|Peters-Lidard, C. - NASA GSFC|
|Mocko, D. - NASA GSFC|
|Garcia, M. - NASA GSFC|
|Santanello JR., J. - NANA GSFC|
|Tischler, M. - US ARMY CORPS OF ENG.|
|Wu, Y. - NASA GSFC|
Submitted to: Water Resources Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: December 12, 2007
Publication Date: May 30, 2008
Citation: Peters-Lidard, C.D., Mocko, D.M., Garcia, M., Santanello Jr., J., Tischler, M., Moran, M.S., Wu, Y. 2008. Role of precipitation uncertainty in the estimation of hydrologic soil properties using remotely sensed soil moisture in a semi-arid environment. Water Resources Research, Vol. 44, W05S18, doi:10.1029/2007WR005884. Interpretive Summary: Soil moisture is a critical component of atmospheric, land-surface, and hydrologic models that impacts weather forecasts on daily to seasonal timescales. However, accurate prediction of the moisture conditions in the soil is limited by insufficient information and representation of soil type and textural properties. In this study, the problem of estimating soil moisture and soil properties is approached from a unique perspective. It is based upon a larger effort where NASA, USDA, and our university partners are helping the US Army Corps of Engineers develop techniques to use microwave satellite measurements to derive soil wetness and properties at very high resolution (tens of meters) for assessment of human and vehicle mobility. The first testbed for this experiment is the Walnut Gulch Experimental Watershed in southeastern Arizona, where 6 daily estimates of near-surface soil moisture across the watershed were derived from passive microwave data using established techniques. Then, a land-surface model was run to determine which soil types and properties are required in the model to simulate the soil moisture conditions that match those from remote sensing. By adjusting the sand, clay, and silt contents (i.e. the properties that control the flow of moisture) of the soil in a physically consistent manner, errors in model simulated versus observed soil moisture were minimized. Overall, this study demonstrates the potential to gain physically meaningful and much-needed soils information at high-resolution using repeated soil moisture data combined with accurate precipitation inputs to the models.
Technical Abstract: The focus of this study is the role of precipitation uncertainty in determining the accuracy and retrieveability of estimated soil texture and hydraulic properties. This work builds on and extends recent work conducted as part of the ongoing development of the Army Remote Moisture System (ARMS), in which it was shown that soil texture and associated hydraulic properties may be estimated based on a combination of multi-temporal microwave remote sensing, land surface modeling and parameter estimation techniques. As in the previous study, the Land Information System (LIS) modeling framework, including the Noah Land Surface Model (LSM) enhanced with pedotransfer functions and the parameter estimation tool (PEST), is applied to the Walnut Gulch Experimental Watershed (WGEW) during the Monsoon ’90 experiment period. Precipitation uncertainty is represented by systematically varying forcing precipitation from the high-density raingage network in WGEW to other lower-quality precipitation sources, from a single raingage in the watershed, to a continental scale reanalysis dataset—the North American Regional Reanalysis (NARR)—to the synoptic raingage closest to the watershed. It is shown that the quality of precipitation data, particularly with respect to detecting convective rainfall events and reproducing the observed rainfall rate PDFs is a critical determinant of whether successive remotely sensed soil moisture images may be used in the ARMS approach to estimate soil texture and hydraulic properties. Poor quality rainfall estimates lead to an inability to estimate soil texture properties, as indicated by non-convergence of PEST. It is also shown that the role of precipitation uncertainty is modulated by other parameters in the Noah LSM, such as the bare soil evaporation factor.