Controls on ERS altimeter measurements over ice sheets: Footprint-scale topography, backscatter fluctuations, and the dependence of microwave penetration depth on satellite orientation

• Main Authors: Arthern, RJ, Wingham, DJ, Ridout, AL
• Format: Article in Journal/Newspaper
• Language: English
• Published: AMER GEOPHYSICAL UNION 2001
• Subjects: Radar altimetry; Surface-properties; Elevation change; Height; Model; Snow; Echo; Antarctic; The Antarctic; Antarc*; Ice Sheet
• Online Access: https://discovery.ucl.ac.uk/id/eprint/75785/1/2001JD000498.pdf; https://discovery.ucl.ac.uk/id/eprint/75785/

We consider the reliability of radar altimeter measurements of ice sheet elevation and snowpack properties in the presence of surface undulations. We demonstrate that over ice sheets the common practice of averaging echoes by aligning the first return from the surface at the origin can result in a redistribution of power to later times in the average echo, mimicking the effects of microwave penetration into the snowpack. Algorithms that assume the topography affects the radar echo shape in the same way that waves affect altimeter echoes over the ocean will therefore lead to biased estimates of elevation. This assumption will also cause errors in the retrieval of echo-shape parameters intended to quantify the penetration of the microwave pulse into the snowpack. Using numerical simulations, we estimate the errors in retrievals of extinction coefficient, surface backscatter, and volume backscatter for various undulating topographies. In the flatter portions of the Antarctic plateau, useful estimates of these parameters may be recovered by averaging altimeter echoes recorded by the European Remote Sensing satellite (ERS-1). By numerical deconvolution of the average echoes we resolve the depths in the snowpack at which temporal changes and satellite travel-direction effects occur, both of which have the potential to corrupt measurements of ice sheet elevation change. The temporal changes are isolated in the surface-backscatter cross section, while directional effects are confined to the extinction coefficient and are stable from year to year. This allows the removal of the directional effect from measurement of ice-sheet elevation change.