Integrating evolutionary, demographic and ecophysiological processes to predict the adaptive dynamics of forest tree populations under global change


Three types of process-based models (PBMs) are traditionally used to predict the response of forest tree populations to global change (GC): (i) ecophysiological models, which simulate carbon and water fluxes in forest ecosystems by explicitly integrating the effects of climate and CO2; (ii) forest dynamics models which simulate forest successions by explicitly linking mortality, growth and regeneration processes; and (iii) evolutionary dynamics models, which simulate the variation and evolution of adaptive traits by explicitly accounting for selection, mutation, gene flow and inheritance rules. The ongoing context of rapid GC, however, questions the boundaries between these types of models. Here, we review different strategies of model integration: (i) physio-demographic PBMs, integrating physiological and demographic processes; (ii) demo-genetic PBMs, integrating demographic and evolutionary processes; and (iii) physio-demo-genetic PBMs, which attempt to integrate these three types of processes. We show that these integrative models allow a better understanding of how different functional traits influence demographic rates (the phenotype-demography map), how the variation in demographic rates influences fitness (the demography-fitness map) and how individual variations of fitness may in turn influence the genetic composition of a population. Our review highlights that accounting for inter-individual variation in ecological processes is increasingly recognized as crucial for modelling the ecosystem response to environmental change. We argue that the effort of integrating these different processes is valuable, both for a basic understanding of their interactive effects on the responses of forests to GC and for applied horizon scanning to support adaptive strategies.