OPTIMIZATION OF SWAT AND HEC-HMS MODELS IN THE EVENT FLOOD IN BANDUNG BASIN

• Main Author: Perdana, Panji
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/61950
• Institution: Institut Teknologi Bandung

Several areas in the Upper Citarum watershed have a high risk of flooding every year, one way that can be used to analyze the discharge from it is by using hydrological modeling. Hydrological models that can be used to simulate discharge include the Soil and Water Assessment Tools (SWAT) and HEC-HMS models. The SWAT model has the best simulation results in a large watershed area, and has a good ability to simulate for a long period of time. While the HEC-HMS model is commonly used in flood events, this is due to the fact that this output model is able to generate discharge for hourly time scales which are good for simulating flood events. However, the HEC-HMS model has limitations in defining the initial soil water content, this limitation will produce results at the beginning of the simulation that tend to underestimate observations. To optimize the performance of the HEC-HMS model due to limitations in defining the initial soil water content at the beginning of the simulation, an initial discharge will be input that will be generated from the HEC-HMS model. This prefix has been calibrated before being input into the HEC-HMS model. Meanwhile, the area used as the research simulation is the Upper Citarum watershed. The simulation results on HEC-HMS using initial discharge input from SWAT model produce a better simulation discharge than the simulation without initial discharge at the beginning of the simulation. At the Nanjung outlet simulation using initial discharge (R2=0.24, and PBIAS=11.09) the simulation results obtained were better than the discharge results without initial discharge from SWAT model (R2=0.22, and PBIAS=34.15).