Solar-induced chlorophyll fluorescence sheds light on global evapotranspiration

The significance of large-scale evapotranspiration (ET) to climate science, water resources management, flood routing, irreversible desertification, and crop yield is not in dispute. Current large-scale ET models combine empirical formulations with a suite of remotely sensed data products that include meteorological variables, vegetation indices and/or soil moisture. In recent years, solar-induced chlorophyll fluorescence (SIF) has been proposed as an indicator of photosynthetic activity but its potential to constrain transpiration (T-r) or ET remains under-explored and frames the scope here. A large-scale terrestrial ET model driven by SIF is developed based on leaf water-carbon exchange complemented with an outcome for intercellular to ambient CO2 concentration derived from optimality theory for stomatal conductance. The model parameters are first calibrated across FLUXNET sites and then extrapolated globally using their dependence on climatic variables and plant functional types. The model, hereafter referred to as ETSIF, requires SIF data, leaf area index, land use type, and basic meteorological variables that include net radiation, air temperature and relative humidity. Global ETSIF estimates computed on a 4-day window for the period spanning 2003 to 2018 was 625 mm yr(-1) in general agreement with other independent global ET estimates, but discrepancy in the spatial distribution still exists implying that global ET estimation remains subject to large uncertainty. ETSIF exhibited a tantalizing positive trend over the same period but this trend was not statistically significant. One of the major advantages of this new approach, is that the model requires few parameters, reduce the parameterization of stomatal conductance and can be immediately used to constrain spatially extended ET estimates.