Deciphering the factors relating sun-induced fluorescence to photosynthesis and transpiration in space and time

Photosynthetic organism slowly transformed Earth from a rocky and bare landscape to a lush, diverse collection of thriving ecosystems. The photosynthetic process also altered the atmosphere by lowering atmospheric CO2 and increasing O2 levels, thus permitting animal life. The synthesis of the organic compounds through photosynthesis (gross primary production, GPP), by terrestrial plants is today the largest global carbon flux and supports human welfare as it is the basis for food, wood and fibers. GPP also provides several critical ecosystem services such as the offsetting of anthropogenic CO2 emissions. The photosynthetic process is mediated by stomata; small openings on plant’s leaves that allow the exchange of CO2 and water vapor. The movement of water through plants and its evaporation from stomata (Transpiration, T) plays a pivotal role in the global water cycle and land-surface energy balance and represents a large fraction of evapotranspiration. Accurate estimations of GPP and T are therefore critical to monitor ecosystems and to quantify the amount of CO2 sequestered from the atmosphere. Remote and proximal sensing techniques offer the possibility to estimate GPP and T in an unintrusive way, while also allowing for high temporal or high spatial resolution measurements. Sun-induced fluorescence (SIF), the radiation emitted by plant’s upon sun’s exposure, is a promising remote sensing tool to estimate GPP and T. SIF contains information on the amount of photosynthetic active radiation absorbed by plants and the efficiency with which it is used to drive photosynthesis. It has also been recently related to T, although the mechanistic link between the two is still unclear. As SIF has been progressively used more extensively in the last decade to estimate GPP, and more recently T, much as been learned on the way SIF may be used to constrain carbon and water fluxes. Still, many research gaps have yet to be investigated, such as: 1) the role of nutrients (such as nitrogen (N) and phosphorus (P)) in shaping the GPP-SIF relationship, 2) how extreme event such as heatwaves can affect GPP-SIF and 3) improving the understanding of how SIF may be used to predict T. This dissertation examines the first objective by investigating a nutrient manipulation experiment (with N and P) where simultaneous GPP and SIF measurements were conducted in a Mediterranean grassland. I uncover the mechanisms that link the fertilization-driven changes in canopy nitrogen and phosphorus concentration to the observed changes in SIF and GPP. Specifically, I find that N addition changed plant community structure and increased canopy chlorophyll content, which jointly lead to changes in absorbed photosynthetic active radiation (APAR), which ultimately affected both GPP and SIF. The changes in plant type abundance driven by N addition lead to changes in structural properties of the canopy such as leaf angles, which ultimately influenced observed SIF by controlling the escape probability of SIF (Fesc). Additionally, the N addition induced changes in the biophysical properties of the canopy that lead to a trade-off between surface temperature, which decreased, and SIF at leaf scale that increased. The P addition lead to a statistically significant increase in light use efficiency of fluorescence emission (LUEf), in particular in plots with also N addition, suggesting a co-limitation of LUEf by N and P. In regard to the second objective I analyzed how the 2018 heatwave, which was characterized by temperatures up to 45 °C, affected the GPP-SIF relationship in a Mediterranean tree-grass ecosystem. I combine canopy scale passive fluorescence (SIF) with leaf scale active fluorimetry, which allows to obtain the amount of heat dissipation (nonphotochemical quenching, NPQ) which is a major dissipation pathway of absorbed energy and an important driver of SIF. I find that the heatwave caused an inversion of the photosynthesis-fluorescence relationship at both canopy and leaf scale. The highly nonlinear relationship was strongly shaped by NPQ. During the extreme heat stress, plants experienced a saturation of NPQ causing a change in the allocation of energy dissipation pathways towards SIF. These innovative results showed that the relationship between GPP and SIF (which has been broadly considered to be linear) can depart from linearity under extreme stress due to physiological regulations. Additionally I show the complex modulation of the relationship NPQ-SIF-GPP at an extreme level of heat stress, which is not fully represented in state-of-the-art coupled radiative transfer and photosynthesis models. Finally for the third objective I predict T using passive and active fluorescence in two sites; a Mediterranean tree-grass ecosystem and a deciduous beech forest. I test the three different methods that have been used so far to relate SIF and T; a) fully empirical, b) hybrid modeling approach and c) water use efficiency (WUE) based approach in order to establish a framework for SIF based T predictions. I find that total SIF had a stronger correlation with T than GPP across sites, as both total SIF and T are driven to larger extent by APAR than GPP. Additionally, I found approach (a) and (b) to have similar predictive power across sites. Finally, the WUE approach had the lowest performance out of the three. In order to better understand the mechanistic relationship between SIF and T I highlight the importance of separating periods in which photosynthesis is stomatal or non-stomatically (i.e. carboxylation) limited from periods of low or no stress. During periods of photosynthetic limitations T was mostly predicted by variables that can be physiologically modulated by plants, such as NPQ and SIF, whereas during periods of no stress I found T to be more energy driven and therefore more strongly predicted by APAR or surface temperature. This thesis advances the understanding of SIF based GPP and T prediction by analyzing the most important factors able to affect the GPP-SIF relationship and by testing SIF based T prediction with different methodologies. SIF is a powerful, yet complex predictor of GPP and T which has the potential to be used to accurately constrain carbon and water fluxes in a changing climate. Future studies should focus on understanding in which ecosystems and under which conditions SIF can be used to predict T, as the link between the two, although increasing more clear, remains understudied