Vulnerability and adaptation of forests to climate change: towards a water-centred approach

Overview & Synopsis

Extreme droughts are having an increasing impact on forests worldwide, yet current models used to predict these effects remain unreliable and often contradict field observations. This is largely due to an overly simplistic representation of water stress, typically based on metabolic or correlative approaches that fail to capture the underlying mechanisms of plant hydraulic function. In response to these limitations, our work aims to better understand and model tree hydraulic functioning under drought conditions. We have contributed to clarifying the sequence of physiological responses triggered during drought. In particular, by developing reliable methods for measuring vulnerability to embolism in tree vascular systems, we demonstrated that stomatal regulation consistently occurs before the onset of xylem embolism. This insight led to the development of SurEau, a Soil–Plant–Atmosphere model that simulates the sequence of plant responses to drought, including beyond stomatal closure, in cases of extreme drought. SurEau enables the evaluation of tree response mechanisms, the exploration of phenotypic variability in the face of climate change, and the anticipation of risks such as forest dieback and vulnerability to wildfires. The model can be enhanced using remote sensing data to improve both spatial and operational predictions of drought stress and disturbances related to drought in forests. Our findings show that incorporating hydraulic functioning into models helps explain some of the effects of tree species mixing on drought response, and allows for the testing of management strategies such as assisted migration or water-efficient silviculture. These approaches offer valuable guidance for the adaptive management of forests under climate change. This work is part of a broader movement toward the structuring of plant hydraulic science, an emerging interdisciplinary field. Despite significant progress, many challenges remain, particularly in scaling models up to large regions or applying them under natural conditions. Our project addresses these challenges through three complementary approaches: - Strengthening large-scale SurEau simulations through data integration and systematic evaluation of predictions on plant water status and mortality using regional and global databases; - Interfacing SurEau with models from other disciplines (e.g., forest dynamics and management, evolutionary biology, hydrology); - Transferring tools to operational stakeholders through collaborative approaches.