Watershed modelling is one of the most important factors in the analysis of rainfall-runoff transformations. As an agent of transformation, the catchment plays an important role in determining flow characteristics, especially peak flow as one of flood hydrograph parameters. The main characteristic of this catchment is expressed in the form of surface topography as the basis for determining network channels and its boundaries. The accuracy of determining these two characteristics is largely determined by the type of baseline data used to model the watershed. This study aims to build a watershed model using medium resolution Digital Elevation Model (DEM) data as a flood simulation means. DEM data is derived from Shuttle Radar Topography Mission (SRTM) data with a resolution of 30 meters and is obtained from USGS Earth Explorer for the Bangga Catchment, as one of the flood-prone areas in Central Sulawesi, Indonesia. Channel network derivation and catchment delineation were performed using Raster (R. *), one of QGIS Processing Toolbox. The result of the catchment analysis using this data was also compared with the low-resolution DEM data of 90 meters to determine the effect of DEM resolution on stream network configuration and catchment boundary. Based on this watershed, flood prediction is carried out using the ITS-2 Model, one of the flood estimation models in the form of synthetic unit hydrographs based on river network configuration. The analysis results indicate that the utilization of DEM data with a higher resolution produces smoother catchment features and influences the accumulative length of drainage channels and the area of the catchment. The hydrology simulation using the watershed with the input of design rainfalls shows that the peak flood discharge in the study area for certain return periods is relatively very large for a catchment of fewer than 100 km2 in area