nerc.ac.uk

The accuracy of ASTER digital elevation models : a comparison with NEXTMap Britain

Hall, M.; Tragheim, D.G.. 2010 The accuracy of ASTER digital elevation models : a comparison with NEXTMap Britain. In: Fleming, C.; Marsh, S.H.; Giles, J.R.A., (eds.) Elevation models for geoscience. Geological Society of London, 43-53. (Geological Society Special Publications, 345).

Before downloading, please read NORA policies.
[img] Text
ASTER_mhall_&_dtragheim_paper_v5.pdf

Download (2465Kb)

Abstract/Summary

In many overseas geological surveying projects an accurate elevation model is often required for analysis, image orthorectification, navigation and the generation of contours. Acquiring an accurate elevation model can be a difficult and expensive task. One possible solution is to generate a digital elevation model (DEM) from Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) satellite imagery. However, to fully understand the potential of ASTER DEMs the accuracy of these models needs to be established. The DEM was created using the Sulsoft ASTER DTM add-on ENVI module. NEXTMap provides an ideal reference dataset for comparison. In this study the accuracy of an ASTER generated DEM was assessed for a 50×50 km area in central Wales. A total of 2.4 million points were compared. Visual and statistical assessments were made, including profile and contour comparisons, allowing the spatial variation in accuracy to be explored. A mean vertical difference of –0.98 m and a standard deviation of c. 9 m were calculated. This suggests that 95% of the ASTER DEM points are within ±20 m of the NEXTMap DEM. Considering these accuracy levels, contours from ASTER can be generated at 40 m intervals.

Item Type: Publication - Book Section
Digital Object Identifier (DOI): 10.1144/​SP345.6
Programmes: BGS Programmes 2010 > Geoscience Technologies
ISBN: 9781862393134
Date made live: 12 May 2011 10:28
URI: http://nora.nerc.ac.uk/id/eprint/14218

Actions (login required)

View Item View Item

Document Downloads

More statistics for this item...