Skip to main content
ARS Home » Research » Publications at this Location » Publication #160596

Title: USING TRMM/TMI TO RETRIEVE SOIL MOISTURE OVER SOUTHERN UNITED STATES FROM 1998 TO 2002

Author
item GAO, H. - PRINCETON UNIV.
item WOOD, E. - PRINCETON UNIV.
item DRUSCH, M. - PRINCETON UNIV.
item MCCABE, M. - PRINCETON UNIV.
item Jackson, Thomas
item BINDLISH, R. - SSAI

Submitted to: European Geophysical Society; American Geophysical Union; European Union of Geosciences
Publication Type: Abstract Only
Publication Acceptance Date: 2/1/2004
Publication Date: 4/23/2004
Citation: Gao, H., Wood, E.F., Drusch, M., McCabe, M., Jackson, T.J., Bindlish, R. 2004. Using TRMM/TMI to retrieve soil moisture over southern United States from 1998 to 2002 [abstract]. European Geosciences Union General Assembly. (6): Paper No. 1607-7962.

Interpretive Summary:

Technical Abstract: Operational soil moisture products from passive microwave satellite remote sensing are expected to improve our understanding of land-atmospheric interactions. The Tropical Rainfall Measuring Mission (TRMM) satellite launched in November, 1997, carries a microwave imager, offering one of two spaceborne sensors sensitive to soil moisture changes. The Advanced Microwave Scanning Radiometer on board the EOS Aqua satellite provides another C- and X-band sensor, however there have been issues related to Radio Frequency Interference (RFI), which affect results from the C-band. In this presentation, a Land Surface Microwave Emission Model (LSMEM) is used to retrieve surface soil moisture over the southern United States from TRMM/TMI 10.65GHz horizontal polarized brightness temperature. Land surface temperatures required for model simulation are derived from validated Variable Infiltration Capacity (VIC) model outputs, driven primarily by the North American Land Data Assimilation System (NLDAS). Other variables and parameters (soil texture, soil salinity, soil surface roughness, vegetation water content, vegetation structure parameter and atmospheric contribution, etc.) necessary for model operation come from operational sources. Soil moisture was estimated over the southern United States from 1998 to 2002, with a sampled resolution of 1/8 degree. The results are compared with soil moisture from predictions based on the VIC land surface model (10cm depth, 1/8th degree spatial resolution), and observations from the Oklahoma Mesonet and USDA ARS SCAN sites. While the three soil moisture datasets have shown consistent patterns in the spatial and temporal domains, TMI soil moisture demonstrates a larger dynamic range compared to the other datasets, as a result of its thin ('0.5cm) sensing depth. The difficulties of validating remote sensing products will be discussed. The final soil moisture product is determined after masking out frozen soil, snow covered area, precipitation and heavily vegetated areas to produce a useful and needed data source.