Ecological and Agricultural Productivity Forecasting Using Root-Zone Soil Moisture Products Derived from the NASA SMAP Mission
Southwest Watershed Research
2012 Annual Report
1a.Objectives (from AD-416):
ARS is interested in developing viable agricultural applications for remotely-sensed surface soil moisture observations. Objective 2 of our Project Plan (5342-13610-010-00D) specifically addresses this interest: “Develop improved watershed model components and decision support systems that can assimilate and utilize remotely sensed data for parameterization, calibration, and model state adjustments” The COOPERATOR (NASA) requires expertise to design the science requirements for an upcoming satellite mission (the Soil Moisture Active/Passive Mission - SMAP) so that they maximize the utility of SMAP observations for agricultural forecasting and monitoring applications (e.g. the detection of agricultural drought).
1b.Approach (from AD-416):
Research will be based on the design and execution of a series of complete end-to-end observing system simulation experiments (OSSE’s) to isolate the added utility of SMAP soil moisture products for agricultural and rangeland forecasting activities. All OSSE experiments will contain three separate components :.
1)an algorithm test-bed facility to generate synthetic SMAP soil moisture data products,.
2)a data assimilation system to integrate these synthetic products into a multi-layered land surface model, and.
3)a crop forecasting system to obtain end-of-season crop yield and rangeland productivity forecasts based the mid-growing season initialization of a crop model with profile soil moisture measurements obtained from a crop growth model. Other cooperators include investigators from NASA, the University of Arizona, USDA ARS ALARC Maricopa, and USDA ARS HRSL Beltsville.
Crop growth simulation models are often used to aid decision-makers, such as the insurance industry, to predict regional crop yields on a seasonal basis. Because of uncertainties in simulations of real-world systems, any model-based estimate of crop yield will be subject to error. We attempted to mitigate this error by assimilating satellite-based observations of crop greenness and soil moisture to improve wheat yield estimates. We found very little improvement in yield predictions by assimilating LAI and soil moisture due to the lack of correlation between leaf and grain growth. Instead of using observations to improve model predictions, we plan to study how we can use observational data to improve our models. This project contributes to objective 1 of the in-house project.