Global quantification of contrasting leaf life span strategies for deciduous and evergreen species in response to environmental conditions

Aim: Species with deciduous and evergreen leaf habits typically differ in leaf life span (LLS). Yet quantification of the response of LLS, within each habit, to key environmental conditions is surprisingly lacking. The aim of this study is to quantify LLS strategies of the two leaf habits under varying temperature, moisture and nutrient conditions, using a global database. We hypothesize that deciduous LLS reflects the length of the growing season, avoiding unfavourable conditions regardless of the cause. Evergreen species adjust to unfavourable periods and amortize lower net carbon gains over several growing seasons, with increasing LLS associated with increasingly short favourable versus unfavourable season lengths. Location: Global. Methods: Data on LLS and environmental variables were compiled from global datasets for 189 deciduous and 506 evergreen species across 83 study locations. Individual and combined effects of measures of seasonality of temperature, water and nutrient availability on length of the growing season and on LLS were quantified using linear mixed models. The best models for predicting LLS were obtained using Akaike's information criterion (AIC) and ΔAIC. Results: The LLS of deciduous and evergreen species showed opposite responses to changes in environmental conditions. Under unfavourable conditions, deciduous LLS decreases while evergreen LLS increases. A measure of temperature alone was the best predictor of the growing season. The LLS of deciduous species was independent of environmental conditions after expressing LLS in relation to the number of growing seasons. Evergreen species, on the other hand, adjusted to unfavourable conditions by increasing LLS up to four growing seasons. Contrary to expectations, variation in LLS was best explained solely by temperature, instead of by combined measures of temperature, moisture and nutrient availability. Shifts in the photosynthesis to respiration balance might provide a physiological explanation. Main conclusions: Temperature, and not drought or nutrient availability, is the primary driver of contrasting responses of LLS for different leaf habit types