The potential of satellite FPAR product for GPP estimation: An indirect evaluation using solar-induced chlorophyll fluorescence

Satellite remotely sensed fraction of photosynthetically active radiation (FPAR) products are widely used in land-surface monitoring and modeling, especially for estimating global terrestrial photosynthetic activity through light use efficiency (LUE) models. PAR absorbed by active chlorophyll (APAR(chl)) is directly linked to vegetation photosynthesis and can be used to estimate ecosystem gross primary production (GPP). Previous studies have demonstrated that solar induced chlorophyll fluorescence has very tight relationship with APAR(chl) at various ecosystems. Therefore, the solar angle normalized SIF (nSIF) is directly related to the fraction of PAR absorbed by chlorophyll (FPAR(chl)). This paper intercompared six space FPAR products from Moderate Resolution Imaging Spectroradiometer (MODIS), Visible Infrared Imaging Radiometer Suite (VIIRS), Copernicus Global Land Service (CGLS), Multi-angle Imaging SpectroRadiometer (MISR), Earth Polychromatic Imaging Camera (EPIC) and Ocean and Land Colour Instrument (OLCI). Their potential relationships with FPAR(chl) were indirectly evaluated with both spaceborne (Orbiting Carbon Observatory-2, OCO-2 and TROPOspheric Monitoring Instrument, TROPOMI) and airborne (Chlorophyll Fluorescence Imaging Spectrometer, CFIS) nSIF data as well as in situ GPP measurements. Our results show that these FPAR products are different in terms of amplitudes and seasonal variations across biomes. Among six FPAR products, OLCI FPAR shows the best relationships with TROPOMI nSIF740, OCO-2 nSIF757, and CFIS nSIF755. The coefficient of determination (R-2) for the relationship between OLCI FPAR and TROPOMI nSIF740 is 0.79 +/- 0.17 on a global average. APAR calculated from OLCI also exhibits the best relationship (R-2 = 0.79) with in situ GPP over 25 flux towers.