Enhancing The USDA Global Crop Assessment Decision Support System Using Satellite-Based Soil Moisture Estimates Obtained From The Soil Moisture Active Passive Mission

Abstract:

The primary goal of the U.S. Department of Agriculture Foreign Agricultural Service (FAS) is to provide timely information on current and expected crop supply and demand estimates. Inter-annual variability in crop condition and crop productivity is largely controlled by the amount of available water to the plants. Thus, knowledge of the root-zone soil moisture is critical for the USDA’s crop analysts. This information is currently provided by the modified Palmer model (PM). The PM is a two-layer, water balance-based hydrologic model that is driven by daily precipitation and daily minimum and maximum temperature observations based on ground meteorological station measurements from the World Meteorological Organization (WMO) and gridded weather data from the U.S. Air Force 557^th Weather Wing (former U.S. Air Force Agency, AFWA). A data assimilation (DA) unit was added to the model to allow the integration of satellite-based soil moisture observations. The DA system was initially developed using retrievals from the Advanced Microwave Scanning Radiometer (AMSR-E), where the AMSR-E soil moisture estimates were ingested into the PM using a 1-D Ensemble Kalman Filter Approach. After the failure of AMSR-E the system was updated and it is currently set to ingest Soil Moisture Ocean Salinity (SMOS)-based retrievals. Operational delivery of the SMOS-based soil moisture product for USDA FAS began in spring, 2014. This talk will demonstrate the added value of assimilating satellite-based data and focus on work that is being done in preparation for updating the system by ingesting soil moisture observations from the Soil Moisture Active Passive (SMAP) mission. Soil moisture estimates derived using data obtained from SMOS and the Advanced Scatterometer (ASCAT) instrument on MetOp have been used as a proxy for the SMAP radiometer and radar products, respectively. The performance of this dual assimilation system would be assessed by examining the lagged rank cross correlation between the Normalized Difference Vegetation Index (NDVI) and the PM soil moisture estimates acquired before and after the assimilation.