Coordinating digital village and arable land green use in China: spatio-temporal evidence in eight economic zones

Abstract

Introduction:

This study investigates the coordinated development between digital village construction and the green utilization efficiency of arable land, both of which play a critical role in advancing sustainable agriculture and rural revitalization in China.

Methods:

Drawing on panel data from 31 provinces from 2018 to 2022, we construct an enhanced coupling coordination degree model to evaluate the interaction between digitalization progress in rural areas and environmentally efficient farmland use. We further examine spatio-temporal evolution through kernel density estimation, identify spatial clustering characteristics using Moran's I and LISA, and decompose regional disparities across eight comprehensive economic zones with Dagum's Gini coefficient.

Results:

The results reveal three major findings. First, the coupling coordination level increased significantly over the study period, rising from a near-dissonant state (0.420) to a low-coordination stage (0.556), although polarization persisted in most regions except the Southern Coastal and Middle Yangtze River Economic Zones. Second, the coordination pattern demonstrates a pronounced east–west gradient, accompanied by stable spatial clustering dominated by clear High-High and Low-Low agglomerations. Third, inter-regional gaps remain the primary source of inequality, contributing an average of 75.29% to overall disparities; notably, the Great Northwest Economic Zone shows the most pronounced divergence relative to other regions.

Discussion:

This study provides both theoretical insight and empirical evidence on how digital rural construction and green land-use efficiency co-evolve, and it offers policy guidance for promoting synchronized digital infrastructure expansion and sustainable agricultural practices, thereby supporting balanced and ecologically resilient rural development.

1 Introduction

Since the beginning of the 21st century, arable land has played a decisive role in safeguarding global food security and ecological stability, serving as a fundamental resource for human survival and socioeconomic development (Li et al., 2017; Su et al., 2019). However, with the continuous acceleration of global industrialization and urbanization, the intensifying heat island effect and the frequent occurrence of extreme high-temperature weather (Shen et al., 2020) are exacerbating soil moisture evaporation and disrupting crop growth cycles. This undoubtedly poses a persistent threat to the cultivated land ecosystems on which agricultural production depends, becoming a prominent bottleneck constraining the sustainable use of green cultivated land (Zhang et al., 2021). Moreover, the long-term overuse of chemical fertilizers, pesticides, and other agricultural inputs has intensified soil degradation and ecological stress (Tang and Chen, 2022), thereby weakening the capacity to supply high-quality agricultural products. The United Nations’ Food and Agriculture Organization (FAO) statistics show that China’s pesticide application increased by 66.4% between 1990 and 2021, rising from 1.1 kg per hectare (kg/ha) to 1.83 kg/ha, and far exceeding the averages of Asia and Europe. Fertilizer use reached 319.70 kg/ha, ranking among the highest globally, while only 31.24% of arable land currently meets high-quality soil standards (Xin et al., 2025). These trends highlight the urgent need to improve the green utilization efficiency of cultivated land to ensure both productivity and ecological resilience.

At the same time, the rapid expansion of digital technologies—such as big data, artificial intelligence, and blockchain—has reshaped global economic and social systems. Yet digital accessibility remains uneven, particularly in rural regions where the digital divide restricts economic participation, technological transformation, and sustainable agricultural development (Foster et al., 2023). Evidence from European rural regions shows that digital exclusion continues to entrench urban–rural disparities (de Clercq et al., 2023). In contrast, China has rapidly advanced its digital economy, and digital rural construction has emerged as a globally recognized model (Setiawan, 2024). National policies increasingly emphasize integrating digitalization with rural revitalization and farmland protection. In 2023, the Chinese government explicitly called for establishing a comprehensive farmland protection system that integrates quantity control, quality improvement, and ecological conservation, underscoring the critical importance of efficient and rational farmland utilization. According to Haken’s synergistic theory, digital rural development and the green utilization efficiency of cultivated land function as interdependent subsystems that influence and reinforce each other through interactive mechanisms (Haken, 2007).

Digital rural development helps narrow the urban–rural digital divide, optimizes resource allocation, enhances agricultural technological innovation, and promotes cleaner energy and production systems. These processes accelerate the transformation from traditional farming toward environmentally sustainable and high-efficiency agricultural models, thereby strengthening green land-use efficiency (Rijswijk et al., 2021). Meanwhile, emerging practices such as precision agriculture, automated monitoring, and real-time digital production data provide solid technological support for greener agricultural transformation. Conversely, farmers’ demand to improve green production efficiency stimulates the upgrading of rural digital infrastructure. To reduce excessive chemical input and enhance precision in irrigation, fertilization, and pest control, farmers increasingly rely on smart agriculture and digital platforms (Ma et al., 2023). As a result, the pursuit of higher green land-use efficiency drives digital innovation, encourages smart community development, and promotes new rural industrial forms. This reciprocal mechanism demonstrates that exploring the coordinated evolution of digital rural development and cultivated land green utilization efficiency is essential for achieving sustainable agricultural modernization.

Analogously, significant regional disparities persist. Unequal digital infrastructure distribution and differentiated technological diffusion create notable provincial differences in green land-use efficiency. Although existing studies have examined digital village development levels and their effects on agricultural production efficiency, technology adoption, and farmer behavior, few have systematically investigated the coupling coordination mechanism between digital rural development and green cultivated land utilization. Moreover, China’s vast spatial heterogeneity means that coordination levels vary widely. Accurately identifying these spatial–temporal patterns is therefore critical for designing differentiated policy strategies. The eight comprehensive economic zones, which constitute the core framework of China’s coordinated regional development strategy, encompass distinct geographical conditions, economic structures, and agricultural development foundations. Studying coordination relationships within this framework can thus provide valuable guidance for optimizing resource allocation and advancing green agricultural transformation.

Within the global context, the dual imperatives of food security and ecological security are increasingly prominent, this study examines the spatio-temporal characteristics and coupling coordination relationship between digital rural development and cultivated land green utilization efficiency across the eight comprehensive economic zones. This research not only provides actionable solutions for addressing global urbanization-related ecological challenges, promoting sustainable rural transformation, and advancing regional pathways for harmonious human-nature development, but also reveals universal mechanisms through which digital technologies empower green rural development. It offers context-transcending insights for advancing rural digitalization and intelligence worldwide and safeguarding cultivated land ecological security, thereby providing key theoretical and practical support for the implementation of the United Nations Sustainable Development Goals. It contributes to the literature in two primary ways. First, it constructs an integrated analytical framework to systematically evaluate the synergistic interaction between digital village development and green land use, thereby addressing a key theoretical and empirical gap. Second, it applies an improved coupling coordination degree model that incorporates subsystem contribution coefficients, enabling more accurate assessment of coordination dynamics and offering robust empirical evidence to support policy formulation for sustainable rural digitalization and ecological agricultural development.

2 Literature review

Several scholars have extensively studied digital village construction and the green utilization efficiency of arable land from multiple perspectives. This review synthesizes existing findings across three focal areas—digital countryside development, green arable land efficiency, and their interrelationship—to identify contemporary research progress, gaps, and the theoretical foundation for this study.

2.1 Digital countryside

China defines the digital village as an endogenous driver of rural modernization, integrating networking, informatization, and digitization into rural economic and social systems while enhancing farmers’ ability to access and apply digital technologies.

Two main themes—development assessment and impact mechanisms—dominate research on digital village construction development. First, several scholars build multidimensional evaluation systems to measure rural digitalization, often emphasizing population structure, agricultural output, and farmers’ participation (Li et al., 2022; Mei et al., 2022; Meng et al., 2022), and employing entropy weighting and spatial analysis techniques (Meng et al., 2022). Notably, several studies commonly apply entropy weighting, spatial analysis, and spatial econometric models and consistently reveal significant spatial disparities: eastern provinces outperform the central and western regions, and coastal urban areas lead mountainous areas in digital readiness (Li et al., 2022; Pick et al., 2024). Spatial distribution maps of composite scores and LISA clustering diagrams reveal that coastal urban areas exhibit markedly higher comprehensive digital rural development levels than mountainous urban areas (Li and Wen, 2023). Analysis through score radar charts, bar comparison diagrams, and spatial distribution maps indicates substantial disparities in digital rural development readiness among townships within counties (Meng et al., 2022). At finer scales, township-level assessments demonstrate striking heterogeneity, indicating that China’s digital countryside development remains uneven and spatially clustered, requiring deeper investigation of regional differentiation.

Second, several researchers have explored the effects and driving mechanisms of digital village construction (i.e., what digital development does rather than merely how it develops). Evidence shows that digital infrastructure stimulates county economies by encouraging entrepreneurship, creating jobs, and improving rural living standards. Moreover, digital technologies also reduce fertilizer and pesticide dependence and lower agricultural carbon emissions. Spatial Durbin and Tobit analyses further identify economic vitality, infrastructure, electricity access, and fiscal capacity as key drivers of digital village success. However, few studies extend these findings to core rural production scenarios, particularly arable land utilization, leaving a critical knowledge gap regarding how digital rural systems contribute to sustainable resource use.

2.2 Green utilization efficiency of arable land

Green utilization efficiency of arable land integrates economic productivity with environmental responsibility, seeking to maximize agricultural output while minimizing ecological damage. Xie et al. (2018) conceptualized it as the optimal economic and ecological benefits achievable during arable land utilization under specific economic and environmental cost constraints (Xie et al., 2018).

Existing studies mainly address three issues. First, indicator design typically considers land, labor, and capital inputs and uses gross agricultural output as the primary desirable output. Recent studies have expanded these systems to include non-desired outputs, such as carbon emissions and soil pollution, to better reflect environmental costs. Second, several scholars employ multiple efficiency assessment tools, including stochastic frontier models, data envelopment analysis (DEA), and, particularly, the super-efficiency Slacks-Based Measure (SBM) model, which effectively incorporates undesired outputs and differentiates decision-making units. New methodological advances, such as the generalized Lu-Nerberg productivity index (GLPI) and its decomposition into technological progress, efficiency change, and scale efficiency, further improve the dynamic evaluation of farmland efficiency. Compared to traditional SBM models, this approach better reflects the dynamic decomposition of efficiency and provides methodological references for multidimensional quantitative analysis of green utilization efficiency of farmland (Guo et al., 2024).

Third, several notable researchers have also analyzed the determinants of green land efficiency. Regional economic development levels, farmland scale and fragmentation, agricultural subsidies, insurance policies, irrigation intensity, natural conditions, and urbanization consistently emerge as influential drivers (Qu et al., 2021). Additionally, some studies have indicated that irrigation intensity, factor inputs, natural conditions, and urbanization rates also exert significant impacts (Koondhar et al., 2018; Saleem et al., 2015). Spatial characteristics of farmland also matter. For instance, ecological constraints on steep-slope farmland require policies that balance ecological protection with production goals, contrasting with the “productivity-centered” logic applicable to plains. More importantly, the spatial land type of farmland (e.g., steep slopes) influences green utilization efficiency through ecological constraints. A global analysis of steep slopes revealed that the synergy between development controls and ecological conservation objectives for steep-slope farmland represents a crucial dimension requiring balanced consideration in green utilization. This contrasts with the “production efficiency-focused” green utilization logic for plain farmland, and the rational optimization of its spatial structure, which can indirectly support improvements in green utilization efficiency (Dai et al., 2023).

2.3 Coordinating digital countryside development and the green use efficiency of arable land

Grounded in Hermann Haken’s Synergetics Theory, this study views rural systems as open and complex structures that can spontaneously evolve from disorder to order through nonlinear interactions among their subsystems (Haken, 2007). Within this framework, digital village development and the green utilization of farmland function as two core open subsystems of the rural socio-ecological system. Through continuous material circulation, energy flow, and information exchange, they establish an intrinsic synergistic evolution mechanism. On the one hand, digital villages act as technology-driven systems that inject data resources and digital tools, reduce information asymmetry in agricultural production, optimize factor allocation, and significantly strengthen the efficiency of green farmland utilization while generating spatial spillover benefits (Tang and Chen, 2022). On the other hand, green farmland utilization provides essential application scenarios, data feedback, and practical testing environments for digital technologies, thereby serving as both the material foundation and ultimate value destination of digital village development (Dai et al., 2023).

Although existing studies have investigated the relationship between these two systems, most retain a “unidirectional linear drive” logic that moves from macro effects to micro mechanisms and then to spatial spillover outcomes. At the macro-performance level, scholars demonstrate that digital innovation directly improves cultivated land utilization efficiency by enhancing input–output performance and boosting green productivity through precision allocation of agricultural resources (Finger, 2023). At the micro-mechanism level, research reveals that digital technologies transform farming behavior by shifting from extensive, large-scale supervision to precise, crop-level management, thus reducing fertilizer and pesticide waste and promoting environmentally efficient land use (Streit and Bellwood, 2023). Spatially, scholars further verify that digital infrastructure generates cross-regional spillover effects; the mobility of information elements enables digital development in one region to promote improvements in neighboring regions’ farmland green utilization efficiency (Tang and Chen, 2022). However, whether analyzing macro performance, internal mechanisms, or spatial spillovers, these studies largely portray digital villages as active drivers and green farmland utilization as passive recipients.

This linear causality does not adequately capture the bidirectional, nonlinear feedback between the two open subsystems. More importantly, it fails to quantify their evolving synergistic states and resonance levels. Coupling theory addresses these shortcomings by offering both a conceptual lens and a methodological system—particularly through the Coupling Coordination Degree Model (CCDM)—to measure complex interactions. International scholarship has progressively advanced from whole-system assessments to key-element analyses and scenario-based applications.

At the system level, the CCDM has quantified the dynamic coordination between digital rural development and farmland utilization efficiency, confirming their tight co-evolution (Zhang and Zhang, 2024). At the element level, studies emphasize the pivotal roles of data and capital, showing how digital inputs reshape farmland resource allocation (Rotz et al., 2019). Scenario-based applications further reveal that CCDM effectively captures coordination between digital monitoring technologies and farmland green productivity in precision agriculture settings (Basso and Antle, 2020). Collectively, these findings demonstrate the strong applicability of coupling coordination models in examining the complex relationship between digital technologies and farmland utilization.

Nonetheless, methodological limitations remain. Most studies continue to rely on traditional CCDM frameworks that assume equal contribution coefficients across subsystems (Greco et al., 2018). This static weighting approach presents two core weaknesses. First, it neglects the asymmetric evolution between systems: digital development progresses rapidly, whereas farmland ecological restoration advances slowly (Yu et al., 2023). Treating them as equivalent obscures their distinct dynamics. Second, it easily produces pseudo-coordination, where high digital development scores conceal weak ecological performance, resulting in flawed sustainability judgments.

Emerging research addresses this gap by proposing improved CCDM models that abandon fixed, equal weighting and instead employ dynamic weight adjustment mechanisms. When a subsystem lags, the model increases its relative weight to highlight weak links (Giller et al., 2021). Empirical studies confirm that this approach mitigates measurement bias caused by heterogeneous system attributes (Reidsma et al., 2020) and effectively identifies ecological lagging risks masked under traditional average-score algorithms (Antle et al., 2014). These advances provide a more scientific and realistic means of evaluating synergy quality.

In summary, current research has clarified the close connection between digital village development and green farmland utilization from macro, micro, and spatial perspectives and has preliminarily explored their synergistic evolution using coupling theory. However, methodological constraints persist because many studies rely on static, equal-weight models that ignore asymmetric evolution and risk misjudging pseudo-coordination. Building on these insights, this study applies an improved coupling coordination model grounded in synergetics theory. Focusing on China’s eight major comprehensive economic zones, it establishes a dynamic weight adjustment mechanism to provide a more objective, rigorous, and precise quantitative basis for evaluating the true synergy level between digital rural development and green farmland utilization efficiency.

3 Indicator system and research methods

3.1 Research area and data sources

3.1.1 Research area

This study examines panel data from 31 provincial-level regions in China (excluding Hong Kong, Macao, and Taiwan) from 2018 to 2022. Following the Development Research Center of the State Council (DRC) regional framework, we group these provinces into eight comprehensive economic zones (Figure 1¹). This classification allows us to capture broad regional dynamics rather than limiting analysis to single-province cases.

FIGURE 1

3.1.2 Data sources

We collected data on digital village construction and green utilization efficiency of arable land from multiple authoritative and nationally recognized sources. Specifically, we extracted the proportion of Taobao villages among administrative villages from the China Taobao Village Research Report (Ali Research Institute—http://www.aliresearch.com/EN/index/). We obtained data on digital financial coverage, depth, and usage levels from the Peking University Digital Inclusive Finance Index (Peking University Digital Finance Research Center—https://en.idf.pku.edu.cn/).

Additionally, we collected information on modern agriculture demonstration projects and agricultural governance decisions from the Ministry of Agriculture and Rural Development (https://english.moa.gov.cn/). Data on digital agribusiness enterprises and online education firms came from the ZJU Carter-Enterprise Research China Agricultural Research Database (CCAD—http://www.card.zju.edu.cn/carden/main.htm/). We sourced statistics on rural e-commerce demonstration counties from the Ministry of Commerce (https://english.mofcom.gov.cn/), while the China Education Statistical Yearbook provided data on networked multimedia classrooms. The China National Knowledge Infrastructure (CNKI—https://oversea.cnki.net/) offered supplementary statistical resources and indicators used to construct the green utilization efficiency index system (2018–2022). When necessary, we addressed missing provincial or temporal data using trend extrapolation and linear interpolation to ensure dataset completeness and continuity.

3.2 Indicator system

We designed the evaluation indicator system according to the principles of scientific validity, systematic integrity, and data operability. The selection process followed three steps. First, we identified primary indicator dimensions based on the theoretical frameworks of “Digital Village” and “Green Utilization of Farmland” as well as national strategy documents such as the Digital Agriculture and Rural Development Plan (2019–2025). Second, we employed literature frequency analysis, drawing on authoritative index systems from Peking University and the China Academy of Information and Communications Technology (CAICT), along with high-impact domestic and international studies, to screen representative secondary and tertiary indicators. Third, we ensured that each indicator possessed measurable meaning, policy relevance, and data continuity.

3.2.1 Construction of the digital village indicator system

To develop the digital village evaluation index system, we constructed the digital village indicator system by synthesizing evidence from the County Digital Rural Index Research Report (2020), the Report on China’s Digital Rural Development (2022), and the Digital Rural Development Report—CAICT (2024). This framework reflects both producer and consumer perspectives and integrates emerging digital trends in rural areas—such as rural e-commerce and smart agriculture—outlined in the Digital Agriculture and Rural Areas Development Plan (2019–2025). The framework consists of four dimensions: digital infrastructure, digitalization of the rural economy, digitalization of rural life, and digitalization of rural governance (Table 1).

Digital infrastructure evaluates network facilities, information services, and digital financial systems. Drawing on the Digital Countryside Development Research Report (2024), we measure network infrastructure using rural broadband subscribers, average rural household computer ownership per 100 rural households, and mobile phone ownership per 100 households. Based on the County Digital Village Index (2020), we capture information service infrastructure using the proportion of Taobao villages in administrative villages. We measure digital financial infrastructure through indicators reflecting both coverage breadth and usage depth.

Digitalization of the rural economy focuses on digital production, finance, marketing, and supply chains. Based on the County Digital Rural Index (2020) Research Report, digital production assessment utilizes national modern agricultural demonstration projects and digital agricultural enterprise inventory. We evaluate digital agricultural production with indicators on national modern agricultural demonstration projects and digital agribusiness inventories. Inclusive digital finance reflects financial digitalization, while e-commerce demonstration counties, rural online transaction volume, and enterprise-level transaction activity indicate digital marketing capacity. We assess supply chain digitalization through rural delivery network coverage and delivery frequency.

Digitalization of rural life includes digital culture, consumption, and education. Following the Digital Agriculture and Rural Development Plan (2019–2025), we use rural radio and television coverage to represent digital cultural access. Furthermore, we approximate digital consumption through per capita rural transport and communication expenditures, which reflect the transition toward online and mobile payment environments. For digital education, we combine the number of networked multimedia classrooms with the presence of online education enterprises to capture access to digital learning resources.

Digital rural governance emphasizes “Internet + Government” services and smart emergency management. Following Bimber (1999), we incorporate an “Internet + Government” indicator that captures interaction intensity between rural residents and administrative departments. We quantify the number of publicly released agricultural and rural administrative decisions on provincial online government platforms to evaluate the scale and efficiency of digital governance. We further assess smart governance capacity using indicators from the China Digital Rural Development Report (2022), including the distribution of seismic and meteorological stations and related observation systems.

3.2.2 Construction of the green utilization efficiency indicator system for cultivated land

To construct the green utilization efficiency indicator system for arable land, we drew on frameworks proposed by Zhang et al. (2024) and Zhou et al. (2022). Guided by a clear definition of green utilization efficiency and aligned with low-carbon, environmentally sustainable development principles, we structured the evaluation system around three dimensions: inputs, desired outputs, and non-desired outputs (Table 2).

We define carbon sequestration as the process through which crops absorb atmospheric carbon dioxide via photosynthesis and store carbon in seeds, stems, leaves, roots, and soil. This phenomenon is calculated using the following formula as shown in Equation 1:

In this equation, represents the carbon uptake of each crop, denotes the number of crop types, represents the carbon sequestration rate, yield, water content, and economic coefficient of the crop, respectively. Table 3 lists the corresponding coefficients.

Carbon emissions from cropland: Agricultural activities generate significant carbon emissions through fertilizer use, irrigation, machinery operation, pesticides, and agricultural films. We calculate cropland carbon emissions as follows as shown in Equation 2:

In this equation, represents the carbon emission from the ith source, and denotes the quantity of each carbon source and its corresponding carbon emission factor, respectively. The correlation coefficients are presented in Table 4².

Cultivated land surface source pollution arises when excessive chemical inputs—such as fertilizers, pesticides, and agricultural films—accumulate and degrade environmental quality. Following Geng et al. (2024), we estimate pollution using the Unit Investigation Method, which evaluates pollutant residues based on material loss mechanisms and applies pollution generation coefficients or loss rates from different sources (Geng et al., 2024). We characterize pollutant emissions through nitrogen and phosphorus fertilizer residues, pesticide residues, and agricultural film residues, as expressed in Equations 3–7:where represents source pollution of arable land, represents nitrogen fertilizer, phosphate fertilizer, pesticide, and agricultural film residues, respectively. Likewise, represent the quantities of nitrogen, phosphate, and compound fertilizer applied, respectively. Similarly, represent the quantities of pesticides and agricultural films used, respectively. represents the production coefficients of TN in nitrogen fertilizers, TP in phosphate fertilizers, and TN and TP in compound fertilizers, respectively. Whereas, represents nitrogen and phosphate fertilizer loss rates, pesticide loss coefficients, and agricultural film residue coefficients, respectively. In these equations, farmland surface pollution depends on the quantities of nitrogen fertilizer, phosphate fertilizer, compound fertilizer, pesticides, and agricultural films, along with their respective production coefficients and loss rates. Based on empirical statistics, we adopt coefficients of 1 for TN in nitrogen fertilizer, 0.44 for TP in phosphate fertilizer, and 0.33 and 0.15 for TN and TP in compound fertilizers. The pesticide loss coefficient is 50%, and the agricultural film residue coefficient is 10%. Following Chen and Xie (2019), Table 5 reports provincial nitrogen and phosphorus fertilizer loss rates.

3.3 Research methods

3.3.1 Entropy method

We employed the entropy method to assign objective weights to digital village development indicators. This method determines weights according to information entropy: indicators with higher variability provide more information and therefore receive larger weights, while indicators with low dispersion are assigned smaller weights. By minimizing subjective interference, the entropy method improves reliability and scientific validity. We calculated the comprehensive digital village development index using standardized indicator values and their entropy-based weights as expressed in Equation 8:where represents the weight of the jth indicator, represents the standardized value of the jth indicator in year i, and m represents the total number of indicators.

3.3.2 Super-efficient SBM model

Traditional DEA models cannot incorporate non-desired outputs, and although the standard SBM model addresses this limitation, it still fails to distinguish among multiple efficient decision-making units. Tone (2002) developed the super-efficient SBM model to resolve this problem (Tone, 2002). The improved model accommodates non-desired outputs, eliminates bias caused by radial and angular measures, and enables accurate ranking of efficient units. We therefore apply the super-efficient SBM model to evaluate the green utilization efficiency of arable land. Equations 9, 10 define the model using the following expressions:where represents the number of decision units, each containing inputs, desired outputs, and non-desired outputs. Whereas, represents the input, desired output, and non-desired output values, respectively. Besides, denotes the slack variables for inputs, desired outputs, and non-desired outputs, respectively, and represents the weight variable. Likewise, reflects the green utilization efficiency of arable land.

3.3.3 Coupled coordination degree model

3.3.3.1 Traditional coupled coordination degree model

We first applied a traditional coupled coordination degree model to evaluate the interaction strength between the two systems. This model calculates the coupling degree (C), the comprehensive coordination index (T), and the final coupled coordination degree (D) using Equations 11–13 as follows:

In this framework, C denotes the coupling degree, which measures the intensity of interaction. T reflects the level of coordinated development, and D synthesizes the overall coupled coordination degree. In addition, the variables and denote the composite indices of the two systems, while and represent their respective contribution coefficients.

3.3.3.2 Improved coupled coordination degree model

Building on Shen et al. (2018), we improved the contribution coefficients to capture disparities between systems more accurately. The revised framework emphasizes that system coordination strengthens when the lagging subsystem improves or when the dominant subsystem moderates its growth. To embody this principle, we assign a greater contribution weight to the underperforming subsystem, thereby recognizing its greater marginal effect on overall coordination. The improved model adopts Equations 14–17, where and represent the redefined contribution coefficients. We then classified coordination outcomes following Chen et al. (2023), as summarized in Table 6, following equations:where and represent the redefined contribution coefficients. Furthermore, we classified the resulting coordination levels using Chen et al.’s (2023) method as shown in Table 6 (Chen et al., 2023).

3.3.4 Kernel density estimation

We employed kernel density estimation, a widely applied non-parametric method, to depict the probability density distribution of the variables. This method converts discrete observations into a continuous distribution surface, enabling us to visualize both spatial distribution characteristics and temporal evolution trends. The estimation follows Equation 18, where denotes the estimated density at position, is the kernel function, indicates bandwidth, represents the number of elements within the bandwidth, is the data dimensionality, and denotes the distance between feature and position. The calculation process follows these steps:where denotes the estimated density value at position ; denotes the kernel function; indicates the bandwidth; represents the total number of features within the bandwidth; while signifies the data dimensionality; and represents the distance between feature and position .

3.3.5 Spatial autocorrelation model

To reveal spatial clustering characteristics and spatial association patterns of the coupled coordination degree between digital village development and arable-land green use efficiency, we applied both global and local Moran’s I indices. Global Moran’s I measures the overall spatial dependence of coordination levels across regions (Equation 19), where and denote the coupling coordination degrees of regions i and j, n is the total number of regions, is the spatial weight matrix, and represents the average coordination value.where and represent the coupling coordination values for regions i and j, respectively. While n denotes the number of study regions, represents the spatial weight matrix, and indicates the average coupling coordination value for all regions.

We constructed the spatial weight matrix using the Queen Contiguity criterion, grounded in the First Law of Geography. Because digital development spillovers and ecological interactions demonstrate strong spatial dependence, adjacent provinces tend to influence each other more significantly (Peng and Mariadas, 2025). Accordingly, when regions i shares a common boundary or vertex with region j (i.e., geographic adjacency), the spatial weight = 1; otherwise, = 0.

To further explore local spatial heterogeneity, we employed the local Moran’s I index (Equation 20), which identifies spatial clusters and outliers by comparing individual regions with their neighboring units.where denotes the variance of the coupled coordination values across all regions.

3.3.6 Dagum Gini coefficient and decomposition

To measure regional disparities in coupling coordination, we calculated the Dagum Gini coefficient in Equation 21:

In this equation, k denotes the number of regions, n represents the number of provinces, is the mean coupling coordination degree across all regions, refers to the coupling coordination degree value of province i(r) in the jth (h) region, and indicates the number of provinces within region j.

We further decomposed the Dagum Gini coefficient into three components: . Where represents intra-regional variance, represents inter-regional variation, and denotes the hypervariable density contribution rates of each component, enabling a clearer understanding of how regional coordination differences form and evolve as expressed in Equations 22-29.where,

4 Empirical findings

4.1 Spatio-temporal evolution of digital village development and green utilization efficiency of arable land

4.1.1 Spatial and temporal evolution of digital village development

Figure 2 illustrates that China’s digital village development steadily increased from 2018 to 2022, with the regional average rising from 0.17 to 0.28. The implementation of policies such as the Digital Rural Development Strategy Outline and the Digital Agriculture and Rural Development Plan (2019–2025) propelled significant progress. These initiatives not only established comprehensive rural information networks but also accelerated intelligent agricultural production and modernized rural governance, supporting urban-rural integration and agricultural modernization. Across the eight comprehensive economic zones, the Eastern Coastal Zone led development, with its composite value rising from 0.26 to 0.48.

FIGURE 2

This growth reflects both its robust economic and technological foundation (Chen et al., 2022) and the impact of pioneering digital village pilot programs in provinces such as Zhejiang and Jiangsu, which created strong demonstration and diffusion effects. The Southern and Northern Coastal Zones maintained relatively high development levels, averaging 0.28 and 0.27, respectively, yet still trailed the Eastern Coastal Zone’s 0.37. The Middle Yangtze River, Greater Southwest, and Middle Yellow River Zones exhibited moderate growth, driven by industrial transfers and policies such as Internet Plus Agriculture. In contrast, the Northeast and Northwest Zones lagged due to weaker industrial structures and underdeveloped digital initiatives (Cheng et al., 2024). These disparities highlight the necessity of implementing coordinated regional strategies, optimizing resource allocation, and promoting high-quality, balanced digital development.

We used ArcGIS to analyze cross-sectional data for 2018, 2020, and 2022. Following Jenks and Caspall (1971), we applied the natural discontinuity method to classify development into three tiers: low (<0.15), medium (0.15–0.25), and high (>0.25). Figure 3 reveals a clear east-to-west gradient: the Greater Northwest consistently remained in the low tier; the Great Southwest and Northeast improved from low to medium; the Middle Yellow River remained medium; the Southern and Northern Coastal Zones and Middle Yangtze River Zone advanced from medium to high; and the Eastern Coastal Zone maintained high development.

FIGURE 3

4.1.2 Spatial and temporal evolution of green utilization efficiency of arable land

Figure 4 shows that national green utilization efficiency initially increased but declined after 2020, suggesting diminishing policy effectiveness over time. Regionally, the Northeast consistently outperformed other zones due to effective black-soil conservation and green practices. The Southern, Eastern, and Northern Coastal Zones exhibited fluctuating but generally declining trends, while the Great Southwest stabilized at 1.06 post-2019. The Middle Yellow River and Great Northwest Zones improved after 2020, reflecting growing ecological prioritization. In contrast, the Middle Yangtze River Zone declined from 0.82 to 0.73, underscoring the need to strengthen farmland protection and optimize resource use.

FIGURE 4

We categorized efficiency into low (<0.85), medium (0.85–1.00), and high (>1.00) levels using the natural discontinuity method. Figure 5 illustrates regional differences: the Eastern, Southern, and Northeastern Zones maintained high efficiency (>1.0) due to advances in water conservation, fertilizer reduction, and green-intensive technologies. The Greater Southwest Economic Zone improved from below 1 in 2018 to high efficiency levels, aided by the State Council’s Three Rural Areas policy. In contrast, the Middle Reaches Yangtze River Economic Zone displayed the lowest efficiency, reflecting limited farmer awareness of land protection and excessive cash crop cultivation, which increased surface pollution and carbon emissions. Other zones remained at medium efficiency. Overall, coastal zones outperform inland areas, while the Middle Yangtze River lags significantly.

FIGURE 5

4.2 Spatio-temporal evolution of the coupled coordination degree of the digital countryside and the green use efficiency of croplands

4.2.1 Time-series variation of the coupling coordination degree

Figure 6 illustrates a steady improvement in coordination between digital village development and green land use efficiency across the eight economic zones, advancing from near-dissonance toward primary and intermediate coordination. The Eastern Coastal Economic Zone progressed from basic coordination in 2018 to intermediate coordination by 2021, driven by strong digital infrastructure and agricultural technology innovation. The Northern and Southern Coastal Zones moved from near-dissonance in 2018 to basic coordination by 2021–2022, eventually reaching primary coordination. The Middle Yangtze River, Yellow River, and Great Southwest Zones also improved from near-dissonance to basic coordination between 2020 and 2021.

FIGURE 6

In contrast, the Northeast Zone declined from mild dissonance to near-dissonance in 2022, and the Northwest Zone remained in Mild imbalance after entering moderate imbalance in 2018. Despite localized setbacks, the overall trend reflects progress enabled by expanding digital infrastructure and enhanced green land use. To further examine the temporal dynamics of coupled coordination between digital countryside development and the green utilization efficiency of arable land across China’s eight economic zones, we applied three-dimensional kernel density estimation to capture the temporal dynamics of coordination across the eight zones (Figure 7). Key observations include.

FIGURE 7

4.2.1.1 Distribution location

The national kernel density curve shifted gradually rightward, indicating a modest but consistent improvement in coordination. This reflects the sustained momentum of macroeconomic policies and underscores the need to further enhance policy precision and systemic coherence to accelerate the coordination process. Regional curves reveal that the Eastern Coastal, Great Northwest, and Great Southwest Zones remained stable, while other zones displayed meaningful rightward shifts, reflecting enhanced coordination.

4.2.1.2 Main peak distribution pattern

The national peak remained stable with a narrower shape, suggesting a reduction in internal disparities. Southern Coastal, Middle Yellow River, and Middle Yangtze River peaks rose, while the Northeast initially increased, then declined. Peak width trends indicate fluctuating dispersion within regions, reflecting uneven progress across provinces. The width of these peaks exhibited an expanding trend, indicating that the dispersion of internal coupling coordination within these regions gradually increased. Firstly, the main peak height in the Northern Coastal and Greater Southwest economic zones increased before decreasing, while the width first narrowed then widened, indicating that the divergence in inter-provincial coupling coordination diminished before growing. Whereas, in the Eastern Coastal and Greater Northwest economic zones, the peak declined before rising, with the width undergoing an expansion-contraction cycle, mirroring an expansion-contraction process in the dispersion of internal coupling coordination within these regions.

4.2.1.3 Distribution ductility

National and Southern Coastal curves exhibited balanced distributions, supporting regional integration policy implementation, indicating a relatively balanced level of coordination between these two regions. This balanced state facilitates the implementation and comprehensive advancement of regional integration policies. Northeast, Middle Yangtze River, Greater Southwest, and Northwest curves showed right-tail features, indicating outlier provinces with higher coordination. Northern Coastal and Middle Yellow River curves leaned left, suggesting concentrations at lower coordination levels. Eastern Coastal exhibited tails on both sides, reflecting pronounced polarization.

4.2.1.4 Number of peaks

National, Southern Coastal, and Middle Yangtze River Zones exhibited unimodal distributions, indicating limited polarization. The other six zones initially displayed bimodal distributions, but most transitioned toward unimodal by 2022, reflecting diminishing polarization and convergence in coordination levels, signaling the initial success of regional integration strategies. This pattern of evolution indicates that the polarization between high-high and low-low provinces within certain regions is diminishing, with coupling coordination levels tending toward convergence.

4.2.2 Spatial divergence of the coupling coordination degree

Between 2018 and 2022 (Figure 8), the eight comprehensive economic zones steadily improved their coupling coordination rankings, revealing a distinct east-to-west decline. The Eastern Coastal Economic Zone consistently achieved the highest coordination levels, maintaining a coordinated stage throughout the period and exemplifying strong synergy between digital village development and green, efficient arable land utilization. By contrast, the Great Northwest and Northeast remained in a dislocated stage through 2022 due to structural constraints: the Great Northwest faced slow digital infrastructure expansion and limited arable land efficiency, while the Northeast experienced delays in implementing digital village initiatives. Other regions progressed from dislocation toward coordination or harmonization stages.

FIGURE 8

Notably, the Eastern Coastal Zone benefited from favorable geographic and environmental conditions—such as a temperate climate, abundant water resources, and fertile soils—that enhanced agricultural productivity and facilitated digital technology integration. Its advanced economic base and ample fiscal resources supported robust investment in digital infrastructure and smart agricultural systems. Local governments demonstrated strong policy implementation capacity, integrating digital village development and land protection into regional strategies. For example, Zhejiang Province became a national leader in digital village pilot programs. In contrast, western regions faced fragile ecological conditions, limiting agricultural potential and technology adoption. Weak economies constrained infrastructure investment and green land-use practices, while minimal policy support, slow technology diffusion, and shortages of technical expertise further hindered progress. These disparities—spanning natural resources, economic development, institutional capacity, and technological readiness—produced marked regional variation in coordination outcomes. Future strategies must adopt tailored approaches aligned with local resource endowments and regional characteristics.

We applied the global Moran’s I index to assess spatial evolution. As Table 7 shows, global Moran’s I values remained positive from 2018 to 2022, confirming persistent positive spatial correlation and agglomeration in the coupling coordination between digital village development and green land use. The steadily increasing Moran index indicates intensifying spatial clustering over time. To examine localized clustering, we calculated the local Moran’s I and generated LISA maps for 2018, 2020, and 2022. Figure 9 shows that high-high clusters—provinces with strong coupling coordination linked to similarly strong neighbors—consistently included Jiangsu (Eastern Coastal), Fujian (Southern Coastal), and Anhui (Middle Yangtze River). These provinces maintain frequent exchanges of technology, resources, and talent. Low-low clusters, reflecting weak coordination and limited interaction, included Xinjiang and Gansu (Northwest) and Jilin (Northeast), constrained by low technology adoption, poor infrastructure, and weak economies.

FIGURE 9

Shandong Province (Northern Coastal) shifted from a high-high cluster to a non-clustered status because neighboring provinces advanced more rapidly, weakening Shandong’s relative position despite absolute growth. Inner Mongolia (Middle Yellow River) moved from non-clustered to low-low, reflecting the slow adoption of digital governance and green agricultural practices. These examples demonstrate that regional coordination depends not only on absolute improvements but also on relative progress and spatial spillover effects. Other zones exhibited minimal clustering changes. Overall, high-high and low-low clusters remained largely stable, indicating persistent spatial differentiation in coupling coordination across China.

Furthermore, Figure 9 reveals that during the study period, the coupling coordination degree between digital villages and green land use efficiency exhibited localized spatial clustering. Jiangsu Province in the eastern coastal region, Fujian Province in the southern coastal region, and Anhui Province in the middle Yangtze River basin consistently remained within the high-high clustering zone. This indicates that these provinces maintain high coupling coordination and close ties with neighboring provinces, characterized by frequent exchanges of technology, resources, and talent. Interestingly, the Xinjiang Uygur Autonomous Region and Gansu Province in the northwest, along with Jilin Province in the northeast, consistently remained in the low-low clustering zone. This was primarily due to weak economic foundations, low technology adoption rates, and inadequate infrastructure. It is worth noting that Shandong Province, located in the Northern Coastal Economic Zone, has transitioned from a high-to-high agglomeration pattern to a non-spatial agglomeration model. This may stem from structural shifts in its relative development momentum. Specifically, while Shandong Province’s own coupling coordination index continued to grow during the study period, its growth rate seems to be moderate.

Concurrently, neighboring provinces within the same high-high cluster zone experienced more rapid improvements in their coupling coordination indices. This led to a weakening of Shandong’s relative competitiveness within the region, consequently reducing its spatial dependency. More so, the Inner Mongolia Autonomous Region in the middle reaches of the Yellow River, influenced by surrounding areas, has transitioned from a non-spatial agglomeration pattern to a low-low agglomeration zone. This reflects limited progress in digital governance and the green transformation of farmland within the region. Together, these two shifts reveal that during the high-quality development phase, regional coordinated development depends not only on improving its own absolute level but is also significantly influenced by the relative pace of development within the region and the nature of spatial spillover effects. Nevertheless, other economic zones have not yet exhibited significant spatial clustering phenomena. Over and above, the spatial clustering pattern of the coupling coordination degree between China’s digital villages and green utilization efficiency of cultivated land did not undergo drastic changes during the study period. In particular, the provincial and municipal members within high-high and low-low clustering zones remained relatively stable.

4.3 Regional differences and sources of coupling coordination

We quantified regional disparities using the Dagum Gini coefficient. The overall Gini coefficient declined from 0.163 in 2018 to 0.148 in 2022, averaging a 1.84% annual reduction (Table 8), reflecting a gradual narrowing of regional differences in coordination. This trend indicates that investments in digital rural infrastructure and green land-use improvements are promoting more balanced and synchronized development.

Within regions (Table 9), the Northern Coastal and Northeast Zones initially experienced rising disparities between 2018 and 2019 (Gini coefficients 0.111→0.145 and 0.073→0.094, respectively) before gradually converging. The Great Southwest Zone displayed an inverse pattern, with its Gini coefficient falling from 0.054 to 0.036 in 2018–2019, then rising in subsequent years. The Great Northwest and Middle Yellow River Zones reached minimal internal disparities in 2020 (0.092 and 0.059) before divergence resumed. The Middle Yangtze River Zone stabilized after peaking in 2020. Eastern and Southern Coastal Zones consistently reduced internal disparities, reflecting effective integration of digital and green land-use strategies.

Across the eight zones, intra-regional disparities narrowed overall. Most zones displayed small Gini coefficients, indicating relatively minor internal variation. The Northern Coastal Zone maintained the largest internal disparities, highlighting the need for targeted interventions to strengthen coordination between digital rural development and green arable land use. Conversely, the Southern Coastal Zone demonstrated minimal internal differences, reflecting balanced development and strong cross-regional synergy (Table 10).

Inter-regionally, disparities widened between the Northeast and both the Southern Coastal and Middle Yangtze River Zones, as well as between the Northern Coastal Zone and the Southern Coastal, Middle Yellow River, and Middle Yangtze River Zones. Other regional pairs showed narrowing gaps, signaling improvements in inter-regional coordination. The Southern Coastal Zone and the Middle Yellow River and Middle Yangtze River Zones maintained low inter-regional differences, while the Great Northwest exhibited significant internal disparities and pronounced coordination gaps with the Northeast, Northern, and Eastern Coastal Zones. Addressing both intra- and inter-regional imbalances in the Great Northwest remains a critical priority.

Decomposition analysis (Figure 10) revealed that inter-regional disparities contributed 72%–77% of the overall variation in coupling coordination, while intra-regional differences accounted for only 6%–7%. This finding indicates that inter-regional differences primarily drive overall variation and aligns with China’s regional coordination and rural revitalization strategies. Strengthening cross-regional coordination mechanisms—through joint infrastructure development, technology diffusion, talent mobility, and ecological compensation—is essential to reduce structural disparities among the eight comprehensive economic zones and promote overall improvements in coordination between digital rural development and green arable land use.

FIGURE 10

5 Discussions

Digital rural development and green land use efficiency interact to drive transformative changes in sustainable agriculture and rural revitalization, particularly under the pressures of digitalization and globalization. Our study examines the coupling and coordination mechanisms between these two systems across China’s eight comprehensive economic zones, showing how their integration enhances green agricultural productivity, optimizes land resource allocation, and strengthens ecological security.

The improved coupling coordination degree model provides a critical tool for analyzing the synergistic interactions between digital village development and arable land’s green utilization efficiency. It offers actionable insights for decision-makers by accurately capturing spatial heterogeneity in coordination patterns. This study, therefore, reveals how the coordination relationship demonstrates significant spatial heterogeneity across regions, aligning with previous research findings (Assunção et al., 2019). Unlike traditional models with equal system contribution coefficients (Kwilinski et al., 2023), which may overlook complex interdependencies, our model assigns greater weight to lagging systems, enabling a more precise assessment of how underdeveloped regions influence overall coordination. Departing from the prevailing view that assumes a simple linear driving relationship (Basso and Antle, 2020; Tang and Chen, 2022), our findings derived from the enhanced model reveal distinct nonlinear constraints. This divergence is particularly pronounced in the Northeast region, which exhibits typical characteristics of asynchronous development. While the region boasts leading green utilization efficiency (Zhang et al., 2024), its coordination level shows a declining trend due to lagging digitalization. This finding challenges the previously held simplistic linear understanding of their relationship. Specifically, our results confirm that the relative lag in digital development imposes significant constraints on overall coordinated progress. As a result, this constraint tends to mask internal structural imbalances within the system, ultimately leading to misjudgments about the true coordination status of the region (Jain and Agnihotri, 2020; Rotz et al., 2019).

By integrating both digital rural development and green land use into a unified framework, our analysis fills a gap in previous research that often considered these systems in isolation (Wang et al., 2023; Yin et al., 2017). Wang et al. (2023), focuses solely on digital rural development as a key driver of rural economic growth (Wang et al., 2023), while Yin et al. (2017) discusses the green utilization efficiency of farmland in isolation within a resource-environmental constraints framework (Yin et al., 2017). Furthermore, while earlier studies predominantly examined provincial units (Li et al., 2023; Wen and Li, 2024), our study evaluates eight major economic zones, revealing that interregional disparities drive most of the observed inequalities in coupling coordination. This regional perspective exposes nuanced heterogeneity, offering deeper insight into spatial dynamics and synergistic development patterns. Therefore, this provides a more granular perspective than the nationwide analysis conducted by Zhao et al. (2024). Our methodology reveals more nuanced regional heterogeneity. Regarding spatial spillover effects, empirical findings confirm the prevailing consensus on positive spillovers (Haldar and Sethi, 2022). However, further decomposition analysis reveals an often-overlooked limitation: the dominant role of interregional disparities (contributing over 70%) indicates that natural resource flows alone are insufficient to bridge the substantial “east-high, west-low” gap (Pick et al., 2020). This implies that relying on market mechanisms for spontaneous adjustment cannot resolve this issue; more proactive, coordinated interventions are essential. High-performing regions emerge as innovation hubs, demonstrating best practices in technology adoption, infrastructure development, and green agriculture. In contrast, underperforming regions require targeted interventions to strengthen digital infrastructure, facilitate technology transfer, and build capacity for sustainable land use. Our findings highlight the importance of balancing absolute improvements with relative development rates to ensure effective regional spillover and convergence.

Despite these contributions, our study has limitations. The current indicator system could be expanded to include dimensions such as digital healthcare, rural tourism, and e-commerce. Additionally, our analysis primarily measures coordination degrees without deeply investigating the drivers behind them. Future research should apply multivariate models to explore influencing factors and provide more nuanced policy guidance for integrated rural development.

6 Conclusion

Using panel data from 31 Chinese provinces (2018–2022), this study evaluates the coordination between digital village development and green arable land utilization. Applying an improved coupling coordination model alongside kernel density estimation, spatial autocorrelation analysis, and the Dagum Gini coefficient, we draw four main conclusions. First, digital village development has improved, with the average index rising from 0.17 to 0.28, and shows a gradual decline from east to west. Integrating digital village indicators and green land use efficiency into a single analytical framework enriches our understanding of regional disparities in digitalization and land management. Second, green land use efficiency peaked around 2020, with coastal regions outperforming the Middle Yangtze River Zone. Using the improved coupling coordination model allowed us to capture interactive effects more accurately, avoiding biases inherent in traditional equal-weight models.

Third, coupling coordination across the eight zones rose steadily, yet most regions—save the Southern Coastal and Middle Yangtze River Zones—showed polarization trends. Examining coordination at the level of integrated economic zones enabled us to identify fine-grained regional heterogeneity. Fourth, an east-west decline and significant spatial agglomeration in coordination were observed, with 75.29% of the total variation explained by inter-regional disparities stemming from differences in developmental, digital, and agricultural land-use factors. Over and above, our findings highlight the critical linkage between digital infrastructure and sustainable land use, providing theoretical and empirical guidance for promoting rural digital transformation and green agricultural practices.

Based on our research findings, we propose the following targeted and actionable policy implications: First, for districts exhibiting high levels of coordination and “high-high” agglomeration characteristics, emphasis should be placed on leveraging their demonstration and radiating effects. Therefore, regions exhibiting high coordination and high-high clustering should serve as innovation hubs. Authorities should disseminate best practices, promote cross-regional knowledge transfer, and strengthen joint infrastructure and technology initiatives with surrounding lower-performing areas to enhance synergistic development. On the one hand, by summarizing mature digital rural development experiences and green farmland utilization models, cross-regional dissemination of technical expertise and management models should be promoted. On the other hand, leveraging regional collaboration mechanisms to strengthen joint digital infrastructure construction and technology sharing with surrounding areas exhibiting lower coordination levels will enhance overall synergistic development.

Second, regions with low coordination and low-low clustering require targeted interventions. Increased fiscal support for digital infrastructure, technical training, demonstration projects, and subsidies should guide farmers toward green production methods, improving the intensive and sustainable use of arable land. Specifically, fiscal support for digital infrastructure construction should be increased to enhance rural area informatization and digital application capabilities. Concurrently, targeted subsidies, technical training, and demonstration projects will guide farmers and agricultural operators toward adopting green production methods, thereby improving the intensive and green utilization potential of arable land resources.

Third, given that inter-regional disparities drive most coordination inequality, authorities should facilitate exchanges, collaborative projects, and resource sharing across economic zones. This approach will help balance development, enhance digital technology diffusion, and promote the coordinated growth of digital rural infrastructure and green land-use efficiency. However, this study faces limitations due to its reliance on secondary data, which may not fully reflect local-level variations in digital adoption or agricultural practices. Future research should integrate longitudinal case studies, qualitative fieldwork, and dynamic indicators—such as digital literacy, climate resilience, and real-time agricultural outputs—to better understand micro-level drivers and barriers to coordination.

References

• 1

  AntleJ. M.StoorvogelJ. J.ValdiviaR. O. (2014). New parsimonious simulation methods and tools to assess future food and environmental security of farm populations. Philosophical Trans. R. Soc. B Biol. Sci.369 (1639), 20120280. 10.1098/rstb.2012.0280

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2

  AssunçãoJ.BragançaA.HemsleyP. (2019). Geographic heterogeneity and technology adoption: evidence from Brazil. Land Econ.95 (4), 599–616. 10.3368/le.95.4.599

  • CrossRef
  • Google Scholar

• 3

  BassoB.AntleJ. (2020). Digital agriculture to design sustainable agricultural systems. Nat. Sustain.3 (4), 254–256. 10.1038/s41893-020-0510-0

  • CrossRef
  • Google Scholar

• 4

  BimberB. (1999). The internet and citizen communication with government: does the medium matter?Polit. Commun.16 (4), 409–428. 10.1080/105846099198569

  • CrossRef
  • Google Scholar

• 5

  ChenQ.XieH. (2019). Temporal-spatial differentiation and optimization analysis of cultivated land green utilization efficiency in China. Land8 (11), 158. 10.3390/land8110158

  • CrossRef
  • Google Scholar

• 6

  ChenY.WangS.WangY. (2022). Spatiotemporal evolution of cultivated land non-agriculturalization and its drivers in typical areas of southwest China from 2000 to 2020. Remote Sens.14 (13), 3211. 10.3390/rs14133211

  • CrossRef
  • Google Scholar

• 7

  ChenX.MengQ.WangK.LiuY.ShenW. (2023). Spatial patterns and evolution trend of coupling coordination of pollution reduction and carbon reduction along the yellow river basin, China. Ecol. Indic.154, 110797. 10.1016/j.ecolind.2023.110797

  • CrossRef
  • Google Scholar

• 8

  ChengK.WangX.LiuS.ZhuangY. (2024). Spatial differences and dynamic evolution of economic resilience: from the perspective of China’s eight comprehensive economic zones. Econ. Change Restruct.57 (2), 73. 10.1007/s10644-024-09665-2

  • CrossRef
  • Google Scholar

• 9

  DaiX.ChenY.ZhangC.HeY.LiJ. (2023). Technological revolution in the field: green development of Chinese agriculture driven by digital information technology (DIT). Agriculture13 (1), 199. 10.3390/agriculture13010199

  • CrossRef
  • Google Scholar

• 10

  de ClercqM.D'HaeseM.BuysseJ. (2023). Economic growth and broadband access: the European urban-rural digital divide. Telecommun. Policy47 (6), 102579. 10.1016/j.telpol.2023.102579

  • CrossRef
  • Google Scholar

• 11

  DubeyA.LalR. (2009). Carbon footprint and sustainability of agricultural production systems in Punjab, India, and Ohio, USA. J. Crop Improv.23 (4), 332–350. 10.1080/15427520902969906

  • CrossRef
  • Google Scholar

• 12

  FingerR. (2023). Digital innovations for sustainable and resilient agricultural systems. Eur. Rev. Agric. Econ.50 (4), 1277–1309. 10.1093/erae/jbad021

  • CrossRef
  • Google Scholar

• 13

  FosterL.SzilagyiK.WairegiA.OguamanamC.de BeerJ. (2023). Smart farming and artificial intelligence in East Africa: addressing indigeneity, plants, and gender. Smart Agric. Technol.3, 100132. 10.1016/j.atech.2022.100132

  • CrossRef
  • Google Scholar

• 14

  GengG.ShenY.DongC. (2024). The impact of green finance on agricultural non-point source pollution: analysis of the role of environmental regulation and rural land transfer. Land13 (9), 1516. 10.3390/land13091516

  • CrossRef
  • Google Scholar

• 15

  GillerK. E.HijbeekR.AnderssonJ. A.SumbergJ. (2021). Regenerative agriculture: an agronomic perspective. Outlook Agric.50 (1), 13–25. 10.1177/0030727021998063

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  GrecoS.IshizakaA.TasiouM.TorrisiG. (2018). On the methodological framework of composite indices: a review of the issues of weighting, aggregation, and robustness. Soc. Indic. Res.141 (1), 61–94. 10.1007/s11205-017-1832-9

  • CrossRef
  • Google Scholar

• 17

  GuoB.YuH.JinG. (2024). Urban green total factor productivity in China: a generalized luenberger productivity indicator and its parametric decomposition. Sustain. Cities Soc.106, 105365. 10.1016/j.scs.2024.105365

  • CrossRef
  • Google Scholar

• 18

  HakenH. (2007). Synergetics. Scholarpedia2 (1), 1400. 10.4249/scholarpedia.1400

  • CrossRef
  • Google Scholar

• 19

  HaldarA.SethiN. (2022). Environmental effects of information and communication technology - exploring the roles of renewable energy, innovation, trade and financial development. Renew. Sustain. Energy Rev.153, 111754. 10.1016/j.rser.2021.111754

  • CrossRef
  • Google Scholar

• 20

  JainA.AgnihotriS. B. (2020). Assessing inequalities and regional disparities in child nutrition outcomes in India using MANUSH - a more sensitive yardstick. Int. J. Equity Health19 (1), 138. 10.1186/s12939-020-01249-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21

  JenksG. F.CaspallF. C. (1971). Error on choroplethic maps: definition, measurement, reduction. Ann. Assoc. Am. Geogr.61 (2), 217–244. 10.1111/j.1467-8306.1971.tb00779.x

  • CrossRef
  • Google Scholar

• 22

  KoondharM. A.QiuL.MagsiH.ChandioA. A.HeG. (2018). Comparing economic efficiency of wheat productivity in different cropping systems of Sindh province, Pakistan. J. Saudi Soc. Agric. Sci.17 (4), 398–407. 10.1016/j.jssas.2016.09.006

  • CrossRef
  • Google Scholar

• 23

  KwilinskiA.LyulyovO.PimonenkoT. (2023). The coupling and coordination degree of digital business and digital governance in the context of sustainable development. Information14 (12), 651. 10.3390/info14120651

  • CrossRef
  • Google Scholar

• 24

  LiY.WenX. (2023). Regional unevenness in the construction of digital villages: a case study of China. Plos One18 (7), e0287672. 10.1371/journal.pone.0287672

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  LiH.WuY.HuangX.SloanM.SkitmoreM. (2017). Spatial-temporal evolution and classification of marginalization of cultivated land in the process of urbanization. Habitat Int.61, 1–8. 10.1016/j.habitatint.2017.01.001

  • CrossRef
  • Google Scholar

• 26

  LiX.Singh ChandelR. B.XiaX. (2022). Analysis on regional differences and spatial convergence of digital village development level: theory and evidence from China. Agriculture12 (2), 164. 10.3390/agriculture12020164

  • CrossRef
  • Google Scholar

• 27

  LiT.WangS.ChenP.LiuX.KongX. (2023). Geographical patterns and influencing mechanisms of digital rural development level at the county scale in China. Land12 (8), 1504. 10.3390/land12081504

  • CrossRef
  • Google Scholar

• 28

  MaW.McKayA.RahutD. B.SonobeT. (2023). An introduction to rural and agricultural development in the digital age. Rev. Dev. Econ.27 (3), 1273–1286. 10.1111/rode.13025

  • CrossRef
  • Google Scholar

• 29

  MeiY.MiaoJ.LuY. (2022). Digital villages construction accelerates high-quality economic development in rural China through promoting digital entrepreneurship. Sustainability14 (21), 14224. 10.3390/su142114224

  • CrossRef
  • Google Scholar

• 30

  MengH.ChenX.WangC.ZhangB.ZhouZ. (2022). Research on the evaluation of digital village development readiness taking changfeng county as an example. Int. J. Educ. Humanit.2 (3), 155–159. 10.54097/ijeh.v2i3.385

  • CrossRef
  • Google Scholar

• 31

  PengL.MariadasP. A. (2025). Research on the low-carbon spatial spillover effect development of the digital economy enabled by new quality productivity. Sustainability17 (4), 1746. 10.3390/su17041746

  • CrossRef
  • Google Scholar

• 32

  PickJ.SarkarA.ParrishE. (2020). The Latin American and Caribbean digital divide: a geospatial and multivariate analysis. Inf. Technol. Dev.27 (2), 235–262. 10.1080/02681102.2020.1805398

  • CrossRef
  • Google Scholar

• 33

  PickJ.RenF.SarkarA. (2024). Digital inequalities in China in 2020: spatial and multivariate analysis. Appl. Sci.14 (13), 5385. 10.3390/app14135385

  • CrossRef
  • Google Scholar

• 34

  QuY.LyuX.PengW.XinZ. (2021). How to evaluate the green utilization efficiency of cultivated land in a farming household? A case study of Shandong Province, China. Land10 (8), 789. 10.3390/land10080789

  • CrossRef
  • Google Scholar

• 35

  ReidsmaP.MeuwissenM.AccatinoF.AppelF.BardajiI.CoopmansI.et al (2020). How do stakeholders perceive the sustainability and resilience of EU farming systems?EuroChoices19 (2), 18–27. 10.1111/1746-692x.12280

  • CrossRef
  • Google Scholar

• 36

  RijswijkK.KlerkxL.BaccoM.BartoliniF.BultenE.DebruyneL.et al (2021). Digital transformation of agriculture and rural areas: a socio-cyber-physical system framework to support responsibilisation. J. Rural Stud.85, 79–90. 10.1016/j.jrurstud.2021.05.003

  • CrossRef
  • Google Scholar

• 37

  RotzS.GravelyE.MosbyI.DuncanE.FinnisE.HorganM.et al (2019). Automated pastures and the digital divide: how agricultural technologies are shaping labour and rural communities. J. Rural Stud.68, 112–122. 10.1016/j.jrurstud.2019.01.023

  • CrossRef
  • Google Scholar

• 38

  SaleemR.AhmadZ.AneesM.RazzaqA.SaleemA. (2015). Productivity and land use efficiency of maize mungbean intercropping under different fertility treatments. Sarhad J. Agric.31, 37–44. 10.5555/20153325514

  • CrossRef
  • Google Scholar

• 39

  SetiawanA. (2024). The role of village government in digital-based community empowerment in tourism villages. J. Gov.9 (3). 461–475. 10.31506/jog.v9i3.28017

  • CrossRef
  • Google Scholar

• 40

  ShenL.HuangY.HuangZ.LouY.YeG.WongS.-W. (2018). Improved coupling analysis on the coordination between socio-economy and carbon emission. Ecol. Indic.94, 357–366. 10.1016/j.ecolind.2018.06.068

  • CrossRef
  • Google Scholar

• 41

  ShenX.LiuY.LiuB. (2020). Urbanization effect on the observed changes of surface air temperature in northeast China. Terr. Atmos. Ocean. Sci.31 (3), 325–335. 10.3319/tao.2019.11.27.01

  • CrossRef
  • Google Scholar

• 42

  StreitR. P.BellwoodD. R. (2023). Moving beyond the one true trait. Trends Ecol. and Evol.38 (11), 1014–1015. 10.1016/j.tree.2023.08.006

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  SuM.GuoR.HongW. (2019). Institutional transition and implementation path for cultivated land protection in highly urbanized regions: a case study of shenzhen, China. Land Use Policy81, 493–501. 10.1016/j.landusepol.2018.11.015

  • CrossRef
  • Google Scholar

• 44

  TangY.ChenM. (2022). The impact mechanism and spillover effect of digital rural construction on the efficiency of green transformation for cultivated land use in China. Int. J. Environ. Res. Public Health19 (23), 16159. 10.3390/ijerph192316159

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 45

  ToneK. (2002). A slacks-based measure of super-efficiency in data envelopment analysis. Eur. J. Operational Res.143 (1), 32–41. 10.1016/s0377-2217(01)00324-1

  • CrossRef
  • Google Scholar

• 46

  WangP.LiC.HuangC. (2023). The impact of digital village construction on county-level economic growth and its driving mechanisms: evidence from China. Agriculture13 (10), 1917. 10.3390/agriculture13101917

  • CrossRef
  • Google Scholar

• 47

  WenR.LiH. (2024). Impact of digital economy on urban land green use efficiency: evidence from Chinese cities. Environ. Res. Commun.6 (5), 055008. 10.1088/2515-7620/ad4514

  • CrossRef
  • Google Scholar

• 48

  WestT. O.MarlandG. (2002). A synthesis of carbon sequestration, carbon emissions, and net carbon flux in agriculture: comparing tillage practices in the United States. Agric. Ecosyst. and Environ.91 (1-3), 217–232. 10.1016/s0167-8809(01)00233-x

  • CrossRef
  • Google Scholar

• 49

  XieH.ChenQ.WangW.HeY. (2018). Analyzing the green efficiency of arable land use in China. Technol. Forecast. Soc. Change133, 15–28. 10.1016/j.techfore.2018.03.015

  • CrossRef
  • Google Scholar

• 50

  XinQ.WuB.ShiY. (2025). The impact of farmers’ digital participation on cultivated land ecological protection. Sustainability17 (13), 6191. 10.3390/su17136191

  • CrossRef
  • Google Scholar

• 51

  YinG.LiuL.JiangX. (2017). The sustainable arable land use pattern under the tradeoff of agricultural production, economic development, and ecological protection—an analysis of Dongting Lake basin, China. Environ. Sci. Pollut. Res.24 (32), 25329–25345. 10.1007/s11356-017-0132-x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 52

  YuY.ZhangJ.ZhangK.XuD.QiY.DengX. (2023). The impacts of farmer ageing on farmland ecological restoration technology adoption: empirical evidence from rural China. J. Clean. Prod.430, 139648. 10.1016/j.jclepro.2023.139648

  • CrossRef
  • Google Scholar

• 53

  ZhangJ.ZhangW. (2024). The impact mechanism of digital rural construction on land use efficiency: evidence from 255 cities in China. Sustainability17 (1), 45. 10.3390/su17010045

  • CrossRef
  • Google Scholar

• 54

  ZhangP.RenG.QinY.ZhaiY.ZhaiT.TysaS. K.et al (2021). Urbanization effects on estimates of global trends in mean and extreme air temperature. J. Clim.34 (5), 1923–1945. 10.1175/jcli-d-20-0389.1

  • CrossRef
  • Google Scholar

• 55

  ZhangF.XieA.JiangC.ChenJ.AnY.YangP.et al (2024). Coupling coordination analysis and spatiotemporal heterogeneity between urban land green use efficiency and ecosystem services in yangtze river economic belt, China. Humanit. Soc. Sci. Commun.11 (1), 1328. 10.1057/s41599-024-03752-5

  • CrossRef
  • Google Scholar

• 56

  ZhaoS.LiM.CaoX. (2024). Empowering rural development: evidence from China on the impact of digital village construction on farmland scale operation. Land13 (7), 903. 10.3390/land13070903

  • CrossRef
  • Google Scholar

• 57

  ZhouM.SunH.KeN. (2022). The spatial and temporal evolution of coordination degree concerning china’s cultivated land green utilization efficiency and high-quality agricultural development. Land12 (1), 127. 10.3390/land12010127

  • CrossRef
  • Google Scholar