Proposition d’une démarche simple pour évaluer la diversité floristique d’une prairie permanente

Proposed simple approach to the assessment of the floristic diversity of permanent pastures

An easy method for the evaluation of a pasture's floristic diversity was developed in Auvergne, making it possible for a non-botanist to assess its bio-diversity. The aim was to make people working in the sphere of agriculture aware of the floristic diversity of ordinary grasslands. It relies on qualitative and quantitative observations made on pasture plots where different vegetation facies are distinguished. In our view, the floristic diversity of a pasture depends on the swards composition, but is also enhanced by the diversity of the landscape; we therefore assessed both these factors in our approach. The selection of indicators of the bio-diversity was based on the study of 126 vegetation facies, evenly distributed over four classes of pastures differing in their type of management (grazing or mowing) and in their environment (upland, lowland with a rich or a poor soil). We made for each facies both a botanical inventory and a simplified description of the vegetation, the fauna and the landscape features. To identify indicators, we looked for linear correlations between bio-diversity, as calculated from the inventories, and the data from the simplified descriptions. Nine indicators were found for the assessment of the herbaceous vegetation's diversity : total and average number of species with a different physiognomy, colours, forms of flowers... and one indicator was based on the diversity in butterflies. These indicators can be obtained on the field by a simplified counting, inside 10 circles of 1 m diameter for the flora, and along a transect for the fauna. They are more or less the same for each class, the main difference being in the identification thresholds, which are adapted to the Auvergne grasslands. The indicators of a given class were gathered in a single grid, in which five levels of bio-diversity were diagnosed. In addition to the assessment of each facies, we also propose a measure of the differences among the facies of the same plot. This point concerns only the number of species of different physiognomy. Separate grids were also designed for the assessment of the diversity of landscapes, as we did not find any correlation between the latter and the diversity of the herbaceous vegetation. Two indicators were used : the number and the types of landscape elements. The grids have also indication thresholds for each class and for the five levels of bio-diversity.
Finally, various ways of interpreting the results are proposed to the users of the method, based on the hypothesis that the diversity depends not only on management and environment, but also on the landscapes. The way landscape diversity can increase and maintain the floristic diversity of grasslands is also emphasized.