Multivariate Assimilation of Satellite-derived Soil Moisture and Evapotranspiration for Drought Monitoring in the CONUS

Abstract:

The aim of this study is to present a framework to jointly assimilate satellite-derived surface soil moisture (SSM) and Evapotranspiration (ET) observations into the Noah-MP land surface model (LSM) in order to improve the effectiveness and usefulness of this model in predicting soil moisture profile together with energy flux in the form of latent heat. We use two remotely sensed products, namely SMOPS (Soil Moisture Operational Product System), and MODIS evapotranspiration (MODIS16 ET) within the proposed framework, which is based on an evolutionary data assimilation technique, to correct the model outputs. We utilize the precipitation estimates from the latest version (V4) of the Global Precipitation Measurement (GPM) mission Integrated Multi-satellitE Retrievals for GPM (IMERG) product as the forcing data for the Noah-MP model. The rest of the model input variables are available from the phase 2 of the North American Land Data Assimilation System (NLDAS-2). Given the multivariate assimilation of SSM and ET, we utilize these two key variables to calculate an integrated drought index for monitoring drought over the Contiguous United States (CONUS). The posterior of SSM and ET are used for calculation of Soil Moisture Deficit Index (SMDI) and Evapotranspiration Deficit Index (EDI), respectively. We combine the Probability Density Functions (PDFs) of these two indices using a Copula function to generate an integrated drought index for better characterization of agricultural drought.