Search

Long-term overgrazing may lead to the degradation of grasslands which are often characterized by an increase in nonpreferred species, especially toxic plants. However, the impact of these toxic nonpreferred species on the restoration processes of degraded grasslands is not well understood, particularly their interactions with soil properties and other plant functional groups. To address this knowledge gap, we conducted an in situ grazing exclusion experiment in a temperate degraded grassland of Inner Mongolia, China. The objective of this study was to investigate how toxic nonpreferred plants influence the recovery of plant diversity and productivity in degraded grasslands and whether these effects can be explained by changes in soil properties. Our findings revealed that Stellera chamaejasme, a toxic nonpreferred species widely distributed in North China, directly altered plant community composition and improved species diversity in degraded grasslands dominated by Asteraceae plants. The presence of S. chamaejasme could inhibit Asteraceae abundance and increase soil copper content in this study area, because Asteraceae plants have a high copper accumulation capacity. Within the communities with S. chamaejasme, the alleviation of soil copper limitation to plants may subsequently enhance the abundance and aboveground productivity of Poaceae and Forbs. Our study demonstrated that the strong direct and indirect interactions of toxic nonpreferred species with other ecosystem components promoted competitive release in terms of biomass accumulation and species diversity. The change of soil limiting microelements content caused by toxic species exerts an important mediation function during the recovery process of degraded grasslands. Thus, these toxic nonpreferred species can act primarily as accelerators for the restoration of community structure and ecosystem function in degraded grasslands., Competing Interests: Declaration of competing interest The authors declare that they have no conflict of interest., (Copyright © 2024 Elsevier B.V. All rights reserved.)