Optimal Estimates of Global Terrestrial Gross Primary Production from Satellite Fluorescence and Dgvms

Determining the spatial and temporal distribution of terrestrial gross primary production (GPP) is a critical step in closing the Earth’s carbon budget. Dynamical global vegetation models (DGVMs) provide mechanistic insight into GPP variability but diverge in predicting the response to climate in poorly investigated regions. Recent advances in the remote sensing of solar-induced chlorophyll fluorescence (SIF) opens up a new possibility to directly measure planetary photosynthesis on spatially resolved scales. Here, we discuss a new methodology to infer the global distribution of GPP and uncertainty from an optimal combination of an ensemble of DGVMs with satellite fluorescence. Prior uncertainty is estimated from the spread of DGVMs and updated through assimilation of SIF. Assimilation of GOSAT SIF with DGVMs leads to a shift of global GPP from northern latitudes (~3.6 Pg C year) to the tropics (3.7 Pg C year), improvements in the structure of seasonal GPP compared to flux towers in N. America, Europe and S. America, and reduction of uncertainty of predicted zonal average GPP of 30% in the annual mean. This methodology provides a novel way to quantify GPP response to climate, evaluate models used in climate projections, and constrain predictions of carbon cycle evolution.