Feasibility of snow water equivalent retrieval by means of interferometric ALOS PALSAR data

Thomas Nagler(1), Helmut Rott(1), Markus Heidinger(1) and Florian Müller(1)

(1) ENVEO, Technikerstrasse 21a, 6020 Innsbruck, Austria


Snow water equivalent (the mass of snow) is a key variable for hydrology of mountain basins, avalanche hazard assessment, and climate research. In mountainous terrain snow water equivalent (SWE) shows high spatial variability, so that remote methods are needed to obtain accurate, spatially detailed maps of snow accumulation. In the L- to C-Band range dry snowpack is highly transparent and backscattering from the snow-ground interface below the snow pack dominates. The phase delay of the radar signal in snow depends on the mass of snow and the propagation path length. Therefore phase changes in time series of coherent SAR images can be used to detect the accumulation of dry snow. In the frame of the EC project GALAHAD we investigated the feasibility for applying this method to map snow accumulation using time series of spaceborne L-Band repeat pass SAR data. The data set for this study consists of a time series of repeat pass ALOS PALSAR images acquired in Full Polarimetric mode over the Eastern Alps of Austria in the period September 2006 to January 2008. The investigation area includes the Inn valley at about 550 m elevation towards the main ridge of the Alps, above 3700 m elevation. Several automatic meteorological stations operated by the regional avalanche warning service are located at different elevations and provide continuous snow information. Though in the L- to C-band range the backscatter contribution of the snow pack itself is small, the interferometric signal is affected by decorrelation due to snow fall, wind erosion and wind deposition even if the snowpack is dry. The main reasons are differential phase shifts within the SAR resolution element. A theoretical model is applied to estimate the temporal decorrelation at L-Band due to changes in the snow mass (accumulation or erosion). The prototype procedure for generating maps of snow water equivalent from repeat pass SAR data includes iterative coregistration with sub-pixel accuracy, generation of coherence and phase images, and topographic phase flattening and local estimation of SWE. For our study we used a high precision digitale elvation model for removing the topographic phase contribution. Examples of coherence and SWE maps of the test area derived from ALOS PALSAR data will be shown. The SWE maps are locally verified with in-situ snow data measured at stations. The capabilities and limitations of this method using ALOS PALSAR will be discussed.


Symposium presentation


  Higher level                 Last modified: 07.05.06