Drivers of thermal tolerance breadth of plants across contrasting biomes

Abstract The climate variability hypothesis (CVH) predicts that species from environments with more variable temperatures should have wider thermal tolerance breadth. This hypothesis has not yet been tested thoroughly across diverse plants. Here, we asked how local climate predictors (including precipitation, mean and extreme temperatures and thermal variability) are associated with species physiological thermal limits. Measures of lower ( T crit‐cold ) and upper ( T crit‐hot ) photosystem II thermal tolerance thresholds were used to determine thermal tolerance breadth (TTB), along with ice nucleation temperature ( T nucleation , freezing tolerance) of 69 plant species sampled from the field across three contrasting biomes: alpine, desert and coastal temperate rainforest. All measured thermal tolerance metrics ( T crit‐cold , T nucleation , T crit‐hot and TTB) differed among biomes. Notably, desert species had the most cold and heat tolerant leaves, and therefore the widest TTB, whereas species in alpine and temperate biomes had similar TTB. For plants in all biomes, TTB exceeded the thermal range of their local climate. Overall, two principal component axes of local climate drivers explained substantial variation in all tolerance metrics. Extreme hot, dry climates improved freezing and heat tolerance. High thermal variability and low minimum temperatures also improved freezing tolerance but were unrelated to heat tolerance or TTB. Species explained a significant amount of variation among all metrics, but this was not due to phylogenetic relatedness. We discuss how the remaining variation could be due to microclimate‐driven plasticity, leaf traits or thermoregulatory mechanisms. Synthesis. Our results provide partial support for the climate variability hypothesis in plants: photosystem thermal tolerance breadth was greatest in more thermally variable biomes. This relationship was largely driven by cold tolerance, with variation in heat tolerance explained better by mean and extreme temperatures. Therefore, we conclude that the CVH alone is not sufficient to explain variation in plant thermal tolerance, with many other aspects of climate, environment and biology being potentially important drivers.