Changes in the species composition and conservation value of limestone grasslands in Northern Switzerland after 22 years of contrasting managements

Abstract

Abstract An experiment was established to assess the long-term effects of six different management types on the species composition of dry limestone grasslands with the aim of finding an alternative to the traditional management (mowing annually in July) which is no longer economically viable. In addition to the traditional management (JU1), the following management types were applied experimentally for 22 years: mowing every 2 years in July (JU2), every 5 years (JU5), or not at all (ABN, abandonment), mowing annually in October (OCT), and controlled burning annually (CBR) at the end of winter. There were two study sites, Grate and Gratental, near Schaffhausen in N Switzerland, each with three replicate 5 m×10 m plots for every management. During the course of the experiment, the vegetation was recorded on 10 occasions, yielding 306 vegetation releves containing a total of 124 species. These data were analysed using correspondence analyses (CA). In addition, changes in the abundance of 16 plant species with high cover, high frequency or particular importance for nature conservation were examined in detail. The CA showed that the vegetation of the two study sites differed considerably, but that in both sites the various management types led to changes in floristic composition which were probably continuing even after 22 years. At the end of the experiment, on the Grate site, species number in comparison to JU1 had declined in the following progressive order: OCT, JU2, JU5, CBR, ABN, where it was down to 70%. At the Gratental site, species number showed no clear pattern, because of the immigration of ‘new’ species from the surrounding hedges and agricultural fields. Interestingly, Bromus erectus , the dominant species with a cover of 40–70% at the beginning, decreased to 5–30% under different managements, mainly because the regeneration of its tufts is affected by abandonment and particularly burning. The rhizomatous Brachypodium pinnatum showed the opposite pattern and increased from 5% to 60% in CBR. Trifolium medium as well as Salvia pratensis , Primula veris and other long-lived forbs of high conservation value tended to be favoured by OCT, CBR and/or ABN because they can store nutrients if they are not annually cut in July. On the other hand, several low growing species like Anthyllis vulneraria and Hieracium pilosella need the better light conditions associated with the annual mowing in July for optimal growth. In every management type, some species of high conservation value decreased whereas others increased! According to our results, the best strategy—for achieving and maintaining a high species diversity—is a mosaic of different management types, i.e. a central part of the grassland managed traditionally, i.e. mown annually in July, and large strips at the margins mown only every second year in July or October. When mown in October, the soil has to be nutrient-poor and the site isolated from the seed rain of forest fringe species, which otherwise could become dominant and lead to a decline in species. The alternative of mowing the whole area every second year in July or annually in October (with the same restrictions for mowing in October) it is not completely satisfactory because several species may be lost in the long term.