Geographic & climatic distances
We extracted georeferenced occurrence records for all 77 investigated tree species from GBIF and BIEN using the respective R packagesrgbif (Chamberlain et al. 2023) and BIEN (Maitner 2023). Both datasets have been combined and occurrence records have been cleaned following the workflow described by Zizka et al. (2019) using the package CoordinateCleaner [removing: ”capitals”, ”centroids”, ”equal”, ”gbif”, ”institutions”, ”outliers”, ”seas”, ”zeros”, “duplicates”].
For each occurrence record, we extracted the BIOCLIM variables mean annual temperature (1), total annual precipitation (12), mean temperature of the warmest quarter (10) and mean temperature of the coldest quarter (11) from CHELSA (v. 2.1, Karger et al. 2017, 2018). We calculated continentality as the temperature difference between the warmest quarter and the coldest quarter (BIOCLIM10 – BIOCLIM11). Similarly, we extracted climate data for the EBG Bayreuth with longitude of 11.57893 and latitude of 49.94782. We scaled all geographic and climatic data to the centre of the botanical garden in Bayreuth [ (x – xBayreuth) / sd(x)]. Thereby, the calculated geographic and climatic distances represent an environmental distance to the botanical garden in Bayreuth.
We calculated hypervolumes of each species geographic (longitude & latitude) and climatic (mean annual temperature, total annual precipitation & continentality) distribution area. Next, we calculated the Euclidean distance between the botanical garden in Bayreuth and the geographic as well as the climatic centroid of each species. Please note, as the geographic distribution area of species can be disjunct, it thereby can have multiple possible centroids. If so, we calculated the mean of the centroids’ distances to the EBG Bayreuth after splitting the hypervolumes. The calculations of hypervolumes and the splitting were conducted using the R-package hypervolume (Blonder et al.2023).