Environmental Heterogeneity Drives the Dark Diversity of Global Plant Communities

Aim Environmental heterogeneity drives biodiversity across biomes and globally. Numerous studies have determined that environmental heterogeneity is the primary driver of the observed diversity of plant communities. Here, we aimed to explore whether environmental heterogeneity could drive the species diversity of plant communities based on the concept of dark diversity (i.e., absent taxa that ecologically and biogeographically fit a site). Location Global. Time Period Present day. Major Taxa Studied Terrestrial plants. Methods First, we used the hypergeometric method to estimate observed diversity, dark diversity, species pool and community completeness based on species pairwise associations. Path analysis was then used to estimate the environmental heterogeneity effects on observed diversity, dark diversity and community completeness through the species pool. Finally, multiple linear regressions with second‐order polynomials were used to account for the relationships observed in the plot data between four diversity metrics and environmental variables (heterogeneity in habitat, climate, soil and topography) based on global and biome‐level scales. Results There were significant relationships of heterogeneity observed in habitat, climate, soil and topography, together with observed diversity, dark diversity, species pool and completeness of plant communities at the global scale; however, the positive or negative effects depend on biome type. The relationships of observed diversity, dark diversity and species pool with evenness and Shannon index were stronger based on habitat heterogeneity than on other factors. Furthermore, environmental heterogeneity affected dark diversity and community completeness through the species pool and depended on the biome type. The diversity–heterogeneity relationships were strongest for dark diversity and species pool in tropical and subtropical coniferous forests, montane grasslands and shrublands, temperate conifer forests, Mediterranean forests, woodlands and scrubs. Conclusions Environmental heterogeneity, particularly habitat heterogeneity, should be used to predict plant species diversity based on the concept of dark diversity, depending on the biome type. We suggest integrating observed and dark diversity frameworks with environmental heterogeneity, particularly habitat heterogeneity, into biodiversity conservation and ecological restoration to enhance the effectiveness of environmental management worldwide.