Evapotranspiration from Airborne Simulators as a Proxy Datasets for NASA’s ECOSTRESS mission – A new Thermal Infrared Instrument on the International Space Station

Surface evapotranspiration (ET) represents the loss of water from the Earth’s surface both by soil evaporation and vegetation transpiration processes. ET is a key climate variable linking the water, carbon, and energy cycles, and is very sensitive to changes in atmospheric forcing and soil water content. The response of ET to water and heat stress directly affects the surface energy balance and temperature which can be measured by thermal infrared remote sensing observations. The NASA ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) will be deployed in 2019 to address critical questions on plant–water dynamics, ecosystem productivity and future ecosystem changes with climate through an optimal combination of thermal infrared measurements in 5 spectral bands between 8-12 µm with pixel sizes of 38×57 m and an average revisit of 5 days over the contiguous United States at varying times of day.

Two instruments capable of providing proxy datasets are the MODIS/ASTER (MASTER) airborne simulator and Hyperspectral Thermal Emissions Spectrometer (HyTES). This study is focused on estimating evapotranspiration using shortwave and thermal infrared remote sensing observations from these instruments. The thermal infrared data from MASTER/HyTES is used as a proxy dataset for ECOSTRESS to demonstrate the capability of the future spaceborne system to derive ET and water stress information from thermal based retrievals of land surface temperature. MASTER and HyTES data collected from 2004 to present over the Western United States at different seasons are used to test and evaluate different ET algorithms using ground-based measurements. Selected algorithms are 1) explicitly based on surface energy budget calculation or 2) based on the Penman-Monteith equation and use information on land surface temperature to estimate the surface resistance to convective fluxes. We use ground data from the Fluxnet and Ameriflux networks, and from permanent validation stations over an agricultural landscape in California operated by the Jet Propulsion Laboratory. The impact of different error sources associated with both the input data or the parameterization of the different models is quantified and used to assess the uncertainty of the future operational spaceborne ET products.