GLOBESCAT

Soil moisture is a key state variable of the global energy and water cycle. Due to the lack of affordable in-situ measurement techniques, soil moisture has long been a research focus in the field of remote sensing. Particularly, sensors operated in the lower part of the microwave spectrum (1-10 cm wavelength) hold a large potential due to their high sensitivity to soil moisture. However, the microwave signal is also affected by vegetation and surface roughness. Since appropriate satellites and retrieval methods have so far not been available, microwave remote sensing is not yet used for operational soil moisture applications. Only recently, a methodological breakthrough has been achieved by our team at the Institute of Photogrammetry and Remote Sensing of the Vienna University of Technology. We have retrieved the first global, multi-year soil moisture data set from ERS scatterometer measurements. It has been found to be of comparable quality as state-of- the-art models and reanalysis data sets. Validation with ground observations has shown that the retrieval error is about 0.05 m 3 m -3 . The usefulness of the data has been tested in a number of pilot applications (agronomy, hydrology, meteorology). These activities clearly demonstrated that the scientific algorithms have reached a level of maturity which makes scatterometer derived soil moisture products attractive for scientific and operational applications. Beside the methodological advances, it is also important to note that the successor instrument of the ERS scatterometer will be flown on an operational satellite platform: The Advanced Scatterometer will be part of the METOP satellite series which starts in 2005 and guarantees a flow of high quality data for the next 14 years. It is the goal of the GLOBESCAT project to demonstrate the operational utility of C-band scatterometers by 1) producing a global, long-term (1991-2007) soil moisture data set, and 2) establishing an automated Internet service to deliver global soil moisture products in near-real time.