Essential Climate Variables for the Ice Sheets from Space and Airborne measurements

The Greenland Ice Sheet is the largest ice mass in the northern hemisphere.Over the past decade, it has undergone substantial changes in e.g. mass balance,surface velocity, and ice thickness. The latter is reflected by surfaceelevation changes, which are detectable with altimetry. Therefore, this studyexploits the advantages of radar and laser altimetry to analyze surface elevationchanges and build a Digital Elevation Model of the ice sheet. Selected advantagesare radar data’s continuity in time and laser data’s higher horizontal andvertical accuracy. Therefore, ESA Envisat and CryoSat-2 radar altimetry dataare used in conjunction with laser data from NASA’s ICESat and airborneATM and LVIS instruments, and from ESA’s airborne CryoVEx campaign.The study is part of the ESA Ice Sheets CCI project. With the release ofREAPER data, one goal is to use the more than two decades of ESA radaraltimetry to develop a long-term surface elevation change product from 1992 topresent. The optimal method is found by comparing ten different solutions submittedby the scientific community across the choice of altimeter and method: A combination of repeat-tracks and cross-overs. The former produces estimatesalong repeat ground-tracks while the latter exploits intersecting ground-tracks.The combination increases the spatial data coverage and reduces topographicerrors. Two results based on Envisat data are presented here: The first repeat track solution (2002 – 2010) of the Greenland Ice Sheet and a merged repeat track and cross-over result from 2006 – 2010.A 2 × 2 km Digital Elevation Model is built from combined radar and laserdata. It is applicable for elevation change detection and correction of topographicerrors. Current models have limitations as they are based on shortobservation periods from one sensor, limiting the spatial data coverage, ormultiple years of data from various sensors, inheriting errors from intermediateelevation changes. The model here consists of Envisat and CryoSat-2 datafrom 2010 merged with ICESat, ATM, and LVIS data. Vertical radar errors are corrected with laser data. Thus, the Digital Elevation Model is referencedto a specific epoch in time and exploits the high spatial coverage of input data. An important finding in the study is disagreeing relocations of radar data dependingon the method. Validation shows the preferred method to be the Pointof Closest Approach with an a-priori Digital Elevation Model to extract thesurface topography. The preferred spatial resolution of the model is 2 × 2 kmfor Envisat and CryoSat-2 LRM data near Jakobshavn Isbræ, i.e. over regionswith both steep and smooth topography.