Deep biogeographic barriers explain divergent global vertebrate communities

Peter J. Williams, Elise F. Zipkin, Jedediah F. Brodie

---

Abstract

Biogeographic history can lead to variation in biodiversity across regions, but it remains unclear how the degree of biogeographic isolation among communities may lead to differences in biodiversity. Biogeographic analyses generally treat regions as discrete units, but species assemblages differ in how much biogeographic history they share, just as species differ in how much evolutionary history they share. Here, we use a continuous measure of biogeographic distance, phylobetadiversity, to analyze the influence of biogeographic isolation on the taxonomic and functional diversity of global mammal and bird assemblages. On average, biodiversity was better predicted by environment than by isolation, especially for birds. However, mammals in deeply isolated regions are strongly influenced by isolation; mammal assemblages in Australia and Madagascar, for example, are much less diverse than predicted by environment alone and contain unique combinations of functional traits compared to other regions. Neotropical bat assemblages are far more functionally diverse than Paleotropical assemblages, reflecting the different trajectories of bat communities that have developed in isolation over tens of millions of years. Our results elucidate how long-lasting biogeographic barriers can lead to divergent diversity patterns, against the backdrop of environmental determinism that predominantly structures diversity across most of the world.

Code

1. 0_DataPrep_AllTaxa.R: Create global 2x2 degree grid cells, calculate environment variables for each grid cell, and create assemblages for each grid cell. Environment variables include mean elevation, elevation range, present-day climate, past climate, distance between present-day climate and past climate, past ice cover, years since significant land conversion, and Human Impact Index. Assemblages are lists of all bird, mammal, or bat species with range maps that overlap a given cell.
2. 1_SR_PD_FR_XXXX.R: Calculate species richness (SR), phylogenetic diversity (PD), and functional richness (FR) for each grid cell. Also calculate functional PCoA. Separate scripts for birds, mammals, and bats.
   • 1_SR_PD_FR_Bird.R
   • 1_SR_PD_FR_Mamm.R
   • 1_SR_PD_FR_Bat.R
3. 2_PhyloB_Bird_XX.R: Calculate phylobetadiversity for birds on subsets of grid cells. Calculating all pairwise comparisions is computationally intractable, so grid cells are split into 5 groups.
   • 2_PhyloB_Bird_12.R: Bird phylobetadiversity for cells in groups 1 and 2.
   • 2_PhyloB_Bird_13.R: Bird phylobetadiversity for cells in groups 1 and 3.
   • 2_PhyloB_Bird_14.R: Bird phylobetadiversity for cells in groups 1 and 4.
   • 2_PhyloB_Bird_15.R: Bird phylobetadiversity for cells in groups 1 and 5.
   • 2_PhyloB_Bird_23.R: Bird phylobetadiversity for cells in groups 2 and 3.
   • 2_PhyloB_Bird_24.R: Bird phylobetadiversity for cells in groups 2 and 4.
   • 2_PhyloB_Bird_25.R: Bird phylobetadiversity for cells in groups 2 and 5.
   • 2_PhyloB_Bird_34.R: Bird phylobetadiversity for cells in groups 3 and 4.
   • 2_PhyloB_Bird_35.R: Bird phylobetadiversity for cells in groups 3 and 5.
   • 2_PhyloB_Bird_45.R: Bird phylobetadiversity for cells in groups 4 and 5.
4. 2_PhyloB_Mamm_XX.R: Calculate phylobetadiversity for mammals on subsets of grid cells. Calculating all pairwise comparisions is computationally intractable, so grid cells are split into 4 groups.
   • 2_PhyloB_Mamm_12.R: Mammal phylobetadiversity for cells in groups 1 and 2.
   • 2_PhyloB_Mamm_13.R: Mammal phylobetadiversity for cells in groups 1 and 3.
   • 2_PhyloB_Mamm_14.R: Mammal phylobetadiversity for cells in groups 1 and 4.
   • 2_PhyloB_Mamm_23.R: Mammal phylobetadiversity for cells in groups 2 and 3.
   • 2_PhyloB_Mamm_24.R: Mammal phylobetadiversity for cells in groups 2 and 4.
   • 2_PhyloB_Mamm_34.R: Mammal phylobetadiversity for cells in groups 3 and 4.
5. 2_PhyloB_Bat_XX.R: Calculate phylobetadiversity for bats on subsets of grid cells. Calculating all pairwise comparisions is computationally intractable, so grid cells are split into 3 groups.
   • 2_PhyloB_Bat_12.R: Bat phylobetadiversity for cells in groups 1 and 2.
   • 2_PhyloB_Bat_13.R: Bat phylobetadiversity for cells in groups 1 and 3.
   • 2_PhyloB_Bat_23.R: Bat phylobetadiversity for cells in groups 2 and 3.
6. 2_FuncB_Bird_XX.R: Calculate functional beta diversity for birds on subsets of grid cells. Calculating all pairwise comparisions is computationally intractable, so grid cells are split into 4 groups.
   • 2_FuncB_Bird_12.R: Bird functional beta diversity for cells in groups 1 and 2.
   • 2_FuncB_Bird_13.R: Bird functional beta diversity for cells in groups 1 and 3.
   • 2_FuncB_Bird_14.R: Bird functional beta diversity for cells in groups 1 and 4.
   • 2_FuncB_Bird_23.R: Bird functional beta diversity for cells in groups 2 and 3.
   • 2_FuncB_Bird_24.R: Bird functional beta diversity for cells in groups 2 and 4.
   • 2_FuncB_Bird_34.R: Bird functional beta diversity for cells in groups 3 and 4.
7. 3_PhyloB_XXXX_full.R: Combine files to create full phylobetadiversity turnover distance matrix, and save NMDS coordinates for 3 axes. Separate scripts for birds, mammals, and bats.
   • 3_PhyloB_Bird_full.R
   • 3_PhyloB_Mamm_full.R
   • 3_PhyloB_Bat_full.R
8. 3_FuncB_Bird_full.R: Combine files to create functional beta diversity turnover distance matrix for birds, then calculate mean functional beta diversity turnover for birds.
9. 3_FuncB_XXXX.R: Calculate functional beta diversity, then calculate mean functional beta diversity turnover. Separate scripts for mammals and bats.
   • 3_FuncB_Mamm.R
   • 3_FuncB_Bat.R
10. 4_Analysis_XXXX.R: Run models for species richness, phylogenetic diversity, functional richness, and mean functional beta diversity turnover. Save variance partitioning results. Calculate differences in residuals and save figures (Fig. 3 and Supplementary Fig. 1). Separate scripts for birds, mammals, and bats.
    • 4_Analysis_Bird.R
    • 4_Analysis_Mamm.R
    • 4_Analysis_Bat.R
11. 5_BarPlots.R: Create barplots of variance partitioning results for birds, mammals, and bats (Fig. 2, Fig. 6a, and Supplementary Fig. 5).
12. 5_DiversityMaps.R: Create maps of global diversity patterns of species richness, phylogenetic diversity, functional richness, and mean functional beta diversity turnover for birds, mammals, and bats (Fig. 4, Fig. 6b, and Supplementary Fig. 3).
13. 5_PhyloB_Figures.R: Create phylobetadiversity figures, including NMDS plots using 2 axes (Fig. 1), NMDS plots using 3 axes (Supplementary Fig. 4), and scree plots (Supplementary Fig. 7).
14. 5_FuncSpace_Figures.R: Create figures showing overlap in functional space (Fig. 5 and Supplementary Fig. 6).
15. 5_Analysis_Realm_XXXX.R: Run models for species richness, phylogenetic diversity, functional richness, and mean functional beta diversity turnover, using realm instead of phylobetadiversity.Calculate differences in residuals and save figures (Supplementary Fig. 2). Separate scripts for birds, mammals, and bats.
    • 5_Analysis_Realm_Bird.R
    • 5_Analysis_Realm_Mamm.R
    • 5_Analysis_Realm_Bat.R
16. 5_PhyloSignal.R: Calculate phylogenetic signal of traits (Supplementary Table 1).
17. 5_AdditionalVariables.R: Calculate additional variance explained by environment variables besides climate (Supplementary Table 2).
18. 5_PresentVsPastClimate.R: Calculate variance explained using climate data from present day, Last Glacial Maximum, and mid-Holocene (Supplementary Table 4).

Data

Raw data

1. newRealms: Zoogeographic realms. Available from the Center for Macroecology, Evolution and Climate at the University of Copenhagen (https://macroecology.ku.dk/resources/wallace).
2. BOTW: BirdLife range map data for birds. Due to the large size, we split the original GDB file into two shapefiles prior to data processing. 'BIRDS_resident.shp' consisted of resident ranges (Seasonal == 1), and 'BIRDS_nonresident.shp' consisted of other ranges (Seasonal != 1). Data downloaded in 2018. Available upon request from BirdLife International (http://datazone.birdlife.org/species/requestdis).
3. MAMMALS: IUCN range map data for mammals. Data downloaded in 2018. Available from the IUCN Red List (https://www.iucnredlist.org/resources/spatial-data-download).
4. BirdFuncDat.csv: EltonTraits 1.0 data for birds. Converted to CSV due to issues with original TXT file. TXT file available from Figshare (https://doi.org/10.6084/m9.figshare.c.3306933.v1)
5. MamFuncDat.txt: EltonTraits 1.0 data for mammals. Available from Figshare (https://doi.org/10.6084/m9.figshare.c.3306933.v1).
6. bird_names_cells.csv: Bird taxonomies for range map data (names_ICUN), trait data (names_elton), and phylogeny (names_phylo), manually aligned.
7. mamm_names_cells.csv: Mammal taxonomies for range map data (names_ICUN), trait data (names_elton), and phylogeny (names_phylo), manually aligned.
8. bird_traits_cat.csv: Trait type for each bird trait.
9. mamm_traits_cat.csv: Trait type for each mammal trait.
10. Stage2_MayrParSho_Hackett_mean phylogeny_ultrametric.tre: Bird phylogeny, ultrametric consensus tree created based on 1,000 credible phylogenies. Credible phylogenies available from BirdTree (https://birdtree.org/subsets/).
11. Upham_mean_phylogeny_ultrametric.tree: Mammal phylogeny, ultrametric consensus tree based on 1,000 credible phylogenies. Credible phylogenies available from VertLife (http://vertlife.org/phylosubsets/).
12. Landmass_area_cells.csv: Landmass area for each cell. Manually entered using the Global Islands dataset available from the United States Geological Survey (https://rmgsc.cr.usgs.gov/gie/gie.shtml).
13. mn30_grd: Global Multi-resolution Terrain Elevation Data (GMTED2010). Available from the United States Geological Survey (https://topotools.cr.usgs.gov/gmted_viewer/gmted2010_global_grids.php).
14. wc2.1_10m_bio: Bioclimatic variables, present day. Available from WorldClim (https://www.worldclim.org/data/worldclim21.html).
15. cclgmbi_10m: WorldClim 1.4 downscaled paleo climate, bioclimatic variables, Last Glacial Maximum. Available from WorldClim (https://www.worldclim.org/data/v1.4/paleo1.4.html).
16. ccmidbi_10m: WorldClim 1.4 downscaled paleo climate, bioclimatic variables, Mid Holocene. Available from WorldClim (https://www.worldclim.org/data/v1.4/paleo1.4.html).
17. KK10.nc: Anthropogenic land cover change over time. Available from PANGAEA (https://doi.pangaea.de/10.1594/PANGAEA.871369).
18. hii_2019-01-01.tif: Human Impact Index, 2019 data. Available from the Wildlife Conservation Society (https://wcshumanfootprint.org/data-access).

Intermediate data products

1. global_cells.shp: 2x2 degree latitude-longitude grid cells used in all analyses.
2. clim_Present_cells.csv: Climate PCA coordinates for 4 axes for each grid cell.
3. elev_cells.csv: Mean elevation and elevation range of each grid cell.
4. XXXX_cell_df.csv: Lists of species in each grid cell. Separate files for birds, mammals, and bats.
   • bird_cell_df.csv: File too large to push to GitHub.
   • mamm_cell_df.csv
   • bat_cell_df.csv
5. XXXX_SR_cells.csv: Species richness of each grid cell. Separate files for birds, mammals, and bats.
   • bird_SR_cells.csv
   • mamm_SR_cells.csv
   • bat_SR_cells.csv
6. XXXX_PD_cells.csv: Phylogenetic diversity of each grid cell. Separate files for birds, mammals, and bats.
   • bird_PD_cells.csv
   • mamm_PD_cells.csv
   • bat_PD_cells.csv
7. XXXX_FR_cells.csv: Functional richness of each grid cell. Separate files for birds, mammals, and bats.
   • bird_FR_cells.csv
   • mamm_FR_cells.csv
   • bat_FR_cells.csv
8. XXXX_mean_fb.csv: Mean functional beta diversity turnover of each grid cell. Separate files for birds, mammals, and bats.
   • bird_mean_fb.csv
   • mamm_mean_fb.csv
   • bat_mean_fb.csv
9. XXXX_pb_NMDS.csv: Phylobetadiversity turnover NMDS coordinates for 3 axes for each grid cell. Separate files for birds, mammals, and bats.
   • bird_pb_NMDS.csv
   • mamm_pb_NMDS.csv
   • bat_pb_NMDS.csv
10. XXXX_pb_dist.RDS: Distance matrix of phylobetadiversity turnover. Separate files for birds, mammals, and bats.
    • bird_pb_dist.RDS
    • mamm_pb_dist.RDS
    • bat_pb_dist.RDS
11. XXXX_func_PCoA.RDS: Functional PCoA coordinates for each species. Separate files for birds, mammals, and bats.
    • bird_func_PCoA.RDS
    • mamm_func_PCoA.RDS
    • bat_func_PCoA.RDS
12. XXXX_fb.RDS: Output from functional beta diversity calculation, including distance matrices for total beta diversity, turnover, and nestedness. Separate files for mammals and bats.
    • mamm_fb.RDS: File too large to push to GitHub.
    • bat_fb.RDS