Estimating Gross Primary Production (GPP) from satellite Solar-Induced chlorophyll Fluorescence (SIF) with a mechanistic model across NEON Ecoregions

Direct measurement of gross primary production (GPP) beyond a single leaf is a core challenge that prevents accurate quantification of global GPP and its spatiotemporal dynamics. Recent advancements in satellite Solar-Induced chlorophyll Fluorescence (SIF) retrieval offer promising opportunities, but so far incorporating satellite SIF to estimate GPP across scales is based solely on empirical linear scaling, an assumption that does not always hold at short timescales and stress conditions. In this study, we employ a process-based model, based on the mechanistic light reaction (MLR) model, to establish the link between SIF, electron transport rate (ETR), and GPP at the canopy scale using SIF retrievals from TROPOspheric Monitoring Instrument (TROPOMI) onboard Sentinel-5p. Our approach is applied across diverse NEON (National Ecological Observatory Network) ecoregions during the growing seasons of 2018-2021. We compare GPP estimates obtained from the conventional linear scaling approach and our mechanistic MLR-based approach with eddy-covariance (EC) flux tower measurements. Additionally, we analyze cross-biome variability in GPP estimates by incorporating ancillary information from hyperspectral reflectance spectra. Our findings highlight the potential of MLR for enabling satellite SIF for global GPP estimation, and the mechanistic advantage of MLR over the widely-accepted linear SIF-GPP scaling.