The combined effects of climatic and land use changes severely affect all ecosystems. However, increasing evidence that high-elevation regions are the most impacted by increasing temperatures (MRI 2015; Wang et al. 2016) suggests that alpine ecosystems are at very high risk. In temperate alpine regions, relatively abundant literature has permitted the building of scenarios for the future of the alpine biodiversity and ecosystem services at these latitudes (e.g. Thuiller et al. 2008; Pauli et al. 2014). A majority of species will need to perform an upward migration to escape the effects of warming, whereas a minority might adapt to their new environment through plasticity, genetic adaptation or by being spatially associated with other species (Anthelme et al. 2014; Lenoir & Svenning 2015). Such knowledge is drastically lacking in the alpine tropics, which represent 10% of the total alpine areas worldwide (Jacobsen 2008), and often are biodiversity hotspots and providers of crucial ecosystem services. Beyond belonging to the same alpine biome constrained by very cold minimum temperatures, temperate and tropical alpine ecosystems are governed by distinctive parameters. The main difference comes from the low effects of seasonality in the tropics, leading to higher temperature extremes between day and night than between summer and winter (Körner 2003). A direct consequence is the (almost) absence of snow cover in the tropics whereas snow cover duration generally reaches 6-10 months in temperate latitudes (Meneses et al. 2015). Snow cover is considered the first driver of plant distribution in high latitude alpine regions (Wipf & Rixen 2010); it is expected to reduce significantly because of climate warming during the next decades (Wipf et al. 2009). Accordingly, we expect tropical alpine and temperate alpine plant communities to: (1) be different in composition, life form and function, and (2) respond differently to current and future warming. In parallel, whereas annual precipitation increase in most case with elevation in temperate regions, they decrease beyond approx. 3000-3500 m in tropical regions, resulting possibly in tropical alpine deserts, which represent an additional constraint for plant development (Crausbay & Hotchkiss 2010, Anthelme & Dangles 2012). A number of studies describing alpine plants observed that life forms in the tropics are more diverse, including tall life forms such as tussock grasses, large cushion-forming plants and giant stemmed rosettes (Hedberg & Hedberg 1979). Tropical alpine environments sometimes harbour trees at extreme alpine elevation, such as Polylepis tarapacana (5200 m, Bolivia; Hoch & Körner 2005) or Pinus hartwegii (4850 m, Mexico; M. A. morales, pers. obs.). Beyond the inherent value of this endemic biodiversity, such life forms may have a significant impact on ecosystem function. We hypothesize that root systems of such plants may have unique form or functions, resulting in differences in carbon storage, as well as affecting soil physico-chemical composition and soil biodiversity. The singular structures of such species may provide enhanced biotic refuges for other plants (nurse plant syndrome), especially against the effects of warming (Anthelme & Dangles 2012, Anthelme et al. 2014). This situation may lead to more facilitative interactions among alpine plants in the tropics than under temperate latitudes, thus changing the functioning of alpine ecosystems. The PhD student will conduct a comparative study between a temperate alpine region (the French Alps, Chamrousse) and a tropical alpine region (Pico de Orizaba, Vera Cruz State, Mexico) with the aim to estimate the effects of climatic changes on plant communities and ecosystem services. In each site, an extended altitudinal gradient from the forest to the upper limit of the alpine belt will be used as a proxy for variations in climate, soil, and the structuring vegetation (2000-3000 m in France, 3000-5000 m in Mexico). Along this gradient at each elevation, he/she will first correlate the duration of snow cover with the composition, abundance and life form of the vegetation. Second, he/she will study the specific characters developed by the plants, especially at leaf level (hairiness, stomatal density) and the influence of the most characteristic plants on the soil and root properties. Third, he/she will quantify at the community scale the frequency of facilitative interactions among plants. In parallel with these three observations, the student will implement an in situ experiment by transplanting blocks of extracted from the forest in the alpine sites at different elevations (tree species in France: Pinus cembro; Mexico: Pinus hartwegii). During two years, he/she will measure the performance of seedlings (functional traits).