Soil Moisture Mapping using ALOS PALSAR Quad-Pol Data

Y S Rao(1), Gulab Singh(1) and G Venkataraman(1)

(1) CSRE, Indian Institute of Technology Bombay, IIT Powai, Mumbai, 400076, India


Several images from ALOS PALSAR in Quad-pol mode acquired over India were processed for soil moisture mapping. Oh et al. model was used with Polsarpro software for the retrieval of soil moisture. Incidence angle map was derived using the polynomial given in the leader file. Boxcar or Lee filters were applied to the data before soil moisture retrieval. Two data sets acquired on March 7 and April 22, 2007 with an interval of 45 days over the same test site clearly show that dry-down nature of soil moisture from March two April. In April scene, the soil moisture was very low. Other data sets (around 5 images) acquired over different test sites show different pattern depending on the agriculture practices. The data acquired in October 2006 show high soil moisture compared to the data acquired in March and May 2007. Unfortunately, we do not have ground truth data synchronous with ALOS passes. We have visited some of the test sites after a year and noted the agriculture practices during the time of ALOS acquisition. It is interesting to note that vegetation and urban areas are clearly discriminated in the PauliRGB images. The retrieval of soil moisture using X-Bragg model may be compared with the retrieval of Oh et al. model as Dubois et al. model does not valid for low incidence angles.



  Higher level                 Last modified: 07.05.06