EGU2016-6692.pdf

An inversion method retrieves soil moisture information from satellite altimetry observations, addressing the terrestrial water cycle and interactions between land and atmosphere. Satellite radar altimeters, such as Jason-1/2 and Envisat, provide high-resolution along-track information on backscatter, which correlates with soil moisture in upper layers, especially in arid and semi-arid regions. A new inversion framework is suggested, using principal component analysis (PCA) and least squares fitting to derive time-invariant spatial patterns and estimate time-variable soil moisture evolution from Jason-2 and Envisat data, tested over Australian arid and semi-arid regions. Preliminary results are compared to high-resolution soil moisture model data, discussing the potential to improve simulation skill in the Global Land Data Assimilation System (GLDAS) over Australia.