Dataset DOI: 10.5061/dryad.66t1g1kdj

The following code and data are included in the discoveries_refine_diversity.zip file.

1. Code

The provided R code (discoveries_refine_diversity.R) reproduces the following:

• Statistical Models:
  To quantify how the accumulation of newly described species over the past century has reshaped our understanding of environment-richness relationships, simultaneous Autoregressive (SAR) regression models of vertebrate species richness were conducted against environmental variables across multiple vertebrate groups and spatial scales.
• Statistical Tests:
  The script performs an adjusted z-test to compare the coefficients for each environmental variable between 1920 and 2020, to evaluate whether the change of environmental drivers is significant.
• Figures: Generation of figures for: New species discoveries, Diversity centers, and Standardized coefficients of environmental variables across globe and biogeographic realms.
  • Figure 1 (New Species Discoveries): Maps and latitudinal diversity gradient plots showing the spatial and latitudinal distribution of new species discoveries for each vertebrate group.
  • Figure 2 (Diversity Centers): Maps showing the shifts in diversity centers (based on total range-weighted rarity) between 1920 and 2020, and plots showing the relationship between historical richness and new discoveries.
  • Figure 3 (Global Environmental Drivers): Plots of standardized coefficients from the SAR models for the global-scale analysis, illustrating temporal trends in the environment-richness relationships.
  • Figure 4 (Realm-Specific Environmental Drivers): Plots of standardized coefficients from the SAR models for each biogeographic realm, showing the contrast in environment-richness relationships between 1920 and 2020.

2. Data

The data is organized into four folders:

• Folder: species_list
  • Description: This is the taxonomy dataset for four vertebrate groups. Data files are named using the pattern [taxonomic_group]_list_[number_of_species].csv.
  • Software: This is a standard comma-separated values (CSV) file. It can be opened and analyzed in any software that supports tabular data, including R and Excel.
  • Column Metadata: A full metadata describing each column's name and definition is provided in the accompanying metadata.csv file.
• Folder: diversity_table
  • Description: This is the diversity dataset for four vertebrate groups, including species richness (sr) and range-weighted rarity (rr). Data files are named using the pattern [taxonomic_group]_[diversity_type].csv.
  • Software: This is a standard comma-separated values (CSV) file. It can be opened and analyzed in any software that supports tabular data, including R and Excel.
  • Column Metadata: A full metadata describing each column's name and definition is provided in the accompanying metadata.csv file.
• Folder: sar_result
  • This folder contains the complete output from the Simultaneous Autoregressive (SAR) regression models used to analyze the relationships between vertebrate species richness and environmental drivers.
    • for_globe/: Contains SAR model results for models fitted at the global scale for each vertebrate group [taxonomic_group]_sar_globe.RData, each .RData file can be opened and analyzed in R by load() and spatialreg::summary() function.
    • for_realm/: Contains SAR model results for models fitted separately for each of the six biogeographic realms (Australian, Neotropical, Nearctic, Palearctic, Ethiopian, Oriental) [taxonomic_group]_sar_[realm].RData, each .RData file can be opened and analyzed in R by load() and spatialreg::summary() function.
    • sar_globe.csv and sar_realm.csv: Summarized model coefficients across multi spatial scales, prepared for figure visualization.  A full metadata describing each column is provided in the accompanying metadata.csv file.
• Folder: shapefile
  • The hexagonal grid geometry (.shp, .dbf, .shx, etc.) required for all spatial mapping can be opened in R and QGIS.

Usage

a. Set the R working directory to the root folder of this repository.

b. Run the discoveries_refine_diversity.R script. It will load the required data from the subfolders and execute the analysis.

c. Ensure all required R packages (e.g., spdep, sf, ggplot2; see R script for complete list) are installed.