Skip to main content
ARS Home » Research » Publications at this Location » Publication #135883

Title: ASSIMILATION OF SURFACE SOIL MOISTURE TO ESTIMATE PROFILE SOIL WATER CONTENT: A FIELD AND MODELING EXPERIMENT

Author
item Heathman, Gary
item Starks, Patrick
item Ahuja, Lajpat
item Jackson, Thomas

Submitted to: Journal of Hydrology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/28/2003
Publication Date: 5/1/2003
Citation: Heathman, G.C., Starks, P.J., Ahuja, L.R., Jackson, T.J. 2003. Assimilation of surface soil moisture to estimate profile soil water content. Journal of Hydrology. 279:1-17.

Interpretive Summary: Data assimilation is a relatively new area of research as related to the integration of remote sensing data and soil water modeling. Most of the related studies reported in the literature have largely been theoretical in nature since field data needed to verify the modeling results were often not available. The objective of this study was to determine the effect of assimilating remotely sensed surface soil water content estimates into a model to estimate root zone soil water content. The Root Zone Water Quality Model was run at four study sites within the Little Washita River Experimental Watershed and updated daily by direct assimilation of soil water content estimates for the June 18 - July 16 1997 study period. The four study sites were selected because of variation in soil texture and vegetation cover and availability of measured meteorological variable required by the model and because of the availability of measured profile soil water content down to a depth of 60 cm. Study results show that assimilation of surface soil water content into the model improved model estimates in the top 30 cm at all sites, but below this level no significant improvement in model estimates was observed. The findings of this research imply that assimilating accurate estimates of surface soil water content derived from remotely sensed data can improve estimates of root zone soil water content when applied at watershed scales.

Technical Abstract: Data assimilation is a relatively new area of research as related to the integration of remote sensing data and soil water modeling. Most of the related studies reported in the literature have largely been theoretical in nature since field data needed to verify the modeling results were often not available. The objective of this study was to determine the effect of assimilating surface soil water content estimates into a model to estimate root zone soil water content. The Root Zone Water Quality Model was run at four study sites within the Little Washita River Experimental Watershed, located in south central Oklahoma, and updated daily by direct assimilation of soil water content estimates for the June 18 - July 16, 1997 study period. The four study sites were selected because of variation in soil texture and vegetation cover and availability of measured meteorological variable required by the model and because of the availability of measured profile soil water content down to a depth of 60 cm. Root mean square error, mean bias error and correlation coefficients from statistical analysis show that assimilation of surface soil water content into the model improved model estimates by 2-6% in volumetric water content in the top 30 cm at all sites, but below this level no significant improvement in model estimates was observed. The findings of this research imply that assimilating accurate estimates of surface soil water content derived from remotely sensed data can improve estimates of root zone soil water content when applied at watershed scales.