From Decoupling to Mitigation Bottlenecks: Spatiotemporal Dynamics and Projections of China’s Crop Production Carbon Emissions
Year: 2026
Type: article
Abstract: China’s efforts to reconcile crop production with decarbonization, as the world’s largest agricultural producer, offer critical lessons for global environmental governance. Using a balanced panel of 31 mainland provincial level regions from 2005 to 2024, we develop an integrated framework combining an input-based carbon inventory, K-means clustering, Tapio decoupling, Gray Relational Analysis (GRA), Kaya-LMDI decomposition, and a Backpropagation Neural Network (BPNN). The inventory covers seven input- and operation-related CO2 sources—chemical fertilizer, pesticides, plastic film, diesel, irrigation, tillage, and machinery power—and reports all results in Mt CO2-eq; soil N2O, soil organic carbon changes, rice CH4, straw burning, and livestock remain outside the system boundary. Three main findings emerge: (1) National emissions followed an inverted-U trajectory, rising from 488.99 Mt CO2-eq in 2005 to a peak of 582.08 Mt CO2-eq in 2015 before declining to 531.52 Mt CO2-eq by 2024 (−8.69%), accompanied by pronounced spatial polarization and a shift from weak to strong decoupling after 2016. (2) Fixed-base Kaya-LMDI decomposition (2005–2024) reveals that agricultural energy efficiency was the dominant mitigating component (contributing approximately −604.3 Mt CO2-eq, or 55.9% of gross cumulative mitigation), reinforced by rural labor out-migration (−366.56 Mt CO2-eq, 33.9%) and declining energy carbon intensity (−110.18 Mt CO2-eq, 10.2%), whereas rising land productivity and per-worker cultivation scale exerted dominant upward accounting pressures on emission trajectories. (3) A BPNN scenario emulator using chronological partitioning (training: 2005–2016; validation: 2017–2019; testing: 2020–2024) projects continued decline through 2040 but a marked deceleration after 2035. Given the 20-year national series, this post-2035 plateau is interpreted as a model conditional warning signal of diminishing returns to conventional input reductions rather than a deterministic forecast, underscoring the need for disruptive energy substitution and spatially differentiated governance.
Source: Agronomy
Authors: Qing Liu(Conceptualization, Data curation, Formal analysis, Investigation, Validation, Visualization, Writing – original draft), Ziwei Jiang(Supervision, Visualization, Writing – review & editing), Xilie Li(Formal analysis, Investigation, Validation, Visualization), Jie Chen(Investigation, Validation), Zhongxia Yu(Conceptualization), Yingce Zhan(Formal analysis, Project administration), Mingcong Zhang(Project administration, Resources, Supervision), Wei Zhou(Resources)
Language: English
Subfield: Environmental Engineering
Field: Environmental Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,137
OpenAlex ID: W7235710068
