Recent advances in remote sensing have shown that soil moisture can be measured by a variety of techniques. However, only microwave technology has demonstrated a quantitative ability to measure soil moisture under a variety of topographic and vegetation cover conditions so that it could be extended to routine measurements from a satellite system [ Engman, 1990].
The major factor inhibiting wide spread use of remotely sensed soil moisture data in hydrology is the lack of data sets and optimal satellite systems. For the most part, scientists have been restricted to data from short duration aircraft campaigns, or analysis of the SMMR and SSM/I passive microwave satellites. Although the available passive systems do not have the optimum wave lengths for soil moisture research has demonstrated that in areas of sparse vegetation that a valuable estimate can be obtained [ Owe et al, 1988]. Historical data from the SSMR passive microwave system is more valuable than the SSM/I data because it had a C-band radiometer which is a better instrument for soil moisture [ Choudhury and Goulus, 1988, Owe et al. 1992]; however, its period of record is limited to 1982 to 1987. In both cases the footprint is rather large, varying from about 25 km for the SSM/I to about 150 km for the C-band SMMR. Experimental passive microwave systems using aperture synthesis such as the Electronically Steered Thinned Array Radiometer (ESTAR) offer hope for higher resolution satellite systems. The airborne ESTAR has been demonstrated in the Walnut Gulch watershed in Arizona [ Jackson et al., 1993].
The SAR systems offer perhaps the best opportunity to measure soil moisture routinely over the next few years. Currently, the European Remote Sensing (ERS-1) C-band and Japanese Earth Resources Satellite (JERS-1) L-band SARs are operating and the Canadian RADARSAT (also C-band) will be launched in 1995. Although it is believed that an L-band system would be optimum for soil moisture, the preliminary results from the ERS-1 demonstrate its capability as a soil moisture instrument. Change detection techniques have been used to detect changes in soil moisture in a basin in Alaska [ Villasenor et al., 1993] However, Merot et al., [1994] have shown that radar data become ambiguous when ponding in variable source areas. One main drawback to the existing SAR systems is that there are no existing algorithms for the routine determination of soil moisture from single frequency, single polarization radars. Oh et al., [1992] have developed a semi-empirical algorithm but it needs multi polarization data. A second limitation comes from their long period between repeat passes; for the most part 35 to 46 days although the RADARSAT should have three day capability for much of the globe in a SCANSAR (wide swath, 500 km) mode.
Soil moisture has many possible applications in hydrology, but the primary areas are in evaporation and runoff modeling. Current applications of remotely sensed soil moisture are discussed in each of these following sections.