A sensitivity analysis of soil moisture retrieval from the τ-ω microwave emission model

The upward microwave emission of soil depends upon the dielectric constant of the top layer of soil, a few centimetres in depth; this in turn is strongly affected by its moisture content. The microwave emission of soil can therefore be used to estimate its moisture content, however when obscured by vegetation account must be taken of its effect. The tau-omega model [(Mo.T. Choudhury, et al., 1982), (Brufeldt, D.R., et al., 1984)] is the predominant model for retrieving the volumetric moisture content of vegetated soil from passive microwave data. It estimates the emitted microwave brightness temperature depending on a number of variables such as look-angle from nadir, soil moisture content, vegetation optical depth, vegetation and soil temperature, the single-scattering albedo of the vegetation and the soil roughness. To retrieve soil moisture from the observations requires some knowledge of these other variables, either gathered from multiangle microwave measurements, or an independent source. We test the theoretical capability of satellite-based single and multiple-angle passive microwave systems to retrieve soil moisture, given limits on the knowledge of these variables, and the effects of measurement noise. We show the strong dependence of a single-angle system on independent knowledge of surface temperature, and the low value of independent measurements of vegetation optical depth