Directly estimating diurnal changes in GPP for C3 and C4 crops using far-red sun-induced chlorophyll fluorescence

- SIF760 tracks the diurnal variations in GPP over C3/C4 crop canopies well.
- SIF760 is linearly correlated with GPP for C3 and C4 crops at the diurnal scale.
- The slope of the GPP-SIF760 relationship is higher for C4 crops.
- The large difference in εP between C3 and C4 crops determines the difference in the slope εPεF.
- The photosynthetic type should be considered when discussing the GPP-SIF relationship.

Accurate estimation of gross primary production (GPP) is of great importance to global change research and also food and fuel security. Previous studies have demonstrated that values of sun-induced chlorophyll fluorescence (SIF) retrieved from hyperspectral data provide a direct measure of ecosystem GPP. However, global analysis of the relationship between satellite SIF and model-based GPP indicates that the relationship between the two parameters is highly dependent on the plant functional type (PFT). The overarching goal of this study is to examine the potential of far-red SIF retrieved at 760 nm (SIF760) to track the diurnal variations in GPP for C3 and C4 crops, and to investigate whether the GPP-SIF relationship is dependent on the type of photosynthesis. GPP values are estimated from flux tower records and daily SIF760 data are derived from ground-based spectral measurements. The results show that GPP and SIF760 have similar diurnal patterns and are linearly correlated for C3 and C4 crops. However, the ratio of εP to εF (the slope of the linear SIF-based GPP model, GPP = εPεF×SIF) for C3 wheat is about 46% of that for C4 maize. The findings from the diurnal variation experiments imply that theεF is weakly sensitive to the photosynthetic pathway type (PsP type) and that the large difference in εP between C3 and C4 crops leads to the difference in the slopes. Our studies confirm the capability of the remotely sensed SIF signals to act as a direct proxy for GPP and suggest that the PsP type should be considered when trying to accurately quantify the ecosystem productivity using the straightforward empirical approach.