Results

The research investigated the impact of climate change on the competitive strength of tree species across Europe using a deep learning-based meta-model. By analysing 135 million simulation-years from 17 process-based forest models, researchers evaluated the competitiveness of nine major tree species. The results indicate that climate change is likely to cause a significant reassembly of European forests, with up to 25% of forest areas potentially experiencing a change in the dominant tree species by the end of the 21st century.

Change in the competitive strength of major European tree species

Under a severe climate change scenario (RCP8.5, 2071–2100), six of the nine studied species showed a decline in competitive strength. Evergreen conifers were particularly affected; the Competitive Strength Index (CSI) decreased for Aleppo pine (−23.9%), Norway spruce (−8.0%), Scots pine (−5.4%), and silver fir (−1.2%). Broadleaved species, including silver birch (−3.9%) and holm oak (−3.2%), also recorded negative CSI values. Conversely, deciduous broadleaved species generally showed increased competitiveness, with pedunculate oak rising by 7.8% and European beech by 5.5%. European larch also showed a slight increase of 0.7%. On average, the competitive strength of broadleaved species responded positively to climate change (+2.3%), while coniferous species saw a decrease (−3.9%).

Spatial patterns of changing competitive strength

Spatial variation in competitive outcomes was high across biomes. Competitive strength generally decreased in warmer, water-limited regions, such as Mediterranean and temperate broadleaved forests, and increased in cold-limited biomes, including boreal forests and tundra ecosystems. For example, European beech saw an overall increase in competitiveness but suffered a 6.8% decrease in the Mediterranean parts of its range. Norway spruce showed a pronounced decrease of 21.2% in temperate broadleaved biomes but gained 14.3% in cold-limited areas like Fennoscandia. Species generally performed better at their cold-induced niche edges compared to their warm niche edges, where competitive strength typically declined.

Shifts in dominant species

A shift in competitive balance can result in a change in the dominant species of a forest stand. The study defined dominance as a species comprising more than 66% of the stand basal area. Norway spruce, holm oak, and silver birch were the species most frequently projected to lose their dominant roles. Under the RCP8.5 scenario, Norway spruce dominance is projected to decrease by 20.2% across 57.6 million hectares. In contrast, European beech and pedunculate oak were overall able to increase their dominance by 9.9% and 9.2%, respectively. In total, 24.8% of Europe’s forest area, corresponding to 96 million hectares, is likely to see a change in dominant tree species due to shifting competitive balances under severe climate change.

Discussion

The findings suggest that evergreen coniferous species, which currently dominate approximately 56% of Europe’s forests, will lose significant competitiveness under climate change. This trend points towards a major shift in the dominant clades of tree species in Europe. These shifts are particularly expected at the warm trailing edges of species distributions. The study highlights that the loss of competitive strength can serve as an early warning indicator for shifts in tree species composition. Such changes have profound implications for biodiversity and ecosystem services, including carbon uptake and resource use efficiency. The researchers suggest that managers should consider these potential changes when adjusting silvicultural concepts to maintain desired forest structures and species mixtures.

Materials and methods overview

The study harnessed a database of 135 million simulation-years from 17 process-based models to train a Deep Neural Network (DNN) of forest state transitions. This meta-model synthesised and scaled local process model results to a continental scale. The model predicted changes in three state variables: canopy height, Leaf Area Index (LAI), and species composition. Competitive strength was quantified using a Competitive Strength Index (CSI), which aggregated indicators for light competition, specifically height growth potential and the ability to shade out competitors (LAI). Projections were performed for current distribution ranges under various Representative Concentration Pathway (RCP) scenarios.