Potential Utility of Remotely-Sensed Surface Soil Moisture for Agricultural Productivity Forecasting

Wade Crow, USDA ARS HRSL, wade.crow@ars.usda.gov (Presenter)
John Bolten, NASA GSFC, john.bolten@nasa.gov
Grey Nearing, University of Arizona, grey@email.arizona.edu
Kelly R Thorp, USDA-ARS, kelly.thorp@ars.usda.gov
Rolf Reichle, NASA GSFC, rolf.h.reichle@nasa.gov

Satellite-based surface soil moisture products have great potential for enhancing our ability to detect the onset and severity of agricultural drought. This, in turns, aids operational efforts to globally forecast inter-annual variability in agricultural productivity and food availability. This poster will describe research attempting to 1) quantify this potential using existing satellite-based soil moisture products and 2) develop appropriate data assimilation tools to fully harness any potential for future satellite products.

The first objective is addressed by evaluating the lagged-correlation between temporal anomalies in MODIS-based enhanced vegetation index (EVI) products and analogous anomalies in root-zone soil moisture predictions derived using the USDA Foreign Agricultural Service (FAS) global water balance model. These lagged EVI/root-zone soil moisture cross-correlations are examined before and after the assimilation of AMSRE-E surface soil moisture retrievals into the water balance model to assess the added impact of such assimilation on our ability to forecast EVI anomalies in global agricultural regions. Results indicate large added value associated with the integration of AMSR-E surface soil moisture products and are strongly supportive of a potential role for satellite-based soil moisture products in operational USDA FAS crop yield and condition forecasts.

The second objective seeks to harness this apparent potential through the development of an appropriate land data assimilation system to directly integrate satellite-based soil moisture products into crop system models capable of forecasting agricultural yields. Using a synthetic experiment design, various data assimilation approaches are evaluated based on their ability to enhance end-of-season yield predictions made with the DSSAT crop system model. Preliminary results suggest that existing data assimilation strategies - designed primarily for application to atmospheric and hydrologic models - require modification before they are suitable for crop yield forecasting.

Presentation Type:  Poster

Session:  Science in Support of Decision Making   (Wed 10:00 AM)

Associated Project(s): 

• Crow, Wade: Ecological and agricultural productivity forecasting using root-zone soil moisture products derived from the NASA SMAP mission. ...details
• Crow, Wade: Enhancing the USDA Global Crop Production Decision Support System with NASA Land Information System and Water Cycle Satellite Observations ...details

Poster Location ID: 131