Front. Sustain. Food Syst. Frontiers in Sustainable Food Systems Front. Sustain. Food Syst. 2571-581X Frontiers Media S.A. 10.3389/fsufs.2026.1774733 Original Research Spatiotemporal evolution and obstacle factors of agricultural green development in China's major grain-producing areas Xiang Huaicheng 1 Writing – original draft Li Zhiling 1 Writing – original draft Zhou Xinhao 2 * Writing – review & editing School of Business Administration, Zhongnan University of Economics and Law, Wuhan, China School of Automation, Wuhan University of Technology, Wuhan, China Correspondence: Xinhao Zhou, 2668823358@qq.com 04 03 2026 2026 10 1774733 24 12 2025 17 01 2026 20 01 2026 Copyright © 2026 Xiang, Li and Zhou. 2026 Xiang, Li and Zhou https://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Agricultural Green Development (AGD) across 13 major grain-producing provinces in China from 2006 to 2022. Drawing on the entropy method, Dagum Gini coefficient, and obstacle degree model, we track spatiotemporal dynamics while identifying the barriers that most constrain AGD. The results show that AGD has improved steadily over time; against this upward backdrop, the Yangtze River Basin and North China Region display a marked catch-up effect, eventually surpassing the Northeast Region. At the same time, spatial disparities do not diminish accordingly, but instead follow a fluctuating upward trend. Decomposition results further suggest a shift in what drives inequality: inter-regional disparity still contributes the most, yet its dominance is weakening, whereas intra-regional inequality and transvariation density are gaining influence. This pattern implies that the green development gap is increasingly shaped by polarization within regions, rather than being explained mainly by the North–South divide. When the constraints are examined more closely, Ecological Conservation stands out as the most binding barrier at the criterion level. At the indicator level, the Proportion of Waterlogging Control Area, the Development Level of Green Food-Labeled Products, and the Proportion of Soil and Water Loss Control Area repeatedly emerge as the key obstacle factors limiting further improvement. These findings indicate that future interventions should shift from uniform guidance toward precision-based governance, with localized ecological restoration prioritized and the supply structure of green agricultural products further optimized.

 agricultural green development China's major grain-producing areas obstacle factors spatiotemporal evolution sustainable agriculture The author(s) declared that financial support was not received for this work and/or its publication. section-at-acceptance Land, Livelihoods and Food Security Introduction

Agriculture underpins both national stability and economic development. During the past four decades of rapid growth in China, agricultural expansion has often depended on intensive inputs—pesticides, chemical fertilizers, and plastic film among them (Deng et al., 2023). Under a traditional quantity-oriented development paradigm, the land's ecological carrying capacity is too easily treated as secondary, while yield maximization is positioned as the overriding objective. The consequences have been visible and cumulative: soil fertility has deteriorated, and non-point source pollution has risen sharply (Huang et al., 2023; Xiong and Zhao, 2024). Against this backdrop, China has strategically pivoted toward Agricultural Green Development (AGD; Liu et al., 2021; Streimikis et al., 2022). AGD differs from conventional approaches in that it does not frame environmental sustainability as an add-on; it also foregrounds the economic benefits associated with development, aiming to steer agricultural growth onto a more efficient, durable path (Koohafkan et al., 2012; Struik and Kuyper, 2017; Liu et al., 2020). Consistent with this logic, our study constructs an evaluation system that jointly accounts for ecological costs, specifically agricultural carbon emissions and economic outputs, thereby moving beyond an exclusive emphasis on yield.

This transition is especially consequential for China's Major Grain-Producing Areas. Since their designation in 2004, the China's Major Grain-Producing Areas have encompassed 13 provinces and autonomous regions (Hebei, Inner Mongolia, Jilin, Liaoning, Heilongjiang, Jiangsu, Anhui, Jiangxi, Henan, Shandong, Hubei, Hunan, and Sichuan). Supported by favorable natural endowments, these regions sustain large-scale grain cultivation and have enabled substantial increases in grain output (Zhou and Wen, 2023). Yet their role as the primary suppliers of commercial grain also places them at the center of the tension between yield maximization and ecological preservation. Reflecting this reality, China in 2022 explicitly positioned AGD as a key component of rural revitalization, advancing policies intended to stabilize yields while tightly constraining non-point source pollution.

Current scholarly research on AGD has largely centered around three core aspects: evaluation indicators, assessment methodologies, and influencing factors. With regard to indicator systems, academic inquiry has witnessed a distinct evolution—from narrowly focused, single-dimensional models to more holistic, integrated frameworks. Initial studies often concentrated on the social, economic, and ecological pillars of AGD (Quintero-Angel and González-Acevedo, 2018). Over time, this framework was extended to include dimensions such as skills and nutrition, reflecting a growing concern for reconciling development with production outcomes (Kanter et al., 2018). More recent contributions have further broadened the scope by incorporating rational management (Chen et al., 2022) and rural development considerations (Jakovljevic et al., 2023; Cheng et al., 2023). Notably, Barbier (2025) contend that in the context of developing countries, attention to rural infrastructure and access to clean energy is essential for ensuring policy relevance and efficacy.

In exploring evaluation techniques, scholars have employed both single-method and composite approaches. The entropy method and the Analytic Hierarchy Process (AHP) remain prominent tools for static evaluations (Liu, 2021; Malik et al., 2024; Liu et al., 2025), owing to their objectivity and adaptability. However, in response to the increasing complexity of spatial dynamics, a growing body of research has turned to multi-method frameworks. For instance, Fu et al. (2024) utilized Moran's I in combination with spatial Durbin models to investigate the role of the digital economy in shaping AGD patterns. Similarly, Ma et al. (2025) integrated entropy-weighted TOPSIS with ArcGIS spatial analysis, offering a visual and quantitative depiction of regional disparities in low-carbon agricultural practices.

When it comes to influencing factors, the existing literature tends to focus on institutional policies, social structures, and individual behavioral mechanisms. At the policy level, one recurring theme is the economic disincentive farmers face when green production costs surpass potential returns in the absence of supportive measures (Uematsu and Mishra, 2012). To address this, researchers advocate for the reorientation of conventional subsidies—which often distort input prices—toward sustainability-oriented alternatives, thereby facilitating the diffusion of green technologies (You and Li, 2023; Reyes-Garcia et al., 2025; Barbier, 2025). On the demand side, targeted initiatives like green consumption subsidies have shown tangible success in offsetting adoption costs (Chen and Gao, 2025). Beyond the realm of formal policy, factors such as organizational affiliation (Sills and Caviglia-Harris, 2015), spillover effects from the digital economy (Hong et al., 2023), and farmers' perceived value of green practices (Li et al., 2020) also exert a significant influence. Importantly, collaborative networks that bring together rule-makers, facilitators, and resource providers are increasingly recognized as critical to fostering localized, context-specific green outcomes (Amaruzaman et al., 2017; Guo et al., 2022).

Despite the valuable insights offered by existing literature, several notable limitations continue to undermine both the analytical precision and policy relevance of current AGD evaluations. One major concern lies in the scope of research and the selection of indicators. Most studies are confined to the national scale or focus selectively on individual provinces (Liu et al., 2020; Shen et al., 2022). As a result, their frameworks often reflect a productivist orientation—overemphasizing economic output while insufficiently accounting for ecological constraints that are particularly salient in grain-producing regions, such as cultivated land quality and agricultural carbon budgets. In addition, while regional disparities have received growing attention, much of the analysis remains at a descriptive level. Conventional metrics may highlight the existence of spatial gaps, yet they fall short of disentangling the structural sources of inequality—especially the relative contributions of intra-regional vs. inter-regional divergence (Zhou and Wen, 2023). This lack of analytical differentiation hampers the formulation of nuanced regional coordination strategies. A further limitation concerns the prevailing emphasis on outcome-oriented assessments at the expense of diagnosing internal constraints. Although recent studies have explored spatial patterns and external drivers such as digitalization (Jiang et al., 2022), the specific endogenous barriers impeding green transformation often remain vague or unexamined. Without uncovering these internal obstacle factors, evaluation results offer limited utility for informing targeted, locally grounded interventions.

In response to these gaps, the present study directs its analytical focus to 13 provinces and autonomous regions that are central to China's grain production. It advances the current literature in three key respects. First, by integrating Agricultural Carbon Emissions and cultivated land quality into the indicator system, the study reconstructs a more balanced evaluation framework—one that moves beyond the conventional emphasis on output value to better reflect the ecological costs of development. Second, the application of the Dagum Gini coefficient facilitates a nuanced decomposition of regional disparities, enabling us to distinguish whether the widening AGD gap stems more from internal imbalances within regions or from inter-regional divergence. This distinction is crucial for designing more effective cross-regional policy mechanisms. Third, rather than merely assessing developmental outcomes, this study shifts analytical attention to the diagnosis of binding constraints. Employing the Obstacle Degree Model, we quantitatively pinpoint the most pressing barriers at both the criterion and indicator levels. This diagnostic approach provides a firmer empirical foundation for overcoming localized bottlenecks in the green transition, offering more actionable insights than traditional descriptive analyses alone.

The remainder of this paper is structured as follows. Section 2 presents the study area, outlines the data sources, and introduces the construction of the evaluation indicator system along with the methodological framework. Section 3 examines the spatiotemporal dynamics and regional disparities in Agricultural Green Development (AGD). Section 4 identifies and analyzes the primary obstacle factors that hinder the green transition. Section 5 interprets the findings and explores their policy implications. The final section summarizes the main conclusions and acknowledges the study's limitations.

 Methods Study area

This research focuses on the 13 major grain-producing provinces designated under China's national food security strategy since 2004. Collectively, these provinces account for more than 75% of the nation's total grain output. Drawing on the regional classification proposed by Ran et al. (2022), the provinces are grouped into three primary functional zones, each distinguished by geographic and agricultural features. The Northeast Region—comprising Heilongjiang, Jilin, Liaoning, and Inner Mongolia—serves as the main production base for corn and soybeans. The North China Region, including Hebei, Henan, and Shandong, functions as the core area for wheat and corn cultivation. Meanwhile, the Yangtze River Basin, encompassing Jiangsu, Anhui, Jiangxi, Hunan, Hubei, and Sichuan, operates as the central hub for rice production. From a biophysical standpoint, these zones benefit from high net primary productivity, a result of favorable hydrothermal conditions (Li et al., 2025), which provide a strong natural basis for large-scale grain production. Yet, each region faces distinct challenges in achieving a green transition. For instance, case studies from the Dongting Lake area—a representative site within the Yangtze River Basin—underscore the difficulty of reconciling intensive agricultural practices with ecological conservation, a dilemma that continues to impede high-quality development (Xiang et al., 2024). Given these regional differences, a spatially differentiated evaluation is not only necessary but also critical to understanding the varied dynamics of AGD across China's grain-producing heartland (see Figure 1).

Spatial distribution of the three major functional zones in China's major grain-producing areas. The map is derived from the standard map service provided by the National Platform for Common Geospatial Information Services [Review Number: GS (2024)0650], with no alterations made to the base map boundaries.

Map of China highlighting three regions: Northeast Region in pink, including Heilongjia, Jilin, Liaoning, and Inner Mongolia; North China Region in blue, covering Heber, Shandong, Henan, Jiangsu, and Anhui; and Yangtze River Basin in green, featuring Sichuan, Hubei, Hunan, and Jiangxi. Legend and scale included.

As summarized in Table 1, the three functional zones represent distinct developmental paradigms, offering unique analytical value for understanding regional heterogeneity. The Northeast Region typifies a resource-dependent model constrained by black soil degradation, necessitating conservation tillage. The North China Region reflects an input-intensive model limited by water scarcity. The Yangtze River Basin, central to the Yangtze River Protection Strategy, exemplifies a development pattern characterized by economic–ecological trade-offs, balancing urbanization with pollution control. Unraveling these specific obstacles allows this study to diagnose region-specific pathways for the green transition.

Characteristics of the three functional zones comprising China's 13 major grain-producing provinces (2022).

Indicators Northeast region North China region Yangtze river basin
Component provinces Heilongjiang, Jilin, Liaoning, Inner Mongolia Hebei, Henan, Shandong Jiangsu, Anhui, Jiangxi, Hunan, Hubei, Sichuan
Primary agricultural role National grain barn Wheat-corn rotation base Rice-rapeseed hub
Major crops Corn, soybean, rice Winter wheat, corn Paddy rice, rapeseed
Grain output share High (26.5% of national total) High (23.6% of national total) High (28.1% of national total)
Economic level (GDP per capita) Relatively low (58,000 CNY) Medium (69,000 CNY) High (92,000 CNY)
Key ecological obstacle Soil degradation (black soil erosion) Water scarcity (groundwater overdraft) Pollution & flooding (agri-non-point pollution)
Resource endowment Abundant land, fertile soil Severe water deficit Abundant Water Network

Data are sourced from the China Statistical Yearbook (2023) and the China Water Resources Bulletin (2022). Economic indicators represent the average values across the provinces within each respective zone.

 Data sources

Using the constructed evaluation indicator system, this study compiles a panel dataset covering the 13 provinces (or autonomous regions) from 2006 to 2022. The data primarily come from the China Statistical Yearbook, China Environmental Statistical Yearbook, China Rural Statistical Yearbook, and the corresponding provincial yearbooks. Main acquisition platforms include the National Bureau of Statistics of China, the EPS database, and other relevant sources. Where minor data gaps exist or certain indicators lack timely updates, linear interpolation is employed to complete the dataset.

 Selection of evaluation indicators

The level of AGD is a relatively comprehensive measure that must reconcile ecological conservation with economic growth, thereby reflecting the dual essence of “green” and “development.” In this sense, AGD extends beyond pollution abatement in the production process and also encompasses the improvement of economic benefits for agriculture, rural areas, and farmers—collectively referred to as “Sannong” issues in China. Grounded in core academic literature (Liu et al., 2020; Shen et al., 2022) and informed by the long-term realities of agricultural development in China's major grain-producing regions, this study develops a comprehensive evaluation index system.

Within this system, the criterion level captures the principal dimensions that merit holistic consideration, covering five broad categories: Resource Conservation, Environmental Friendliness, Ecological Conservation, Supply Security, and Economic Growth. Beneath it, the indicator level specifies 17 sub-indicators that are subject to quantitative assessment, and the complete structure of the evaluation index system is reported in Table 2. The rationale for indicator selection is outlined below.

(1) Resource Conservation: This dimension evaluates the efficiency and intensity of resource utilization. To represent agricultural water-use efficiency, Water Consumption per 10,000 CNY of Agricultural Output (negative) and the Proportion of Water-Saving Irrigation Area (positive) are selected. Meanwhile, the Multiple-Cropping Index of Cultivated Land and the Level of Agricultural Mechanization are treated as negative indicators in this framework. From a conservation standpoint, an excessively high Multiple-Cropping Index of Cultivated Land often signals high-intensity exploitation of land resources and potential soil exhaustion (Huang et al., 2023). A similar logic applies to mechanization: when mechanization intensity becomes excessive, it is frequently accompanied by heightened fossil fuel dependency (“petroleum agriculture”) as well as an increased risk of soil compaction.

(2) Environment-Friendly: This dimension captures the environmental pressure arising from agricultural inputs and outputs. Accordingly, chemical fertilizer use intensity, pesticide use intensity, Agricultural Film Use Intensity, and Agricultural Carbon Emissions (calculated using the method detailed in Section 2.4.1) are incorporated as negative indicators. Deng et al. (2023) emphasize that the excessive application of agrochemicals constitutes the primary driver of non-point source pollution; in this regard, curbing input intensity becomes a central task in advancing the green transition.

(3) Ecological Conservation: This dimension evaluates ecosystem resilience as well as the region's capacity for ecological governance. The Forest Coverage Rate, Proportion of Soil and Water Loss Control Area, and Proportion of Waterlogging Control Area are therefore adopted as positive metrics. Together, these indicators reflect the ability to maintain ecological stability, enhance resistance to natural disasters, and mitigate soil erosion.

(4) Supply Security: For major grain-producing areas, safeguarding food security remains the bottom line (Ran et al., 2022). With this constraint in mind, Grain Yield per Unit Area, Land Productivity, and the development level of green-labeled food products are included as positive indicators, ensuring that progress toward greener production does not come at the expense of fundamental supply capacity or product quality.

(5) Economic Growth: This dimension represents the internal driving force of AGD. To capture improvements in rural welfare and the pursuit of common prosperity, the Per Capita Disposable Income of Rural Residents and the Contribution Level of Agriculture to the Economy are selected as positive indicators. By contrast, the Engel Coefficient of Rural Residents is treated as a negative indicator, since a lower value typically corresponds to a higher standard of living and a more optimized consumption structure.

Evaluation index system for agricultural green development in major grain-producing areas.

Primary indicator Secondary indicator Indicator definition Unit Attribute
Resource conservation Multiple-cropping index of cultivated land Sown area of crops/cultivated land area – Negative −
Level of agricultural mechanization Total power of agricultural machinery/sown area of crops W/ha Negative −
Proportion of water-saving irrigation area Water-saving irrigation area/cultivated land area % Positive +
Water consumption per 10,000 CNY of agricultural output Agricultural water consumption/total agricultural output value ton/10,000 CNY Negative −
Environmental friendliness Pesticide use intensity Pesticide use/total sown area of crops kg/ha Negative −
Chemical fertilizer use intensity Chemical fertilizer use/total sown area of crops kg/ha Negative −
Agricultural film use intensity Agricultural film use/total sown area of crops kg/ha Negative −
Agricultural carbon emissions Statistical data 10,000 tons Negative −
Ecological conservation Forest coverage rate Statistical data % Positive +
Proportion of soil and water loss control area Area under soil and water loss control/land area % Positive +
Proportion of waterlogging control area Waterlogging-control area/land area % Positive +
Supply security Grain yield per unit area Grain output/grain sown area ton/ha Positive +
Land productivity Agricultural output value/sown area of crops 10,000 CNY/ha Positive +
Development level of green food-labeled products Number of green food-labeled products/cultivated land area items/10,000 ha Positive +
Economic growth Contribution level of agriculture to the economy Agricultural output value/GDP % Positive +
Per capita disposable income of rural residents Statistical data CNY Positive +
Engel coefficient of rural residents Statistical data % Negative −

Indicators were selected in accordance with the principles of scientific validity and data availability. The symbols “+” and “–” denote positive and negative indicators, respectively.

 Measurement method Calculation of agricultural carbon emissions

This study explicitly incorporates Agricultural Carbon Emissions as a core indicator within the “Environmental Friendliness” dimension. In line with the mainstream accounting approaches recommended by the IPCC and widely adopted in authoritative literature (Li et al., 2011; Oak Ridge National Laboratory, 2004), Agricultural Carbon Emissions are estimated from six major input sources, chemical fertilizer, pesticides, agricultural film, agricultural diesel oil, irrigation, and tillage, thereby capturing the carbon pressure embedded in agricultural production activities. The corresponding calculation formula is presented as follows:

C=∑Ci=∑Ti•βi

In Equation 1, C represents the total agricultural carbon emissions; Ci denotes the carbon emissions from the i-th carbon source; Ti refers to the usage amount of the i-th carbon source; and βi is the corresponding carbon emission coefficient. The specific coefficients used in this study are listed in Table 3.

Carbon emission coefficients and data sources for major agricultural inputs.

Carbon source Coefficient Reference source
Chemical fertilizer 0.89 kg C/kg Oak Ridge National Laboratory, 2004
Pesticides 4.93 kg C/kg Oak Ridge National Laboratory, 2004
Agricultural film 5.18 kg C/kg Li et al., 2011
Agricultural diesel 0.59 kg C/kg IPCC, 2013
Agricultural irrigation 266.48 kg C/ha Li et al., 2011
Agricultural tilling 312.60 kg C/ha Li et al., 2011
Entropy method

To determine indicator weights in a data-driven manner while minimizing subjective interference, this study employs the Entropy Method. The procedure is implemented through the following steps:

(1) Data Standardization:

To eliminate the influence of differing units and scales, the range method is applied to standardize the original data. In this evaluation index system, Equation 2 is used for standardizing positive indicators, whereas Equation 3 is employed for standardizing negative indicators.

Yij=Xij-Xj minXi max-Xi min Yij=Xi max-XijXi max-Xi min

In Equations 2 and 3, Yij represents the normalized value of the j-th indicator for the i-th sample (i = 1, 2, …, m; j = 1, 2, …, n). Xij denotes the original data of the j-th indicator for the i-th province (or autonomous region), while Xi max and Xi min represent the maximum and minimum values of the j-th indicator within the original data set, respectively.

(2) Determination of Weights:

Calculation of the proportion of the j-th indicator for province (or autonomous region) i:

Pij=Yij∑i=1mYij

In Equation 4, Pij denotes the proportion of the j-th indicator for the i-th province (or autonomous region), and m represents the sample size.

Calculation of the information entropy of the j-th indicator:

Ej=-1ln(m)∑i=1mPijln Pij

In Equation 5, Ej is the entropy value of the j-th indicator, where m is the sample size.

Calculation of the redundancy of the j-th indicator:

Dj=1-Ej

In Equation 6, Dj represents the redundancy of the j-th indicator. The smaller the variability in Dj, the less influence the indicator has on the evaluation results, and accordingly, the smaller the weight it holds.

Calculation of the weight of the j-th indicator (see Table 4):

Wj=Dj∑j=1nDj

Weights of criteria and indicator layers.

Primary indicator Code Secondary indicator Meaning of indicator Weight
Resource conservation 0.1661 C1 Multiple-cropping index of cultivated land Sown area of crops/cultivated land area 0.0371
C2 Level of agricultural mechanization Total power of agricultural machinery/sown area of crops 0.0252
C3 Proportion of water-saving irrigation area Water-saving irrigation area/cultivated land area 0.0873
C4 Water consumption per 10,000 CNY of agricultural output Agricultural water consumption/total agricultural output value 0.0165
Environmental friendliness 0.1734 C5 Pesticide use intensity Pesticide use/total sown area of crops 0.0429
C6 Chemical fertilizer use intensity Chemical fertilizer use/total sown area of crops 0.0670
C7 Agricultural film use intensity Agricultural film use/total sown area of crops 0.0248
C8 Agricultural carbon emissions Statistical data 0.0387
Ecological conservation 0.3056 C9 Forest coverage rate Statistical data 0.0590
C10 Proportion of soil and water loss control area Area under soil and water loss control/land area 0.0935
C11 Proportion of waterlogging control area Waterlogging-control area/land area 0.1531
Supply security 0.1782 C12 Grain yield per unit area Grain output/grain sown area 0.0187
C13 Land productivity Agricultural output value/sown area of crops 0.0574
C14 Development level of green food-labeled products Number of green food-labeled products/cultivated land area 0.1021
Economic growth 0.1767 C15 Contribution level of agriculture to the economy Agricultural output value/GDP 0.0623
C16 Per capita disposable income of rural residents Statistical data 0.0861
C17 Engel coefficient of rural residents Statistical data 0.0283

The weights were calculated using the entropy method to ensure objective evaluation. Data source: Relevant statistical yearbooks and the EPS database. Unless otherwise specified, the data sources for all subsequent tables and figures are consistent with this table. C1–C17 represent the codes for secondary indicators.

In Equation 7, Wj denotes the weight of the j-th indicator within the overall indicator system.

(3) Composite Index:

After the weight of each indicator is determined, the composite score for AGD in the 13 provinces (autonomous regions) within China's major grain-producing areas is calculated using Equation 8.

Zi=∑j=1nWjYij

In Equation 8, Zi denotes the composite AGD score of the i-th province (autonomous region). A larger Zi indicates a higher level of AGD in that province (autonomous region), and vice versa.

(4) Measurement Results:

Using the constructed indicator system, this study calculates the composite AGD scores for the 13 provinces (autonomous regions) in China's major grain-producing areas over 2006–2022 (detailed results are reported in Supplementary Table S1).

(5) Sensitivity Analysis:

To test the robustness of the evaluation results obtained from the entropy method, a sensitivity analysis was conducted using a combined weighting framework (CRITIC-Entropy-TOPSIS). The results indicate strong consistency between the comprehensive scores produced by the two approaches, with a Pearson correlation coefficient of 0.9863 (p < 0.001) and a Spearman's rank correlation coefficient of 0.9795 (p < 0.001). This suggests that the findings are robust and are not sensitive to the weighting technique adopted. Detailed results are provided in Supplementary Figure S1.

 Dagum Gini coefficient

To identify the sources of regional disparities in AGD, this study employs the Dagum Gini coefficient. The overall Gini coefficient G can be decomposed into three components: intra-regional differences in AGD levels Gw, inter-regional differences Gb, and transvariation (overlapping) effects among regionsGt, satisfying G = Gw+Gb+Gt (Dagum, 1997). The specific formula is as follows:

G=∑j-1k∑h-1k∑i-1nj∑r-1nh|yji-yhr|2nj2y¯ Gjj=∑i=1nj∑j=1nj|yji-yjr|2nj2y¯ Gjh=∑i-1nj∑r-1nh|yji-yhr|njnh(y¯j+y¯h)

Where G represents the overall Gini coefficient; k = 3 and n = 13, where k and n denote the number of subregions and provinces, respectively. yji represents the level of AGD in province i within region j, while yhr denotes the level in province r within region h. y¯ refers to the average level of AGD in China's major grain-producing areas. Gjj and Gjh are the Gini coefficients within region j and between regions j and h, respectively.

 Obstacle degree model

The obstacle degree model is designed to identify and diagnose the factors constraining AGD by drawing on three measures—factor contribution, indicator deviation, and obstacle degree—thereby quantifying the extent to which each indicator affects the overall evaluation objective. The corresponding formulas are as follows:

Pij=1-Yij Nij=Pij•Wij∑i=1nPij•Wij Mij=∑Nij

Here, Wij denotes factor contribution, namely the extent to which each indicator in the evaluation system influences the level of AGD; these values are the weights of each evaluation indicator obtained from the evaluation of AGD levels in Chapter 3. Pij represents indicator deviation, reflecting the gap between the observed indicator value and the optimal target. Nij is the indicator obstacle degree, indicating the extent to which each evaluation indicator acts as an obstacle to AGD. Mij refers to the criterion-level obstacle degree, which captures the extent to which each criterion layer constitutes an obstacle to AGD.

 Spatiotemporal evolution and regional differences of agricultural green development in major grain-producing areas

Drawing on the composite scores of AGD in China's major grain-producing areas (see Supplementary Table S1), this section provides an integrated assessment of temporal and spatial dynamics. Visual charts are used to present the temporal evolution of AGD for the overall major grain-producing region, its subregions, and each individual province or autonomous region. Regional disparities are further examined through the Dagum Gini coefficient and its decomposition, which helps to clarify the spatial characteristics of development. In addition, cluster analysis is employed to identify distinct development types and to profile the features associated with each category.

Temporal evolution of agricultural green development in major grain-producing areas

As shown in Figure 2, the average AGD level in China's major grain-producing areas followed a sustained upward trajectory from 2006 to 2022. The composite score rose from 0.309 in 2006 to 0.500 in 2022, corresponding to an average annual growth rate of about 3.04%, and indicating notable progress in the transformation of agricultural development patterns. Overall, the process can be described as a steady growth phase: while minor fluctuations occurred in certain years, the long-term trend points to continuous optimization of the agricultural system.

Temporal evolution of agricultural green development in China's major grain-producing areas (2006–2022). The line represents the average comprehensive score of the 13 major grain-producing provinces.

Line graph showing the overall score from 2006 to 2022, with a gradual increase from approximately 0.30 in 2006 to 0.50 in 2022. The scores are consistently rising each year.

Figure 3 further illustrates the temporal changes across the five dimensions of AGD, all of which show growth to varying extents. Environmental Friendliness and Ecological Conservation consistently recorded the highest contribution scores over the study period, forming the fundamental pillars of green development. Supply Security and Economic Growth, meanwhile, exhibited the fastest acceleration—particularly after 2012—suggesting parallel advances in safeguarding food security and improving farmers' income. By comparison, Resource Conservation increased more slowly, implying that resource-use efficiency remains a key bottleneck requiring further improvement.

Temporal evolution of each dimension of agricultural green development.

Line graph showing trends from 2006 to 2022 for five categories: Resource Conservation, Supply Security, Economic Growth, Environmental Friendliness, and Ecological Conservation. Resource Conservation and Supply Security remain stable, while others increase gradually.

According to Figure 4, the three functional zones followed distinct evolutionary trajectories. In 2006, the Northeast Region occupied the leading position, largely attributable to its advantageous natural resource endowment, yet its growth momentum has eased in recent years. By contrast, the Yangtze River Basin and the North China Region displayed a pronounced catch-up effect, overtaking the Northeast Region around 2018 and emerging as the new leaders in green development. This shift points to a reconfigured regional landscape, in which southern and central areas are advancing their green transition more rapidly than the traditional grain bases in the north.

Temporal evolution of agricultural green development in each region (2006–2022).

Line graph depicting region-specific scores from 2006 to 2022 for Northeast China, North China, and the Yangtze River Basin. Northeast China generally shows higher values, followed by North China, then the Yangtze River Basin. All regions show fluctuations over time.

To substantiate these spatiotemporal dynamics, Figure 5 depicts the geographical distribution of AGD levels across six representative years. The progressive deepening of map color from 2006 to 2022 provides intuitive confirmation of broad-based improvement in green development. Spatial differentiation is also apparent: the Yangtze River Basin moved rapidly into high-level clusters after 2015, whereas the Northeast Region improved at a comparatively slower pace. This visual pattern is consistent with the North–South divergence examined in the subsequent inequality analysis. Provincial scores and rankings for the selected years are reported in Table 6 (Section 3.3.2).

Spatiotemporal evolution of agricultural green development levels in China's major grain-producing areas (2006–2022). The map is based on standard map GS (2024)0650.

Series of six maps of China from 2006 to 2022 showing regional changes in color gradient from light to dark green. Colors represent data ranges: 0.27-0.34, 0.35-0.42, 0.43-0.49, 0.50-0.57, and 0.58-0.64. Overall increase in darker greens over time, indicating changes in the measured data. Spatial disparities and decomposition of green agricultural development in China's major grain-producing areas

To further examine disparities in AGD across subregions within China's major grain-producing areas, this study employs the Dagum Gini coefficient to evaluate the balance of development. Beyond identifying whether disparities exist, the Dagum Gini coefficient enables decomposition and helps clarify the sources of inequality. By addressing sample overlap, it also provides a more intuitive and accurate representation of intra-regional and inter-regional differences than the traditional Gini coefficient. The detailed results for within-group and between-group Gini coefficients, together with the contribution rates of each component, are reported in Supplementary Tables S2, S3.

Overall difference

As shown in Figure 6, the overall Gini coefficient of AGD in China's major grain-producing areas followed a fluctuating upward trend from 2006 to 2022, increasing from 0.053 in 2006 to 0.061 in 2022, with an average annual growth rate of 0.86%. Although the coefficient declined in some years (e.g., 2010), the long-run trajectory suggests that spatial inequality in AGD has gradually intensified. This implies that, even as the overall development level improves, the gap between advanced and lagging provinces has not narrowed and has instead expanded slowly.

Temporal evolution of the overall and intra-regional Gini coefficients of agricultural green development in China's major grain-producing areas (2006–2022).

Line graph showing the Dagum Gini Coefficient from 2006 to 2022 for four regions: Overall, Northeast, North China, and Yangtze River Basin. Each region's coefficient fluctuates, with noticeable variations among them. The Overall line remains consistently higher, while others vary, especially between 2006 and 2022. Intra-regional differences

Intra-regional differences across the three functional zones exhibited distinct evolutionary patterns. Figure 6 indicates that the Northeast Region experienced the most pronounced rise in internal inequality, increasing from 0.039 in 2006 to 0.057 in 2022, which points to a substantial widening of the development gap among its provinces. The Yangtze River Basin also showed an upward trend, reaching 0.055 in 2022. By comparison, the North China Region maintained the lowest internal inequality (0.043 in 2022), suggesting a relatively more balanced development pattern among its provinces than in the other two regions.

 Inter-regional differences

Figure 7 illustrates the evolution of inter-regional differences. In contrast to the expansion of intra-regional disparities, the inter-regional Gini coefficient followed a fluctuating downward trend, declining from 0.035 in 2006 to 0.023 in 2022. This pattern suggests that the overall development gap among the three functional zones—the Northeast, North China, and the Yangtze River Basin—has been narrowing. The reduction is largely associated with the catch-up effect observed in the Yangtze River Basin and the North China Region, as discussed in Section 3.1. Even so, despite this downward movement, the absolute level of inter-regional difference remains notable.

Changes in inter-regional differences in agricultural green development level.

Line graph depicting the Dagum Gini Coefficient from 2005 to 2022 for three regions: Northeast-North China, Northeast-Yangtze River Basin, and North China-Yangze River Basin. The graph shows variations in economic inequality over time, with distinct trends for each region. Sources of difference and contribution rates

Figure 8 reports the sources of spatial inequality and their contribution rates, revealing a clear structural shift over time. In 2006, inter-regional difference constituted the dominant driver, accounting for 65.14% of total inequality; by 2022, its contribution had fallen sharply to 38.16%. Over the same period, the contribution of intra-regional difference increased from 24.35 to 32.04%, while transvariation density rose from 10.51 to 29.80%. These changes indicate that, although inter-regional disparity remains the largest single source of inequality, its dominance has weakened. The rising shares of intra-regional difference and transvariation density imply that inequality is increasingly shaped by internal polarization within regions, together with the overlap of development levels across regions. Because this study uses full-population data for the 13 designated provinces, the decomposed contribution rates should be interpreted as structural parameters of the study area rather than probabilistic estimates; accordingly, attention is placed on their magnitude and temporal evolution.

Sources of differences in agricultural green development level.

Bar chart showing contribution rates from 2006 to 2022, divided into regional (red), inter-regional (orange), and super variable density (yellow). The regional rate is highest in 2006 and 2007. Super variable density increases notably from 2014 onwards. Spatiotemporal differences in agricultural green development among provinces in major grain-producing areas Temporal evolution of agricultural green development among provinces in major grain-producing areas

As shown in Figure 9, all 13 provinces (autonomous regions) in China's major grain-producing areas experienced rising AGD levels from 2006 to 2022. In terms of provincial growth dynamics, Shandong, Sichuan, and Jiangsu were at the forefront of the transition. Jiangsu, in particular, recorded growth exceeding 100%, reflecting strong momentum in AGD. By contrast, Jilin exhibited the slowest increase over the study period, at 24.60%. This pronounced heterogeneity underscores the need for targeted revitalization policies that respond to the specific developmental bottlenecks of lagging grain-producing hubs.

Temporal evolution of agricultural green development in each province of major grain-producing areas. The color gradient from blue to red indicates the level of agricultural green development from low to high.

Heatmap showing changes in data across various Chinese provinces from 2006 to 2022. The Y-axis lists provinces, and the X-axis covers years. Colors range from blue (lower values) to red (higher values), with Jiangsu showing significant red in 2019-2022, indicating higher values. Classification evaluation based on systematic cluster analysis

To evaluate the overall AGD level of each province (autonomous region) across the full study period—while accounting for differences in resource endowments and other influencing factors—this study applies the sum of squared deviations method in SPSS to conduct a systematic cluster analysis. The results classify the 13 provinces (autonomous regions) into three groups: high-level regions, medium-level regions, and low-level regions, as reported in Table 5.

Cluster analysis of agricultural green development in major grain-producing areas.

Category Region
High-level regions Jiangsu Province, Shandong Province
Medium-level regions Liaoning Province, Jiangxi Province, Heilongjiang Province, Hubei Province, Hebei Province, Sichuan Province
Low-level regions Anhui Province, Hunan Province, Henan Province, Jilin Province, Inner Mongolia Autonomous Region

The classification is derived from the sum of squared deviations method using SPSS software.

Building on Table 6, which reports AGD rankings for each province within China's major grain-producing areas, and Table 7, which provides rankings by each AGD dimension in 2022, this section examines the development characteristics of each category. Provinces in the first category lead in both agricultural and economic development. Those in the second category show no pronounced advantages, yet also exhibit no major weaknesses, indicating a relatively balanced development profile. Provinces in the third category, by contrast, display clear deficiencies across multiple dimensions of AGD.

Provincial rankings of agricultural green development in major grain-producing areas.

Province 2006 Rank 2012 Rank 2019 Rank 2022 Rank
Hebei 0.3318 4 0.4093 3 0.4620 7 0.5269 5
Inner Mongolia 0.3023 9 0.3170 13 0.4232 9 0.4235 12
Liaoning 0.3526 2 0.4134 2 0.4955 2 0.5326 4
Jilin 0.3329 3 0.3450 9 0.3850 12 0.4148 13
Heilongjiang 0.3670 1 0.4020 4 0.4860 4 0.5027 7
Jiangsu 0.3041 8 0.4344 1 0.5844 1 0.6392 1
Anhui 0.3061 7 0.3543 7 0.4103 10 0.4782 9
Jiangxi 0.3148 5 0.3610 6 0.4697 5 0.5342 3
Shandong 0.2802 11 0.3866 5 0.4881 3 0.5350 2
Henan 0.3109 6 0.3207 12 0.3843 13 0.4390 11
Hubei 0.2701 12 0.3539 8 0.4488 8 0.4918 8
Hunan 0.2682 13 0.3366 11 0.4038 11 0.4692 10
Sichuan 0.2823 10 0.3432 10 0.4654 6 0.5074 6

Rankings are determined by the composite scores (Zi) of each province.

Rankings of provincial scores in each dimension of agricultural green development in 2022.

Province Resource conservation Environmental friendliness Ecological conservation Supply security Economic growth
Hebei 0.1299 (1) 0.1043 (9) 0.1149 (6) 0.0871 (9) 0.0906 (6)
Inner Mongolia 0.1162 (3) 0.1307 (4) 0.0308 (13) 0.0529 (13) 0.0929 (5)
Liaoning 0.0996 (5) 0.1031 (10) 0.1402 (3) 0.1006 (7) 0.0892 (10)
Jilin 0.0736 (9) 0.1106 (6) 0.0805 (11) 0.0597 (11) 0.0904 (7)
Heilongjiang 0.0805 (8) 0.1520 (1) 0.0812 (10) 0.0596 (12) 0.1294 (1)
Jiangsu 0.1213 (4) 0.0951 (11) 0.1464 (2) 0.1692 (1) 0.1072 (2)
Anhui 0.0686 (11) 0.1046 (8) 0.1089 (7) 0.1153 (4) 0.0808 (13)
Jiangxi 0.0719 (10) 0.1432 (3) 0.1468 (1) 0.0896 (8) 0.0828 (12)
Shandong 0.1081 (3) 0.0782 (12) 0.1362 (4) 0.1154 (3) 0.0971 (3)
Henan 0.0812 (7) 0.0755 (13) 0.1081 (8) 0.0782 (10) 0.0960 (3)
Hubei 0.0609 (12) 0.1061 (7) 0.1319 (5) 0.1032 (6) 0.0895 (9)
Hunan 0.0482 (13) 0.1219 (5) 0.0862 (9) 0.1230 (2) 0.0899 (8)
Sichuan 0.0964 (6) 0.1445 (2) 0.0754 (12) 0.1063 (5) 0.0848 (11)

The values represent the evaluation scores for each criterion layer in 2022, with the corresponding provincial rankings shown in parentheses.

For provinces in the third category, where AGD levels remain relatively low, development priorities should be differentiated and aligned with their main constraints. Anhui Province needs to place greater emphasis on agricultural economic growth, with particular attention to raising the per capita disposable income of rural residents and improving living standards. In Hunan Province, resource conservation should take precedence through the implementation of a cultivated land rotation and fallow system, upgrades to water conservancy infrastructure, and the expansion of water-saving irrigation. Henan Province, by contrast, faces a more pressing need to strengthen environmental protection by reducing chemical fertilizer and pesticide use, promoting the recycling and reuse of agricultural plastic film, and lowering agricultural carbon emissions. For Jilin Province, efforts should concentrate on supply security by improving land productivity so as to raise grain yield per unit area, while also placing stronger emphasis on developing pollution-free, green, organic, and geographically indicated agricultural products to expand the supply of green and healthy food products. In the Inner Mongolia Autonomous Region, ecological conservation should be the central focus, including advancing afforestation to increase forest coverage, strengthening wetland protection, controlling land desertification and soil erosion, and improving drainage capacity.

 Study on obstacle factors to agricultural green development in major grain-producing areas

To further identify the main factors restricting AGD in China's major grain-producing areas, this study applies the obstacle degree model to calculate the obstacle degree at both the criterion and indicator levels within the index system. On this basis, the main obstacle factors influencing AGD are identified, and targeted countermeasures suitable for promoting AGD in these areas are proposed in light of the diagnosed constraints.

Diagnosis of obstacle factors Obstacle factors at the criterion level

(1) Average Obstacle Degree at the Criterion Level:

Figure 10 presents the temporal evolution of the average obstacle degrees for the five criterion layers over the study period. From 2006 to 2022, the average obstacle degrees of the criterion-level factors affecting AGD in China's major grain-producing areas are as follows: Ecological Conservation (34.90%), Supply Security (19.38%), Economic Growth (18.86%), Resource Conservation (13.97%), and Environmental Friendliness (12.88%). These results indicate that Ecological Conservation and Supply Security represent the most significant constraints, exerting a decisive influence on AGD levels in these regions.

(2) Evolution of Obstacle Degrees at the Criterion Level:

Average obstacle degree of each criterion layer in agricultural green development.

Bar chart depicting obstacle degrees for different categories: Resource Conservation (15%), Environmental Friendliness (15%), Ecological Conservation (35%), Supply Security (25%), and Economic Growth (20%). Ecological Conservation has the highest degree.

Over time, the obstacle degrees associated with Resource Conservation and Ecological Conservation show an upward trend, whereas those for Supply Security and Economic Growth display a downward trajectory. The obstacle degree for Environmental Friendliness rises in the early stage and then declines. Taken together, these patterns suggest that future efforts to promote sustainable agricultural development in China's major grain-producing areas should place greater emphasis on Resource Conservation and Ecological Conservation (Table 8).

Obstacle degree of criterion layer (2006–2022; unit: %).

Year Resource conservation Environmental friendliness Ecological conservation Supply security Economic growth
2006 13.34 10.78 32.21 22.82 20.85
2007 13.23 11.42 32.16 22.39 20.80
2008 13.35 11.64 32.15 22.01 20.85
2009 13.63 12.16 31.84 21.93 20.44
2010 13.75 12.76 32.02 21.12 20.35
2011 13.96 13.02 32.22 20.79 20.01
2012 14.32 13.86 33.51 18.11 20.20
2013 14.15 13.57 34.65 18.66 18.97
2014 14.22 13.88 34.91 18.74 18.25
2015 14.03 13.84 35.26 18.86 18.01
2016 13.54 13.93 35.52 19.14 17.87
2017 13.55 13.84 35.62 19.20 17.79
2018 13.68 13.55 36.50 18.55 17.72
2019 14.01 13.05 37.35 18.39 17.20
2020 14.39 12.55 38.18 17.47 17.41
2021 15.07 12.79 39.45 15.41 17.28
2022 15.34 12.32 39.75 15.88 16.71

The values indicate the percentage (%) of constraint contributed by each dimension to the overall green development.

The results above indicate that Ecological Conservation currently constitutes the most significant obstacle to AGD. Since 2006, its obstacle degree has consistently remained above 31.84%, underscoring the urgency of improving the ecological environment and enhancing cultivated land quality in order to raise the level of green development in these areas (Zhang et al., 2024). Supply Security ranks as the second major constraint, with even its lowest obstacle degree exceeding 15.41%, highlighting the continuing need to improve both the quantity and quality of agricultural output (Argyropoulos et al., 2013). At the same time, the steady year-on-year decline in this obstacle degree suggests that notable progress has been achieved in strengthening Supply Security in recent years.

It is also notable that, aside from Supply Security and Economic Growth, the obstacle degrees of the remaining criterion-layer factors either rise and then fall or increase continuously. This pattern suggests that, to some extent, improvements in agricultural supply have been accompanied by pressures on ecological integrity, higher resource input intensity, and aggravated agricultural non-point source pollution.

In summary, meaningful progress in AGD depends on improving the ecological environment and enhancing cultivated land quality. An uncritical pursuit of higher agricultural output and economic growth may constrain advancement in other dimensions of green development, thereby weakening the broader objective of agricultural green transformation.

 Obstacle factors at the indicator level

Across 2006–2022, the top 10 indicators with the highest obstacle degrees influencing AGD in China's major grain-producing areas remained highly stable. Table 9 reports the top 10 obstacle factors for each year. Among them, C11 (Proportion of Waterlogging Control Area), C14 (Development Level of Green Food-Labeled Products), C10 (Proportion of Soil and Water Loss Control Area), and C15 (Contribution Level of Agriculture to the Economy) consistently appeared among the leading positions. C11, in particular, remained the most prominent obstacle throughout the study period, reaching 23.18% in 2022. C14 ranked second with an obstacle degree of 10.96%, followed by C10 at 10.14% and C15 at 9.86%. The prominence of these indicators indicates that ecological shortcomings and supply-structure imbalances continue to represent the primary bottlenecks.

Obstacle degree of indicator layer (2006–2022; unit: %).

Year Indicator Indicator ranking
1 2 3 4 5 6 7 8 9 10
2006 Obstacle factor C11 C14 C16 C3 C10 C13 C15 C6 C9 C5
Obstacle degree 17.88 14.00 12.17 10.20 9.22 7.30 6.19 5.13 5.11 3.02
2007 Obstacle factor C11 C14 C16 C3 C10 C13 C15 C6 C9 C5
Obstacle degree 17.91 13.87 11.90 10.01 9.11 7.00 6.37 5.41 5.13 3.20
2008 Obstacle factor C11 C14 C16 C3 C10 C13 C15 C6 C9 C5
Obstacle degree 18.03 13.88 11.64 10.11 8.95 6.77 6.62 5.52 5.17 3.28
2009 Obstacle factor C11 C14 C16 C3 C10 C15 C13 C6 C9 C5
Obstacle degree 18.15 13.82 11.55 10.23 8.84 6.78 6.63 5.81 4.85 3.32
2010 Obstacle factor C11 C14 C16 C3 C10 C15 C13 C6 C9 C5
Obstacle degree 18.31 13.58 11.23 10.37 8.79 6.88 6.14 6.03 4.91 3.53
2011 Obstacle factor C11 C14 C16 C3 C10 C15 C6 C13 C9 C5
Obstacle degree 18.52 13.80 10.73 10.52 8.73 7.16 6.22 5.75 4.98 3.42
2012 Obstacle factor C11 C14 C3 C16 C10 C15 C6 C13 C9 C5
Obstacle degree 19.32 11.30 10.69 10.68 8.96 7.46 6.57 5.59 5.23 3.62
2013 Obstacle factor C11 C14 C10 C16 C3 C15 C6 C13 C9 C1
Obstacle degree 19.02 12.37 10.49 9.95 8.30 7.31 6.40 5.16 5.14 3.76
2014 Obstacle factor C11 C14 C10 C16 C3 C15 C6 C13 C9 C1
Obstacle degree 19.42 12.38 10.46 9.64 8.25 7.61 6.57 5.12 5.03 3.81
2015 Obstacle factor C11 C14 C10 C16 C3 C15 C6 C9 C13 C1
Obstacle degree 19.81 12.73 10.34 9.34 7.99 7.76 6.53 5.12 4.97 3.82
2016 Obstacle factor C11 C14 C10 C16 C15 C3 C6 C9 C13 C1
Obstacle degree 20.07 12.92 10.23 8.99 8.09 7.85 6.62 5.22 5.01 3.76
2017 Obstacle factor C11 C14 C10 C16 C15 C3 C6 C9 C13 C1
Obstacle degree 19.96 12.73 10.27 8.59 8.58 7.73 6.60 5.39 5.35 3.75
2018 Obstacle factor C11 C14 C10 C15 C16 C3 C6 C9 C13 C1
Obstacle degree 20.51 12.09 10.35 9.01 8.22 7.66 6.54 5.63 5.26 3.82
2019 Obstacle factor C11 C14 C10 C15 C3 C16 C6 C9 C13 C1
Obstacle degree 21.07 12.24 10.42 9.12 7.76 7.64 6.32 5.86 5.04 3.92
2020 Obstacle factor C11 C14 C10 C15 C3 C16 C6 C9 C13 C1
Obstacle degree 21.83 11.56 10.35 9.14 7.93 7.19 6.08 6.01 4.72 4.07
2021 Obstacle factor C11 C10 C14 C15 C3 C9 C6 C16 C13 C1
Obstacle degree 22.58 10.49 9.97 9.84 8.27 6.37 6.29 6.23 4.30 4.29
2022 Obstacle factor C11 C14 C10 C15 C3 C9 C6 C16 C1 C13
Obstacle degree 23.18 10.96 10.14 9.86 8.42 6.43 6.02 5.52 4.30 3.74

C1–C17 correspond to the indicator codes listed in Table 4. The values represent the obstacle degree (%) of the top hindering factors.

From the perspective of Resource Conservation, Figure 11 indicates that the most significant barrier to AGD in China's principal grain-producing areas is C3 (Proportion of Water-Saving Irrigation Area). In 2022, its obstacle degree reached 8.42%, substantially higher than C1 (Multiple-Cropping Index of Cultivated Land, 4.30%), C2 (Level of Agricultural Mechanization, 2.20%), and C4 (Water Consumption per 10,000 CNY of Agricultural Output, 0.42%). As a critical input to agricultural production, water places a hard constraint on green transition, and advancing AGD requires a shift away from a cost-ignorant model toward one that emphasizes high input–output efficiency. To achieve long-term sustainability, major grain-producing areas must maintain the stability of water resources (Aeschbach-Hertig and Gleeson, 2012). Where water-saving irrigation infrastructure is inadequate, agricultural water use tends to be excessive and inefficient, contributing to groundwater overextraction and unlawful diversion of surface water; these practices degrade the ecological environment and, in turn, impede the progress of AGD (Liu et al., 2023).

Obstacle degree of resource conservation indicators.

Line graph showing obstacle degree percentage from 2006 to 2022 for four categories labeled C1 to C4. C2 starts at 12% and declines to around 9%, while C1 begins near 4% and increases to about 8%. C3 and C4 remain near 1% or below, with C3 slightly rising in 2013.

Regarding the environmental friendliness dimension, Figure 12 shows that C6 (Chemical Fertilizer Use Intensity) remains the most prominent obstacle to green agricultural development in China's main grain-producing areas, with an obstacle degree of 6.02% in 2022. It is followed by C5 (Pesticide Use Intensity, 2.52%), C8 (Agricultural Carbon Emissions, 2.44%), and C7 (Agricultural Film Use Intensity, 1.33%). Since the implementation of the “zero growth of pesticide and fertilizer usage” policy in 2015, the obstacle degree associated with fertilizer application intensity has generally declined; nevertheless, it continues to exert substantial pressure on AGD and remains comparatively high within this dimension. Excessive fertilizer use can induce soil compaction, nutrient imbalance, and elevated concentrations of toxic elements, thereby hindering the transition toward more sustainable practices (Li et al., 2023). Meanwhile, the obstacle degrees of Agricultural Film Use Intensity and Agricultural Carbon Emissions have trended upward. By 2022, the obstacle degree for Agricultural Carbon Emissions had nearly converged with that for pesticide use intensity, underscoring the increasing importance of reducing agricultural carbon output and improving the recycling and reuse of agricultural films.

Obstacle degree of environmentally friendliness indicators.

Line graph showing the obstacle degree percentage from 2006 to 2022 for four categories: C5, C6, C7, and C8. C6 has the highest values, peaking around 2012, while C8 has the lowest.

In the Ecological Conservation dimension (Figure 13), C11 (Proportion of Waterlogging Control Area) constitutes the dominant obstacle, reaching 23.18% in 2022, followed by C10 (Proportion of Soil and Water Loss Control Area) at 10.14% and C9 (Forest Coverage Rate) at 6.43%. The exceptionally high obstacle degree of C11 suggests that weaknesses in farmland drainage infrastructure markedly reduce the capacity to withstand natural disasters such as floods. At the same time, the rise in C10 indicates that soil erosion control remains a weak link in the broader ecological defense system and warrants urgent attention.

Obstacle degree of ecological conservation indicators.

Line graph showing obstacle degrees for C9, C10, and C11 from 2006 to 2022. C11 (orange) increases from about 20% to over 25%. C10 (triangle markers) remains around 15%. C9 (circle markers) stays near 10%.

Regarding the Supply Security dimension, Figure 14 shows that C14 (Development Level of Green Food-Labeled Products) constituted the key constraint, with an obstacle degree of 10.96% in 2022. This level far exceeded that of C13 (Land Productivity, 3.74%) and C12 (Grain Yield per Unit Area, 1.18%). The relatively high obstacle degree of C14 suggests that the development of green food-labeled products lags behind market demand. By contrast, the low obstacle degrees for C12 and C13 imply that basic grain production capacity is relatively secure, and that the primary tension has shifted toward the quality and branding of supply.

Obstacle degree of supply security indicators.

Line graph showing obstacle degrees from 2006 to 2022 for C12, C13, and C14. C12 remains near zero. C13 gradually decreases from around 11.5% to 9%. C14 starts near 14%, dips sharply in 2014, and then stabilizes around 12%.

In the Economic Growth dimension, Figure 15 indicates that, prior to 2017, the main constraint on AGD in China's major grain-producing areas was C16 (Per Capita Disposable Income of Rural Residents), whereas after 2017 the binding factor shifted to C15 (Contribution Level of Agriculture to the Economy). The sustained decline in the obstacle degree of rural per capita disposable income points to a steady improvement in rural living standards. A similar downward trend is observed for C17 (Engel Coefficient of Rural Residents) before 2020, further indicating that income is no longer the primary constraint on AGD in these regions. In contrast, the obstacle degree associated with the agricultural economic contribution has continued to rise and has become the most salient economic barrier. A low contribution level from the agricultural sector suggests inadequate attention and investment in its development. At the same time, knowledge dissemination related to green agricultural practices and environmental protection remains limited. Since many farmers do not view agriculture as their principal occupation, land resources are often underutilized or wasted, which in turn constrains the advancement of green agriculture in these regions.

Obstacle degree of economic growth indicators.

Line chart showing obstacle degree percentages from 2006 to 2022 for three categories: C15 (red circles), C16 (red triangles), and C17 (orange triangles). C15 remains steady, C16 decreases, while C17 slightly increases. Discussions Spatiotemporal evolution and regional disparities

During the sample period, AGD in China's major grain-producing regions followed a clear upward trajectory. This sustained improvement is consistent with the national trend reported by Shen et al. (2022), who linked the progress to the cumulative effects of China's “Zero Growth” action plan for fertilizer and pesticide use. Yet the upward movement in aggregate scores does not fully resolve the underlying tension identified in this study. A sustainability paradox remains evident: in some high-yield areas, ecological costs continue to function as a binding constraint, mirroring the broader global challenge of reconciling food security with planetary boundaries.

The Dagum Gini decomposition further indicates that internal disparities across the main grain-producing regions have evolved in a more nuanced way than the conventional narrative of uniformly widening gaps would suggest. Inter-regional disparity still represents the largest source of inequality, although its contribution rate has weakened over time. Meanwhile, intra-regional inequality and transvariation density have intensified. This structural shift is in line with Zhou and Wen (2023) and implies that the “North-South Divergence” is no longer adequately captured by a simple divide between zones; instead, it is increasingly characterized by polarization within regions. Viewed from a global perspective, this trajectory offers an instructive case: securing national food supply does not automatically translate into a regionally balanced green transition. Future governance therefore needs to address both cross-regional compensation and intra-regional coordination so as to curb the risks associated with deepening polarization.

 Classification and obstacle diagnosis

Building on the findings of Wang et al. (2024), our cluster analysis further identifies distinct development types. Jiangsu and Shandong, supported by strong economic foundations, exemplify the economic–ecological coupling mechanism described by Jiang et al. (2022). By contrast, low-level regions such as Inner Mongolia are constrained by pronounced ecological pressures. This heterogeneity suggests that a one-size-fits-all approach is unlikely to be effective; instead, differentiated strategies—aligned with the UN Sustainable Development Goals, particularly SDG 2 (Zero Hunger) and SDG 15 (Life on Land) are needed across clusters.

With respect to specific barriers, our results show that ecological conservation has become the most consequential constraint, as evidenced by its consistently rising obstacle degree. This pattern is consistent with the warning raised by Zhang et al. (2024), who noted that the deterioration of foundational ecological conditions, including cultivated land quality, is increasingly acting as a binding limitation. Against this backdrop, the obstacle-factor analysis points to a structural shift in which the dominant bottleneck has moved from economic insufficiency to ecological rigidity. In practical terms, the marginal returns to traditional input-driven growth appear to be weakening, while long-term sustainability increasingly hinges on restoring ecosystem services rather than simply pursuing higher yields.

 Policy implications under the new context

Building on the obstacle diagnosis and the observed North–South Divergence, policy interventions should move away from uniform guidance and toward precision-based governance.

In the Northeast Region, this shift implies stricter enforcement of the Black Soil Protection Law. Since soil exhaustion remains a primary constraint, local governments could institutionalize a compensation mechanism for conservation tillage, thereby internalizing the externalities of soil protection and, in doing so, addressing the decline in land quality while safeguarding farmers' long-term interests.

For the Yangtze River Basin, attention needs to center on water-saving irrigation (C3) and waterlogging control capacity (C11). Under the Yangtze River Protection Law, investment priorities should favor climate-resilient high-standard farmland so as to mitigate flood risks; meanwhile, incorporating an environmental early warning system into the River Chief System would strengthen the capacity to monitor non-point source pollution risks in a dynamic manner.

Closing the regional gap also calls for a cross-regional green payment transfer mechanism. High-performing provinces such as Jiangsu can provide technological and financial support to ecologically fragile areas like Inner Mongolia through the East–West Collaboration framework, ensuring that the costs of ecological protection are shared more equitably and avoiding a situation in which grain-exporting regions bear disproportionate environmental burdens.

Regarding green food development (C14), endogenous market power is as crucial as regulation. This, in turn, depends on broad-based societal synergy: producers can be encouraged to enhance product branding and quality certification, while consumers may be incentivized via green consumption subsidies to choose certified green food. As such a production–consumption feedback loop takes shape, the transition from exogenous policy dependency to self-sustaining green development becomes more attainable.

 Conclusions

Based on the constructed evaluation index system and the analysis of AGD in China's major grain-producing areas from 2006 to 2022, this study draws three main conclusions:

The overall level of AGD across the 13 provinces has exhibited a steady upward trend, reflecting the effectiveness of China's agricultural transition from quantity-oriented to quality-oriented development. At the same time, the Dagum Gini decomposition shows that spatial heterogeneity persists. Although inter-regional disparity remains the dominant source of the overall gap, its influence has weakened, while intra-regional inequality has intensified. This suggests that the green development gap is no longer confined to a simple North–South divide but is increasingly expressed as polarization within regions, highlighting the need for more localized and precise governance.

The obstacle diagnosis further reveals a shift in binding constraints, from economic insufficiency toward ecological rigidity. At the criterion level, Ecological Conservation continues to constitute the primary barrier. At the indicator level, specific factors—most notably waterlogging control capacity (C11) and green food development (C14)—have emerged as critical bottlenecks hindering high-quality development.

Despite its rigorous design, this study remains subject to potential bias. The estimation of agricultural carbon emissions draws on standardized coefficients reported in the literature rather than field-based monitoring; although this macro-level approach is widely used, it may obscure micro-level heterogeneity in farming practices. Data availability also limited the inclusion of certain soft indicators, such as farmers' environmental awareness, which may introduce selection bias.

In terms of contribution and comparative context, the overall upward trend is consistent with domestic evidence reported by Shen et al. (2022). At the same time, the identified structural shift in the sources of inequality provides an additional perspective. In an international comparison, and in contrast to the subsidy-driven green transition in the European Union described by Streimikis et al. (2022), our findings suggest that hard infrastructure constraints—particularly drainage and water-saving facilities—continue to constitute the primary bottleneck in developing regions.

Regarding generalizability, the sustainability paradox highlighted here, in which high grain yields are accompanied by high ecological costs, is not unique to China. Similar tensions are likely to emerge in other intensive agricultural zones in the Global South, including India and Brazil. In this sense, the evaluation framework developed in this study offers a transferable tool for reconciling supply security with ecological conservation under climate change pressures.

 Data availability statement

This data can be found at: publicly available datasets were analyzed in this study. The data are sourced from the China Statistical Yearbook, China Environmental Statistical Yearbook, and China Rural Statistical Yearbook. These data can be accessed through the National Bureau of Statistics of China (http://www.stats.gov.cn/english/) and the EPS database (http://www.epsnet.com.cn/).

 Author contributions

HX: Writing – original draft. ZL: Writing – original draft. XZ: Writing – review & editing.

 Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

 Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

 Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

 Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fsufs.2026.1774733/full#supplementary-material

 References Aeschbach-Hertig W. Gleeson T. (2012). Regional strategies for the accelerating global problem of groundwater depletion. Nat. Geosci. 5, 853–861. doi: 10.1038/ngeo1617 Amaruzaman S. Leimona B. Van Noordwijk M. Lusiana B. (2017). Discourses on the performance gap of agriculture in a green economy: a Q-methodology study in Indonesia. Int. J. Biodivers. Sci. Ecosyst. Serv. Manag. 13, 233–247. doi: 10.1080/21513732.2017.1331264 Argyropoulos C. Tsiafouli M. A. Sgardelis S. P. Pantis J. D. (2013). Organic farming without organic products. Land Use Policy 32, 324–328. doi: 10.1016/j.landusepol.2012.11.008 Barbier E. B. (2025). Greening agriculture for rural development. World Dev. 191:106974. doi: 10.1016/j.worlddev.2025.106974 Chen L. Gao Y. (2025). How to implement the government subsidy policy in promoting the green development of agriculture in Hebei Province? J. Clean. Prod. 496:145141. doi: 10.1016/j.jclepro.2025.145141 Chen Z. Sarkar A. Rahman A. Li X. Xia X. (2022). Exploring the drivers of green agricultural development (GAD) in China: a spatial association network structure approaches. Land Use Policy 112:105827. doi: 10.1016/j.landusepol.2021.105827 Cheng L. Gao Y. Dai X. (2023). Spatio-temporal comprehensive measurement of China's agricultural green development level and associated influencing factors. PLoS ONE 18:e0288599. doi: 10.1371/journal.pone.028859937540681 Dagum C. (1997). A new approach to the decomposition of the Gini income inequality ratio. Empir. Econ. 22, 515–531. doi: 10.1007/BF01205777 Deng H. Zheng W. Shen Z. Streimikiene D. (2023). Does fiscal expenditure promote green agricultural productivity gains: an investigation on corn production. Appl. Energy. 334:120666. doi: 10.1016/j.apenergy.2023.120666 Fu L. Min J. Luo C. Mao X. Liu Z. (2024). The impact of digitalization on agricultural green development: evidence from China's provinces. Sustainability 16:9180. doi: 10.3390/su16219180 Guo Z. Chen X. Zhang Y. (2022). Impact of environmental regulation perception on farmers' agricultural green production technology adoption: a new perspective of social capital. Technol. Soc. 71:102085. doi: 10.1016/j.techsoc.2022.102085 Hong M. Tian M. Wang J. (2023). The impact of digital economy on green development of agriculture and its spatial spillover effect. China Agric. Econ. Rev. 15, 708–726. doi: 10.1108/CAER-01-2023-0004 Huang C. Ge Y. Liu X. Wang Y. Li Y. (2023). The effects of partial substitution of fertilizer using different organic materials on soil nutrient condition, aggregate stability and enzyme activity in a tea plantation. Plants 12:3791. doi: 10.3390/plants1222379138005686 IPCC R. R. (2013). Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge and New York, NY: Cambridge University Press. Jakovljevic M. Chang H. Kumagai N. (2023). Editorial: global excellence in health economics: Asia and Australasia. Front. Public Health 11:1172632. doi: 10.3389/fpubh.2023.117263237033066 Jiang Q. Li J. Si H. Su Y. (2022). The impact of the digital economy on agricultural green development: evidence from China. Agriculture 12:1107. doi: 10.3390/agriculture12081107 Kanter D. R. Musumba M. Wood S. L. Palm C. Antle J. Balvanera P. . (2018). Evaluating agricultural trade-offs in the age of sustainable development. Agric. Syst. 163, 73–88. doi: 10.1016/j.agsy.2016.09.010 Koohafkan P. Altieri M. A. Gimenez E. H. (2012). Green agriculture: foundations for biodiverse, resilient and productive agricultural systems. Int. J. Agric. Sustain. 10, 61–75. doi: 10.1080/14735903.2011.610206 Li B. Qiao Y. Yao R. (2023). What promote farmers to adopt green agricultural fertilizers? Evidence from 8 provinces in China. J. Clean. Prod. 426:139123. doi: 10.1016/j.jclepro.2023.139123 Li B. Zhang J. Li H. (2011). Research on spatial-temporal characteristics and affecting factors of agricultural carbon emission in China. Energy Procedia 5, 2245–2253. Li M. Liang L. Huang Z. Song H. Wang S. Wang Q. Sun Y. (2025). Spatiotemporal dynamics and driving factors of net primary productivity in Asian terrestrial ecosystems. Ecol. Modell. 510:111322. doi: 10.1016/j.ecolmodel.2025.111322 Li M. Wang J. Zhao P. Chen K. Wu L. (2020). Factors affecting the willingness of agricultural green production from the perspective of farmers' perceptions. Sci. Total Environ. 738:140289. doi: 10.1016/j.scitotenv.2020.14028932806378 Liu D. (2021). Application and research of analytic hierarchy process in green GDP development planning of smart city. J. Urban Plan. Dev. 147:04020051. doi: 10.1061/(ASCE)UP.1943-5444.0000646 Liu D. Zhu X. Wang Y. (2021). China's agricultural green total factor productivity based on carbon emission: an analysis of evolution trend and influencing factors. J. Clean. Prod. 278:123692. doi: 10.1016/j.jclepro.2020.123692 Liu F. Wang L. Gao J. Liu Y. (2025). Study on the coupling coordination relationship between rural tourism and agricultural green development level: a case study of Jiangxi Province. Agriculture 15:874. doi: 10.3390/agriculture15080874 Liu Y. Cao H. Du C. Zhang Z. Zhou X. Yao C. . (2023). Novel water-saving cultivation system maintains crop yield while reducing environmental costs in North China Plain. Resour. Conserv. Recycl. 197:107111. doi: 10.1016/j.resconrec.2023.107111 Liu Y. Sun D. Wang H. Wang X. Yu G. Zhao X. (2020). An evaluation of China's agricultural green production: 1978–2017. J. Clean. Prod. 243:118483. doi: 10.1016/j.jclepro.2019.118483 Ma Z. Wen J. Huang Y. Zhuang P. (2025). Analyzing spatiotemporal variations and influencing factors in low-carbon green agriculture development: empirical evidence from 30 Chinese districts. Agriculture 15:1853. doi: 10.3390/agriculture15171853 Malik M. Y. Tiwari A. Hemat W. (2024). Heterogeneous effects of agricultural insurance on agricultural green development: evidence from sub national economies of India. Discov. Sustain. 5:472. doi: 10.1007/s43621-024-00687-4 Oak Ridge National Laboratory R. R.. (2004). Carbon Dioxide Information Analysis Center. Oak Ridge, TN: Oak Ridge National Laboratory, U.S. Department of Energy. Quintero-Angel M. González-Acevedo A. (2018). Tendencies and challenges for the assessment of agricultural sustainability. Agric. Ecosyst. Environ. 254, 273–281. doi: 10.1016/j.agee.2017.11.030 Ran D. Zhang Z. Jing Y. (2022). A study on the spatial-temporal evolution and driving factors of non-grain production in China's major grain-producing provinces. Int. J. Environ. Res. Public Health 19:16630. doi: 10.3390/ijerph19241663036554509 Reyes-Garcia V. Villasante S. Benessaiah K. Pandit R. Agrawal A. Claudet J. . (2025). The costs of subsidies and externalities of economic activities driving nature decline. Ambio 54, 1128–1141. doi: 10.1007/s13280-025-02147-340019715 Shen Z. Wang S. Boussemart J.-P. Hao Y. (2022). Digital transition and green growth in Chinese agriculture. Technol. Forecast. Soc. Change 181:121742. doi: 10.1016/j.techfore.2022.121742 Sills E. O. Caviglia-Harris J. L. (2015). Evaluating the long-term impacts of promoting ‘green' agriculture in the Amazon. Agric. Econ. 46, 83–102. doi: 10.1111/agec.12200 Streimikis J. Yu Z. Zhu N. Balezentis T. (2022). Achievements of the European Union member states toward the development of sustainable agriculture: a contribution to the structural efficiency approach. Technol. Forecast. Soc. Change 178:121590. doi: 10.1016/j.techfore.2022.121590 Struik P. C. Kuyper T. (2017). Sustainable intensification in agriculture: the richer shade of green. A review. Agron. Sustain. Dev. 37:39. doi: 10.1007/s13593-017-0445-7 Uematsu H. Mishra A. K. (2012). Organic farmers or conventional farmers: where's the money? Ecol. Econ. 78, 55–62. doi: 10.1016/j.ecolecon.2012.03.013 Wang J. Han X. Liu W. Ni C. Wu S. (2024). Comprehensive assessment system and spatial difference analysis on development level of green sustainable agriculture based on life cycle and SA-PP model. J. Clean. Prod. 434:139724. doi: 10.1016/j.jclepro.2023.139724 Xiang H. Zhou C. Cuidong Song H. (2024). High-quality agricultural development in the central China: empirical analysis based on the Dongting Lake area. Geomatica 76:100010. doi: 10.1016/j.geomat.2024.100010 Xiong C. Zhao X. (2024). Impacts of chemical fertilizer reduction on grain yield: a case study of China. PLoS ONE 19:e0298600. doi: 10.1371/journal.pone.029860038452115 You F. Li S. (2023). Environmental regulations, green development of agriculture, and residents' health: empirical analysis of Yangtze River Economic Belt in China. PeerJ 11:e15717. doi: 10.7717/peerj.15717 Zhang J. Zhang P. Liu Y. Lu S. Wu G. (2024). An analysis of the spatiotemporal evolution and driving force of cultivated land green utilization in karst region of Southwest China. J. Clean. Prod. 434:140002. doi: 10.1016/j.jclepro.2023.140002 Zhou F. Wen C. (2023). Research on the level of agricultural green development, regional disparities, and dynamic distribution evolution in China from the perspective of sustainable development. Agriculture 13:1441. doi: 10.3390/agriculture13071441

Edited by: Antonio Santoro, University of Florence, Italy

Reviewed by: Xiaoming Li, Yangzhou University, China

Huaxiang Song, Hunan University of Arts and Science, China