Submitted to: American Geophysical Union
Publication Type: Abstract only
Publication Acceptance Date: 10/15/2003
Publication Date: 12/8/2003
Citation: Cosh, M.H., Jackson, T.J., Starks, P.J., Heathman, G.C., Bindlish, R. 2003.Satellite soil moisture validation using in situ point sources in the southern great plains during SMEX03 [abstract]. EOS Transactions. 84(46):618. Interpretive Summary:
Technical Abstract: A number of studies have been conducted in the area of large-scale soil moisture observation, including Washita' 92, SGP97, and SMEX02. These experiments led to the development of satellite concepts specifically designed to retrieve estimates of surface soil moisture from space. The validation of satellite soil moisture products, such as those provided by the AMSR and AMSR-E instruments requires accurate estimates of large-scale surface soil moisture across a variety of landforms and climates. For validation, short-term experiments are important, but long term monitoring of surface soil moisture is also a necessity. It is not possible to gravimetrically monitor the surface at scales sufficient for validation, but an innovative statistical analysis technique may allow a small number of points to replace more extensive efforts. This concept is called temporal stability and it may hold the key to effective satellite product validation. Large-scale soil moisture monitoring networks provide a method of validation assuming that the networks accurately capture the variability of the surface. The Little Washita River Watershed in southwestern Oklahoma has been a focal point of soil moisture research in recent years. With 19 soil moisture continuous monitoring sensors distributed throughout the watershed, this is an ideal watershed for validation work. Soil moisture is measured half-hourly at a depth of 5 cm, which is closely related the depth needed for satellite validation. The Soil Moisture Experiment 2003 (SMEX03), an intensive operational period, was designed to calibrate this network for large-scale soil moisture estimates. Comparing the watershed record to detailed field sampling and larger regional sampling conducted during the experiment allows multi-scale spatial and temporal stability analysis.