Evaluation of the Digital Elevation Model from the Shuttle Radar Topography Mission (SRTM) on the Papaloapan Macro-Basin, Mexico, using LiDAR as benchmark

Abstract

The Digital Elevation Model (DEM) from the Shuttle Radar Topography Mission (SRTM) is evaluated using a LiDAR DEM from INEGI as benchmark in the Papaloapan Basin (~58,000km2) in Mexico. Three representative regions are selected: 1) a hilly region with strong slopes and elevations over 3,000m, 2) a transitional region with relatively smoother slopes and elevations around 1,000m, and 3) a floodplain with flat terrain and elevations below 100m. The straight comparison of both datasets shows very similar elevation values at the hilly and transitional regions. However, in the floodplain, the relationship has a parabolic shape, and errors are relatively higher, in terms of the elevation range. This is probably due to systematic errors in SRTM being very close to the actual low elevations. Maps of errors suggest strong association with water bodies and the aspect. For example, in the transitional region, most negative errors are found on slopes facing east, while positive errors are found on slopes facing west. Three-dimensional histograms of errors vs. topographic features (elevation, slope, and aspect) are estimated. The histograms suggest a systematic error, which means SRTM could be improved with a simple calibration at least in these cases. Evaluations of public DEMs from different sources in Mexico are considered necessary for the identification of their strengths and weaknesses. We believe these evaluations might provide the grounds for the creation of improved MEDs in the future either by either a simple calibration or through composite MEDs from multiple sources.