This paper presents the assimilation of solar-induced chlorophyll fluorescence (SIF) into a terrestrial biosphere model to estimate the gross uptake of carbon through photosynthesis (GPP). We use the BETHY-SCOPE model to simulate both GPP and SIF in a process-based manner, going beyond a simple linear scaling between the two. We then use satellite SIF data from the Orbiting Carbon Observatory-2 (OCO-2) for 2015 in the data assimilation system to constrain model GPP. The assimilation results in considerable improvement between model and observed SIF, despite difficulties in simulating large SIF values due partly to uncertainties in the prescribed LAI. SIF-optimized global GPP increases by 7 % to 137 ± 6 PgCyr<sup>−1</sup> and shows improvement in its global distribution relative to independent estimates. This change in global GPP is driven by an overall decline in APAR and increase in the light-use efficiency of photosynthesis across almost all ecosystems. This process-based data assimilation opens up new pathways to the effective utilization of satellite SIF data that will improve our understanding of the global carbon cycle.