Early changes of orthopteran assemblages after grassland restoration: a comparison of space-for-time substitution versus repeated measures monitoring

Although grasslands harbour significant biodiversity and their restoration is common in biodiversity conservation, we know very little about how such interventions influence arthropod groups. Here we compared orthopteran assemblages in croplands, natural grasslands and one to four-year-old grasslands restored in a large-scale programme in Hortobágy National Park (East Hungary). We sampled orthopterans by standardized sweep-netting both in a repeated measures design from Year 0 (croplands) to 4 and in a space-for-time substitution (chronosequence) design in 2009. Species richness, abundance and Shannon diversity of orthopterans decreased in Year 1 following restoration, but increased afterwards. By Year 4, species richness doubled and abundance increased almost ten-fold in restored grasslands compared to croplands. Species composition diversified compared to croplands and progressed towards natural grasslands. Local restoration conditions (last crop, seed mixture) and landscape configuration (proportion of natural grasslands) did not influence the above patterns in either study design, whereas time since restoration affected almost all community variables. We found that ubiquitous generalist species were the first to appear in restored grasslands and that species characteristic to the target natural grasslands colonised gradually in later years. The qualitative and quantitative properties of the orthopteran assemblages in restored fields did not yet reach those of natural grasslands, therefore, our study suggests that the full regeneration of the orthopteran assemblages takes more than four years. We also concluded that the repeated-measures design was more sensitive to subtle changes and was thus more effective than the chronosequence design at detecting post-restoration changes in orthopteran assemblages.