Assessment Of ASTER GDEM And SRTM Performance By Comparing With Survey Control Points And Icesat/GLAS Data

Topographic data plays a critical role in water resources modeling with determining watershed hydrologic characteristics from raster-based digital elevation models (DEM). Raster-based DEMs have been widely used to derive topographic attributes used in hydraulic and hydrologic modeling such as slope, stream network, basins boundary and area. Accurate models of floodplain topography are essential for having accurate output of hydrologic models. The Shuttle Radar Topography Mission (SRTM) provides near-global topographic coverage of the Earth's surface with unprecedented consistency and accuracy with the resolution of 1-3 arc sec. A new Global Digital Elevation Model (GDEM) from optical stereo data acquired by the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) was recently released with the resolution of 1 arc sec. This study the performance of the ASTER GDEM and SRTM DEMs is assessed by comparing with ground-based survey control points and point data from ICESat/GLAS (Ice, Cloud, and land Elevation Satellite/Geoscience Laser Altimeter System) for a 7x7degree area in Queensland Australia. This area contains large and remote river basins, where these datasets provide an invaluable resource from which river floodplain inundation can be measured and modelled. Our study focuses on both data and datum issues, which are required to provide a realistic assessment of the achievable relative and absolute ground topographic accuracies. We assessed ground-based survey control points' with ICESat satellite altimetry points and in turn assessed accuracy of the ASTER GDEM and SRTM DEMs over the study area. ICESat provides globally-distributed elevation data of high accuracy (2.47.3 m horizontal error and 0.040.13 m (per degree of incidence angle) vertical error)We applied the GLA14 elevation products (Land/Canopy elevations) for the Laser 3a observation period and Release 31 to compare with 5000 ground survey control points in study area.Our ICESat-control point, ICESat-DEM and DEM-DEM analysis shows that SRTM DEMs are more suited than ASTER GDEM data for large scale hydrologic studies in our basins, even with the lower spatial resolution. The ASTER GDEM shows inland water noises and other noises like straight lines which are caused by the methods that are used to produce them that affect replication of hydrological processes. We were also able to correct absolute elevation errors through datum adjustment by analyzing the difference of the SRTM DEM and ICESat elevations, with corrections made to the DEM for hydrological modeling applications.