Revisiting vegetation carbon and water flux representation during drought events in the ORCHIDEE land surface model
- 1Laboratoire des Sciences du Climat et de l’Environnement, LSCE/IPSL, CEA-CNRS-UVSQ, Université Paris-Saclay, 91191 Gif-sur-Yvette, France
- 2College of Engineering, Mathematics and Physical Sciences, University of Exeter, Exeter, UK
- 3Science Partners, 42 quai de Jemmapes, 75010 Paris, France
Climate change is projected to increase the frequency and intensity of droughts, inducing further water stress on vegetation. These water stress conditions impact both vegetation carbon and water exchanges that are regulated through stomatal diffusion. Land surface models (LSMs) have been developed to simulate the amount of carbon taken up by vegetation through photosynthesis (GPP), and the emission of water into the atmosphere through plant transpiration. However, LSMs struggle to accurately simulate vegetation response to drought, which contributes to the strong uncertainty on GPP and plant transpiration estimates and is critical for future projections. Moreover, LSMs represent different vegetation types by grouping plants with similar characteristics in terms of structure, behavior, and climatic conditions, therefore not accounting for possible differences in vegetation response to drought due to the diversity of environmental conditions within the same biome. In the ORCHIDEE LSM, we used in situ GPP and latent heat flux (LE) estimates at more than 40 sites from the FLUXNET Warm Winter 2020 network, which captures the recent drought years over Europe, to evaluate and refine the simulated vegetation response to soil water stress. We performed multisite data assimilation experiments based on the dominant vegetation type at these sites to optimize the parameters involved in vegetation response to soil water stress using the in situ GPP and LE estimates. We found that the optimized values of the coefficient that determines the speed of vegetation response to soil water stress can be defined as a function of the mean annual vapor pressure deficit (VPD). This new function enables to consider the environmental conditions on site through VPD in vegetation response to soil water stress, instead of having a response that only depends on the vegetation type. During a drought event, this soil water stress function induces vegetation growing under low VPD to close its stomata faster than vegetation acclimated to higher VPD conditions. Finally, regional simulations were performed to evaluate the impact of including this dependency on VPD in vegetation response to water stress over the recent European drought years.
How to cite: Abadie, C., Maignan, F., Peylin, P., Vuichard, N., Raoult, N., and Bastrikov, V.: Revisiting vegetation carbon and water flux representation during drought events in the ORCHIDEE land surface model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5665, https://doi.org/10.5194/egusphere-egu24-5665, 2024.