Résumé
Rapid warming is a major threat for the alpine biodiversity but, at the same time, accelerated glacial retreat constitutes an opportunity for taxa and communities to escape range contraction or extinction. We explored the first steps of plant primary succession after accelerated glacial retreat under the assumption that the first few years are critical for the success of plant establishment. To this end, we examined plant succession along a very short post-glacial chronosequence in the tropical Andes of Ecuador (2–13 years after glacial retreat). We recorded the location of all plant individuals within an area of 4,200 m2 divided into plots of 1 m2. This sampling made it possible to measure theresponses of the microenvironment, plant diversity and plants traits to time since the glacial retreat. It also made it possible to produce species-area curves and to estimate positive interactions between species. Decreases in soil temperature, soil moisture, and soil macronutrients revealed increasing abiotic stress for plants between two and 13 years after glacial retreat. This increasing stress seemingly explained the lack of positive correlation between plant diversity and time since the glacial retreat. It might explain the decreasing performance of plants at both the population (lower plant height) and the community levels (lower species richness and lower accumulation of species perarea). Meanwhile, infrequent spatial associations among plants indicated a facilitation deficit and animal-dispersed plants were almost absent. Although the presence of 21 species on such a small sampled area seven years after glacial retreat could look like a colonization success in the first place, the increasing abiotic stress may partly erase this success, reducing species richness to 13 species after 13 years and increasing the frequency of patches without vegetation. This fine-grain distribution study sheds new light on nature’s responses to the effects of climate change in cold biomes, suggesting that faster glacial retreat would not necessarily result in accelerated plant colonization. Results are exploratory and require site replications for generalization.