环境变化
变化(天文学)
生物扩散
复制
草原
生态学
随机建模
气候变化
环境科学
生态系统
生物
数学
统计
人口学
人口
社会学
物理
天体物理学
作者
Ximei Zhang,Eric R. Johnston,Wei Liu,Linghao Li,Xingguo Han
摘要
Abstract Both ‘species fitness difference’‐based deterministic processes, such as competitive exclusion and environmental filtering, and ‘species fitness difference’‐independent stochastic processes, such as birth/death and dispersal/colonization, can influence the assembly of soil microbial communities. However, how both types of processes are mediated by anthropogenic environmental changes has rarely been explored. Here we report a novel and general pattern that almost all anthropogenic environmental changes that took place in a grassland ecosystem affected soil bacterial community assembly primarily through promoting or restraining stochastic processes. We performed four experiments mimicking 16 types of environmental changes and separated the compositional variation of soil bacterial communities caused by each environmental change into deterministic and stochastic components, with a recently developed method. Briefly, because the difference between control and treatment communities is primarily caused by deterministic processes, the deterministic change was quantified as (mean compositional variation between treatment and control) – (mean compositional variation within control). The difference among replicate treatment communities is primarily caused by stochastic processes, so the stochastic change was estimated as (mean compositional variation within treatment) – (mean compositional variation within control). The absolute of the stochastic change was greater than that of the deterministic change across almost all environmental changes, which was robust for both taxonomic and functional‐based criterion. Although the deterministic change may become more important as environmental changes last longer, our findings showed that changes usually occurred through mediating stochastic processes over 5 years, challenging the traditional determinism‐dominated view.
科研通智能强力驱动
Strongly Powered by AbleSci AI