Advancing Remotely Sensed Evapotranspiration Estimates Using the MODIS Soil Moisture-Evapotranspiration (MOD-SMET) Model

Accurate estimation of evapotranspiration (ET) is critical for understanding local and regional water budgets and for the management of agricultural and water resources, especially in water-stressed regions. ET accounts for approximately 60% of terrestrial precipitation globally and approaches 100% of annual rainfall in arid ecosystems. ET is a particularly difficult hydrologic flux to measure due to its variation in both time and space. The spatiotemporal variation of ET requires data with adequate coverage, making remote sensing an ideal source of observations from which to calculate ET. The recent development of the Moderate Resolution Imaging Spectroradiometer (MODIS)-based Soil Moisture-Evapotranspiration (MOD-SMET) model has improved the temporal resolution of ET estimates to a daily time step while maintaining a spatial resolution of 1 km, similar to the widely-used Simplified Surface Energy Balance operational model (SSEBop) and the MODIS Global Evapotranspiration Project (MOD16) products. By calculating potential evapotranspiration (PET) from MODIS data using the Priestley-Taylor equation and downscaling soil moisture observations from the Soil Moisture and Ocean Salinity (SMOS) satellite, MOD-SMET provides a simple ET model based on scaling PET using a soil moisture function. The current formulation of the MOD-SMET model requires in situ data observations of soil texture, which are not available in all regions of the country or globally. The current presentation details new methods for scaling of MOD-SMET PET and the development of scaling functions using remote sensing indices, including Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Normalized Difference Water Index (NDWI), and Green Vegetative Fraction (GVF). Validation is performed using eddy flux towers covering multiple vegetation types and climate classifications. By utilizing these new scaling methods, the need for in situ data is removed, allowing for global application and the ability to obtain moderate spatial resolution daily ET estimates using MOD-SMET. These estimates allow for better understanding of the changing dynamics of hydrologic fluxes and the partitioning of water after land cover disturbances and fragmentation as well as the ability to resolve statistically significant trends after perturbations.