Abstract
The conservation and restoration of forest ecosystems require detailed knowledge of the native plant compositions. Here, we map global forest tree composition and assess the impacts of historical forest cover loss and climate change on trees. The global occupancy of 10,590 tree species reveals complex taxonomic and phylogenetic gradients determining a local signature of tree lineage assembly. Species occupancy analyses indicate that historical forest loss has significantly restricted the potential suitable range of tree species in all forest biomes. Nevertheless, tropical moist and boreal forest biomes display the lowest level of range restriction and harbor extremely large ranged tree species, albeit with a stark contrast in richness and composition. Climate change simulations indicate that forest biomes are projected to differ in their response to climate change, with the highest predicted species loss in tropical dry and Mediterranean ecoregions. Our findings highlight the need for preserving the remaining large forest biomes while regenerating degraded forests in a way that provides resilience against climate change.
Introduction
While forests are recognized as critical pillars for carbon sequestration and global biodiversity, our understanding of tree distributions has remained surprisingly fragmented. Previous mapping efforts were often restricted by regional boundaries, focused on a few well-known families, or lacked the resolution necessary to capture local nuances. This study addresses this gap by developing a high-resolution global map of 10,590 tree species, revealing that every forest on Earth possesses a "near-unique" biodiversity signature. By quantifying the impacts of historical land-use change and future climate scenarios, the research provides a critical baseline for effective conservation and restoration targets.
Methodology and Data Integration
To map the global forest at a 1-km resolution, researchers constructed a massive environmental niche modeling (ENM) pipeline using Google Earth Engine. The study aggregated nearly 30 million observations from 13 major databases, including GBIF and BIEN, focusing on species with at least 90 records.
An ensemble of machine learning models (Random Forests and Gradient Tree Boost) was used to predict ecological suitability based on nine climatic and soil variables, such as temperature seasonality, soil pH, and net primary productivity. To ensure accuracy, projections were constrained by geographic range polygons based on reported native countries. The resulting models were validated against independent datasets like the sPlot database, achieving a high average Area Under the Curve (AUC) of 0.93.
The Global Biodiversity Signature
A central finding of the study is the complexity of global tree composition. Using non-metric multidimensional scaling (NMDS) for taxonomic composition and evolutionary Principal Component Analysis (evoPCA) for phylogenetic lineages, the researchers demonstrated that almost every 100km cell on the map has a unique identity.
As shown in the ordination maps, lineage-based (phylogenetic) patterns often exhibit sharper boundaries than species-based (taxonomic) ones. For instance, while specific species may vary across Northern Canada and Alaska, these regions share a remarkably homogeneous evolutionary history. This "biodiversity signature" underscores that there is no such thing as a "generic" forest; local history and ecology create distinct fingerprints globally.
Impacts of Historical Forest Loss and Climate Change
By comparing potential habitat suitability to actual remaining forest cover, the study identified a median range reduction of 22% across all species. This range contraction is most severe in Mediterranean biomes, which have suffered a staggering 67.2% median reduction due to historical land use. This indicates that many species are currently squeezed into small fractions of their natural ecological niches.
Looking toward the future (2071–2100) under the extreme SSP 5.85 climate scenario, the models predict significant shifts:
- Climate Losers: Mediterranean and tropical dry forests are identified as high-risk hotspots, with some ecoregions predicted to lose over 50% of their currently suitable species.
- Climate Winners: Boreal forests are projected to see massive latitudinal shifts (median 6.00°) and may potentially gain species as suitable habitats move northward.
Limitations and Future Directions
The authors acknowledge several limitations, particularly the exclusion of rare or under-sampled species due to data thresholds. Furthermore, as correlative models, these projections assume a tree's current climate niche is its only possible niche, without accounting for biological plasticity, competition, or the impact of extreme events like mega-fires and droughts. Future research integrating deep learning and mechanistic models could better account for how trees adapt, disperse, and compete in a rapidly changing world.
Ultimately, these findings emphasize the urgent need to preserve remaining large forest biomes while regenerating degraded areas to bolster resilience against the shifting climate horizon.
Summary updated