Warming Drives Large-Scale Shifts in Post-Disturbance Vegetation Dynamics and Expansion of Deciduous Trees in the Boreal Forest in a Dynamic Vegetation Model

Abstract

The intensification of disturbance regimes due to climate change is a growing concern in boreal forests, yet the long-term consequences for forest recovery remain poorly understood, as boreal successional dynamics unfold over a century or longer and exceed the length of available observational records. To overcome this research gap we use the dynamic vegetation model LPJ-GUESS to investigate future climate change impacts on boreal successional dynamics over the 21st and 22nd centuries. We first confirm that LPJ-GUESS reproduces historic patterns of post-disturbance recovery. We then show that under future warming, deciduous transients become more frequent, longer, and more pronounced due to a temperature-mediated shift in competitive balance emerging in LPJ-GUESS. Using a random forest framework, we find that the long-term recovery pathway can be predicted with 90% accuracy from vegetation and environmental conditions in the first decade after disturbance alone. This supports the hypothesis that the changes in post-disturbance recruitment currently observed in boreal forests are an indicator of more long-lasting changes in vegetation composition in the future. While we find no evidence for permanent shifts toward deciduous tree cover in LPJ-GUESS, the combination of prolonged deciduous transients and more frequent disturbances makes an increase in deciduous tree cover a likely future scenario.