Removal of invasive iceplant Carpobrotus on a Mediterranean island: 10-year monitoring results on the beetle and spider assemblage dynamics
Résumé
Mediterranean island biodiversity is threatened by invasive plants, especially by Carpobrotus species. Invasive plant control is a major management tool for the restoration of invaded areas, but the effect of Carpobrotus removal on arthropods has not been studied yet. In this study, we evaluate how beetle and spider assemblages are influenced by Carpobrotus removal on a Mediterranean island. A 0.5-ha Carpobrotus patch was manually uprooted followed by litter removal between November 2011 and February 2012. A nearby site with native matorral vegetation and no management was also studied. Beetle and spider communities were sampled using pitfall traps two years before removal (2010-2011), and then every two years after, between 2013 and 2019. Vegetation at the Carpobrotus site changed from a dense mat of Carpobrotus to a diverse halophilous grassland with some chamaephytes, while it remained stable in the matorral. Beetle and spider species richness increased significantly after removal and significant changes in assemblage composition were observed between pre-and post-removal years with a high turnover rate, while beetle and spider assemblages remained relatively stable in the matorral. Detritivore beetles (e.g. Asida dejeani and Arthrolips spp.) and predators (e.g. Oligota muensteri and Harpalus attenuatus) were the species most enhanced by the restoration. Netocia oblonga, a floricolous phytophage, was negatively impacted. Litter-dwelling spiders with trap strategy (e.g. Oecobius navus) were the most reduced by Carpobrotus removal, while ground-dwelling spiders with active hunting strategy (e.g. Xysticus spp.) benefited most likely from the diversification of plant microhabitats. The increase in bare ground cover favored Zodarion elegans, Nomisia celerima, and Linyphiids, which are characterized by an affinity for dry sunny environments. This rapid change in beetle and spider assemblages within seven years of invasive plant removal showed that these two groups can be useful bioindicators of restoration processes.