Global Patterns of Species Richness Across Terrestrial Biomes

Abstract

Understanding global patterns of species richness across terrestrial biomes is fundamental to macroecology and conservation biogeography, yet the relative importance of energy, climate stability, area, evolutionary history, and biotic interactions in determining large-scale diversity gradients remains debated. This study analysed vertebrate and vascular plant species richness patterns across all 14 major terrestrial biomes (Olson et al. 2001 classification) using species richness data from GBIF (2,840,000 occurrence records cleaned and aggregated to 1 degree grid cells) and nine biome-level environmental predictors compiled from ERA5 reanalysis, Copernicus Land Cover, and WorldClim v2.1. Species richness was highest in tropical moist broadleaf forests (TMBF; vertebrate richness mean 484 +/- 84 species per 1 degree cell; vascular plant richness 2,840 +/- 484 per cell) and lowest in tundra (vertebrate 48 +/- 12; plants 284 +/ 84). Structural equation modelling (SEM) confirmed that mean annual temperature (path coefficient -0.748; directly reflecting latitudinal energy gradient) and precipitation seasonality (-0.384) were the strongest biome-level richness predictors, collectively explaining 72.4% of the variance in vertebrate richness across biomes. Biome area contributed a significant but weaker independent effect (partial r = +0.48; p < 0.001). Evolutionary age (estimated from molecular phylogeny crown ages) explained an additional 14.4% of variance in biome richness beyond climate variables -- highest in TMBF, where longer uninterrupted evolutionary time has accumulated species richness through diversification. A Biome Richness Vulnerability Index (BRVI) combining current richness, climate velocity, and species thermal sensitivity identified TMBF and montane grasslands as the two highest-priority biomes for biodiversity monitoring and conservation under climate change.