PALSAR sensitivity to surface soil moisture in the Volta basin

Daniel Sabel(1), Bernhard Bauer-Marschallinger(1), Annett Bartsch(1) and Wolfgang Wagner(1)

(1) Vienna University of Technology, Gusshausstrasse 27-29, 1040 Vienna, Austria


A method for retrieving 1 km surface soil moisture information from the C-band ENVISAT ASAR Global Monitoring Mode (GM) has been developed at Vienna University of Technology. The transfer of this method to the L-band PALSAR is of high interest. The sensitivity of PALSAR Fine Beam (FB) mode to surface soil moisture over the Volta region was therefore analysed and compared with results from ASAR GM. The regions of Ejura and Tamale within the Volta Basin were chosen as study sites due to their climatic characteristics with distinct dry and wet seasons. Both sites are covered more than 90% with savanna type land cover, with a higher vegetation density in Ejura than in Tamale. Two PALSAR FB images were selected for each of the sites, with acquisition dates during the peaks of dry and wet seasons. It is assumed that the image pairs were representative of very dry and very wet surface conditions respectively. Assuming that the temporal backscatter dynamics are dominated by the variations in soil moisture, the difference between the image pairs gives an indicator of the site-specific sensitivity to soil moisture. The sensitivity for ASAR GM was generated using time series analysis based on approximately 200 GM images per study site, acquired between December 2004 and April 2008. A land cover analysis of the sensitivity for both sensors was carried out. The mean sensitivity was higher in Tamale than Ejura with a difference for C-band of about 2.5 dB. Preliminary results shows that the PALSAR FB has slightly higher sensitivity than the ASAR GM, indicating a high potential for soil moisture retrieval.



  Higher level                 Last modified: 07.05.06