Decadal region-wide and glacier-wide mass balances derived from multi-temporal aster satellite digital elevation models. Validation over the mont-blanc area

Abstract

Since 2000, a vast archive of stereo-images has been built by the Advanced Spaceborne Thermal Emission and Reflection (ASTER) satellite. Several studies already extracted glacier mass balances from multi-temporal ASTER digital elevation models (DEMs) but they lacked accurate independent data for validation. Here, we apply a linear regression to a time series of 3D-coregistered ASTER DEMs to estimate the rate of surface elevation changes (dh/dtASTER) and geodetic mass balances of Mont-Blanc glaciers (155km2) between 2000 and 2014. Validation using field and spaceborne geodetic measurements reveals large errors at the individual pixel level (>1m a−1) and an accuracy of 0.2–0.3m a−1 for dh/dtASTER averaged over areas larger than 1km2. For all Mont-Blanc glaciers, the ASTER region-wide mass balance [–1.05 ± 0.37m water equivalent (w.e.) a−1 ] agrees remarkably with the one measured using Spot5 and Pléiades DEMs (–1.06 ± 0.23m w.e. a−1) over their common 2003–2012 period. This multi-temporal ASTER DEM strategy leads to smaller errors than the simple differencing of two ASTER DEMs. By extrapolating dh/dtASTER to mid-February 2000, we infer a mean penetration depth of about 9 ± 3m for the C-band Shuttle Radar Topographic Mission (SRTM) radar signal, with a strong altitudinal dependency (range 0–12m). This methodology thus reveals the regional pattern of glacier surface elevation changes and improves our knowledge of the penetration of the radar signal into snow and ice.