Eradication of invasive Carpobrotus sp.: effects on soil and vegetation

International audienceInvasive species management (eradication or control) can be used to promote native plant restoration. The objective of this study is to evaluate different treatments to guide the selection of future modalities for the eradication (i.e. elimination of all individuals in a population) of Carpobrotus sp. from a strict nature reserve. Two removal methods were tested: (1) living Carpobrotus removal; (2) living Carpobrotus and litter removal. To assess the effectiveness of each treatment, we studied the recolonization of native vegetation, the recolonization of Carpobrotus, and soil erosion and compared these metrics to those taken in native vegetation and in patches of intact Carpobrotus. We also tested the capacity of a 50‐cm‐wide Carpobrotus strip to retain soil. The removal of Carpobrotus together with its litter led to high rates of soil erosion. The Carpobrotus strips were found to retain the soil rather well. Removing live Carpobrotus while leaving its litter in place reduced soil erosion and led to higher native plant species recolonization. The composition of the vegetation 10 months after applying the treatments was biased in favor of native pioneer species. These are typically the first species to establish (Aetheoriza bulbosa and Arisarum vulgare resprouted, Frankenia hirsuta and Lotus cytisoides germinated, and Sonchus sp. benefited from long‐distance dispersal). Few weedy species were recorded (e.g. Sonchus asper asper). Whatever the treatment, the risk of reinvasion from the seed bank or from resprouting stems is nonnegligible, so long‐term monitoring is vital to the ultimate success of the eradication program