Climate, soil resources and microbial activity shape the distributions of mountain plants based on their functional traits

International audienceWhile soil ecosystems undergo important modifications due to global change, the effect of soil properties on plant distributions is still poorly understood. Plant growth is not only controlled by soil physico-chemistry but also by microbial activities through the decomposition of organic matter and the recycling of nutrients essential for plants. A growing body of evidence also suggests that plant functional traits modulate spe-cies' response to environmental gradients. However, no study has yet contrasted the importance of soil physico-chemistry, microbial activities and climate on plant species distributions, while accounting for how plant functional traits can influence species-specific responses. Using hierarchical effects in a multi-species distribution model, we investigate how four functional traits related to resource acquisition (plant height, leaf carbon to nitrogen ratio, leaf dry matter content and specific leaf area) modulate the response of 44 plant species to climatic variables, soil physico-chemical properties and microbial decomposition activity (i.e. exoenzymatic activities) in the French Alps. Our hierarchical trait-based model allowed to predict well 41 species according to the TSS statistic. In addition to climate, the combination of soil C/N, as a measure of organic matter quality, and exoenzymatic activity, as a measure of microbial decomposition activity, strongly improved predictions of plant distributions. Plant traits played an important role. In particular, species with conservative traits performed better under limiting nutrient conditions but were outcompeted by exploitative plants in more favorable environments. We demonstrate tight associations between microbial decomposition activity, plant functional traits associated to different resource acquisition strategies and plant distributions. This highlights the importance of plant-soil linkages for mountain plant distributions. These results are crucial for biodiversity modelling in a world where both climatic and soil systems are undergoing profound and rapid transformations