Abstract:
Satellite radar altimetry is the most important data source for ice sheet elevation modeling but it is well established that the accuracy of such data from satellite borne radar altimeters degrade seriously with increasing surface slope and level of roughness. A significant fraction of the slope-correlated noise can be effectively removed by the so-called relocation error correction method. The adjustment, however, produces a different spatial sampling of the data, which introduces a non-negligible slope related bias to the computation of digital elevation models. In this paper we incorporate high-precision airborne laser profiling data from the so-called Arctic Ice Mapping project as a tool to determine that bias and to calibrate the satellite altimetry. This is achieved by a simple statistical analysis of the airborne laser profiles, which defines the mean amplitude of the local surface undulations as a linear function of surface slope. This linear correspondence is in turn tested as a model for adjusting the satellite altimetry data for the observed slope correlated bias. The adjustment is shown to have a significant effect in terms of reducing the bias, thus improving the modeling accuracy of the data.
