|Holifield Collins, Chandra|
|Goodrich, David - Dave|
Submitted to: First Interagency Conference on Research in the Watersheds
Publication Type: Proceedings
Publication Acceptance Date: 8/20/2003
Publication Date: 9/15/2003
Citation: Bryant, R., Thoma, D., Moran, M.S., Holifield, C.D., Goodrich, D.C., Keefer, T.O., Paige, G., Williamd, D., Skirvin, S.M. 2003. Evaluation of hyperspectral, infrared temperature and radar measurements for monitoring surface soil moisture. Proceedings First Interagency Conference on Research in the Watersheds. Oct. 27-30, 2003, Benson, AZ., pp. 528-533.
Interpretive Summary: Resource managers need information about soil moisture status to make land use decisions regarding cross-country mobility, irrigation scheduling, pest management, biomass production and soil erosion. However, soil moisture is difficult to measure over large areas by sampling on the ground due to fine scale variations in soil moisture and the large areas that must be monitored. Remote sensing has advantages for monitoring surface soil moisture, by monitoring large areas in a timely and efficient manner that cannot be achieved by other means. This project demonstrated that reflectance, thermal and radar measurements can all be used to measure soil moisture accurately. But reflectance measurements require calibrations for different soil types and both reflectance and thermal techniques only measure the first few millimeters of soil. However, radar offers the potential for directly measuring soil moisture without the need to derive soil-specific calibrations, and it measures moisture to depths of several centimeters. The ability of radar to measure moisture regardless of soil type and without the need for calibration makes it potentially very useful in regions where calibrations are not possible, thus greatly extending the monitoring capability and information available for resource management decision making
Technical Abstract: Remote sensing techniques for monitoring soil moisture were tested by comparing hyperspectral reflectance and spectral indexes; surface temperature (Ts) and thermal indexes; and normalized radar backscatter to soil moisture. A laboratory study indicated that hyperspectral reflectance and Ts were sensitive to surface soil moisture (r2 range from 0.72 to 0.96). However, Ts was the only optical measurement that appeared insensitive to soil type. An index derived from differences between measurements of dry and wet soils ('-index) was presented and tested on the optical data as well as on data collected from two radar field studies at the United States Department of Agriculture ' Agricultural Research Service (USDA-ARS) Walnut Gulch Experimental Watershed (WGEW). Using the '-index, radar backscatter measured by different satellite sensors was merged into a single relationship with surface soil moisture. Furthermore, the radar '-index may be physically related to surface soil moisture such that field-based empirical relationships may be unnecessary in sparsely vegetated environments.