Accuracy validation and comparison of global digital elevation models over Croatia

Abstract

Accuracy assessment of digital elevation models (DEMs) plays an important role in facing their use in geoscience applications. This study investigates the vertical accuracy of most recently published versions of global DEMs over Croatia: 1 arc-minute global relief model (ETOPO1), Global 30 Arc-Second Elevation (GTOPO30), SRTM30+, Global Multi-resolution Terrain Elevation Data 2010 (GMTED2010), Altimeter Corrected Elevations 2 DEM (ACE2 DEM), Shuttle Radar Topography Mission GDEM (SRTM GDEM), and Advanced Spaceborne Thermal Emission and Reflection GDEM (ASTER GDEM). Geodetic ground control points (benchmarks) allowed the comparison in terms of vertical accuracy. Additionally, the differences between models in the identical points were determined and analysed after adjusting the models to the same resolution. The results have yielded information about the overall accuracy of models; the accuracy depending on height, land cover, and slope; presence of large and systematic errors; and mutual agreement between the models.Overall vertical accuracies according to the root mean square error over Croatia are: ETOPO1 27.6 m, GTOPO30 21.6 m, SRTM30+ 21.3 m, GMTED2010 7.4 m, ACE2 DEM 4.5 m, SRTM GDEM 3.8 m, and ASTER GDEM 7.1 m. The highest accuracy was shown by SRTM GDEM version 3, which is far better than the previously released versions. All models have proved to be worse in the rough and forest areas, whereas in flat and barelands, new-generation global DEMs SRTM GDEM, and ASTER GDEM are highly accurate. IntroductionIn the past decade, global digital elevation models (global DEMs) have become invaluable sources of information in geosciences (e.g. Raaflaub and Collins 2006;Toutin 2008). Global DEMs have been significantly improved in comparison with those first released in the mid1990s, Global 30 Arc-Second Elevation (GTOPO30) and 1 arc-minute global relief model (ETOPO1), with the new-generation model releases, the Shuttle Radar Topography Mission GDEM (SRTM GDEM) in 2007, with a resolution of 3 arc seconds, and the Advanced Spaceborne Thermal Emission and Reflection GDEM (ASTER GDEM) in 2009, with a resolution of 1 arc second. As all models deviate from reality, global DEMs are usually used without an estimation of their accuracy and reliability and without considering errors that can affect results (Wechsler 2003). The goals of this work were to quantify and compare vertical accuracy, assess dependence between vertical accuracy and topographic characteristics (land cover and slope), and assess the differences between all global DEMs that were published from the mid-90s over Croatia. Testing of global DEMs was done for most recent versions of

Improving the positional accuracy of drainage networks extracted from Global Digital Elevation Models using OpenstreetMap data

“…Thus, the accuracy assessment of the GDEMs is an issue well covered in literature. Several studies have been made to assess the accuracy of the GDEMs in different regions of the world (e.g., Cook et al, 2012;Eckert et al, 2005;Mukherjee et al, 2013;Varga and Bašić, 2013;Weydahl et al, 2007;Yadav and Indu, 2016) and also the accuracy of the topographic parameters extracted from them (Ebenezer, 2015;Lin et al, 2005;Liu et al, 2015;Monteiro et al, 2015;Mukherjee et al, 2013;O'Callaghan and Mark, 1984;Tarboton et al, 1991). These studies enable to conclude that the vertical accuracy of DEMs is affected by morphologic terrain characteristics (Holmes et al, 2000), by the land use and land cover, but also by the techniques used for data acquisition and processing.…”

Models of geomorphometric parameters and drainage networks for the territory of Republic of Croatia

“…Another aim is to study each step of the development process in order to get insights into the influence of all input variables and models on the resulting models. The basis of the developed models are digital elevation models, which provide basic quantitative information of the topographic relief (Varga and Bašić, 2013).…”

“…The major landcover types, calculated from the Advanced Very High-Resolution Radiometer Global Land Cover Model (AVHRR GLCM) were forests (about 30%), scrublands and bushlands (about 30%), croplands (about 25%), and grasslands (about 10%). The defined study area has an elevation range of about 1700 m, with an average slope of approximately 6° (Varga and Bašić, 2015).…”

Accuracy Assessment of Alos W3d30, Aster Gdem and Srtm30 Dem: A Case Study of Nigeria, West Africa

“…It is true that DEMs have become very useful sources of data for a range of applications in Earth and environmental sciences [16] but despite their usefulness, there are many sources of errors inherent in them [17]. Owing to this fact, and as a result of recent improvement and release of newer versions of the Advanced Land Observing Satellite World 3D Digital Surface Model version 2.1 (ALOS W3D30); the Shuttle Radar Topography Mission Digital Elevation Model version 3.0 (SRTM30) and the Advanced Spaceborne Thermal Emission and Reflection Radiometer Global DEM version 2.0 (ASTER GDEM2), it is very important to assess and compare the quality of these data in order to see how well the DEMs fit the locally available or acquired data.…”