A Coupling Coordination Analysis for Natural Gas Production: A Perspective from the Energy Trilemma

Abstract

The natural gas sector, as a pivotal transition fuel, is fundamentally constrained by the “Energy Trilemma”—the intertwined and often competing goals of energy security, affordability, and sustainability. Current research predominantly focuses on the demand side, leaving a significant gap in understanding the synergistic dynamics within production regions, which are critical to resolving this trilemma at its source. To address this gap, this study constructs a “Safety–Economy–Green” (S-E-G) evaluation framework aligned with the trilemma’s dimensions. Utilizing panel data (2011–2021) from four major Chinese natural gas production regions (Sichuan, Chongqing, Shaanxi, and Shanxi). By integrating the Entropy Weight Method, a Coupling Coordination Model, and Kernel Density Estimation, it delineates the system’s synergistic dynamics from both temporal and regional perspectives. The key findings are as follows: (1) Significant disparities and polarization are observed in the S and G dimensions, while the E dimension shows a narrowing gap, with its peak height increasing by 177.8% and bandwidth shrinking by 64.2%. G has emerged as a constraint on overall system coupling coordination. The persistently high coupling degree—rising from 0.87 in 2011 to 0.97 in 2021 while consistently exceeding the coordination degree, which increased from 0.45 to 0.62—underscores the continued need for improvement in synergistic development. (2) The coupling coordination degree of the S-E-G system underwent a three-stage evolution: rapid improvement (2011–2013, from 0.36 to 0.58 at 7.3% annually), fluctuating adjustment (2014–2017, between 0.58 and 0.66), and finally high-level stability (2018–2021, stabilizing at 0.76–0.80). (3) Obvious regional differentiation exists: Sichuan achieved a moderate level of 0.76 by 2021, Shaanxi maintained primary coupling coordination (0.6–0.7), while Chongqing and Shanxi remained marginal, fluctuating between 0.4 and 0.6. Enhancing subsystem coordination and implementing differentiated pathways are therefore essential for these regions’ sustainable development. The study suggests promoting the sustainable development of natural gas production regions by enhancing subsystem coordination and exploring differentiated pathways, thereby providing practical guidance for the energy transition of resource-based regions.

1. Introduction

The global energy transformation is accelerating in response to the dual objectives of “carbon peaking and carbon neutrality”, alongside the broader trajectory of the energy system revolution [1,2,3,4]. Against this backdrop, China’s energy sector continues to grapple with the longstanding challenge of the “Impossible Triangle”, the inherent difficulty of simultaneously achieving energy security, economic efficiency, and environmental sustainability. This triadic dilemma has become increasingly prominent [5], a significant barrier to high-quality energy development. As a key transitional fuel and a representative of green energy, natural gas plays a pivotal role in China’s energy strategy. It serves as a cleaner alternative to coal and acts as a stabilizing force amid the shift toward renewable energy sources [6]. The Chinese government has thus prioritized the development and efficient utilization of natural gas resources to safeguard energy security during the transition and to support national carbon-emission reduction targets [7,8,9]. However, the natural gas industry itself is not immune to the constraints of the “Impossible Triangle”. Striking a balance between safety (S), economic efficiency (E), and green (G) low-carbon production has proven challenging, hindering the sector’s coordinated, sustainable development. Natural gas’s strategic significance extends beyond its role as an energy source. It is closely tied to national economic security, the optimization of the energy structure, the enhancement of supply capacity, and the broader goals of sustainable social and economic progress. Consequently, there is a pressing need to develop an “S-E-G” evaluation framework that serves as a critical tool to navigate this trilemma and enable the strategic balance essential to the high-quality, sustainable transformation of China’s natural gas sector.

This persistent trilemma in the energy sector has been vividly termed the “Impossible Triangle,” a concept that initially emerged in financial theory through the Mundell-Fleming model of the 1960s [10,11]. It has been widely adopted across disciplines to explore trade-offs among three competing objectives. For example, Ref. [12] analyzed the trade-offs between exchange rate stability, financial market openness, and monetary policy independence. Ref. [13] modeled the economic development trilemma in the Nile River Basin using indicators of GDP growth, poverty, and inequality. Ref. [14] developed a framework to assess zero-energy building (ZEB) performance based on energy-related trilemma indicators, while Ref. [15] investigated the tensions between carbon market integration, national sovereignty, and policy flexibility in China. Additional studies have examined resilience dilemmas [16], trade-offs between logistics and the environment [17], and other trilemma dynamics in complex systems. In the energy field, the “Impossible Triangle” was formally conceptualized by the World Energy Council (WEC) in 2011 [18]. It encapsulates the persistent difficulty of achieving energy security, affordability, and environmental sustainability simultaneously. Decision-makers must strike a balance among them, transforming the “impossible triangle” into a “possible triangle”. Since then, numerous studies have sought to quantify this dilemma through multidimensional analytical methods, such as entropy weighting, principal component analysis, and spatial modeling [19,20,21]. For example, Ref. [22] evaluated national energy trilemma performance using an interval decision matrix and PCA across energy security, equity, and environmental dimensions. Ref. [23] developed a coupling coordination model based on energy, economy, and environmental metrics. In contrast, Ref. [24] employed global principal component analysis to evaluate China’s regional energy-economic security levels. Emerging approaches also apply geographic information system (GIS) tools to map the spatiotemporal evolution of energy trilemma indices [25,26,27]. Meanwhile, research on the “impossible triangle” concerning specific fossil energy categories has also been conducted. Ref. [28], based on the energy trilemma, proposed an innovative “three-gate process” for oil production enterprises. Ref. [29] examined the “trilemma” among fossil energy rents (coal, oil, etc.), economic growth, and energy efficiency in Africa. Ref. [30] analyzed the structural dilemma faced by India’s coal-dominated power system during its energy transition from the triangular dimensions of “accessibility, security, and sustainability.”

Despite these advances, a critical gap remains: while the classic “Energy–Environment–Economy” (3E) framework focuses on consumption, the dynamics among safety, economy, and environmental sustainability in the production lifecycle are critically underexplored, especially for natural gas. Existing production-side studies remain largely safety-centric. For instance, although references such as Reference [31] (assessing U.S. underground hydrogen storage potential) and Reference [32] (a European techno-economic evaluation of hydrogen storage) link “production-storage” safety to geomechanics and leakage risks, they primarily address physical storage potential. They have yet to delve deeply into the complex coupling feedback mechanisms among “Safety–Economy–Green” (S-E-G) across the entire production chain (extraction, processing, storage, and transportation), leaving research in this area notably scarce. This study addresses this gap by shifting focus to the production side and proposing a dedicated “S-E-G” coupling analysis framework. Unlike the broader 3E model, this framework directly captures the systemic characteristics and interactions within production regions, enabling a finer-grained analysis of synergies and trade-offs in natural gas production.

This gap underscores the need to expand the “Impossible Triangle” framework to natural gas production regions, particularly in China, where most supply originates from central and western areas. These regions support local demand and serve as the backbone of interprovincial transmission projects, including the West-to-East and Sichuan-to-East gas pipelines. However, as demand surges, the pressure to produce natural gas safely, economically, and sustainably is mounting, necessitating a more integrated and regionally coordinated approach to production [33]. In this context, the natural gas production zones of Sichuan, Chongqing, Shaanxi, and Shanxi offer a representative case for examining the dynamics of the “Possible Triangle”. This study operationalizes the “Possible Triangle” concept by developing a multidimensional evaluation framework for safety, economic performance, and green performance. Its innovation lies in developing a production-centered “S-E-G” assessment framework that integrates the entropy weight method, kernel density method, and a coupling coordination degree model. Through focused empirical analysis of China’s natural gas production regions, it reveals the spatiotemporal evolution patterns of the coupling coordination degree. The goal is to generate actionable insights that can inform strategic planning and policymaking for the high-quality development of the natural gas sector. This study contributes to advancing China’s energy transition and the global discourse on sustainable energy development by offering replicable frameworks and approaches for overcoming the energy production system’s trilemma.

2. Definition of the “Possible Triangle” of Regional Natural Gas Industry

Under the “Impossible Triangle” paradigm of energy systems, the inherent tension among security, economy, and environmental sustainability arises from the exclusive competition for resources. This contradiction is acute in China’s natural gas industry, driven by vast geographic disparities and rising energy demand. Integrating synergy theory [34] and the energy trilemma [18], this study reframes the traditional “Impossible Triangle” and proposes a “Possible Triangle” for natural gas production regions.

To empirically ground this conceptual shift, a coupling coordination degree model is employed to quantify the synergistic interactions within the “Possible Triangle.” This framework integrates three critical subsystems: Safety (S) [35], Economy (E) [24], and Green (G) [36]. Safety ensures reliability and a stable supply throughout the gas production chain, avoiding disruptions and sudden risks for continuous output. Economy focuses on optimizing resource allocation and costs to maximize economies of scale, efficiency, profitability, and market competitiveness. Green promotes ecological coordination by controlling emissions, increasing investment in pollution control, and enhancing regional environmental sustainability.

Although distinct, as shown in Figure 1, these subsystems are interdependent in a complex coupling system: S is the foundation for E and G. Stable supply prevents economic losses and enables environmental protection; reliable gas can replace polluting fuels. E drives S and G: Economic benefits attract investment, spur innovation, improve efficiency, and fund safety and environmental measures, creating a positive cycle. G is the goal for S and E: Development must serve ecological sustainability, require emission control and minimize environmental impact, aiming for harmony between energy extraction and nature. This coupling coordination mechanism creates a complementary, dynamically balanced system. It achieves “Overall Emergence” [37], in which synergistic benefits exceed the sum of the individual parts, thereby realizing the “Possible Triangle” in natural gas production regions.

3. Method

The technical framework of this study, outlined in Figure 2, employs a multi-stage process to assess the S-E-G coupling coordination in four major natural gas production regions. The research begins by constructing a comprehensive evaluation index system and collecting the corresponding data. This is followed by applying a coupling coordination degree model to evaluate the system’s state. The resulting data are then analyzed to elucidate spatiotemporal evolution patterns, ultimately informing practical strategies for realizing the “Possible Triangle” in production regions.

3.1. Construction of the Evaluation Index System

Guided by the core “possible triangle” framework for natural gas production regions, this study identified the most suitable evaluation indicators through a comprehensive review of relevant literature and statistical yearbooks. The proposed evaluation framework encompasses the three dimensions of safety, economy, and green aspects in natural gas production, as outlined in Table 1.

Traditional subjective weighting methods (such as AHP), rely on expert judgment to quantify the importance of criteria. However, they often overlook the biases and uncertainties inherent in such judgments, which may fail to fully reflect practical realities [38]. The core objective of this study is to reveal the objective differences and evolutionary information contained within the evaluation indicator data themselves. Therefore, the entropy weight method, an objective weighting approach, was selected. This method derives weights entirely based on the degree of dispersion of each indicator’s data within the sample through mathematical derivation. This process eliminates human subjective judgment, ensuring that weight allocation is entirely data-driven, thereby enhancing the transparency and reproducibility of the evaluation foundation [39].

(1)

Data normalization with min-max scaling [40].

Normalization of positive indicators:

$$x_{imn}^{\prime }={\displaystyle \frac{x_{imn}-\min \left(x_{imn}\right)}{\max \left(x_{imn}\right)-\min \left(x_{imn}\right)}}$$

(1)

Normalization of negative indicators:

$$x_{imn}^{\prime }={\displaystyle \frac{\mathrm{man}\left(x_{imn}\right)-x_{imn}}{\max \left(x_{imn}\right)-\mathrm{m}\mathrm{i}\mathrm{n}\left(x_{imn}\right)}}$$

(2)

where $x_{imn}$ is the raw data of the $n$th indicator of province $m$ in year $i$; $x_{imn}^{\prime }$ is the indicator data after standardization of the raw data; $\max \left(x_{imn}\right)$ is the maximum value of the $n$th indicator; $\mathrm{m}\mathrm{i}\mathrm{n}\left(x_{imn}\right)$ is the minimum value of the $n$th indicator. The standardized results indicate that all indicators are distributed within the range of 0 to 1 after standardization, with no significant skewness, meeting the requirements for modeling.

(2)

Data translation X_imn:

$$X_{imn}=x_{imn}^{\prime }+\sigma$$

(3)

where $\sigma$ is the magnitude of the translation, where σ is the magnitude of the translation, for the later elimination of the effect of 0 generated by the normalization of the data. This study selects the value of $\sigma$ as 0.0001 to minimize the impact of the translation on the original data, which aligns with common practices in entropy weight applications to ensure computational stability without altering the intrinsic data structure [41].

(3)

Indicator proportion determination $y_{imn}$:

$$y_{imn}={\displaystyle \frac{X_{imn}^{\prime }}{{\sum }_{i=1}^I{\sum }_{m=1}^M{x}_{imn}^{\prime }}}$$

(4)

where $I$ is the total number of years, and $M$ is the total number of provinces.

(4)

Determination of indicator weights:

Determine the entropy of the $n$th metric:

$$e_n=-k\sum _{i=1}^I\sum _{m=1}^M{y}_{imn}\times \ln y_{imn}$$

(5)

where $k=-{\displaystyle \frac{1}{\ln \left(IM\right)}}$, $k>0$, $e_n>0$.

Determine the coefficient of variation for the $n$th indicator:

$$g_n=1-e_n$$

(6)

Determine the weight of the $n$th indicator:

$${\omega }_n={\displaystyle \frac{g_n}{{\sum }_n^N{g}_n}}$$

(7)

where $N$ is the number of indicators.

(5)

Calculation of composite score:

$$U_m=\sum _{n=1}^N{\omega }_n{x}_{imn}^{\prime }$$

(8)

Table 1. Evaluation index system of “S-E-G” in natural gas production regions.

Primary Indicators	Secondary Indicators	Interpretation and Calculation of Indicators	Units	Weight	Reference
Safety	Level of natural gas production	Total natural gas production	108 m3	0.159	[42]
	Natural gas export level	natural gas export volume	108 m3	0.206	\
	Scientific and technical inputs to natural gas extraction	Research and development (R&D) expenses of natural gas extraction	108 CNY	0.267	[43]
	Natural gas transportation capacity	Natural gas pipeline length	km	0.133	[44]
	Natural gas extraction capacity	Number of people employed in natural gas extraction	104 people	0.235	\
Economy	Natural gas gate station price	The sum of ex-factory and transportation prices of natural gas	CNY/km3	0.135	[45]
	Industrial cost margins in the natural gas extraction industry	Total profit/total costs	%	0.283	[46]
	Investment in the natural gas extraction industry	Reflecting the intensity of investment in the construction of natural gas extraction facilities	108 CNY	0.282	[47]
	Natural gas loss ratio	Gas losses/gas production	%	0.097	\
	Growth rate of total industrial output value of natural gas exploitation industry	Reflecting the economic development of the natural gas extraction industry	%	0.203	\
Green	Carbon emissions from natural gas production and supply	Carbon dioxide from natural gas during the supply phase of production	t	0.075	\
	Industrial SO2 emission intensity	Industrial sulfur dioxide emissions/Area	t/km2	0.108	[48]
	Number of green inventions	Reflecting green and low-carbon development capacity	\	0.279	\
	Area covered by greenery	Reflecting urban greening	ha	0.253	[49]
	Total investment in environmental pollution control	Reflecting the level of environmental pollution control	108 CNY	0.285	[50]

Table 1. Evaluation index system of “S-E-G” in natural gas production regions.

Primary Indicators	Secondary Indicators	Interpretation and Calculation of Indicators	Units	Weight	Reference
Safety	Level of natural gas production	Total natural gas production	108 m3	0.159	[42]
	Natural gas export level	natural gas export volume	108 m3	0.206	\
	Scientific and technical inputs to natural gas extraction	Research and development (R&D) expenses of natural gas extraction	108 CNY	0.267	[43]
	Natural gas transportation capacity	Natural gas pipeline length	km	0.133	[44]
	Natural gas extraction capacity	Number of people employed in natural gas extraction	104 people	0.235	\
Economy	Natural gas gate station price	The sum of ex-factory and transportation prices of natural gas	CNY/km3	0.135	[45]
	Industrial cost margins in the natural gas extraction industry	Total profit/total costs	%	0.283	[46]
	Investment in the natural gas extraction industry	Reflecting the intensity of investment in the construction of natural gas extraction facilities	108 CNY	0.282	[47]
	Natural gas loss ratio	Gas losses/gas production	%	0.097	\
	Growth rate of total industrial output value of natural gas exploitation industry	Reflecting the economic development of the natural gas extraction industry	%	0.203	\
Green	Carbon emissions from natural gas production and supply	Carbon dioxide from natural gas during the supply phase of production	t	0.075	\
	Industrial SO2 emission intensity	Industrial sulfur dioxide emissions/Area	t/km2	0.108	[48]
	Number of green inventions	Reflecting green and low-carbon development capacity	\	0.279	\
	Area covered by greenery	Reflecting urban greening	ha	0.253	[49]
	Total investment in environmental pollution control	Reflecting the level of environmental pollution control	108 CNY	0.285	[50]

3.2. Verification of Collinearity Among Indicators

Calculation of the Pearson Correlation Coefficient Matrix:

$$r_{pq}={\displaystyle \frac{{\sum }_{i=1}^I{\sum }_{m=1}^M{(x}_{imp}^{\prime }-{\overline{x}}_p^{\prime }\left)\right(x_{imq}^{\prime }-{\overline{x}}_q^{\prime })}{\sqrt{{\sum }_{i=1}^I{\sum }_{m=1}^M{{(x}_{imp}^{\prime }-{\overline{x}}_p^{\prime })}^2{(x_{imq}^{\prime }-{\overline{x}}_q^{\prime })}^2}}}$$

(9)

where $r_{pq}$ is the correlation between indicator $p$ and indicator $q$. $x_{imp}^{\prime }$ and $x_{imq}^{\prime }$ are the standardized values of indicator $p$ and indicator $q$ for province $m$ in year $i$. ${\overline{x}}_p^{\prime }$ and ${\overline{x}}_q^{\prime }$ are the overall means of the standardized indicators $p$ and $q$ across all provinces and all years.

$$\mathrm{R}=\left[\begin{array}{cc}\begin{array}{cc}{r}_{11}& r_{12}\\ r_{21}& r_{22}\end{array}& \begin{array}{cc}\cdots & r_{1N}\\ \cdots & r_{2N}\end{array}\\ \begin{array}{cc}⋮& ⋮\\ r_{N1}& r_{N2}\end{array}& \begin{array}{cc}\ddots & ⋮\\ \cdots & r_{NN}\end{array}\end{array}\right]$$

(10)

where $\mathrm{R}$ is the correlation matrix.

The calculation results show that the absolute values of correlation coefficients between the vast majority of indicators are below 0.6 (a moderate level), indicating the absence of strong collinearity [51]. More importantly, the entropy weight method adopted in this study is fundamentally different from regression models: its weights are entirely determined by the dispersion (information entropy) of the indicator data and do not involve the estimation of inter-variable regression relationships. Therefore, the potential collinearity among indicators does not pose a substantial threat to the core conclusions of this research, and the model evaluation results are robust.

3.3. Coupled Coordination Degree Model

The “possible triangle” concept in natural gas production regions reflects the complementarity and coordination among safety, economy, and green dimensions [52]. Coupling coordination theory effectively captures the interactions and dependencies between these subsystems. While the coupling degree measures the strength of their interconnection, it does not assess balance. The coordination degree reflects the harmony in development goals, speed, scale, and quality, helping identify shortcomings. The coupling coordination degree model combines both strengths, offering a comprehensive evaluation of system effectiveness, development quality, and optimization paths [53,54]. Ref. [55] examines the degree of coupling and coordination among security, economic feasibility, and environmental sustainability in natural gas production areas of Northwest China, which demonstrates that the coupling coordination degree model provides an effective tool for evaluating the synergistic development level among the “Safety–Economy–Green” subsystems in natural gas production regions.

(1) “Safety–Economy–Green” Coupled Coordination Degree Model:

$$C_{jt}=3\sqrt[3]{S_{tjk}{E}_{tjk}{G}_{tjk}}/(S_{tjk}+E_{tjk}+G_{tjk})$$

(11)

$$T_{jt}=\alpha S_{tjk}+\beta E_{tjk}+\gamma G_{tjk}$$

(12)

$$D_{jt}=\sqrt{T_{jt}{C}_{jt}}$$

(13)

where $t$ is time; $j$ is region; $k$ is the corresponding secondary indicators under the primary indicators; $C_{jt}$ is the coupling degree; $D_{jt}$ is the coupling coordination degree; $T_{jt}$ is the comprehensive evaluation index; $S_{tjk},E_{tjk},G_{tjk}$ are the composite scores of the safety, economy, and green subsystems, respectively; $\alpha ,\beta ,\gamma$ are the corresponding weights. According to the energy ternary paradox, its equal weight is assigned here, and its value is 1/3 [18].

(2) Dual-System Coupled Coordination Degree Model:

$$C_{jt1}=2\sqrt{S_{tjk}{E}_{tjk}}/(S_{tjk}+E_{tjk})$$

(14)

$$T_{jt1}={\alpha }_1{S}_{tjk}+{\beta }_1{E}_{tjk}$$

(15)

$$D_{jt1}=\sqrt{T_{jt1}{C}_{jt1}}$$

(16)

where $C_{jt1}$, $D_{jt1}$ and $T_{jt1}$ respectively correspond to the coupling degree, coupling coordination degree, and comprehensive evaluation index; ${\alpha }_1$ and ${\beta }_1$ are the corresponding weights, whose values are 0.5 and 0.5, respectively.

The coupling coordination level is classified using the uniform distribution function method [56], which provides a systematic and widely adopted threshold scheme for evaluating the degree of synergy within multi-dimensional systems, as shown in

Table 2. Coupling coordination level scale.

Coupling Coordination Degree	Level	Type of Coordination
[0, 0.1]	Extreme dysfunction	Dysfunction and decline category
(0.1, 0.2]	Severe dysfunction
(0.2, 0.3]	Moderate dysfunction
(0.3, 0.4]	Mild dysfunction
(0.4, 0.5]	On the verge of dysfunction	Transition category
(0.5, 0.6]	Barely coordination
(0.6, 0.7]	Initial coordination
(0.7, 0.8]	Moderate coordination	Coordinated development category
(0.8, 0.9]	Good coordination
(0.9, 1]	High coordination

.

3.4. Kernel Density Analysis Method

The kernel density estimation method is employed to analyze the temporal evolution of individual indicators, subsystem performance, and the overall coupling coordination level across the study area. The results are visualized for further analysis and interpretation [57].

$$f\left(x\right)={\displaystyle \frac{1}{Nh}}\sum _{j=1}^{4}K\left({\displaystyle \frac{D\left(X_j\right)-D\left(x\right)}{h}}\right)$$

(17)

where $f\left(x\right)$ is the density function; $h$ is the kernel density bandwidth [58]; $K(·)$ is the stochastic kernel function; $D\left(X_j\right)$ is the coupling coordination degree of each province; $D\left(x\right)$ is the average value of the coupling coordination degree.

4. Study Area and Data

This study utilizes panel data from Sichuan, Chongqing, Shaanxi, and Shanxi provinces covering the period from 2011 to 2021. To ensure the objectivity and authenticity of the data, the primary data for all indicators were sourced from authoritative databases, including the National Bureau of Statistics (NBS) database, CEAD database, China Statistical Yearbook [59], China Urban Construction Statistical Yearbook [60], China Energy Statistical Yearbook [61], China Science and Technology Statistical Yearbook [62], and Statistical yearbooks of the four provinces. Missing data were addressed using linear interpolation [63] to ensure a complete dataset for analysis. This method is suitable for panel data with stable time series and continuous trends, particularly matching the gradual short-term evolution characteristic of energy economic indicators. After interpolation, no significant deviations were observed upon trend comparison.

All findings in this study are derived from raw statistical data through a standardized, objective computational process—comprising data normalization, entropy weight method-based weighting, and coupling coordination model calculation. This approach ensures the robustness of the coupling coordination degree results: all conclusions are grounded in the variational characteristics of the actually observed data, rather than relying on predefined subjective weights. Consequently, the results truthfully reflect the objective status and dynamic patterns of the synergistic development within the “Safety–Economy–Green” system in the study regions.

5. Results

5.1. Characteristics of the Time-Series Evolution of “S-E-G” Coupling Coordination Degree

5.1.1. Time-Series Characteristics of the Safety, Economic, and Green Indices

This study uses kernel density estimates from four provinces to examine the time-series characteristics of the “S-E-G” index for 2011, 2016, and 2021. As shown in Figure 3, the safety index demonstrates a distinct transition from unimodal to bimodal distribution, with a 16.2% narrowing in width, revealing emerging multi-level clustering patterns. This bimodal characteristic directly reflects regional disparities in key indicators: Sichuan and Chongqing concentrate R&D investment at 480 million CYN, significantly exceeding the 150 million CYN level of Shaanxi and Shanxi, while natural gas employment clusters around 27,000 practitioners in Sichuan and Shaanxi versus merely 4000 in Chongqing and Shanxi. Concurrently, the economy index exhibits a strongly right-skewed unimodal distribution, marked by a remarkable 177.8% increase in peak height and a substantial 64.2% contraction in bandwidth. This pattern indicates synchronized economic improvement alongside significant regional convergence.

This economic optimization is reflected in the significant increase in Sichuan Province’s extraction investment, which reached 10.2 billion CNY. This growth was driven by the natural gas pipeline transportation price mechanism reform post-2016 and the rising demand under the “Dual Carbon” goals. The green index presents contradictory dynamics: although the overall rightward shift suggests enhanced low-carbon performance, the emergence of bimodality, coupled with a 22.9% bandwidth expansion, reveals deepening regional polarization. This environmental challenge is evidenced by several key disparities: Shaanxi’s carbon emissions reached 5.1 t in 2016, dramatically higher than the below 1 t recorded in other provinces; the green patent outputs of Sichuan and Shaanxi exceeded 1000 in 2021, far surpassing the fewer than 500 in Chongqing and Shanxi; and Sichuan’s exceptional 160,000 hectares of green space contrasted sharply with the clustered 76,000 hectares shared by the other three regions.

5.1.2. Time-Series Characterization of Comprehensive Evaluation Index, Coupling Degree, and Coupling Coordination Degree

The time evolution characteristics of the comprehensive evaluation index, coupling degree, and coupling coordination degree in the “S-E-G” system of natural gas production regions from 2011 to 2021 are shown in Figure 4. Analysis reveals an overall improvement in the comprehensive development level, evidenced by the rightward shift and rising peak of its kernel density curve. Synergy among the three dimensions increased over time, mirroring the evolution of the coupling coordination degree. A core finding is the “high interaction, low coordination” relationship, in which coordination consistently lags behind coupling, indicating strong interactions but weaker overall coordination among safety, economy, and environment. Regional disparities narrowed significantly, with the distribution width of the coupling degree and coupling coordination degree contracting by 82.4% and 39.7%, respectively. The coupling degree distribution became highly concentrated with a sharper peak and steeper right tail, reflecting a substantial increase in highly coupled regions. The average coupling degree across the four provinces rose from 0.87 in 2011 to 0.97 in 2021, with Shaanxi and Sichuan maintaining levels above 0.95 for years. The coupling coordination degree also shifted to the right and became more concentrated, entering a medium-high level overall, with its average rising from 0.45 in 2011 to 0.62 in 2021. Sichuan recorded the most significant growth, advancing from 0.53 to 0.76, while Shanxi, starting from a lower base of 0.33 in 2011, demonstrated notable catch-up by reaching 0.53 in 2021.

5.1.3. Time-Series Characterization of Coupled Coordination Between the Overall and Internal Subsystems of “S-E-G”

The evolution of the comprehensive coupling coordination degree of the “S-E-G” composite system in Sichuan, Chongqing, Shaanxi, and Shanxi from 2011 to 2021 is shown in Figure 5. Overall, the system showed continuous optimization, with the coupling coordination degree registering an increase from 0.36 to 0.80 at an average annual rate of 4%, indicating significantly enhanced regional multi-objective synergy. The evolution can be divided into three periods:

(1)

2011–2013: Period of rapid improvement

The coupling coordination degree jumped from 0.36 to 0.58, registering an annual growth of 7.3%, which was primarily driven by a sharp rise in the “S-E” coupling coordination from 0.30 to 0.58. This was supported by a 13.8% increase in production, a 29.2% increase in transport capacity, >108% in cost-profit margin, and a 34.1% increase in R&D. Early efforts under the 12th Five-Year Plan and the 2013 Air Pollution Action Plan initiated environmental improvements. However, their impact was constrained by the relatively low number of green patents—657—limiting the development of “S-G” synergy.

(2)

2014–2016: Period of fluctuation and adjustment

The coupling coordination degree fluctuated within the 0.58–0.66 range, with its most notable gain observed in the “S-G” dimension, where it rose from 0.54 to 0.68. However, the economic subsystem faltered due to international energy price shocks, with key indicators such as the gate station price falling by 29.1% and profits declining by 14.18%. This prompted firms to sharply reduce environmental spending, with pollution control investment falling to 6.488 billion CNY by 2016. This created a paradox where substantial “S-G” gains of 33% were effectively offset by weakened economy-environment linkages.

(3)

2017–2021: Period of high stability

The coupling coordination degree stabilized at a high level, ranging from 0.76 to 0.80, though its growth plateaued. Meanwhile, employment in gas production plunged from 137,900 to 63,200, signaling potential long-term capacity constraints. Growth rates for output and R&D slowed. The 2020 “dual carbon” goals intensified systemic tensions, triggering a surge in green technology investment—4467 green patents were filed in 2021—while also causing a 72.33% plunge in pollution control investment. A 143% surge in carbon emissions from large-scale gas exploitation highlighted the conflict between energy supply and the green transition, making tripartite coordination more difficult.

In summary, the concurrent improvement in coupling coordination degree and fluctuations in green performance reveal a temporary, superficial “decoupling” between environmental performance and economic-security activities. This essentially reflects the “transitional challenges” faced by the system under the “Dual Carbon” goals. The intrinsic cause of the current “high coupling but low coordination” state lies precisely in the split-response mechanism within the green subsystem—where “ end-of-pipe decline” and “source innovation emergence” coexist—as well as its fundamental conflict with the objectives of safety and economy. The improvement in coordination primarily reflects a transitional process of intensified interaction and search for balance under multiple pressures. Achieving genuine sustainable development necessitates driving the system to complete a fundamental paradigm shift from relying on “ end-of-pipe treatment” to focusing on “source prevention.”

5.2. Spatial Distribution Characteristics of Comprehensive Evaluation Index, Coupling Degree, and Coupling Coordination Degree

The geospatial distribution of natural gas production areas in Sichuan, Shaanxi, Shanxi, and Chongqing in 2011, 2016, and 2021 demonstrates notable interregional spatial disparities.

As shown in Figure 6, the comprehensive evaluation index of all systems improved significantly, with values rising from approximately 0.2 in 2011 to between 0.6 and 0.7 in 2021 across all provinces. Among them, Sichuan Province started with a high baseline and experienced rapid growth, achieving the highest index values across all systems by 2021. Although most systems in Shaanxi Province initially led, their growth slowed in later stages, with the “S-E” and “S-G” subsystems declining by 2021. In contrast, Chongqing and Shanxi started with the lowest values but showed consistent growth over time. Overall, the evolutionary trend of the comprehensive evaluation index aligns closely with the coupling coordination degree, serving as the fundamental driver for its improvement.

The geospatial distribution of the coupling degree is shown in Figure 7. With the exception of Shanxi Province, the coupling degrees of all systems in the other three regions remained at high values between 0.9 and 1 from 2011 to 2021. In Shanxi Province, the coupling degrees of the “S-E-G”, “S-E”, and “S-G” systems increased year by year, rising from values between 0.5 and 0.6 in 2011 to levels similar to those in the other three regions by 2021. The “E-G” subsystem performed the best, with its coupling degree consistently similar to that of the other three regions. The coupling degree acts as both a “stabilizer” and an “amplifier” for the coupling coordination degree, and its high and stable values provide a fundamental guarantee for the coupling coordination degree.

The coupling coordination degree synthesizes the developmental patterns of both the comprehensive evaluation index and the coupling degree. Its spatial distribution is presented in Figure 8. By 2021, all coupling coordination degrees reached “Moderate Coordination”. The “S-E” and “S-G” dimensions showed similar coupling coordination levels, while “E-G” was consistently slightly higher. The improvements in coupling coordination degree were positively correlated with time. Sichuan generally led in coupling coordination degree across dimensions, followed by Shaanxi, with Chongqing and Shanxi showing comparable performance. Sichuan’s initial lowest level was “On the Verge of Dysfunction,” advancing to “Moderate Coordination” in all dimensions by 2021. Shaanxi maintained stable “Primary Coordination”. While its coupling coordination degree outperformed Sichuan’s in 2011 and 2016, Sichuan surpassed it by 2021. Chongqing improved from an imbalance to “Primary Coordination” by 2021. Shanxi trailed slightly behind Chongqing, with its “S-G” coupling coordination degree peaking only at “On the Verge of Dysfunction” and “S-E” at “Barely Coordinated”, resulting in the poorest overall performance.

Four primary factors contribute to the spatial disparities in natural gas development among Sichuan, Shaanxi, Shanxi, and Chongqing.

Resource Endowment Divergence: Sichuan’s preeminence is anchored in possessing 44.9% of China’s technically recoverable gas reserves. Conversely, Shanxi’s energy structure, characterized by coal accounting for over 80% of its energy mix, systematically weakens incentives for gas-sector development.

Policy Intervention and Infrastructure: Sichuan reaped synergistic benefits from national strategies during the 12th Five-Year Plan (e.g., the Third West–East Gas Pipeline and the Sichuan-Chongqing production base), reinforced by regulations such as the Air Pollution Prevention and Control Action Plan (2013) and the Technical Guidelines for Green Exploitation of High-Sulphur Gas Fields (2014). In contrast, Shanxi’s coal-prioritizing policy framework led to disproportionately lower investment in gas infrastructure. Shaanxi’s policy efficacy was attenuated by the Ordos Basin’s complex geology, while Chongqing exhibited “policy idling”—a planning-implementation disconnect evidenced by inadequate financial matching and administrative delays.

Market Mechanisms and Technological Innovation: The NDRC’s Measures for the Management of Natural Gas Pipeline Transportation Prices (2016) enabled Sichuan’s gate station price linkage mechanism and market-based shale gas pricing by 2017. Dedicated R&D under the Sino-US “2 + 2” Clean Energy Cooperation Framework (2017–2021) catalyzed desulfurization advancements. Shanxi’s coal subsidies and Shaanxi’s under-specified innovation regulations led to lower patent commercialization.

Market Structure Heterogeneity: Sichuan’s efficient cost transmission contrasts sharply with Shanxi’s coal-conditioned, distorted market.

In summary, inter-provincial disparities are not attributable to any single factor. Sichuan’s leading position stems from a self-adaptive system formed by its resource base, coordinated policies, and dynamic innovation. The remaining provinces, however, are constrained by distinct types of system constraints: Shaanxi faces rigid constraints imposed by geography and technology. Chongqing is hindered by implementation constraints due to policy inefficiency, and Shanxi is entrenched in a structural lock-in caused by coal dependency. The nature of these spatial differences reflects varying regional system configurations, whose driving factors—such as policy effectiveness and resource conversion models—exhibit strong endogeneity and path dependency, making it difficult to disentangle them as independent quantitative variables. Therefore, understanding these disparities requires moving beyond simple factor decomposition and adopting a holistic systems perspective coupled with qualitative comparative logic. This is precisely the deeper mechanism revealed through the spatiotemporal analysis in this study, which also points to the need for tailored, place-based systemic intervention strategies.

6. Discussion

Balancing safety, economic growth, and environmental sustainability—the “possible triangle”—poses a critical challenge for China’s natural gas industry. This study summarizes the main coupling coordination states, their policy and real-world drivers, and targeted recommendations in

Table 3. Coupling Coordination Status and Policy Recommendations.

Coupling Coordination Status	Policy Diagnostics/Reality Drivers	Targeted Recommendations
(1) The pervasive characteristic of the entire system is high coupling but low coordination.	During the 2014–2016 period of international energy price volatility, the economic subsystem contracted, weakening its coordination with other dimensions. The green dimension remained a systemic bottleneck: under the “Dual Carbon” goals, increased natural gas extraction raised emissions while pollution control investment was cut during economic downturns. Meanwhile, green technology uptake lagged, resulting in unstable environmental performance and undermining overall coordination.	Regions with strong green technology foundations (e.g., Sichuan, Shaanxi) should establish regional technology promotion centers, supported by fiscal and tax incentives, to accelerate deployment of CCUS, low-pollution drilling fluids, and methane monitoring. Lagging regions (e.g., Shanxi, Chongqing) must increase investment in pollution control, safeguarding environmental expenditures against economic fluctuations to prevent erosion of subsystem synergy.
(2) Progression from Coordination to Optimization (Sichuan)	It leads in the synergy of resources, technology, and policies, and must not only maintain but also further optimize this favorable condition in the future.	Sichuan should leverage its advantages to pioneer intelligent shale gas extraction and digital transformation, establishing a national green extraction demonstration zone.
(3) Moderate Coordination with Stagnant Growth (Shaanxi)	The complex geological conditions, such as those in the Ordos Basin, lead to high extraction costs and difficulties in improving efficiency, which constrain the economic subsystem’s ability to support safety and green initiatives, resulting in a developmental plateau.	Establish a special fund for complex oil and gas reservoir extraction technologies, encourage joint industry-academia research initiatives, and improve recovery rates and economic efficiency.
(4) Low coordination level (Chongqing)	Chongqing exhibits a disconnect between policy planning and implementation, with insufficient supporting funding and administrative efficacy, leading to diluted policy outcomes.	Chongqing must enhance policy implementation efficiency and deepen synergy with Sichuan for shared technology and infrastructure.
(5) Lagging coordination (Shanxi)	Shanxi’s coal-dominated energy structure and entrenched interests create a lock-in effect, squeezing out the investment, market, and policy space needed for natural gas development.	Shanxi needs a structural shift away from coal dependency toward integrated “coal-to-gas—CCUS—hydrogen energy” models to increase gas consumption and enhance tripartite coordination.

:

Differentiated interventions are fundamental for enhancing the quality of regional development. However, to systematically narrow gaps and achieve overall synergy, it is essential to build a cross-regional collaborative governance system on the basis of differentiated strategies. To this end, efforts should be made to promote cross-regional cooperation and learning: establish a “Sichuan-Shaanxi Shale Gas Technology Sharing Platform” for joint research and development, and set up a “Shanxi-Chongqing CCUS and Abandoned Mine Utilization Pilot Zone” to replicate successful models. Through policy integration (such as inter-provincial price linkage and emergency allocation mechanisms), collaboration on major projects, shared infrastructure, and data and training exchanges, market resilience and systemic coordination can be strengthened, thereby enhancing the overall level of coupling coordination. Achieving the “possible triangle” is a nonlinear process that requires flexible, region-specific strategies. China’s evolving experience demonstrates that a balanced and sustainable future for natural gas is attainable through targeted innovation, cooperation, and policy support.

7. Conclusions

This study investigates the development trajectory of the “possible triangle”—balancing safety, economy, and environmental sustainability—across China’s major natural gas production regions (Sichuan, Chongqing, Shaanxi, and Shanxi) from 2011 to 2021. The principal findings are summarized as follows:

(1)

Spatiotemporal Evolution of Subsystems and Coupling Coordination: Subsystem development exhibited asynchrony and multiploidization. Regional disparities were significant and multipolar for safety and green dimensions. While the economic dimension’s regional gap narrowed, overall polarization intensified. The coupling degree consistently exceeded the coupling coordination degree, indicating strong interactions but insufficient overall synergy. Regional gaps in coupling and coupling coordination narrowed over time, concentrating at high levels. The green dimension emerged as the key bottleneck constraining system coordination.

(2)

Trends in Comprehensive Coupling Coordination: The comprehensive coupling coordination degree demonstrated phased growth, rising at an average annual rate of 4%, with significantly enhanced subsystem synergy. The evolution comprised three distinct phases: Rapid Increase (2011–2013), dominated by “safety-economy” synergy; Fluctuation and Adjustment (2014–2016), where international energy price shocks caused economic indicator volatility and contracted environmental investment; and High Stabilization (2017–2021), where environmental investment rose notably, but system synergy entered a plateau period under the “dual carbon” target, highlighting the conflict between green transformation and production increase/supply security.

(3)

Spatial Heterogeneity and Regional Differentiation: Significant spatial variation in coupling coordination levels was observed. Sichuan Province led, achieving “moderate coordination (0.7–0.8)” due to superior resource endowment and effective policy implementation. Shaanxi maintained a stable “initial coordination (0.6–0.7)”. Chongqing and Shanxi lagged, remaining at “barely coordinated (0.5–0.6)” or “on the verge of dysfunction (0.4–0.5)”. These disparities stemmed from varied resource conditions, infrastructure investment intensity, policy support/effectiveness, and energy structures.

With these findings, targeted policy recommendations are proposed to enhance regional coordination of coupling and foster integrated security, economic, and environmental development. This study thus offers valuable, adaptable insights for the global energy transition. Other resource-based regions can adapt its core principles to explore context-specific, balanced, and sustainable development pathways, facilitating the shift from the “impossible” to the “possible triangle”. Of course, this study has some limitations, such as a limited regional sample size, failure to quantify exogenous shocks like international energy prices, and the omission of social dimensions. Future research could introduce VAR models to analyze subsystem interactions, extend comparisons to international natural gas production regions, and construct a comprehensive “S-E-G-S” framework that incorporates the social dimension.