Author(s): A. Md. Ali; G. Di Baldassarre; D. P. Solomatine
Linked Author(s):
Keywords: Topographic contour; ASTER; SRTM; LiD AR; One-dimensional model
Abstract: Digital elevation models (DEMs) used as geometric input of hydraulic models can be acquired through numerous sources of topographic information, originating from remote sensing techniques or ground surveys. Currently, many space-borne DEMs are available at a global scale and low resolution, such as the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and Shuttle Radar Topography Mission (SRTM). At river reach scale, and high resolution, DEMs can be acquired through the highly accurate and precise Light Detection and Ranging (Li DAR). In this study, the effects of using different sources of DEM on the results of hydraulic modelling of floods were analyzed. Different DEMs of a 30-km reach of the Johor River, in Malaysia were generated from four sources of topographic information: traditional ground survey, ASTER, SRTM and LiD AR. One-dimensional (1D) models were built by using these DEMs as topographic input data and HEC-RAS as the model code. The hydraulic models were set up and calibrated using hydrometric data of the 2006 flood event, and then validated against the 2007 flood event. The hydraulic models were investigated particularly on the following important criteria: (i) the sensitivity of each hydraulic models to changes in the Manning’s n roughness coefficients; (ii) the results of water surface elevation derived from each hydraulic model; and (iii) the accuracy of the simulated flood extent. The results of the study revealed interesting insights of practical importance on: (i) the impact on hydraulic modelling when different DEM sources is used as the input data, (ii) the potentials and limitations of HEC-RAS when combined with DEMs that are globally and freely available (ASTER, SRTM). In particular, the research works showed that the source of topographic information has serious implications to the outcomes of flood inundation models.
Year: 2014