Introduction: In the Alpine countries, many summer pastures have been under-exploited or abandoned, thus favouring the development of encroachment and secondary succession. These processes remain poorly documented, and the awareness of an ongoing evolution occurs often too late for reversing the process with simple management measures. The present work, which is the first part of a long-term survey, aims at describing the different types of early succession, with particular emphasis on plant species that could serve as indicators. We made two hypotheses: (i) the type of succession is strongly dependent on the ecological factors, especially aspect and soil depth, and (ii) the pastoral value of those early-successional vegetation types is low. Materials and methods: The study was conducted on 17 summers pastures located in the western part of the Swiss Alps. Overall, we carried out 45 relevés within 25 m2 areas (i.e. two or three relevés for each pasture) according to the method of Braun-Blanquet. Soil depth, slope, aspect, elevation, mean annual temperature and rainfall, type of soil (lithology) were determined. Based on species abundance, we calculated the mean values of Landolt ecological indices, Shannon-Weaver index, as well as pastoral values (calculated using the indices of specific quality of Cavallero et al., 2007). A Redundancy Analysis was run in order to identify the ecological factors significantly associated to botanical composition. Finally, based on Jaccard/Ward clustering, we proposed a classification of the main vegetation types. Results: In total, 268 species were recorded. Table 1 presents the classification in 5 vegetation types resulting from the clustering. As expected, we found different plant communities that primarily relate to environmental factors. The RDA analysis (data not shown) indicated that factors such as aspect, slope, elevation, and precipitation were significant, whereas depth of soil was not. Furthermore, pH (i.e. Landolt’s Reaction value) strongly determined botanical composition. The pastoral values were low, in accordance with our hypothesis. Conclusion: this study confirmed that the vegetation in the early successional stages is very diverse but of low pastoral value. In absence of precise data on the management, it is difficult to gain a complete understanding of the secondary succession processes. A second survey in a few years is planned, in order to link the different spatial and temporal scales and gain a better insight into the process.
