Accuracy assessment of global digital elevation models in alpine area

Digital Elevation Models (DEMs) are the most important tool to memorize elevation data and are very useful for many environmental and territorial applications. Depending on the used source data and on the needed application, DEMs with different resolution and accuracies can be constructed. Four Global DEMs that cover almost the entire Earth surface: Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), Shuttle Radar Topography Mission (SRTM), Consortium for Spatial Information of the Consultative Group for International Agricultural Research (CGIAR-CSI) SRTM and Global Multi-resolution Terrain Elevation Data 2010 (GMTED) are downloaded and compared with a Local model HELI-DEM with better accuracy, that covers alpine area between Italy and Switzerland, and was considered as the “truth” in this work. The goal is to assess the accuracy of Global DEMs for alpine area. Software GRASS GISS 7.0 is used to carry out most of the work: resampling Global DEMs to a common grid, finding the differences between Global models and HELI-DEM and their statistics, as well as visually presenting the process. From obtained results it is seen that SRTM and GMTED accuracy matches with their nominal accuracy, whereas CGIAR-CSI SRTM and ASTER have poorer accuracy than the nominal one. Additional analysis is done in order to locate and extract the areas causing problems in ASTER model, and they turned out to be random biases. Finally, the distribution of the errors of Global models is examined with chi square test, and it was found that they are normally distributed.