Spaceborne scatterometers for change detection over land

Studies of the Earth's land surface involving scatterometers are becoming an increasingly important application field of microwave remote sensing. In this respect, the global and real time access to geophysical parameters such as soil moisture or biomass proves to be necessary and often critical in many hydrological and climatological applications. A typical example of geophysical parameter retrieval using spaceborne scatterometers is the recent method of deriving soil moisture developed at the Institute of Photogrammetry and Remote Sensing (I.P.F.) of Vienna University of Technology. The derived soil moisture time series are based on data from the C-band scatterometers onboard the European Remote Sensing Satellites ERS-1 and ERS-2. An interesting aspect of this work is the operational continuity the ers scatterometers have found in the Advanced Scatterometers (ASCAT) flown onboard the Metop series of satellites. In addition to the above instruments, the SeaWinds scatterometer onboard the QuikSCAT satellite offers an additional wavelength (Ku-band) and a 360o continuous azimuthal viewing direction range. Any attempt to use these instruments to retrieve (preferably global) geophysical parameters over land is confronted with a series of complications and pitfalls arising from the very special regime of backscatter behaviour for many land cover types and climates. The present dissertation has several objectives. Firstly, it describes the basics of radar remote sensing in general and scatterometers in particular, in terms of instrument design, functionality, data coverage, processing, calibration and validation, as well as the synergies between the ERS, ASCAT and SeaWinds instruments. Secondly, two of the soil moisture algorithm issues currently requiring improvements are addressed: the selection of a new discrete geodetic grid for accommodating ascat data in the soil moisture processor and the characterisation and correction of azimuthal effects observed in scatterometer backscatter, in order to improve soil moisture retrieval in certain problematic regions. The third objective consists in presenting the soil moisture retrieval method itself after incorporating the above changes, as well as a new processor that has been developed for NRT (near real-time) ASCAT processing. Also, some first results with ascat soil moisture are presented. The fourth goal is to assess the absolute and relative calibration of the two instrument generations (ERS and ASCAT), very important for establishing long-term, uniform and consistent datasets of backscattering coefficients and soil moisture. (author)