Description of Soil Moisture Retrieval Algorithm for ADEOS II AMSR

To a large degree, the research in microwave remote sensing of soil moisture has focused on the forward modeling problem. This is the process of predicting the brightness temperature from soil properties using radiative transfer theory. When measuring soil moisture we must be concerned with inversion of this model. This is more difficult than forward modeling. There are five steps involved in extracting soil moisture using passive microwave remote sensing. These are; normalizing brightness temperature to emissivity, removing the effects of vegetation, accounting for the effects of soil surface roughness, relating the emissivity measurement to soil dielectric properties, and finally relating the dielectric properties to soil moisture. In our approach, soil moisture retrieval is based upon an algorithm developed by Jackson (1993). Brightness temperature for a single AMSR channel (6.9 GHz H) is converted to emissivity using a surrogate for the physical temperature of the emitting layer. This emissivity is corrected for vegetation and surface roughness to obtain the soil emissivity. The Fresnel equation is then used to determine the dielectric constant. Finally, a dielectric mixing model is used to obtain the soil moisture. The theory describing this follows. Fundamental basis for a smooth bare soil. The measurement provided is the brightness temperature, TB, that includes contributions from the atmosphere, reflected sky radiation, and the land surface. Atmospheric contributions are negligible at frequencies <6 GHz. Galactic and cosmic radiation contribute to sky radiation and have a known value that varies very little in the frequency range used for soil water content observations (Tsky ~4 K). The brightness temperature of a surface is equal to its emissivity (e) multiplied by its physical temperature (T). Based upon the above, the equation for TB is skyB TeeTT]1 [-+ = (1) The second term of equation 1 will be on the order of 2 K and will be dropped for computational purposes. For inversion equation 1 is rearranged as follow