生态系统
生产力
初级生产
环境科学
生态学
草原
碳循环
生态稳定性
全球变化
扰动(地质)
气候变化
降水
脆弱性(计算)
初级生产力
抗性(生态学)
营养循环
生物量(生态学)
作物生产力
农林复合经营
营养物
农学
草地生态系统
全球变暖
陆地生态系统
环境变化
生物多样性
交替稳态
碳纤维
生态系统生态学
土壤碳
氮气循环
作者
Zhang Ze,Liu Hong-Yan,Li, Zidong,LIANG Boyi,Qi JingHua,Li JiaMei,Hautier Yann
标识
DOI:10.6084/m9.figshare.29609360.v3
摘要
Grassland ecosystems are central regulators of global carbon cycling and biodiversity, yet it remains unclear whether belowground productivity consistently maintains greater stability compared to aboveground productivity across diverse environmental perturbations. Previous studies have predominantly focused on aboveground productivity stability, potentially overestimating ecosystem vulnerability by neglecting critical belowground processes. By synthesizing 1,513 experimental observations from 113 studies spanning 85 grasslands worldwide, we quantify the responses of productivity, temporal stability, and carbon allocation to nine global change drivers, including nutrients enrichment, precipitation shifts, elevated CO₂, warming, mowing and grazing. Our results reveal that belowground net primary productivity (BNPP) consistently exhibits high temporal stability across all global change drivers, highlighting an intrinsic resistance of belowground processes to environmental fluctuations. In contrast, aboveground net primary productivity (ANPP) stability is significantly influenced by climatic variability, particularly precipitation and aridity. These distinct patterns suggest fundamental ecological asymmetry: rapid, climate-sensitive aboveground processes contrast markedly with slower, soil-buffered belowground dynamics. Notably, shifts in carbon allocation toward belowground components emerge as a critical ecological mechanism enhancing and sustaining belowground stability. Our findings challenge the notion of unified top-down control of grassland functioning and provide a mechanistic framework for predicting ecosystem vulnerability and designing resilience-based grassland management under global environmental changes.
科研通智能强力驱动
Strongly Powered by AbleSci AI