Climatic marginality: a new metric for the susceptibility of tree species to warming exemplified by Fagus sylvatica (L.) and Ellenberg’s quotient

In the face of climate warming, maps of potential tree species distribution can support forest management planning at coarse scales. For evaluating future suitability, conditions at the rear edge, i.e. at the meridional and lower altitudinal limits of species distribution, are of particular importance. Therefore, we present the concept of climatic marginality (distance to the rear edge) as a metric for the susceptibility against climate warming. Using a statistic niche model fitted to observed and potential beech occurrence in ICP Forests Level I monitoring plots and WorldClim data, we evaluate the modelled xeric limit of European beech based on the Ellenberg’s climate quotient involving thresholds suggested by Ellenberg and other authors. The applicability of the marginality index was tested with independent study sites. Despite the limitations of niche modelling, estimated climatic thresholds of beech were well in accordance with textbook knowledge and recent research. The regional patterns of climatic marginality were plausible and more meaningful with respect to the rear edge compared to conventional niche model outputs. In terms of climatic marginality, most regions in Central Europe are far from the xeric limit of beech. Evaluation based on independently sampled sites indicated that inclusion of soil and topography (microclimate) may permit implications at the local scale, e.g. growth potential estimations.     «

In the face of climate warming, maps of potential tree species distribution can support forest management planning at coarse scales. For evaluating future suitability, conditions at the rear edge, i.e. at the meridional and lower altitudinal limits of species distribution, are of particular importance. Therefore, we present the concept of climatic marginality (distance to the rear edge) as a metric for the susceptibility against climate warming. Using a statistic niche model fitted to observed a...     »