You must have a javascript-enabled browser and javacript and stylesheets must be enabled to use some of the functions on this site.
 
 

 

Quality assessment of X- and C-SRTM with ERS-tandem DEMs over 4 European CEOS WGCV test sites.

Jan-Peter Muller(1) and Dietmar Backes(1)

(1) University College London, Gower Street, London WC1E 6BT, United Kingdom

Abstract

The CEOS working Group on Calibration/Validation (WGCV) Sub-group on "Terrain mapping from satellites" (of which the first author is Chair) recently selected 4 test sites in Europe for a comparative assessment of different DEM extraction techniques.

Results will be presented on the quality assessment of X- and C-SRTM for these sites in the UK (North Wales), Spain (Barcelona), France (Aix-en- Provence) and Germany (Bavaria). Independent DTMs obtained from national mapping sources or via the SPOT-5 HRS assessment campaign are used for validation and LANDSAT-5 or -7 data used for the interpretation of height differences. For the UK, ERS-tandem DEMs from the LANDMAP project (Muller et al., 1999) are also assessed and kinematic GPS data collected as part of LANDMAP employed for assessing all 3 DEMs.

Results indicate that for all 4 test sites, both X- and C-SRTM DEMs are not co- registered with the available "ground truth" DEMs and not co-registered with each other. For one site, a simple translation was required (UK) whilst the other 3 sites required that a rubber-sheet warping using a 1st order affine transformation was required. Most height differences remaining are due to forest cover with a small effect in very rugged areas due to SAR look angle in steep sided valleys. The SRTM DEMs in the UK have an overall mean difference of around 1m and a standard deviation in the range 4-6m. The SRTM results are poorer for the other 3 European continental sites and an analysis of the reasons for this are discussed. For LANDMAP in rugged areas the standard deviation is affected by atmospheric effects resulting in values around 20m whilst in flat regions the standard deviation is around 1-3m.

Preliminary results will be shown on the detection of outliers in these DEMs using novel statistical methods for outlier detection and an early indication shown on how these could be used to plug gaps and replace outliers in the fusion of different DEM sources.

Acknowledgements X-SRTM DEM data were kindly provided by DLR under Principal Investigator 121. C-SRTM DEM data was kindly provided by JPL. DTMs of Barcelona and Bavaria were kindly provided by Dr. Wolfgang Kornus (ICC) and Prof. Dr. Ing. Manfred Schroeder (DLR-IMF) respectively as part of the SPOT-HRS study.

Reference cited Muller, J.-P., J.G. Morley, A. Walker, J. Barnes, P.A. Cross, I.J. Dowman, K. Mitchell, A. Smith, K. Chugani, and K. Kitmitto, The LANDMAP project for the creation of multi-sensor geocoded and topographic map products for the British Isles based on ERS-tandem interferometry., in Proc. Second International Workshop on ERS SAR Interferometry on "'Advancing ERS SAR Interferometry from Applications towards Operations'", pp. 9pp, ESA, 10 – 12 November 1999, Palais des Congrès, Liège, Belgium, 1999

 

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry