You must have a javascript-enabled browser and javacript and stylesheets must be enabled to use some of the functions on this site.


Snow Mass Retrieval by means of SAR Interferometry

Helmut Rott(1) and Rolf Scheiber(2)

(1) University of Innsbruck, Innrain 52, A-6020 Innsbruck, Austria
(2) German Aerospace Centre (DLR), Microwaves and Radar Institute, Oberpfaffenhofen, Germany


The accumulated mass of snow (the snow water equivalent, SWE) is a key variable for hydrology of alpine and high latitude basins and for climate research. In complex terrain and in regions with strong winds SWE is characterized by high spatial variability, so that conventional point measurements are not suitable for describing the mass of snow in a drainage basin. Whereas satellite microwave radiometry enables large scale retrievals of SWE for low-topography terrain, no suitable techniques are presently available for SWE mapping in mountain areas. A method for retrieving SWE with high spatial resolution was proposed by T. Guneriussen (IEEE TGRS, Oct. 2001), based on the interferometric phase shift in repeat pass SAR images which is proportional to the mass of snow deposited in this time interval. In the frame of an ESA study, aimed at new possibilities to retrieve bio- and geophysical variables by means of InSAR, we investigate the feasibility for applying this method to SWE mapping in alpine regions. Backscatter model calculations show that at L-and C-band the signal of ground covered by dry snow is clearly dominated by backscatter at the snow/ground interface, and the decorrelation in the snow volume is very small. However, changes of the snowpack depth due to snow fall, wind erosion and wind deposition may cause rapid decorrelation. We applied a theoretical model to estimate the temporal decorrelation due to snow accumulation or erosion, assuming a Gaussian probability distribution for changes in the propagation path through the snowpack at sub-pixel scale. Two scales of snow surface roughness, affecting the path length in the snowpack, were considered. The simulations show that at C-band comparatively moderate accumulation may cause complete decorrelation, versus at L-band the phase is much better preserved. This is confirmed by the analysis of ERS InSAR data over different time spans. In case of dry snow fall 35-day repeat pass data over the Alps decorrelate completely because the snowpack changes also if not melting. One-day repeat pass data show significant decorrelation only in case of strong snowfall or wind redistribution. In order to study the phase shifts due to snow fall we analyzed ERS-1 3-day repeat pass data from a cold winter period in the Austrian Alps, when temporal changes of the atmospheric phase delay were insignificant. Differential processing was applied to subtract the topographic phase. Preliminary results indicate that the residual phase is related to the spatial pattern of snowfall within the interferometric time span. Investigations are going on with a longer time series to assess the suitability of the method for various snowfall events.


Full paper

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry