GDRarity

An R package for the Global Model of Discretized Rarity

Overview

The Global Model of Discretized Rarity (GDR) integrates geographic, functional, and phylogenetic dimensions of rarity at regional and local scales. GDRarity is an R package that characterizes the GDR and its 63 restrictions.

Its webpage is at https://anytko.github.io/GDRarity/; its source code is at https://github.com/anytko/GDRarity.

Installation

You can install the development version of GDRarity like so:

devtools::install_github("anytko/GDRarity")

Features

• Automates restriction selection based on available data.
• Characterizes the rarity of species using multiple restrictions simultaneously.
• Tests various methods of calculating geographic, functional, and phylogenetic dimensions of rarity and rarity thresholding.
• Analyzes restriction performance in explaining a response variable (i.e. biological processes).
• Accepts additional, user specified dimensions of rarity to create or augment new restrictions of the GDR.

Caveats

• Only pulls plant trait data from the LEDA and BIEN plant trait databases.
• Certain methods, such as calculating regional geographic rarity using extent of occurrence (range size) requires an internet connection and can be computationally intensive.
• Does not incorporate Rabinowitz Rarity as a GDR restriction, but habitat specificity can be added as an additional, user specified dimension of rarity.

Example

The 63 restrictions of the GDR can be applied using phylogeny, species, abundance, and trait data within one function.

library(GDRarity)

species_70_path <- system.file("extdata", "species_70.csv", package = "GDRarity")
abundance_70_path <- system.file("extdata", "abundance_70.csv", package = "GDRarity")

species_70 <- read.csv(species_70_path)
abundance_70 <- read.csv(abundance_70_path)

all_restrictions <- gdrare_pipeline(
      species_df = species_70,
      abundance_df = abundance_70,
      trait_cols = c("SLA", "seed_mass", "canopy_height"),
      geo_rarity_method = "taxonomic",
      fun_rarity_method = "min_distance")

Results of Restriction GRFLPR

all_restrictions$classified_df %>%
  select(species, GR_raw, FL_raw, PR_raw, GRFLPR) %>%
  head(6) %>%
  mutate(across(where(is.numeric), ~ round(., 3))) %>%  
  knitr::kable()

species	GR_raw	FL_raw	PR_raw	GRFLPR
Acer_campestre	0.656	0.409	2.193	GR-FL+PR-
Alnus_glutinosa	0.447	0.237	2.442	GR-FL-PR-
Alopecurus_myosuroides	0.825	0.144	-1.178	GR-FL-PR-
Anthemis_cotula	0.871	0.134	-0.008	GR-FL-PR-
Arum_maculatum	0.565	0.423	3.105	GR-FL+PR+
Calluna_vulgaris	0.446	0.129	-0.564	GR-FL-PR-