The spatio-temporal dynamics of the relationship between gross primary productivity and solar induced chlorophyll fluorescence across the agricultural ecosystem of India

Food security is dependent on agriculture as such prediction and monitoring of crop yield is of essential importance. Crop productivity primarily depends upon its potential to transmute light energy to sugar through photosynthesis and the total amount of carbon fixed by this process is coined as Gross Primary Production (GPP). Therefore, a reliable estimation of GPP is a vital step toward crop monitoring. Typically, accurate quantification of GPP at various spatial and temporal scales through the light use efficiency models remains challenging. Novel remote sensing methods such as Solar Induced chlorophyll Fluorescence (SIF) is a direct probe into the photosynthetic machinery and has been demonstrated to be a much better indicator of primary production compared to traditional vegetation indices. SIF has also been demonstrated to be a proxy for GPP as well as estimating crop productivity. However, most studies focus on homogeneous crop areas across the globe by using satellite or ground-based SIF concurrently with flux tower GPP. In this study, we examine how well satellite-based SIF products can monitor the GPP and crop productivity across the heterogeneous cropping system of India. Linear correlation analysis is carried out to analyse the relationship between FLUXCOM GPP with Global Ozone Monitoring Experiment-2 (GOME-2) SIF and TROPOspheric Monitoring Instrument (TROPOMI) SIF at different spatio-temporal scales. The results indicate a significant pixel-wise correlation at 8 daily and monthly scales across the crop area of India. However, a weak linear correlation is found between GPP and SIF at yearly scale. From the analysis of TROPOMI SIF and GOME-2 SIF, we find that TROPOMI SIF has a higher potential to predict GPP across the crop area of India. To explore the spatial and temporal variability in GPP and SIF relationship, we used GPP/SIF ratio as an indicative parameter. The maximum GPP/SIF values occurred in September and October. We found the seasonal pattern of GPP/SIF ratio following the seasonal dynamics of Leaf area index (LAI, canopy structural metric). During peak growing season GPP/SIF was positively corelated to short wave radiation and moisture availability, but during the early growing season it mostly dependent on soil moisture. Our results will enhance the understanding of the mechanisms of the link between GPP and SIF.