Data Assimilation As an Effective Approach of Downscaling Coarse-Resolution Remotely-Sensed Solar-Induced Chlorophyll Fluorescence

Most current long-term spaceborne Solar-Induced chlorophyll Fluorescence (SIF) observations have a relatively coarse spatial resolution (i.e. over 40km), and often contain a lot of gap that limit their use for studying terrestrial ecosystems functions and physiology as they directly relate to cycling of carbon in Earth system. Data assimilation offers a great potential for overcoming these limitations by combining the power of advanced biosphere-atmosphere models with space-based observations to develop gap-free and high-resolution SIF products with high accuracy. We used the available remotely-sensed SIF observations from the Global Ozone Monitoring Experiment (GOME-2) and an Adaptive Spatial Average Ensemble Kalman Filter (ASA-EnKF) together with a SIF model to downscale the raw level-2 SIF product. We accounted for the effects of viewing geometry with a satellite-based escape ratio for SIF, to develop the high-resolution (i.e., 1km, hourly, gap-free) SIF over the Southwest region of United States. We found the resulting SIF product to be in good agreement with independent estimates of terrestrial ecosystems gross primary productivity (GPP), and with more regional details. The resulting dataset is suitable for study of ecosystem’s response to environmental stresses, e.g. droughts, and how their contribution to carbon cycle is affected at seasonal, interannual and regional level.