Does delaying the first mowing date benefit biodiversity in meadowland?

Background Meadows are regularly mown in order to provide fodder or litter for livestock and to prevent vegetation succession. However, the time of year at which meadows should be first mown in order to maximize biological diversity remains controversial and may vary with respect to context and focal taxa. We carried out a systematic review and meta-analysis on the effects of delaying the first mowing date upon plants and invertebrates in European meadowlands. Methods Following a CEE protocol, ISI Web of Science, Science Direct, JSTOR, Google and Google Scholar were searched. We recorded all studies that compared the species richness of plants, or the species richness or abundance of invertebrates, between grassland plots mown at a postponed date (treatment) vs plots mown earlier (control). In order to be included in the meta-analysis, compared plots had to be similar in all management respects, except the date of the first cut that was (mostly experimentally) manipulated. They were also to be located in the same meadow type. Meta-analyses applying Hedges’d statistic were performed. Results Plant species richness responded differently to the date to which mowing was postponed. Delaying mowing from spring to summer had a positive effect, while delaying either from spring to fall, or from early summer to later in the season had a negative effect. Invertebrates were expected to show a strong response to delayed mowing due to their dependence on sward structure, but only species richness showed a clearly significant positive response. Invertebrate abundance was positively influenced in only a few studies. Conclusions The present meta-analysis shows that in general delaying the first mowing date in European meadowlands has either positive or neutral effects on plant and invertebrate biodiversity (except for plant species richness when delaying from spring to fall or from early summer to later). Overall, there was also strong between-study heterogeneity, pointing to other major confounding factors, the elucidation of which requires further field experiments with both larger sample sizes and a distinction between taxon-specific and meadow-type-specific responses.