Modélisation de la composition isotopique des cernes d'arbres (13C et 18O) et des transferts de COS entre l'atmosphère et la biosphère continentale pour quantifier les flux bruts de carbone

摘要

In the context of global climate change, the behavior of the terrestrial biosphere can be durably affected by the increased frequency and intensity of extreme climatic events, which can decrease the photosynthetic assimilation of carbon and/or increase the respiration rate of the ecosystems. Therefore, quantifying the carbon storage capacity of the ecosystems and predicting their sensitivity to climate changes strongly rely on our capacity to separately estimate the photosynthesis and respiration rates at different scales. The gross primary productivity (GPP) is however not directly measurable. Indirect approaches have been proposed to estimate the biospheric gross fluxes (GPP and respiration), combining for instance stable isotopologues of CO2 (13C and 18O), and, more recently, the measure of carbonyl sulfide (COS) concentrations in the atmosphere. In this context, my PhD work followed two complementary approaches. In the first approach, isotopic measurements and tree-ring widths were used, because both of them are linked to the photosynthetic activity. The inter-annual variations of the photosynthetic fluxes simulated with the ORCHIDEE continental biosphere model were evaluated and compared with in situ measurements. The second approach consisted in using atmospheric measurements of OCS concentrations and in exploring their potential to constrain the current estimates of the GPP in dynamic global vegetation models (DGVM), by (1) establishing a new global budget of sources and sinks of this gas, (2) optimizing the source and sink terms of this cycle and (3) estimating the potential of this new tracer to validate/invalidate the simulated GPP when using current DGVMs.