Aim: Intraspecific trait variation (ITV) within natural plant communities can be large, influencing local ecological processes and dynamics. Here, we shed light on how ITV in vegetative and floral traits responds to large‐scale abiotic and biotic gradients (i.e., climate and species richness). Specifically, we tested whether associations of ITV with temperature, precipitation and species richness were consistent with any of four hypotheses relating to stress tolerance and competition. Furthermore, we estimated the degree of correlation between ITV in vegetative and floral traits and how they vary along the gradients.Location: Global.Time period: 1975–2016.Major taxa studied: Herbaceous and woody plants.Methods: We compiled a dataset of 18,401 measurements of the absolute extent of ITV (measured as the coefficient of variation) in nine vegetative and seven floral traits from 2,822 herbaceous and woody species at 2,372 locations.Results: Large‐scale associations between ITV and climate were trait specific and more prominent for vegetative traits, especially leaf morphology, than for floral traits. The ITV showed pronounced associations with climate, with lower ITV values in colder areas and higher values in drier areas. The associations of ITV with species richness were inconsistent across traits. Species‐specific associations across gradients were often idiosyncratic, and covariation in ITV was weaker between vegetative and floral traits than within the two trait groups.Main conclusions: Our results show that, depending on the traits considered, ITV either increased or decreased with climate stress and species richness, suggesting that both factors can constrain or enhance ITV, which might foster plant‐population persistence in stressful conditions. Given the species‐specific responses and covariation in ITV, associations can be hard to predict for traits and species not yet studied. We conclude that consideration of ITV can improve our understanding of how plants cope with stressful conditions and environmental change across spatial and biological scales.

A modelagem preditiva tem sido aplicada para analisar a distribuição geográfica de espécies, a partir de extrapolações das características ambientais dos locais conhecidos de ocorrência. O interesse por esse tipo de modelagem deve-se à necessidade de respostas rápidas e fundamentadas para as ameaças que as espécies têm enfrentado, devido à perda de habitat, invasão de espécies exóticas, mudanças climáticas, entre outros. Este artigo oferece uma visão geral dos avanços recentes no campo da modelagem e visa incentivar a discussão e aplicação desse método, que pode auxiliar tanto na aquisição de conhecimento básico sobre a biologia das espécies, quanto na análise e formulação de políticas para sua conservação.