Incorporating stochasticity from extreme climatic events and multi-species competition relationships into single-tree mortality models

Abstract

Extreme climatic events, such as windstorms and drought, and competition are some of the main factors underlying tree mortality in Atlantic and Central European forest stands. However, current empirical tree mortality models are not adapted to separately consider the different causes of mortality. In addition, these approaches do not distinguish between intra- and inter-specific competition. In this study, we present a comprehensive empirical single-tree mortality model that incorporates all of the aforementioned features. On the one hand, extreme events can enter the model via fixed effects and random effects. The latter allows taking the stochasticity of the process into consideration. The distribution of these random effects can be seen as an extreme event severity distribution, which is of great interest for carrying out stochastic simulations on tree mortality. On the other hand, intra- and inter-specific competition is taken into account through linear interactions of species-specific competition indexes and species factors. In order to test this approach, we selected the beech–oak mixture in France as a case study. Beech–oak mixed forests are a common type in France which provides owners with significant economic benefits. Our findings confirm that drought and, especially, windstorm occurrence are major causes of tree mortality in these forests. The model was able to capture the stochasticity of windstorm events by means of random effects. In terms of competition, the probability of mortality in beech understories was expected to decrease in the presence of oak overstories with respect to pure beech stands. This result reveals the importance of complementarity processes in tree mortality.