Investigating added value of improved DEM, river geometry and model parameter definitions on flood inundation mapping

Digital Elevation Model (DEM) constitutes an essential topographic dataset, which have been used to derive hydrological features forming inputs to various models. Elevation datasets are available from several major sources (remote sensing and land survey based techniques) and at different spatial resolutions. In particular, the resolution of DEM plays an important role in securing the expected accuracy of flood plain inundation maps as one of the major determinants. In general, most of floodplain inundation maps are created using high resolution topographic datasets such as the DEMs derived from Light Detection and Ranging data. Although these datasets provide high accuracy (≈ 1.5 m grid size), they are relatively more expensive and not fully available in Turkey covering greater areas. Therefore, coarser topographic datasets with less accuracy are used to create the floodplain inundation maps. This study applied a methodology framework in order to determine and quantify the "loss of attribute" issue when reducing the resolution of DEM as well as to understand the impact of changing DEM resolution on the floodplain hydraulic model outputs. In the study, HEC-RAS was used for the Iyidere Basin and the results clearly indicated that varying the resolution of DEMs would make significant impacts on both the depth and the extent characteristics observed in the corresponding floodplain for the river. Keywords: Floodplain modeling, HEC-RAS, DEM resolution, iyidere River