Development of a Flood Model Based on Globally-Available Satellite Data for the Papaloapan River, Mexico

摘要: Flood inundation modelling is highly dependent on an accurate representation of floodplain topography. These remotely sensed data are often not available or expensive, especially in developing countries. As alternative, freely Digital Elevation Models (DEMs), such as the near-global Shuttle Radar Topography Mission (SRTM) data, have come into focus flood modellers. To what extent these low-resolution can be exploited for hydraulic still open research question. This benchmarking study investigated potentials and limitations SRTM set example Papaloapan River, Mexico. Furthermore effects vegetation signal removal from DEM Baugh et al. (2010) were tested. A reference model based a light detection ranging (LiDAR) was up with code LISFLOOD-FP run two events. Test models DEMs output extents compared to by applying measure fit. fit, which binary wet/dry maps both outputs, gave information how well test simulated giving percentage performance theoretically 0 100 %. SRTM-based could reproduce promising results previous studies. mostly underestimated commonly flooded areas almost exclusively made out main channel surface. One reasons this likely much steeper slope opposed LiDAR where water probably conducted faster though channel. Too high bank cells generally more pronounced elevation differences throughout whole another problem preventing simulations. Vegetation successful certain degree improving fit about 10 However, realistic shape due too big pixel sizes used canopy height set. Also, conditioned overestimated increasing removal, rendering some useless comparison, leaving domain accounted showed modeling approximation river accurately captured topography crucial outcome. has been shown potentially useful but should rather applied densely covered catchments.