Effect of Data Resolution on Topographic Index and Performance of the Simi-Distributed Model: TOPMODEL

Digital Elevation Models (DEMs) are one of the most important inputs in most rainfall -Runoff models and also in deriving watersheds geomorphological characteristics. One of the most important issues that should be taken into account when using DEMs in rainfall runoff modeling is the effects of DEM source on the results of the models. At present time, radar based DEMs are attracting a considerable attention in many earth related fields. Currently, there are several sources such as: Shuttle Radar Topography Mission (SRTM) data and the advanced space thermal emission radiometer (ASTER) that due to ease of access and free of charge, have an important role in hydrological modeling and the extraction of geomorphological parameters of catchments. This paper addressed the effects of different DEM sources: ASTER, SRTM, and 1:50000 topographic based DEM on the topographic index and performance of a semi- distributed model, called TOPMODEL. The obtained results indicated that by increasing DEM cell size, disregarding its source, the topographic index inceased. Also, it was illustrated that for all cell sizes other that 30 m there was no difference between the results of ASTER and SRTM DEM. Such a coincidence was observed for TOPO DEM for cell sizes larger than 100 m, as well. Results showed that the effect of data resolution on the average daily discharge, average saturation deficit of basin and the average distance to water table level is not meaningful. In daily discharge simulation, the performance of TOPMODEL when using the TOPO DEM ,based on the NASH- SUTCLIFFE efficiency index, was the highest and for the SRTM DEM was the lowest. In flood simulation, upto 200 m resolution, TOPMODEL efficiency was constant disregarding the DEM source, while its efficiency reduced for cell sizes greater than 200 m in all conditions. It was illustrated that by increasing the cell size, surface runoff contribution to total runoff, in contrast to subsurface flow, increased. Furthermore, For all of the sources, increasing the DEM cell size, incresesd the surface runoff contribution and decreased subsurface flow. The percent of surface flow simulated by the model when implementing the DEM of SRTM is the highest. According to achieved results, the percentage of simulated surface runoff from SRTM DEM was higher than the other two sources. It was observed that the number of cells of low slope in SRTM DEM is greater than the other sources, which was concluded as the main reason for this issue. Reducing the slope of a cell could reduce its potential to carry subsurface flow, which in turn could increase the cell potential to be saturated. It was concluded that DEM different sources effects on average saturation deficit and water table is negligible, and these ouputs of TOPMODEL is mainly affected by DEMs cell size. Finally, this important conclusion was drawn from the present research that the information content of ASTER DEM is nearly similar to TOPO DEM and higher than SRTM DEM; therefore, between satellite based DEMs, it is more convenient to use ASTER DEM for rainfall runoff modeling.