How mowing restores montane Mediterranean grasslands following cessation of traditional livestock grazing

Abstract

Traditional land-use cessation allows the spread of invasive tall-grass species and thus leads to a reduction in grassland biodiversity. We hypothesized that long-term multi-year mowing of invaded grassland fosters the recovery of its taxonomic diversity and functional composition by reducing the dominance of the tall grass species Brachypodium rupestre, and that this change increases the pasture feed value. In 2010, we fenced part of a grassland, abandoned for about 30 years, invaded by Brachypodium rupestre (1 ha) in the central Apennine ridge (1,000–1,300 m a.s.l.), and cut it twice a year for six years; another area (1 ha) was fenced and left unmown. Before the experiment started, we recorded species cover in 30 random sampling units (0.5 × 0.5 m) in the experimental area and in the unmown area. The sampling was repeated for the experimental area, every year for six years, except 2013, while it was done for the unmown area only at the end of the six years. We investigated the effect of the reduction of Brachypodium rupestre over time on taxonomic diversity indices, functional traits and pastoral value, using species accumulation curves and generalized linear mixed-effect modelling.Twice-yearly mowing was effective in reducing the abundance of B. rupestre (mean cover 50.7 ± 19.8 % Standard deviation to 9.0 ± 7.2 %), and in increasing significantly species richness (16.6 ± 3.0–26.8 ± 3.2), exponential Shannon (8.8 ± 2.8–14.4 ± 2.7), but not Shannon evenness (0.53 ± 0.1 to 0.54 ± 0.1). At the end of the treatment quite stable richness and diversity values were reached. Reduction of B. rupestre cover increased the pastoral value of the system (39 ± 12 %–64 ± 8 % of pabular species relative cover) and promoted the presence of species without vegetative propagation (17.8 ± 12.3–28.2 ± 10.9), with pleiocorms (25.4 ± 14.3–49.6 ± 17.9), rosettes (5.6 ± 5.6–16.6 ± 12.7) and hemirosettes (32.1 ± 15.8–47.7 ± 13.6), but caused a decrease in species whose clonal growth organs have prevalent vertical spread (123.9 ± 25.7–97.0 ± 23.3), reptant (60.3 ± 20.9–20.2 ± 10.4), and late flowering species (67.4 ± 19.9–46.3 ± 17.2). The decrease of B. rupestre and the increase in mowing frequency reduced the abundance of species with clonal growth organs with prevalent horizontal spread (66.0 ± 19.5–17.9 ± 10.7), caespitose species (96.7 ± 20.9–43.6 ± 21.9), and grasses (94.0 ± 19.5–43.3 ± 22.0). Recurrent twice-yearly mowing seems to be appropriate in the restoration of sub-Mediterranean grasslands invaded by competitive species, facilitating the re-introduction of traditional grazing.