There is compelling experimental evidence and theoretical predictions that temporal stability of productivity, i.e., the summation of aboveground biomass growth of surviving and recruitment trees, increases with succession. However, the temporal change in productivity stability in natural forests, which may undergo functional diversity loss during canopy transition, remains unclear. Here, we use the forest inventory dataset across the eastern United States to explore how the temporal stability of forest productivity at multi-spatial scales changes with stand age during canopy transition. We find that productivity stability decreases with stand age at the local and metacommunity scales. Specifically, consistent declines in local diversity result in less asynchronous productivity dynamics among species over succession, consequently weakening local stability. Meanwhile, increasing mortality and the transition from conservative to acquisitive species with succession weaken species and local stability. Successional increases in species composition dissimilarity among local communities cause more asynchronous productivity dynamics among local communities. However, the decline in local stability surpasses the rise in asynchronous productivity dynamics among local communities, resulting in lower metacommunity stability in old forests. Our results suggest lower productivity stability in old-growth forests and highlight the urgency of protecting diversity at multiple spatial scales to maintain productivity stability.

Forest soil CO₂ efflux (FCO₂) is a crucial process in global carbon cycling; however, how FCO₂ responds to disturbance regimes in different forest biomes is poorly understood. We quantified the effects of disturbance regimes on FCO₂ across boreal, temperate, tropical and Mediterranean forests based on 1240 observations from 380 studies. Globally, climatic perturbations such as elevated CO₂ concentration, warming and increased precipitation increase FCO₂ by 13% to 25%. FCO₂ is increased by forest conversion to grassland and elevated carbon input by forest management practices but reduced by decreased carbon input, fire and acid rain. Disturbance also changes soil temperature and water content, which in turn affect the direction and magnitude of disturbance influences on FCO₂. FCO₂ is disturbance‐ and biome‐type dependent and such effects should be incorporated into earth system models to improve the projection of the feedback between the terrestrial C cycle and climate change.

Responses of the terrestrial biosphere to rapidly changing environmental conditions are a major source of uncertainty in climate projections. In an effort to reduce this uncertainty, a wide range of global change experiments have been conducted that mimic future conditions in terrestrial ecosystems, manipulating CO₂, temperature, and nutrient and water availability. Syntheses of results across experiments provide a more general sense of ecosystem responses to global change, and help to discern the influence of background conditions such as climate and vegetation type in determining global change responses. Several independent syntheses of published data have yielded distinct databases for specific objectives. Such parallel, uncoordinated initiatives carry the risk of producing redundant data collection efforts and have led to contrasting outcomes without clarifying the underlying reason for divergence. These problems could be avoided by creating a publicly available, updatable, curated database. Here, we report on a global effort to collect and curate 57,089 treatment responses across 3644 manipulation experiments at 1145 sites, simulating elevated CO₂, warming, nutrient addition, and precipitation changes. In the resulting Manipulation Experiments Synthesis Initiative (MESI) database, effects of experimental global change drivers on carbon and nutrient cycles are included, as well as ancillary data such as background climate, vegetation type, treatment magnitude, duration, and, unique to our database, measured soil properties. Our analysis of the database indicates that most experiments are short term (one or few growing seasons), conducted in the USA, Europe, or China, and that the most abundantly reported variable is aboveground biomass. We provide the most comprehensive multifactor global change database to date, enabling the research community to tackle open research questions, vital to global policymaking. The MESI database, freely accessible at doi.org/10.5281/zenodo.7153253, opens new avenues for model evaluation and synthesis‐based understanding of how global change affects terrestrial biomes. We welcome contributions to the database on GitHub.

Interactions between plants and soil microorganisms in the rhizosphere are vital for maintaining the nutrient cycle and stability of terrestrial ecosystems. Nitrogen, closely related to carbon (C) cycling and ecosystem productivity, undergoes transformation by soil microorganisms into forms that can be assimilated by plants. However, the mechanism by which plants impact nitrogen (N) cycling in acid subtropical forest soil by governing the rhizosphere microbial community remains unclear. This study examined the impact of host phylogeny, root traits, and soil properties on microbial communities and N cycles in the rhizospheres of Fagaceae and Lauraceae, the two dominant families in the subtropical forest of Heishiding Natural Reserve in southern China. Our findings revealed that host phylogeny dominates the variation in the richness and composition of the rhizosphere microbial community. Furthermore, antibacterial exudates from Lauraceae might reduce the stability of microbial network structure and enrich fatty acid biosynthesis function for antimicrobial resistance. Net nitrification rates (NNR) and nitrifying enzyme activities (NEA) of arbuscular mycorrhizae trees of Lauraceae were significantly higher than those of ectomycorrhizae trees of Fagaceae. The positive relationships among NNR, NEA, and root nitrogen content suggest a close correlation between soil nitrification activity and N uptake efficiency by plants. Complete ammonia oxidizer (comammox) dominated the ammonia‐oxidizing communities of these acid soils, followed by ammonia‐oxidizing archaea and ammonia‐oxidizing bacteria. The niche differentiation among these groups was strongly influenced by soil properties and plant traits, which in turn affect plant N uptake efficiency. In conclusion, we conducted a comprehensive analysis of the root‐mediated plant–microbe interactions and the regulated microbial N cycles in the rhizosphere soils of an acid subtropical forest. These findings are crucial for elucidating the biological mechanisms guiding microbial assembly in the rhizosphere and refining biological strategies to mitigate N loss in forest ecosystems. Read the free Plain Language Summary for this article on the Journal blog.