AbstractGross primary production (GPP) models driven by fine resolution remote sensing data characterize the spatial and temporal heterogeneities in plant photosynthesis, which is largely dependent on biome-specific maximum photosynthetic capacity. The red-edge reflectance, sensitive to leaf chlorophyll content, is a good proxy of maximum photosynthetic capacity. More importantly, studies show that the red-edge reflectance-related chlorophyll content index (CIr) multiplied by the incident photosynthetic active radiation (PARin) strongly correlates to GPP estimated at carbon flux towers (GPPflux). Yet, to the best of our knowledge, there is no systematic study investigating the general relationship between fine spatial resolution CIr and GPP among biomes and the relationship between CIr and maximum photosynthetic capacity in GPP models. To provide an overview on incorporating space-borne CIr into a GPP model, we applied fine resolution Sentinel-2-derived CIr and GPPflux over 57 flux sites representative of 10 biomes. We investigated the relationship between CIr and GPPflux, and the spatio-temporal relationship between CIr and ecosystem maximum photosynthetic capacity indicated by the potential ecosystem light use efficiency (LUEpot). We also evaluated the relationship between other five vegetation indices (VIs) and GPPflux. Results showed that the CIr multiplied by PARin has a higher agreement (R2 > 0.5) with GPP than other VIs. A universal relationship exists between the CIr multiplied by PARin and GPP, except for forest biomes. The CIr also strongly (R2 > 0.5) relates to the LUEpot during the peak of the growing season. The CIr has a low spatial variance (CV = 0.25) among biomes, highlighting that CIr can be a proxy of maximum photosynthetic capacity in GPP models that do not require biome-dependent coefficients. This study provides insight for incorporating CIr into GPP models and better quantifying global terrestrial photosynthesis.
Published: January 26, 2023