The distribution of tree species and the composition of forests are highly dependent on climate conditions. Therefore, changes in climate will alter which trees can grow where, creating new habitats for species to grow in some places, and making existing habitats elsewhere unsuitable for particular species and eventually causing local extinction events (so-called, range erosion).
Knowing current species ranges and predicting when and where they will change over the course of the 21st century is critical to mitigate the impact of species range shifts on Europe’s forest ecosystems. Since the vast majority of European trees have a relatively low dispersal ability, it seems unlikely that species will be able to compensate for range erosion through natural dispersal into new habitats. Especially in Europe, where forests are far from naturalness (Strona et al. 2016) and hardly any primary or old-growth forest are left (Sabatini et al 2018, Barredo et al. 2021), forest management, which can include strict protection and assisted migration, may play a key role to minimize biodiversity loss due to climate change.
As the science and knowledge service of the European Commission, the JRC has studied the ecology, use, and fate of European forest tree species for many years (e.g., Casalegno et al. 2011, Caudullo et al. 2017, Mauri et al. 2017, Beck et al. 2020, Camia et al. 2020, Forzieri et al. 2021). In 2016, the JRC published the European Atlas of Forest Tree Species (San-Miguel Ayanz et al. 2016). It provides concise and essential information about the most important and widespread tree species in Europe. The Atlas also includes species distribution and enviromental suitability maps modelled from bio-climatic and topographic predictors using a tree species occurrences data set derived from National Forest Inventories as reference. The maps in the European Atlas of Forest Tree Species are available on this website.
EU-Trees4F is a new data set describing the current and future potential distribution of 67 tree species in Europe, and is the result of a collaboration between the JRC and University of Helsinki (Mauri et al. 2022). EU-Trees4F maps areas in Europe where the climate and soil are currently suitable for particular tree species to grow. It also models where climate change could, in theory, allow these tree species to grow in the future and which of these areas the trees could reach if they dispersed only naturally. The current period (1991-2020) and the three future period (2021-2050, 2051-2080, and 2081-2110) projections were derived from 11 regional climate model simulations sourced from EURO-CORDEX using two representative concentration pathways (RCP 4.5 and RCP 8.5). The models were calibrated using Biomod2 platform (Thuiller et al. 2009) with the tree species occurrences from the EU-Forest data set (Mauri et al. 2017), which is a comprehensive database of tree species occurrences for Europe, made up of data from 21 National Forest Inventories, compiled and harmonised by ENFIN, as well as data from Forest Focus (Hiederer et al. 2011) and BioSoil (Huston Durrant et al. 2011).
EU-Trees4F is available in GeoTIFF format at a resolution of 5 arc-minutes (ca. 10 square kilometers) in the WGS84 reference system (EPSG:4326) and at resolution of 10 square kilometres in the LAEA reference system (EPSG:3035). The projections into future climates were created using two approaches:
1) A climatic ensemble mean (ENS_CLIM) that projects the consensus model into future conditions using the average of the 11 regional climate models.
2) A Species Distribution Model ensemble mean (ENS_SDMS) that projects the consensus model for every single regional climate model and a posteriori averaged.
The complete EU-Trees4F data set is downloadable as a single compressed ZIP file from this website and from Figshare. Should you not want the entire data set, you find below the data for individual trees in ZIP files organized in 4 main folders:
/ENS_CLIM: the layers derived from the climatic ensemble mean approach. The ensemble binary layers (“bin” code name) are GeoTIFF raster files in the Lambert Azimuthal Equal Area (EPSG:3035) reference system with integer number values 0 and 1. The ensemble probability layers (“prob” code name) are GeoTIFF raster files in the WGS84 (EPSG:4326) reference system with integer number values from 0 to 1000.
/SDMS_CLIM: the layers derived from the SDM ensemble mean approach. The ensemble binary layers (“bin” code name) are GeoTIFF raster files in the Lambert Azimuthal Equal Area (EPSG:3035) reference system with integer number values 0 and 1. The ensemble probability layers (“prob” code name) with integer number values from 0 to 1000 and the ensemble standard deviation layers (“std” code name), both are GeoTIFF raster files in the WGS84 (EPSG:4326) reference system.
/SINGLE_MODELS: the output maps derived from Biomod2 using the 11 RCM. For each of the models, the potential distribution maps are available as probabilities (“prob” code name), and in binary format (“bin” code name) along with the associated coefficient of variation (“cv” code name). All maps are in GeoTIFF raster files in the WGS84 (EPSG:4326) reference system.
/PNGS: the output maps of the tree species potential distributions in PNG image format for both ensemble approach (“ens-clim" and “ens-sdms") and two emission scenarios ("rcp45" and “rcp85”). The “future potential dispersal” maps show in green and blue the potential distribution range (according to ensemble SDM projections). The expected distribution range as simulated using a dispersal model (Migclim) is shown in green, and the blue area represents the potential suitable range that is not occupied by the species due to dispersal limitations. The "migclim categorized" maps show the potential distribution range using a dispersal model (Migclim) modelled until the end of the century: in green the ‘Stable presence’ areas that will remain suitable habitat from the present until the end of the century; in red the ‘Decolonized’ areas that will become climatically unsuitable by the end of the century; in blue the ‘Suitable but not occupied’ areas that will become climatically suitable by the end of the century but will not be naturally colonized due to dispersal limitations; in grey ‘Always absent’ where the species will remain absent.
Please cite the EU-Trees4F data set as:
Mauri A., Girardello M., Strona G., Beck P. S. A., Forzieri G., Caudullo G., Manca F. & Cescatti A., 2022. EU-Trees4F, a dataset on the future distribution of European tree species. Scientific Data 9, 37. DOI: 10.1038/s41597-022-01128-5
|- Complete EU-Trees4F data set||(ca. 2 Gb)||EU-Trees4F_dataset.zip (also on Figshare)|
|- Climatic ensemble mean layers only||(ca. 70 Mb)||EU-Trees4F_ens-clim.zip|
|- SDM ensemble mean layers only||(ca. 130 Mb)||EU-Trees4F_ens-sdms.zip|
|- Single model layers only||(ca. 1.2 Gb)||EU-Trees4F_single-models.zip|
|- PNG images only||(ca. 500 Mb)||EU-Trees4F_pngs.zip|
|- Layers by species:|