Abstract
Currently available Digital Elevation Models(DEMs) of the Greenland Ice Sheet (GrIS) were originally derived from radar altimetry data, e.g. Bamber (Bamber et al., 2001) and later improved by photoclinometry to fill the regions between orbits (Scambos and Haran, 2002). The elevation error of these DEMs is a few meters in the higher part (above 2000 m) of the ice sheet, but it can be as much as 50-100 meters in marginal regions. The relatively low resolution and accuracy poses a problem, especially for ice sheet modeling. Although accurate elevation data have been collected by airborne and spaceborne laser altimetry (airborne: Airborne Topographic Mapper (ATM) (1993-present), Laser Vegetation Imaging Sensor(LVIS) (2007,2009 and 2011); spaceborne: Ice, Cloud, and land Elevation Satellite (ICESat) (2003-2009)) and DEMs have been derived from stereo satellite imagery (e.g., SPOT (40 m), ASTER (15 m)), a high resolution, consistent DEM of GrIS is not yet available. This is due to various problems, such as different error sources in the data and different dates of data acquisition.
In order to overcome these difficulties, we generated a multi-resolution DEM of GrIS, reflecting June 2008 conditions, by fusing a photoclinometry DEM, SPOT and ASTER DEMs as well as elevations from ICESat, ATM and LVIS laser altimetry. The new multi-resolution DEM has a resolution of 40 m x 40 m in the marginal ice sheet regions and 250 m elsewhere. The ice sheet margin is mapped from SPOT and Landsat imagery and SPOT DEMs are used to cover the complex topography of ice sheet marginal regions. The accuracy of SPOT DEMs is approximately $\pm 6$ m except in the areas covered by clouds regions, where the SPOT elevations were replaced by ASTER DEMs. The ASTER DEMs were checked and improved by the DEM derived from aerial photography from the 1980s.
A new photoclinometry DEM, derived from Advanced Very High Resolution Radiometer (AVHRR) and Moderate Resolution Imaging Spectroradiometer (MODIS) imagery provide a 250-meter resolution base topography at the higher elevations. ICESat and ATM and LVIS laser altimetry observations are used to improve the absolute elevation accuracy by computing the difference between the base DEM and point elevations, followed by ‘distributing’ the differences over the DEM using kriging interpolation. Adding these interpolated differences to the elevations at the grid posts results in an updated DEM. Finally, all elevations were corrected using elevation changes determined by SERAC (Surface Elevation Reconstruction And Change detection), to achieve a common reference date. Airborne laser altimetry elevations are used to evaluate the accuracy of the new GrIS DEM.
In order to overcome these difficulties, we generated a multi-resolution DEM of GrIS, reflecting June 2008 conditions, by fusing a photoclinometry DEM, SPOT and ASTER DEMs as well as elevations from ICESat, ATM and LVIS laser altimetry. The new multi-resolution DEM has a resolution of 40 m x 40 m in the marginal ice sheet regions and 250 m elsewhere. The ice sheet margin is mapped from SPOT and Landsat imagery and SPOT DEMs are used to cover the complex topography of ice sheet marginal regions. The accuracy of SPOT DEMs is approximately $\pm 6$ m except in the areas covered by clouds regions, where the SPOT elevations were replaced by ASTER DEMs. The ASTER DEMs were checked and improved by the DEM derived from aerial photography from the 1980s.
A new photoclinometry DEM, derived from Advanced Very High Resolution Radiometer (AVHRR) and Moderate Resolution Imaging Spectroradiometer (MODIS) imagery provide a 250-meter resolution base topography at the higher elevations. ICESat and ATM and LVIS laser altimetry observations are used to improve the absolute elevation accuracy by computing the difference between the base DEM and point elevations, followed by ‘distributing’ the differences over the DEM using kriging interpolation. Adding these interpolated differences to the elevations at the grid posts results in an updated DEM. Finally, all elevations were corrected using elevation changes determined by SERAC (Surface Elevation Reconstruction And Change detection), to achieve a common reference date. Airborne laser altimetry elevations are used to evaluate the accuracy of the new GrIS DEM.
| Original language | English |
|---|---|
| Publication date | Nov 2011 |
| Number of pages | 1 |
| Publication status | Published - Nov 2011 |
| Event | American Geophysical Union Fall Meeting 2011 - San Francisco, United States Duration: 5 Dec 2011 → 9 Dec 2011 |
Conference
| Conference | American Geophysical Union Fall Meeting 2011 |
|---|---|
| Country/Territory | United States |
| City | San Francisco |
| Period | 05/12/2011 → 09/12/2011 |