Altimetry landed - Digital Elevation Data from ICESat, SRTM and Surveying
Georgia Fotopoulos(1) , Alexander Braun(2) , and Vidyavathay Renganathan(2)
University of Toronto,
35 St George Street,
(2) University of Calgary, 2500 University Dr NW, T2N1N4, Calgary, Canada
NASA's laser altimetry mission ICESat launched in 2003 has provided high-accuracy and high-resolution elevation data over all surface types including land. Due to its small footprint of about 70 meters, ICESat waveform processing resulted in accurate elevation retrieval over multiple land types (e.g. bare rock, vegetation, mountainous areas) and even over steep terrain. While such data is too sparse to derive a digital elevation model (DEM), its high accuracy allows for the calibration of independent DEMs. The Shuttle Radar Topography Mission (SRTM) has provided homogeneous and highly accurate data for Digital Elevation Models (DEM) in February 2000. The accuracy of the resulting DEM depends on various factors including the roughness and slope of the terrain, type of land cover and the determination of the antenna positions on-board the shuttle. Both C-band and X-Band DEMs show intrinsic errors of about 5-15 meters, limiting the accuracy of absolute height determination in numerous applications. A comparison and a potential calibration using more precise, but point-wise, height information is conducive to improve the SRTM derived DEMs. Survey control monuments (SCM) situated across Canada are used as an independent data source for comparison. The analysis focuses on both data and datum issues which must be addressed in order to provide a realistic comparison and assessment of the achievable accuracy. The discussion highlights the seasonal considerations that account for changes in vegetation and snow cover over time. Preliminary results suggest that laser altimetry provides an excellent means to calibrate less accurate DEMs (e.g. SRTM DEM) after the differences due to penetration depth have been accounted for. Utimately, ICESat can potentially be used to derive error maps of the published SRTM DEM worldwide.