Yuchuan Fan

Agricultural landscapes sit at the center of two coupled crises: accelerating degradation of soil functions and persistent impairment of surface and groundwater quality

Leading agribusiness firms are pivotal to modernizing agricultural supply chains, yet evidence on how digital transformation translates into high-quality development remains fragmented. Using a 2024 sample of 30 Chinese national agribusiness leaders and the technology–organization–environment (TOE) framework, we integrate grey relational analysis with DEMATEL to quantify interdependencies among conditions, and combine fuzzy-set QCA with necessary condition analysis to identify both configurational pathways and binding constraints. The results of the analysis indicate that high-quality development rarely stems from a single driver; it emerges from complementary bundles linking digital technologies and R&D investment with organizational readiness (e.g., talent and governance) under supportive external conditions (e.g., policy incentives and market pressure). The findings provide a configurational explanation of digital upgrading in agribusiness and inform differentiated digital strategies for managers and policymakers.

Grasslands harbor high biodiversity and regulate continental carbon and nitrogen cycling, yet rising anthropogenic nitrogen (N) inputs are reshaping their structure, function, and stability. Synthesizing recent evidence, we show that in N-limited systems moderate N addition tends to raise both ANPP and BNPP by elevating leaf N, optimizing canopy structure, and rebalancing carbon allocation. However, once ecosystem-specific thresholds are exceeded, gains plateau or reverse, coinciding with biodiversity loss, functional-trait homogenization, declines in root-associated mutualists, and soil acidification. N effects are context dependent: thresholds shift lower in dry–hot or semi-arid grasslands and under intense grazing, while soil pH, available phosphorus, and microbial assemblages act as proximal controls that determine whether short-term productivity gains convert into long-term carbon sequestration. We propose a management-ready indicator framework organized along three axes—N dose × water–energy balance × P availability—and paired with field diagnostics (pH, available P, leaf N:P, microbial diversity and key enzyme activities, N ₂ O fluxes) to detect early transitions from “moderate” to “excessive” N addition. Priorities include long-term, multifactor experiments and observation–remote sensing–model integration that jointly track plant traits, microbial dynamics, and coupled C–N processes to improve cross-scale prediction and provide actionable guidance for N application and grazing management.

While climate is known to regulate forest productivity, the mechanistic contribution of soil microbial communities—and whether it differs between natural and plantation forests—remains poorly quantified at broad scales. Here, we provide a synthesis-level, unified analysis that jointly evaluates climate, edaphic conditions, and soil microbes to compare mechanistic pathways underlying productivity divergence between forest types. We synthesized 237 observations across China and integrated productivity metrics—gross primary productivity (GPP) and net primary productivity (NPP)—with microbial diversity, dominant taxa, and soil drivers to compare natural and plantation forests within the current environmental coverage. Plantation productivity showed nonlinear responses to microbial diversity and appeared more sensitive than natural forests. Natural forests exhibited higher bacterial Shannon and Chao1 but lower fungal Chao1 and were characterized by taxa such as Nitrobacter, Bradyrhizobium, and Cortinarius. In contrast, plantations were characterized by taxa often associated with disturbance tolerance and opportunistic life-history strategies (e.g., Sphingomonas, Fusarium, Gemmatimonas), consistent with potential functional simplification. Structural equation models identified climate as the strongest correlate of productivity, while soil properties showed contrasting associations with microbial diversity across forest types. Random forest models further highlighted cation-exchange capacity and total nitrogen as key predictors of microbial diversity in plantations. Overall, our results indicate that soil microbial communities are differentially associated with forest productivity across forest types and environmental contexts and underscore the need for future climate-comparable designs and management-intensity information to more robustly isolate microbial contributions.