放牧
生物多样性
草原
保护性放牧
生态系统
生物量(生态学)
多样性指数
生态学
濒危物种
放牧压力
环境科学
物种多样性
生物
物种丰富度
栖息地
作者
Yue Yan,Jingyi Ding,Jing Wang,Shurong Zhang,Ao Zhou,Yi Han,Xuan Gao,Wenwu Zhao
摘要
Abstract Biodiversity loss by grazing pressure has threatened multiple ecosystem functions simultaneously (ecosystem multifunctionality, EMF) in grasslands. Therefore, detecting the relation between EMF and biodiversity under human intervention would provide guidance on the optimization of sustainable grassland management. However, little is known about how grazing effects grassland EMF by mediating biodiversity under different grazing intensities. Here, species and functional diversity were investigated along a longitudinal gradient in Inner Mongolia grassland, and its EMF was quantified comprehensively on six ecosystem functions by an averaging approach, threshold approach and multiple threshold approach to assess the impact of grazing intensity on EMF. There was a significantly lower average EMF index and above‐ground biomass under heavy grazing (HG) than under light grazing (LG). The threshold approach‐based EMF and average EMF index were positively related to species diversity and functional diversity, and species diversity was a stronger driver of EMF than functional diversity. Compared with LG, the slope of the positive relation between EMF and species diversity significantly weakened by 40%–86% under moderate grazing (MG) and decreased by 30%–100% under HG, and reversed to a negative relation between functional diversity and EMF under HG. Additionally, species diversity was more negatively influenced by soil pH and sand content, while functional diversity was more negatively influenced by grazing intensity, indicating that functional diversity was more sensitive to intensive grazing. This study highlights the role of grazing intensity for driving multiple ecosystem functions, and reinforces the significance of managing both species diversity and functional diversity for sustaining productive functional grasslands.
科研通智能强力驱动
Strongly Powered by AbleSci AI