Topography Mission (SRTM) DEM with a resolution of 30 m. This topographic elevation data is one of the free medium-resolution DEMs produced by the US National Geospatial-Intelligence Agency (NGA) and National Aeronautics and Space Administration (NASA). Two flood hydrograph models based on catchment morphometry—GAMA I and ITS-2—were evaluated by quantifying their performance using root mean squared error (RMSE) criteria. This performance indicator was measured based on predicted and measured unit hydrograph in one of sub-watersheds of Palu, Sulawesi, Indonesia, namely Bangga Catchment. Predicted unit hydrograph is interpreted from morphometric and fractal catchment parameters, while the observed unit hydrograph is derived from several flood hydrograph data. The evaluation results showed that the two hydrograph models that were examined showed very good performance with an RMSE below 1. This indicates that the 30 m SRTM DEM used to derive catchment parameters is still relatively good in defining the characteristics of the catchment. However, for applications to other catchments using the same DEM type, the rainfall characteristic must be carefully identified in order to represent the rainfall distribution pattern across the catchment.