European forest vulnerability to hydraulic failure: an ecohydrological approach

The current acceleration of climate change in Europe makes it essential to assess spatially the impact of drought and heat waves on forest disturbances risk (mortality, wildfire risk, etc…). Recent studies have shown that hydraulic failure is a key driver of forest disturbances. Hydraulic failure can be modelled with state-of-the-art plant hydraulic models that are driven by climate data and different traits including (i) hydraulic traits (such as xylem cavitation resistance and stomatal regulation), (ii) leaf area index and (iii) total soil water capacity. Among these traits soil water capacity is highly sensitive, but is poorly available at large scale.
In this study we used the process based plant hydraulic model SUREAU (Cochard et al., 2021; Ruffault et al 2022) to estimate hydraulic failure risk for forest at the European scale for the last 3 decades. To initialize the model we used spatialized  climate (ERA5), LAI data (from Copernicus remote sensing) and land cover (ESA CCI). Species hydraulic traits for major European species were extracted from global databases. In order to initialize the total soil water capacity at European scale and compensate the lack of soil water data, we developed an algorithm of model inversion based on ecohydrological assumption. The ecohydrological assumption is that forest adjust their total available water capacity through rooting depth, for a given climate, traits combination and Leaf area index to maintain a low embolism rate under normal conditions (excluding extreme drought). Our simulation approach simulations allowed to spatialized forest vulnerability to drought and to map total soil water capacity under forest stands.