M-TRAITMarie Curie FellowshipProject duration: Feb 2018 to Feb 2020The response of forests to climate change driven increase in droughts is still very uncertain. Part of this is due to the calibration of current vegetation models for current growing conditions made necesary by their heavy parameterization. The M-TRAIT project aims at improving the representation of drought response of forests by using a traits approach to simulate water-carbon interactions in trees. Functional traits such as leaf mass per area or leaf hydraulic conductance allow the description of plant-level growth strategies and response to environmental changes.IntroductionThe changes in precipitation patterns that are projected for the coming decades are expected to increase the frequency and intensity of droughts, especially in the Mediterranean region. Despite the potentially dramatic impacts on wood and forest-based ecosystem services, the effect of these changes in the hydrological cycle on European forests is still highly uncertain. One reason is that the predictive power of Land Surface Models is hampered by their heavy parameterization per broad plant type and calibration under current conditions, contradictory to the ecological processes of acclimation and selection.Plant traits are tree characteristics that define plant growth strategies. They vary as much within species as between species, as individuals and communities adapt to their environmental conditions. M-TRAIT will use a traits-based approach to model large-scale forest response to droughts. The traits determinant to the drought response of trees will be determined from analysis of worldwide trait databases and the relationships between traits and between traits and the environment will be formulated. The key traits will then be integrated into a forestry-enhanced LSM with an individuals-sampling approach in which a large number of plant growth strategies are initialized and allowed to interact.