Identifying the key factors influencing agricultural carbon emissions and accurately predicting future trends are essential for achieving carbon peak and carbon neutrality goals. This study aims to assess carbon emissions in agriculture from 1997 to 2022, construct an accurate model to identify the key influencing factors, and predict carbon emissions in agriculture from 2023 to 2030 with an intelligent prediction system to discuss risk management. Additionally, the Dagum method was employed to explore regional differences in agricultural carbon emissions across China. The results reveal that China’s agricultural carbon emissions exhibited a fluctuating trend from 1997 to 2022, peaking in 2015, followed by a period of decline and a moderate rebound in recent years. Elastic Net Regression identified eleven key variables, including Agricultural Machinery Level (MA), Numbers of Agricultural Tools (AT), and Agricultural Industrial Structure Upgrading (AICE), as major determinants of agricultural carbon emissions. Furthermore, the RF-PSO method demonstrated the highest predictive accuracy, forecasting a minor peak in agricultural carbon emissions in 2027, followed by stabilization. Regionally, imbalances in emissions were observed, with the intensity of transvariation accounting for 37.078% of the disparity. Therefore, the Chinese government is advised to implement region-specific strategies for controlling agricultural carbon emissions, cultivate new high-quality agricultural productivity, and promote advanced technologies.

Reducing agricultural carbon emissions is key to promoting the sustainable development of agriculture. Carbon sources play a significant role in the carbon emissions of China’s planting industry. Researching the principles of evolutionary trends of carbon sources regarding carbon emissions in China’s planting industry helps formulate scientific policies to control such emissions in the industry. This paper adopted an emission factor approach from the IPCC to estimate the CO₂ emissions of all kinds of carbon sources in China’s planting industry from 1997 to 2017. On the basis of the data, the principles of dynamic evolution in China’s planting industry and six carbon sources were analyzed by the kernel density estimation approach. Notably, the study discovered that carbon emissions peaked in 2015. In terms of the contributions of various carbon sources to the carbon emissions of the planting industry, sorted by chemical fertilizers, agricultural diesel oil, agricultural films, pesticides, agricultural irrigation, and seeding, their contribution rates were 60.82%, 13.95%, 12.88%, 9.83%, 1.88%, and 0.64%. At the same time, the kernel density results show that there was an increasing trend in carbon emissions across the whole of China’s planting industry and six kinds of carbon sources nationwide, with apparent “multipolarization”. From the perspective of various regions, the carbon emissions of chemical fertilizers, diesel oil, films, and pesticides in China’s planting industry had an evolutionary trend of multipolarization in central regions, while there was an evolutionary trend of monopolarization in eastern and western regions. The carbon emissions of seeding and irrigation had a similarly evolutionary trend in eastern, central, and western regions. Basically, they all had a double increase pattern in carbon emissions and regional differences. Therefore, China’s government needs a target to set up long-term mechanisms to ensure a stable and orderly reduction in carbon emissions in the planting industry, leading its development from the traditional planting industry to a climate-smart planting industry.

Agricultural total factor productivity (ATFP) serves as the core driver and a critical benchmark for high-quality agricultural development. Utilizing panel data from 31 Chinese provinces (2012–2023), this study examines the impact, mechanisms, and heterogeneity of agricultural digital infrastructure construction (ADIC) on ATFP. Results show that digital infrastructure significantly promotes ATFP with a distinct non-linear acceleration. Specifically, once development surpasses a critical threshold (1.5416), marginal contributions increase substantially, marking a transition of digital dividends from “accumulation” to “explosive release.” Mediation analysis identifies green technological innovation and farmland scale management as key pathways, accounting for 11.4% and 12.5% of the total effect, respectively. This confirms ADIC enhances ATFP through a technical innovation effect and a scale operation effect. Regional heterogeneity is significant: digital infrastructure yields significant positive impacts on main food consumption and balanced areas, whereas results for main grain-producing areas remain statistically insignificant. Based on these conclusions, it is recommended that policymakers maintain strategic focus by proactively deploying rural digital infrastructure to surpass critical thresholds, implementing differentiated regional strategies, and effectively translating digital dividends into endogenous momentum for high-quality agricultural development through the integrated application of green technologies and the improvement of farmland transfer service systems.

A precise understanding of the carbon dynamics of agricultural land use is essential for advancing China’s “dual carbon” goals and promoting sustainable rural development. Drawing on panel datasets for 31 Chinese provinces over the period 1997–2022, this study comprehensively analyzes the spatiotemporal evolution, regional disparities, and transition dynamics of agricultural carbon capture and emissions. Using a combination of the emission factor method, the Dagum Gini coefficient, kernel density estimation, and Markov chain models, the study finds that China’s total agricultural carbon capture has continued to increase, yet regional disparities are widening, with the central region leading and the northeastern region lagging. Meanwhile, agricultural carbon emissions exhibit a “strong west, weak east” spatial pattern and demonstrate a high degree of club convergence. Club convergence refers to the phenomenon where regions with similar initial levels converge to the same steady-state over the long run, while remaining persistently different from other regions. The net carbon effect exhibits a dual structure of carbon surplus zones and carbon deficit zones: 23 provinces act as carbon surplus zones, while 8 provinces are carbon deficit zones, primarily located in ecologically fragile or special-function regions. These findings highlight the spatial heterogeneity, path dependence, and policy sensitivity of carbon effects from agricultural land use. Accordingly, the study proposes differentiated policy recommendations, including region-specific carbon management strategies, the establishment of a unified agricultural carbon trading system, and the integration of technological and institutional innovations to achieve a balanced and low-carbon agricultural transformation.

Comprehensive utilization of crop straw (CUCS) is a critical pathway toward sustainable agricultural development, synergizing food security and carbon neutrality goals. However, there remains a lack of systematic empirical evidence regarding its macro-level productivity associations and the conditions under which they materialize. Based on China’s provincial panel data from 2011 to 2023, this paper takes the CUCS pilot policy launched in 2016 as a quasi-natural experiment and employs the difference-in-differences (DID) model to examine the association between CUCS and grain yield, along with its moderating factors and environmental co-benefits. This study yields four main findings. First, CUCS is associated with higher grain yield in pilot regions, and this finding remains robust after a series of endogeneity and robustness checks. Second, the positive association between CUCS and grain output appears to be moderated by fiscal support and innovation–entrepreneurship. The relationship is more pronounced in regions with higher fiscal expenditures on agriculture and environmental protection, as well as more agricultural patents and agricultural enterprises. Third, heterogeneity analysis suggests that the CUCS–grain output association tends to be stronger in regions with richer groundwater resources and more agricultural meteorological observation stations. Fourth, extended analysis indicates that CUCS is also associated with lower particulate matter and agricultural carbon emissions, a pattern consistent with synergistic environmental benefits. By integrating economic and environmental dimensions into a unified analytical framework, this study provides empirical evidence on the contribution of comprehensive straw utilization to grain output and highlights the enabling role of fiscal and innovation environments. These findings offer integrated evidence from China for the policy evaluation of climate-smart agriculture and contribute to the broader sustainable development agenda.