Carbon Decoupling of the Mining Industry in Mineral-Rich Regions Based on Driving Factors and Multi-Scenario Simulations: A Case Study of Guangxi, China

Abstract

The mining industry (MI) in mineral-rich regions is pivotal for economic growth but is challenged by significant pollution and emissions. This study examines Guangxi, a representative region in China, in light of the country’s “Dual Carbon” goals. We quantified carbon emissions from the MI from 2005 to 2021, employing the generalized Divisia index method (GDIM) to analyze the factors driving these emissions. Additionally, a system dynamics (SD) model was developed, integrating economic, demographic, energy, environmental, and policy variables to assess decarbonization strategies and the potential for carbon decoupling. The key findings include the following: (1) Carbon accounting analysis reveals a rising emission trend in Guangxi’s MI, predominantly driven by electricity consumption, with the non-ferrous metal mining sector contributing the largest share of total emissions. (2) The primary drivers of carbon emissions were identified as economic scale, population intensity, and energy intensity, with periodic fluctuations in sector-specific drivers necessitating coordinated policy adjustments. (3) Scenario analysis showed that the Emission Reduction Scenario (ERS) is the only approach that achieves a carbon peak before 2030, indicating that it is the most effective decarbonization pathway. (4) Between 2022 and 2035, carbon decoupling from total output value is projected to improve under both the Energy-Saving Scenario (ESS) and ERS, achieving strong decoupling, while the resource extraction shows limited decoupling effects often displaying an expansionary connection. This study aims to enhance the understanding and promote the advancement of green and low-carbon development within the MI in mineral-rich regions.

1. Introduction

In the context of accelerating global climate governance, the low-carbon development of mineral-rich regions has become a critical factor in achieving the temperature control targets outlined in the Paris Agreement [1]. China, as the world’s largest developing economy, faces mounting pressure to balance resource exploitation and ecological protection while pursuing its “Dual Carbon” goals [2]. The nation’s heavy reliance on mineral resources is evident, with 80% of industrial raw materials and 93% of the energy supply derived from mining activities [3], contributing 38% of total industrial carbon emissions [4]. Notably, 48% of resource-based cities emit over 100 million tons of CO₂ annually [5]. These regions often grapple with the “resource curse” dilemma: while the mining industry (MI) generates 32–67% of local fiscal revenue [6], its carbon intensity per GDP unit exceeds the national average by 1.8 times [7], rendering current development models unsustainable for long-term socioeconomic progress.

To advance its “Dual Carbon” goals, China has introduced policies such as the 14th Five-Year Plan for Promoting High-Quality Development in Mineral-Rich Regions, targeting energy conservation and carbon reduction. As a high-emission sector, the MI in these regions faces significant low-carbon development challenges [8]. Between 2005 and 2021, China’s MI carbon emissions grew at an annual rate of 2.14%, totaling 279 million tons by 2021 [3]. While some northern resource-rich provinces, such as Shanxi and Inner Mongolia, have successfully reduced carbon intensity through industrial agglomeration and optimization [9], southern mineral-rich regions like Guangxi and Yunnan struggle with resource extraction and CO₂ emissions [10]. These regions pose major obstacles to China’s carbon peaking efforts, often exhibiting the “three highs” characteristic: a high intensity of mineral resource exploitation [11], a high dependence on energy consumption [12], and high costs for implementing emission reduction policies [13].

Guangxi, a mineral-rich region in southern China (Figure 1), serves as a strategic mineral reserve base and a key node in China’s New Western Land–Sea Corridor, as well as a vital gateway to the Association of Southeast Asian Nations (ASEAN) [14]. The region’s economic growth is heavily reliant on resource exploitation, with mining-related industries accounting for 40% of its total industrial output [15]. However, Guangxi’s carbon intensity is 36% higher than the national average [16], and it lags behind other coastal or economically advanced regions in China, exhibiting significantly higher carbon emissions alongside slower economic growth. This dual challenge of driving economic growth while reducing emissions underscores the need for sustainable development in the MI to unlock Guangxi’s growth potential.

Guangxi has introduced various decarbonization policies following the adoption of China’s “Dual Carbon” goals. However, these efforts have fallen short of expectations [17]. Firstly, the current carbon accounting system lacks the precision to accurately reflect intra-industry variations, resulting in poorly targeted policies. Secondly, the strong correlation between resource endowments and industrial development continues to drive carbon emissions under the traditional factor-driven growth model. This study aims to identify effective strategies that promote both economic growth and carbon reduction by addressing several critical questions: 1. How can carbon emissions from the MI and its sub-sectors be accurately quantified? 2. What are the primary drivers of carbon emission increases in the MI? 3. Under which development scenarios could the MI achieve a carbon peak by 2030? 4. How will the relationships between economic growth, resource utilization, and carbon emissions evolve in mineral-rich regions like Guangxi?

This study examines the carbon emission characteristics and their influencing factors within Guangxi’s MI and its sub-sectors, elucidating the mechanisms of these factors and the decoupling effects of carbon emissions in mineral-rich regions. The research includes a review of the existing literature on carbon emissions, an assessment of Guangxi’s MI carbon emissions from 2005 to 2021, and the application of the extended generalized Divisia index method (GDIM) to identify the primary drivers. Furthermore, a system dynamics (SD) model is employed to project carbon emission trajectories under various scenarios from 2022 to 2035, investigating the evolution of carbon decoupling effects. The structure of the paper is organized as follows: Section 2 reviews and synthesizes the pertinent literature; Section 3 describes the data sources and outlines the methodology; Section 4 details and discusses the findings; and Section 5 offers conclusions, key insights, and policy recommendations.

2. Literature Review

Current research on carbon reduction in mineral-rich regions primarily concentrates on policy adjustments, green transitions, and performance evaluations [18,19]. Conversely, studies pertaining to the MI predominantly focus on carbon emission accounting, the analysis of influencing factors, the development of predictive models, and the exploration of low-carbon development pathways [20,21].

Carbon emission accounting is a crucial aspect of MI research, and the emission factor method is widely adopted due to its simplicity and adaptability across various dimensions [22]. For instance, Aramendia et al. [23] combined the emission factor method with Monte Carlo simulations, concluding that MI accounts for approximately 1.7% of global energy consumption. Their projections indicate that global mining energy consumption could increase by two to eight times by 2060. Similarly, Xu et al. [24] and Chen et al. [25] employed the emission factor method to estimate China’s rising carbon emissions, identifying the electricity, industrial, and transportation sectors as key contributors. However, few studies comprehensively capture emissions across the entire industry. To address this, Cao et al. [26] and Li et al. [27] applied a life cycle assessment (LCA) approach to estimate carbon emissions in China’s MI, revealing that electricity and coal consumption are the primary drivers. Additionally, Shao et al. [28] utilized nighttime light data to simulate and estimate energy consumption and carbon emission trends in mining cities in northern China, noting an expanding trajectory. Although these studies provide valuable emission data at various scales, insights into sector-specific and energy-type-specific emissions remain limited.

In the literature on carbon emission drivers, studies predominantly concentrate on economic, demographic, and energy-related factors [29,30], among which factor decomposition and structural decomposition are the most widely applied [31]. For instance, Li et al. [4] utilized the logarithmic mean Divisia index (LMDI) decomposition method to examine the drivers of carbon emissions in China’s manufacturing industry, identifying output scale, energy intensity, and industrial scale as key determinants. Building on this approach, the GDIM incorporates both explicit and implicit carbon emission drivers, effectively addressing the interdependencies among the influencing factors. Shao et al. [32] applied the GDIM, emphasizing energy utilization efficiency and carbon intensity as primary drivers of emissions. Similarly, Amira et al. [33] employed structural decomposition analysis, pointing to demand scale as a central driver of carbon emissions. Beyond decomposition methods, researchers have also explored other analytical approaches, including spatial panel econometric models, geographically weighted regression, and models accounting for undesirable outputs [34,35]. For example, Wei et al. [36] employed the stochastic impacts by population, affluence, and technology (STIRPAT) model, finding that energy intensity is the most significant factor influencing carbon emissions in the manufacturing industry. Furthermore, Cui et al. [37] used a super-efficiency model to investigate the drivers of carbon emissions in mining cities along China’s Yellow River Basin, revealing significant positive impacts from both economic and environmental factors. These findings underscore the existence of regional variations in the dominant drivers of carbon emissions. Nonetheless, the majority of studies continue to focus on macro-level factors such as economic, policy, and energy-related determinants.

Scholars have extensively investigated carbon reduction pathways in mineral-rich regions using diverse methodological approaches. Several studies have concentrated on the influence of individual factors, e.g., economic, policy, or environmental variables on carbon emissions [38,39]. In contrast, others have embraced a multifactorial interaction framework, employing advanced techniques such as machine learning algorithms, gray prediction models, SD, and scenario analysis. These methods are utilized to project carbon emission trends and explore potential reduction strategies under various developmental scenarios [26,40]. Among these methodologies, SD stands out for its capacity to incorporate feedback loops among factors, thereby enhancing the flexibility of simulations [41]. For example, Jie et al. [42] and Andewi et al. [43] utilized the SD model to identify critical strategies for mitigating carbon emissions, including reducing energy intensity, moderating economic growth, and adjusting the industrial structure. Similarly, Han et al. [44] and Lu et al. [45] applied machine learning algorithms to forecast carbon emissions in various regions, highlighting the roles of technological innovation and energy efficiency improvements in achieving early carbon peak targets. While most studies on carbon reduction pathways involve adjusting a range of parameters to model emission trends, the effectiveness of these measures varies significantly without appropriate constraints. Moreover, the rationality of parameter settings requires further verification.

Decoupling analysis is a valuable approach for understanding how an economy can grow without increasing resource consumption or causing detrimental environmental impacts. From the perspective of carbon decoupling potential, excessive emphasis on emission reductions in mineral-rich regions may exert economic pressures [46]. Liu et al. [47] and Hickel et al. [48] have highlighted the strong correlation between future carbon emission growth, economic development, and resource utilization. Tong et al. [49] found that resource decoupling plays a predominant role in influencing carbon decoupling in China’s resource-based regions, although most areas display negative decoupling. In the mining sector, Chen et al. [50] analyzed China’s MI and identified an inverted U-shaped relationship between carbon emissions and economic growth, i.e., emissions increase with growth initially but decline over time. Similarly, Asif et al. [51] emphasized the adverse impact of mineral extraction on carbon decoupling. However, it was also noted that improving energy efficiency can lead to short-term marginal emission reductions, while industrial restructuring and energy consumption control are more significant drivers of long-term decoupling [52]. Although most of these studies assess decoupling by analyzing the rate of change between two variables over time, the multi-factor synergistic effects on decoupling potential are often overlooked. Therefore, the complex decoupling relationships among economic growth, resource utilization, and carbon emissions in mineral-rich regions deserve further exploration.

This study develops a scenario simulation model for carbon emission decoupling effects, integrating GDIM decomposition and the SD model. The scenario simulation model provides a robust framework for analyzing mineral-rich regions, with a specific focus on the MI and its sub-sectors, offering a precise foundation for local governments to design carbon reduction policies for regional development. The main innovations of this study include three aspects: First, in addition to the research on national or industry-level carbon emission analysis, this study investigates a specific resource-based region, exploring the heterogeneity of carbon emissions across sub-sectors. Second, conventional studies typically rely on single-factor analytical methods that fail to capture the nonlinear dynamic relationships between influencing factors, thus this study examines the driving effects of multiple factors across different time periods, clarifying how these factors interact and shape the evolution of carbon emissions. Third, unlike previous studies that primarily depend on historical data for decoupling analyses or carbon emission forecasting, this study extends the analysis by simulating future decoupling relationships between economic growth, resource utilization, and carbon emissions under different scenarios. Additionally, it investigates the alignment between short-term decoupling patterns and long-term trends, providing a more comprehensive understanding of emission dynamics.

In this study, the GDIM is utilized to identify the driving factors of carbon emissions, followed by scenario modeling to address the strong subjectivity and limited applicability of traditional SD models. Furthermore, decoupling analysis is employed to clarify the relationship between coordinated development and emission reduction. The main contributions of this study are as follows: (1) Carbon emissions for Guangxi’s MI and its five sub-sectors are quantified. The GDIM model is extended by refining carbon emission factors into ten specific categories—economic scale, energy consumption scale, technological level, population scale, economic intensity, energy intensity, technological intensity, population intensity, energy efficiency output, and per capita output—thereby providing a more detailed analysis of their respective driving effects. (2) A system dynamics model incorporating economic, population, energy, environmental, and policy factors is developed to explore the interactions among these elements in shaping carbon emissions. Building upon this framework, this study examines future carbon emission trends and decoupling effects under various policy scenarios, aiming to elucidate the mechanisms driving carbon emissions in the MI and identify optimal reduction pathways.

3. Materials and Methods

Based on the industry classification from the National Bureau of Statistics of China, the mining industry (MI) is divided into the following five sub-sectors: coal mining and dressing (CMD), petroleum and natural gas extraction (PNGE), ferrous metal ore mining and dressing (FMMD), non-ferrous metal ore mining and dressing (NFMD), and non-metallic mineral mining and dressing (NMMD). The research framework of the model is shown in Figure 2. As mentioned earlier, this study develops a comprehensive model analyzing the carbon emission status and driving factors of Guangxi’s MI and its five sub-sectors from 2005 to 2021, while incorporating scenario simulations and decoupling analysis for 2022 to 2035. The model uses energy consumption data from different MI sub-sectors for carbon emission accounting and applies the GDIM model to classify ten influencing factors into three main categories. Based on these factors, the SD model with five interrelated subsystems is built to simulate the MI’s low-carbon development pathways under various scenarios. By embedding decoupling analysis into these simulations, the model offers an in-depth examination of the relationships among economic growth, resource extraction, and carbon emissions.

3.1. Data Source

This study utilizes data spanning 2005–2021, including socio-economic statistics for Guangxi and energy consumption and economic development indicators from the MI. The datasets were obtained from official Chinese statistical yearbooks and databases, with the detailed sources cataloged in

Table 1. Data source statistics.

Category	Data	Reference	Source
Guangxi region	Socio-economic	Guangxi Statistical Yearbook	http://tjj.gxzf.gov.cn/tjsj/tjnj/ (accessed on 27 December 2023)
	Energy	China Energy Statistical Yearbook	https://www.shujuku.org/china-energy-statistical-yearbook.html (accessed on 16 May 2022)
Guangxi’s MI	Environment	China Environmental Statistical Yearbook	https://www.shujuku.org/china-environment-statistical-yearbook.html (accessed on 20 January 2021)
	Resources	China Natural Resources Statistical Yearbook	https://www.mnr.gov.cn/sj/ (accessed on 14 March 2025)
	Energy	CEADs Database	https://www.ceads.net.cn/ (accessed on 1 February 2025)
	Industrial economy	China Industrial Statistical Yearbook	https://www.shujuku.org/china-industry-statistical-yearbook.html (accessed on 19 January 2021)

.

3.2. Carbon Emission Accounting

This study estimates the carbon emissions of Guangxi’s MI using the carbon emission factor method [53]. The selection of energy types is based on the actual energy consumption structure of Guangxi’s MI, as well as previous studies and the availability of basic data [24,36]. Specifically, the analysis selects energy consumption data for coal (including raw coal, clean coal, briquette coal, other washed coal, coke, and coke oven gas), oil products (such as crude oil, gasoline, kerosene, diesel, fuel oil, other petroleum products, and liquefied petroleum gas), natural gas, and electricity. The carbon emissions are then calculated by multiplying the corresponding carbon emission factors for each energy type. The relevant carbon emission factor data for each energy source are presented in

Table 2. Parameters related to carbon emission factors.

Energy Type (g)	Low Level Net Calorific Value (NCVg) (GJ/T) or (GJ/103 m3)	Carbon Content per Unit Calorific Value (αg) (TC/TJ)	Carbon Oxidation Rate (βg) (%)	Emission Factors	Unit
Raw Coal	20.908	25.8	0.899	1.7781	t·t−1
Clean Coal	26.344	25.8	0.899	2.2404	t·t−1
Coal Type	17.584	33.56	0.900	1.9474	t·t−1
Other Washed Coal	9.409	25.8	0.899	0.8002	t·t−1
Coke	28.435	29.2	0.970	2.9531	t·t−1
Coke Oven Gas	17.981	12.1	0.990	0.7898	t/103 m3
Crude Oil	41.816	20.0	0.980	3.0052	t·t−1
Petrol	43.070	18.9	0.980	2.9251	t·t−1
Paraffin	43.070	19.5	0.980	3.0179	t·t−1
Diesel oil	42.652	20.2	0.980	3.0959	t·t−1
Fuel Oil	41.816	21.1	0.980	3.1705	t·t−1
Natural Gas	38.931	15.3	0.900	1.9656	t/103 m3
Other Petroleum Products	41.031	20.0	0.980	2.9487	t·t−1
Liquefied Petroleum Gas	50.179	17.2	0.99	3.133	t·t−1

.

According to the carbon emission accounting system of the IPCC [53], carbon emissions are calculated as shown in Equation (1). Since electricity is a secondary energy source, the marginal emission factor for the Southern Regional Grid electricity is taken as 0.8042 t·MWh⁻¹ [54].

$$C_{eng}={\displaystyle \sum _{g=1}^{11}(E_g}\times NC{V}_g\times {\alpha }_g\times {\beta }_g)\times {\displaystyle \frac{44}{12}}$$

(1)

where C_eng represents the total carbon emissions of the MI; E_g is the consumption of energy from MI type g; NCV_g is the average net calorific value of energy type g; α_g is the carbon content per unit calorific value of energy type g; β_g is the oxidation rate of energy type g; 44/12 is the molecular weight ratio of carbon dioxide to carbon; and NCV_g·α_g·β_g·44/12 represents the emission factors of energy type g.

3.3. Factor Decomposition Model

To explore the development trends and driving factors of carbon emissions in the MI of Guangxi, this study adopts the generalized Divisia index decomposition method (GDIM) proposed by Vaninsky [55]. A factor decomposition model for Guangxi’s MI and its five sub-sectors is constructed to dynamically examine the driving factors of carbon emissions. Based on the GDIM, the carbon emission expressions for the five sub-sectors of the MI are shown in Equations (2)–(4), with symbol explanations provided in

Table 3. Explanation of variables in Equations (2)–(4).

Variables	Definition
C	Total carbon emissions
Ci	Carbon emissions of sector i
Gi	Total assets of sector i
Ei	Total energy consumption of sector i
Ti	Profit of sector i
Pi	Average number of employees in sector i
CGi = Ci/Gi	Carbon emissions per unit of economic output in sector i
CEi = Ci/Ei	Carbon emissions per unit of energy consumption in sector i
CTi = Ci/Ti	Carbon emissions per unit of profit in sector i
CPi = Ci/Pi	Per capita carbon emissions in sector i
EGi = Ei/Gi	Energy consumption per unit of economic output in sector i
GPi = Gi/Pi	Per capita economic output in sector i
i	1. CND; 2. PNGE; 3. FMMD; 4. NFMD; 5. NMMD

.

$$\begin{array}{ll}\hfill C& ={\displaystyle \sum _{i=1}^5{C}_i={\displaystyle \sum _{i=1}^5\frac{C_i}{G_i}}}{G}_i={\displaystyle \sum _{i=1}^5\frac{C_i}{E_i}}{E}_i={\displaystyle \sum _{i=1}^5\frac{C_i}{T_i}}{T}_i={\displaystyle \sum _{i=1}^5\frac{C_i}{P_i}}{P}_i\hfill \\ \hfill & ={\displaystyle \sum {\mathrm{C}}_i}={\displaystyle \sum C{G}_i\times G_i}={\displaystyle \sum C{E}_i\times E_i}={\displaystyle \sum C{T}_i\times T_i}={\displaystyle \sum C{P}_i\times P_i}\hfill \end{array}$$

(2)

$$G/P={\displaystyle \sum _{i=1}^5{G}_i}/{\displaystyle \sum _{i=1}^5{P}_i}={\displaystyle \frac{{\displaystyle {\sum }_{i=1}^5{C}_i}}{{\displaystyle {\sum }_{i=1}^5{P}_i}}}/{\displaystyle \frac{{\displaystyle {\sum }_{i=1}^5{C}_i}}{{\displaystyle {\sum }_{i=1}^5{G}_i}}}$$

(3)

$$E/G={\displaystyle \sum _{i=1}^5{E}_i}/{\displaystyle \sum _{i=1}^5{G}_i}={\displaystyle \frac{{\displaystyle {\sum }_{i=1}^5{C}_i}}{{\displaystyle {\sum }_{i=1}^5{G}_i}}}/{\displaystyle \frac{{\displaystyle {\sum }_{i=1}^5{C}_i}}{{\displaystyle {\sum }_{i=1}^5{E}_i}}}$$

(4)

To apply the GDIM model, Equations (2)–(4) are transformed as shown in Equations (5) and (6).

$$C=C{G}_i\times G_i$$

(5)

$$\begin{array}{l}C{G}_i\times G_i-C{E}_i\times E_i=0\\ C{G}_i\times G_i-C{T}_i\times T_i=0\\ C{G}_i\times G_i-C{P}_i\times P_i=0\\ G_i-G{P}_i\times P_i=0\\ E_i-E{G}_i\times G_i=0\end{array}$$

(6)

According to Equation (6), by taking the first-order partial derivative and using the carbon emission factor X and its contribution function Z(X), the Jacobian matrix is obtained as follows:

$${\Phi }_X={\left[\begin{array}{cccccccccc}C{G}_i& G_i& -C{E}_i& -E_i& 0& 0& 0& 0& 0& 0\\ C{G}_i& G_i& 0& 0& -C{T}_i& -T_i& 0& 0& 0& 0\\ C{G}_i& G_i& 0& 0& 0& 0& -C{P}_i& -P_i& 0& 0\\ 1& 0& 0& 0& 0& 0& -G{P}_i& 0& 0& -P_i\\ -E{G}_i& 0& 1& 0& 0& 0& 0& 0& -G_i& 0\end{array}\right]}^T$$

(7)

According to the GDIM decomposition principle, the contributions of the above factors are aggregated to represent the overall carbon emission change (ΔC) in Guangxi, as shown in Equation (8).

$$\Delta C=\left[X\left|\Phi \right.\right]={\displaystyle {\int }_t\nabla C^T\left(I-\Phi {\Phi }^{+}\right)}dX$$

(8)

where t represents the time span $\Delta C={\left(C{G}_i{ \mathrm{G}}_i 0 0 0 0 0 0 0 0\right)}^T$; I denotes the identity matrix; and “+” indicates the generalized inverse matrix. If the column vectors of the Jacobian matrix Φ(X) are linearly independent, then ${\Phi }_X^{+}={\left({\Phi }_X^T{\Phi }_X\right)}^{-1}{\Phi }_X^T$.

The change in carbon emissions can be decomposed into the sum of ten effects, as shown in Equation (9).

$$\Delta C^t=C^t-C^0=\Delta C_{G_i}^t+\Delta C_{C{G}_i}^t+\Delta C_{E_i}^t+\dots +\Delta C_{G{P}_i}^t$$

(9)

where $\Delta G_i^t, \Delta E_i^t, \Delta T_i^t \mathrm{and} \Delta P_i^t$ the absolute factors of economic scale, energy consumption scale, technological level, and population size influence changes in carbon emissions; $\Delta C{G}_i^t, \Delta C{E}_i^t, \Delta C{T}_i^t \mathrm{and} \Delta C{P}_i^t$ the relative factors of economic intensity, energy intensity, technological intensity, and the level of low-carbon development in population intensity impact changes in carbon emissions; and $\Delta E{G}_i^t \mathrm{and} \Delta G{P}_i^t$ represent the effects of other factors, such as energy efficiency output and per capita output, on the change in carbon emissions.

3.4. SD Model

The carbon emission system of the MI is a complex system, as highlighted by Song et al. [56]. It is closely related to factors such as economic growth, population size, energy structure, carbon intensity, environmental constraints, and carbon mitigation factors, with strong interdependencies among these variables. Therefore, this study employs the SD model to address the dynamic interactions of multiple factors within the carbon emission system of Guangxi’s MI. The SD model is developed using VENSIM 6.3 software with a one-year time step. The model covers a time span from 2005 to 2035, where data from 2005 to 2021 represents historical actual data, and data from 2022 to 2035 is forecasted based on the model.

From a system perspective, the relationship between various factors and carbon emissions is analyzed by dividing the carbon emission system of the MI in Guangxi into five core components: economy, population, energy, environment, and policy subsystems. The system boundaries and structure are shown in Figure 3. The population subsystem fuels economic growth through labor supply, which in turn escalates energy demand. Energy systems directly generate emissions, while environmental systems quantify resultant pressures on ecosystems. Policy mechanisms mediate these interactions through low-carbon policies and technological regulation, establishing dynamic equilibrium. These five subsystems are interconnected and interact with each other, collectively influencing the carbon emission levels of the MI in Guangxi.

3.4.1. Causality Setting

Figure 4 presents the causal relationships governing carbon emissions in Guangxi’s MI through five interacting subsystems. The economic subsystem (the red part shown in Figure 4) includes factors such as total GDP, industrial structure, and total mining output. Changes in industrial structure and economic scale drive mining output growth [57], leading to increased energy consumption and carbon emissions, while financial support and technological advancements help mitigate emissions. Additionally, changes in the economic subsystem influence the population subsystem and provide feedback to other subsystems.

The population subsystem (the blue part shown in Figure 4) comprises factors such as total population, mining employment, and the number of enterprises. It drives energy consumption and resource extraction through population growth, increasing carbon emissions, while simultaneously mitigating emissions through improvements in labor efficiency and enterprise optimization [58].

The energy subsystem (the green part shown in Figure 4) encompasses resource extraction, energy consumption, and carbon emissions, directly affecting carbon output through energy consumption patterns and indirectly impacting other subsystems via environmental degradation.

The environmental subsystem (the pink part shown in Figure 4) includes factors such as land degradation, waste emissions, and carbon sinks. It reflects the pressure exerted by carbon emissions on resources and ecosystems, prompting the government to implement environmental policies and control measures to alleviate system stress and improve environmental quality [59].

Lastly, the policy subsystem (the orange part shown in Figure 4) consists of factors such as government expenditures, technological investment, and education funding. By regulating economic growth, energy supply, and environmental protection through policy instruments, this subsystem plays a crucial role in balancing the interactions among the other subsystems.

3.4.2. Flow and Stock Settings

The carbon emission system model of Guangxi’s MI consists of 64 parameter variables, including 2 state variables, 2 flow variables, and 60 auxiliary variables, as illustrated in Figure 5.

The model incorporates various types of parameter variables, including constants, table functions, and level variables. Constants represent fixed values, while table functions describe the nonlinear relationships between different variables [60]. Level variables are initialized using values from 2005. Furthermore, historical trends of key variables (e.g., population and GDP) are utilized to develop regression equations that serve as references for model validation and simulation analysis.

3.4.3. Carbon Emission Scenario Settings

Based on previous studies [41,43] and Guangxi’s carbon reduction policies, carbon emission scenarios are established from five key perspectives: population, GDP, energy consumption, industry structure, and resource extraction. Population is a fundamental driver of economic development, directly correlating with economic activity and resource use. Demographic expansion typically elevates resource demands and corresponding carbon emission levels [61]. GDP reflects the economic development level of the region and industry, influencing carbon emissions through multiple channels. Energy consumption has a direct causal relationship with carbon emissions, as energy use is a primary source of carbon output. Industry structure shapes the sector’s overall energy consumption patterns. An optimized industrial structure can reduce the reliance on energy-intensive industries and facilitate low-carbon development [62]. Resource extraction directly reflects the scale of mining activities, which often involves a high energy consumption and multiple processing stages, all contributing to increased carbon emissions [63].

For each of the five key indicators, three parameter levels are established, high, medium, and low, as detailed in Table 4, along with corresponding references. The population parameter is determined based on demographic planning and projected growth trends. The GDP parameter is formulated concerning economic development plans and outlook reports. The energy consumption parameter is set by considering historical growth rates, development strategies, and energy consumption targets. The industry structure parameter is derived from statistical yearbooks, development guidelines, and industrial revitalization action plans. Finally, the resource extraction parameter is defined according to the master plan for mineral resource development, resource utilization strategies, and the scarcity of resources. These parameters provide a multidimensional basis for assessing the future trends of carbon emission drivers in Guangxi’s MI.

Based on the future development trends of carbon emission drivers in Guangxi’s MI and previous studies [12,64], this study establishes five major scenario categories comprising a total of twelve sub-scenarios, with detailed parameters listed in Table 5. The Baseline Scenario (BS) assumes moderate growth across all variables, reflecting a continuation of the current social development trajectory. The Energy-Saving Scenario (ESS), including ESS1 and ESS2, prioritizes emission reduction at the expense of economic growth. These sub-scenarios reflect different changes in population size and industrial structure caused by sacrificing economic growth. The Rapid Development Scenario (RDS), consisting of RDS1, RDS2, and RDS3, assumes rapid socio-economic growth without deliberate carbon emission control. The differences among these sub-scenarios lie in how economic expansion influences other emission-driving factors. The Green Development Scenario (GDS), consisting of GDS1, GDS2, and GDS3, maintains overall development levels while adjusting the industrial structure to regulate the share of the secondary sector, thereby addressing carbon emission changes. These sub-scenarios consider possible differences in the depth and pace of industrial adjustment. The Emission Reduction Scenario (ERS), including ERS1, ERS2, and ERS3, assumes stable development levels while controlling resource extraction and energy consumption to optimize the energy consumption structure and reduce carbon emissions. These sub-scenarios capture the potential variability in emission outcomes under different resource and energy control intensities.

Table 4. The five carbon emission drivers adopted in this study and their definitions.

Driver	Setting Model	Setting Parameter	Remarks
Population	High	−0.10%	The National Population Development Plan (2016–2030) states that the population will peak around 2030 [65].
	Medium	−0.06%	The outline mentions that Guangxi’s average population growth rate is 0.83 percent, which has continued to decrease over the past five years [66].
	Low	−0.05%	Guangxi’s population growth rate remains above the national average [67].
GDP	High	6.50%	Guangxi’s plan aims to maintain stable and rapid economic growth [68].
	Medium	5.50%	The outline targets 5.5% average annual GDP growth over 2024–2028 [66].
	Low	4.50%	Future national economic growth is expected to slow [69].
Energy consumption	High	5.50%	China’s energy consumption has been increasing along with economic growth and is significantly positively correlated [70].
	Medium	3%	Guangxi’s energy consumption grew 3% annually on average in 2013–2023.
	Low	−2.50%	Guangxi aims to reduce energy intensity by 13% by 2025 [68].
Industrial structure	High	−0.65%	Guangxi’s industrial structure improved 1.8% annually in 2016–2020 [71].
	Medium	−0.55%	Guangxi will prioritize tourism and agriculture in development [66].
	Low	−0.45%	Recent industrial revitalization policies highlight continued support for secondary industry growth [72].
Resource extraction	High	3.10%	Guangxi’s mineral reserves forecast >one billion tons by 2025 [73].
	Medium	1.50%	The outline promotes comprehensive development and utilization of mineral resources to ensure steady growth [66].
	Low	−1.10%	Mineral resource depletion and sustainability concerns are making extraction increasingly difficult [74].

Table 4. The five carbon emission drivers adopted in this study and their definitions.

Driver	Setting Model	Setting Parameter	Remarks
Population	High	−0.10%	The National Population Development Plan (2016–2030) states that the population will peak around 2030 [65].
	Medium	−0.06%	The outline mentions that Guangxi’s average population growth rate is 0.83 percent, which has continued to decrease over the past five years [66].
	Low	−0.05%	Guangxi’s population growth rate remains above the national average [67].
GDP	High	6.50%	Guangxi’s plan aims to maintain stable and rapid economic growth [68].
	Medium	5.50%	The outline targets 5.5% average annual GDP growth over 2024–2028 [66].
	Low	4.50%	Future national economic growth is expected to slow [69].
Energy consumption	High	5.50%	China’s energy consumption has been increasing along with economic growth and is significantly positively correlated [70].
	Medium	3%	Guangxi’s energy consumption grew 3% annually on average in 2013–2023.
	Low	−2.50%	Guangxi aims to reduce energy intensity by 13% by 2025 [68].
Industrial structure	High	−0.65%	Guangxi’s industrial structure improved 1.8% annually in 2016–2020 [71].
	Medium	−0.55%	Guangxi will prioritize tourism and agriculture in development [66].
	Low	−0.45%	Recent industrial revitalization policies highlight continued support for secondary industry growth [72].
Resource extraction	High	3.10%	Guangxi’s mineral reserves forecast >one billion tons by 2025 [73].
	Medium	1.50%	The outline promotes comprehensive development and utilization of mineral resources to ensure steady growth [66].
	Low	−1.10%	Mineral resource depletion and sustainability concerns are making extraction increasingly difficult [74].

Table 5. Carbon emission scenarios in Guangxi’s MI.

Scenarios	Sub-Scenarios	Population	GDP	Energy Consumption	Industrial Structure	Resource Extraction
BS	BS	Medium	Medium	Medium	Medium	Medium
ESS	ESS1	Medium	Low	Low	Medium	Medium
	ESS2	Low	Low	Low	Low	Medium
RDS	RDS1	High	High	High	High	High
	RDS2	Low	High	Medium	Low	High
	RDS3	Medium	High	High	Medium	Medium
GDS	GDS1	Medium	Medium	Medium	High	Low
	GDS2	High	Medium	Low	High	High
	GDS3	Low	Medium	Medium	High	Medium
ERS	ERS1	High	Medium	Low	Low	Low
	ERS2	Medium	Medium	Low	Medium	Low
	ERS3	Low	Low	Low	Low	Low

Table 5. Carbon emission scenarios in Guangxi’s MI.

Scenarios	Sub-Scenarios	Population	GDP	Energy Consumption	Industrial Structure	Resource Extraction
BS	BS	Medium	Medium	Medium	Medium	Medium
ESS	ESS1	Medium	Low	Low	Medium	Medium
	ESS2	Low	Low	Low	Low	Medium
RDS	RDS1	High	High	High	High	High
	RDS2	Low	High	Medium	Low	High
	RDS3	Medium	High	High	Medium	Medium
GDS	GDS1	Medium	Medium	Medium	High	Low
	GDS2	High	Medium	Low	High	High
	GDS3	Low	Medium	Medium	High	Medium
ERS	ERS1	High	Medium	Low	Low	Low
	ERS2	Medium	Medium	Low	Medium	Low
	ERS3	Low	Low	Low	Low	Low

3.5. Carbon Emission Decoupling Effect Model

As illustrated in Figure 3, energy consumption is identified as the key driver of carbon emissions in the MI, closely linked to resource extraction. The scale and methods of resource extraction directly determine energy consumption levels, which in turn significantly impact carbon emissions [75]. Furthermore, economic development plays a crucial role in the decoupling effect of carbon emissions, as highlighted by Zhao et al. [76] and Xie et al. [77]. To comprehensively analyze the carbon decoupling effect in Guangxi’s MI, this study explores two aspects: (1) the elasticity variation between carbon emissions and total mining output, (2) the elasticity variation between carbon emissions and resource extraction. Drawing upon elasticity-based decoupling models, a computational framework is developed to assess the sensitivity of carbon emissions. The classification criteria for decoupling based on Dong et al. [78] are presented in

Table 6. Decoupling elasticity index criteria.

Decoupling State		δΔC	δΔG or δΔY	D	Meaning
Decoupling	Strong	<0	>0	Dt < 0	Output/extraction increases while carbon emissions decrease.
	Weak	>0	>0	0.8 > Dt > 0	Output/extraction increases while carbon emissions grow slowly.
	Recessive	<0	<0	Dt > 1.2	Output/extraction decreases while carbon emissions drop rapidly.
Negative decoupling	Expansive	>0	>0	Dt > 1.2	Output/extraction increases while carbon emissions grow rapidly.
	Strong	>0	<0	Dt < 0	Output/extraction decreases while carbon emissions grow.
	Weak	<0	<0	0.8 > Dt > 0	Output/extraction decreases while carbon emissions decline slowly.
Connection	Recessive	<0	<0	1.2 > Dt > 0.8	The rate of output/extraction decline is smaller than the rate of carbon emission decline.
	Expansive	>0	>0	1.2 > Dt > 0.8	The rate of output/extraction growth exceeds the rate of carbon emission growth.

, with specific formulas detailed in Equations (10) and (11).

$$D{G}^t={\displaystyle \frac{\delta \Delta C_t}{\delta \Delta G_t}}={\displaystyle \frac{(C_t-C_{t-1})/C_{t-1}}{(G_t-G_{t-1})/G_{t-1}}}$$

(10)

$$D{R}^t={\displaystyle \frac{\delta \Delta C_t}{\delta \Delta R_t}}={\displaystyle \frac{(C_t-C_{t-1})/C_{t-1}}{(R_t-R_{t-1})/R_{t-1}}}$$

(11)

where t and t − 1 represent the terminal period and the base period, respectively; DG^t and DR^t denote the decoupling indices between the carbon emissions of Guangxi’s MI on the total output value and resource extraction, respectively; δΔC, δΔG, and δΔR, respectively, represent the total value added of carbon emissions, total output value, and resource extraction in Guangxi’s MI.

4. Results and Analyses

4.1. Analysis of Carbon Emissions in Guangxi’s MI

Based on the sectoral energy consumption data of Guangxi’s MI and Equation (1), the carbon emissions of the MI and its five sub-sectors were calculated, as illustrated in Figure 6. Between 2005 and 2021, the total carbon emissions of Guangxi’s MI exhibited a continuous upward trend, rising from 1.3921 to 4.6262 million tons, with an average annual growth rate of 8.7%, highlighting the sector’s ongoing dependence on energy consumption. Regarding the sub-sector differentiation, carbon emissions from coal mining and dressing (CMD), ferrous metal ore mining and dressing (FMMD), and non-ferrous metal ore mining and dressing (NFMD) grew significantly, reaching 0.7789, 1.6932, and 1.7246 million tons, respectively, by 2021. The non-metallic mineral mining and dressing (NMMD) displayed a steady growth pattern, increasing from 0.1714 to 0.4083 million tons. In contrast, petroleum and natural gas extraction (PNGE) consistently maintained the lowest carbon emissions, with the smallest growth in emissions over the period.

From 2005 to 2021, the carbon emissions of the five sub-sectors continuously increased, ranked from highest to lowest as follows: non-ferrous metal ore mining and dressing (NFMD) > ferrous metal ore mining and dressing (FMMD) > coal mining and dressing (CMD) > non-metallic mineral mining and dressing (NMMD) > petroleum and natural gas extraction (PNGE) (shown in Figure 6). This development trend can be divided into three stages. From 2005 to 2011, carbon emissions in Guangxi’s MI and its five sub-sectors rose rapidly, fueled by accelerated industrialization and intensified resource extraction. From 2012 to 2017, the growth rate slowed, mainly due to the gradual implementation of energy-saving and emission reduction policies, though emissions did not decline significantly. Between 2018 and 2021, carbon emissions surged again, driven by increased resource demand from economic growth, particularly as the NFMD and FMMD sub-sectors intensified production efforts, resulting in a significant rise in emissions. Therefore, Guangxi’s MI has experienced continuous growth in carbon with an increasingly differentiated emissions structure, underscoring the ongoing challenge of managing future carbon emission trends.

Figure 7 shows the carbon emission distribution among the five sub-sectors in Guangxi’s MI. The results reveal that NFMD and FMMD collectively dominate emissions (70%), reflecting the province’s role as China’s largest non-ferrous metal base and a black metal production hub. This profile drives extensive mineral extraction, resulting in substantial emissions. CMD contributes approximately 15%, with the main energy consumption in Guangxi still coming from coal. Despite national clean energy mandates, coal persists due to infrastructural inertia and transitional challenges [27]. NMMD accounts for about 10%, which is mainly due to its simpler production processes and lower energy consumption. PNGE emissions remain minimal (1%), constrained by Guangxi’s limited oil and natural gas reserves and small-scale extraction operations.

The extraction processes, production flows, and energy usage methods vary across different mining sectors, leading to differences in carbon emissions and energy consumption structures among industries [79]. The energy consumption structure of carbon emissions in Guangxi MI and its sub-sectors is shown in Figure 8.

In the energy consumption structure of the MI (Figure 8a), coal, oil products, and electricity account for an average of 13.28%, 6.98%, and 79.74%, respectively, with the proportion of electricity consumption gradually increasing. In coal mining and dressing (CMD) (Figure 8b), ferrous metal ore mining and dressing (FMMD) (Figure 8c), and non-ferrous metal ore mining and dressing (NFMD) (Figure 8d), the share of electricity-related carbon emissions averages around 80%. This is primarily attributed to the complex geological conditions in Guangxi’s mining regions and the varying ore grades. The production process typically involves deep mining, complex ore beneficiation, and smelting, all of which heavily depend on electrically driven machinery and advanced processing techniques, resulting in a high concentration of electricity consumption.

In petroleum and natural gas extraction (PNGE), electricity consumption accounts for the entire energy consumption, which is related to resource scarcity in Guangxi and incomplete statistical data. In non-metallic mineral mining and dressing (NMMD) (Figure 8e), the proportion of electricity-related carbon emissions also reaches around 60%, showing a gradually increasing trend. Historically, coal’s contribution to carbon emissions was approximately 30%, while oil accounted for 20–30%. However, recent years have seen a decline in emissions from these sources. This transition reflects the impact of technological progress and the enforcement of policies aimed at energy conservation and emission reduction. As a result, electricity is increasingly supplanting coal and oil as the predominant source of carbon emissions in NMMD, thereby facilitating a shift toward more sustainable and lower-carbon production processes.

While carbon emissions reflect the scale of carbon emissions from an industry or region, the carbon emission intensity, calculated as the ratio of total carbon emissions to total output value in the mining industry, measures the amount of carbon emissions per unit of output value. This is a more accurate indicator of the industry’s progress toward low-carbon development. Using the data from 2005 to 2021, the results are shown in Figure 9. During this period, Guangxi’s MI carbon emission intensity exhibited a fluctuating decline, plummeting 83% from 2.12 to 0.36 tCO₂/10⁴ CNY between 2005 and 2017. This reduction aligns with policy initiatives like the “State Council’s Several Suggestions on Further Promoting the Economic and Social Development of Guangxi,” which stimulated economic growth and technological innovation investments to enhance energy efficiency [71]. However, carbon emission intensity gradually increased from 2018 to 2021, likely due to the expansion of production scale leading to a rebound in carbon emissions.

Among the sub-sectors (shown in Figure 9), petroleum and natural gas extraction (PNGE) exhibits the lowest carbon emission intensity. Between 2005 and 2021, the carbon emission intensity of non-metallic mineral mining and dressing (NMMD) decreased significantly from 1.26 to 0.16 tCO₂/10⁴ CNY, driven by sustained improvements in energy efficiency and structural industrial optimization. The sector has enhanced production value while systematically lowering energy consumption per unit output, supported by Guangxi’s strategic initiatives to modernize building material manufacturing through digitalization and smart mining technologies [80]. Currently, NMMD has developed a relatively comprehensive industrial system, which plays a crucial role in the effective reduction of carbon emission intensity.

The carbon emission intensity of coal mining and dressing (CMD) and non-ferrous metal ore mining and dressing (NFMD) followed a U-shaped trajectory from 2005 to 2021 (shown in Figure 9). This was mainly due to Guangxi’s earlier expanding production strategy that prioritized rapid economic growth, which effectively lowered carbon emission intensity. However, after reaching their lowest points in 2013 (0.38 tCO₂/10⁴ CNY) for CMD and in 2015 (0.37 tCO₂/10⁴ CNY) for NFMD, this trend started to reverse again. During this period, traditional industries accounted for nearly 80% of Guangxi’s industrial value-added output. As industrialization decelerated and developmental momentum waned, the consumption of conventional energy sources (e.g., coal) increased incrementally. Concurrently, the escalating market demand prompted further expansion in the production scale of mineral resources, notably non-ferrous metals. Nonetheless, obsolete production technologies lagged behind, failing to enhance production efficiency, which consequently led to a significant increase in carbon emission intensity.

From 2005 to 2021, ferrous metal ore mining and dressing (FMMD) maintained a notably high carbon emission intensity of 2.16 tCO₂/10⁴ CNY (shown in Figure 9), primarily due to its fragmented mining structure dominated by small-scale operations. These operations exhibited low mineral utilization efficiency, leading to substantial resource waste and elevated energy consumption per production unit. Although the carbon emission intensity temporarily decreased to 1.84 tCO₂/10⁴ CNY in 2019, subsequent years saw a marked resurgence. The fluctuations during this period were influenced not only by the pandemic, which sharply reduced the demand for manganese ore and other black metals such as iron and chromium, but also by declining resource prices and industry overcapacity [81].

Overall, the total carbon emissions of Guangxi’s MI have consistently shown an upward trend, while the carbon emission intensity has followed a trajectory of initial decline followed by a gradual rebound. This pattern reflects the adverse effects of production scale expansion on overall carbon emissions. A sectoral analysis indicates that carbon emissions have increased across all sub-sectors, with NFMD and FMMD experiencing particularly significant growth. Regarding carbon emission intensity, there are notable disparities among sub-sectors: NMMD has seen a continuous decrease, PNGE has remained stable, while other sub-sectors have shown an initial decrease followed by a rebound.

4.2. Analysis of Carbon Emission Drivers in Guangxi’s MI

The decomposition of carbon emission drivers for Guangxi’s MI from 2006 to 2021 was conducted using R software.

Table 7. Contribution rate of Guangxi’s MI change rate of carbon emissions.

Year	△G	△E	△T	△P	△CG	△CE	△CT	△CP	△GP	△EG
2005–2006	0.0718	−0.0455	0.0603	−0.0147	−0.0581	0.0596	−0.0480	0.0269	0.0048	−0.0023
2006–2007	0.1601	0.0337	0.0952	0.0452	−0.1051	0.0121	−0.0512	0.0004	0.0081	0.0024
2007–2008	0.0754	0.0066	−0.0050	0.0023	−0.0425	0.0258	0.0373	0.0302	0.0005	0.0009
2008–2009	0.0062	−0.0165	−0.0119	−0.0083	0.0161	0.0388	0.0340	0.0302	0.0001	−0.0003
2009–2010	0.1313	0.0304	0.1904	0.0878	−0.0813	0.0173	−0.1568	−0.0392	0.0043	−0.0028
2010–2011	0.1024	−0.0176	0.0075	−0.0071	−0.0797	0.0364	0.0109	0.0255	0.0050	0.0009
2011–2012	0.0429	−0.0188	0.0168	0.0086	−0.0358	0.0263	−0.0100	−0.0018	0.0015	−0.0015
2012–2013	0.0511	0.0182	0.0235	0.0082	−0.0433	−0.0110	−0.0164	−0.0012	−0.0002	0.0006
2013–2014	0.0230	−0.0016	0.0057	−0.0190	−0.0147	0.0097	0.0025	0.0276	−0.0010	0.0005
2014–2015	0.0074	0.0064	−0.0458	−0.0028	−0.0040	−0.0029	0.0492	0.0063	−0.0001	0.0000
2015–2016	−0.0033	0.0190	−0.0270	−0.0136	0.0017	−0.0197	0.0254	0.0126	−0.0005	−0.0010
2016–2017	0.0141	0.0586	0.0289	−0.0475	0.0046	−0.0331	−0.0103	0.0752	−0.0108	−0.0061
2017–2018	−0.0968	−0.0038	−0.0868	−0.0623	0.1103	0.0339	0.1146	0.1079	0.0007	−0.0186
2018–2019	−0.0287	0.0281	−0.0853	−0.0263	0.0554	0.0006	0.1104	0.0575	0.0001	−0.0049
2019–2020	0.0035	−0.0214	−0.0572	−0.0179	0.0126	0.0374	0.0726	0.0337	0.0000	−0.0002
2020–2021	0.0242	−0.0016	0.1229	−0.0183	0.0081	0.0335	−0.0943	0.0510	−0.0010	0.0004

details the specific decomposition results, including the contribution rates of the top ten influencing factors on carbon emission changes, while Figure 10a displays the corresponding contribution values. The primary factors driving the increase in carbon emissions were economic scale (G), population intensity (CP), and energy intensity (CE), which had cumulative contribution rates of 58.45%, 44.29%, and 26.49%, respectively. From 2006 to 2015, Guangxi experienced a decline in carbon emissions due to rapid economic growth coupled with relatively low increases in carbon emissions. Economic intensity (CG) emerged as the principal factor contributing to this reduction. From 2016 to 2021, through technological advancements and industrial restructuring, Guangxi achieved a reduction in energy consumption per unit of output. Concurrently, a declining workforce in the mining sector further facilitated reductions in carbon emissions, with population size (P) and energy efficiency output (EG) becoming the predominant factors. Overall, the influence of carbon emission drivers in Guangxi’s MI exhibited dynamic shifts across different periods, reflecting progress in economic development and adjustments in the energy structure. Despite these achievements, challenges in further reducing carbon emissions persist.

The carbon emission contribution values of the MI and its five sub-sectors are shown in Figure 10, revealing the contribution of the top ten influencing factors to carbon emission changes and the overall CO₂ fluctuation. The specific performance of each sub-sector is as follows: For non-ferrous metal ore mining and dressing (NFMD), which had the largest carbon emissions (see Figure 10e), the main factors driving the increase in carbon emissions are population intensity (CP), economic scale (G), and energy intensity (CE), which cumulatively contribute 0.6069, 0.4529, and 0.3854 million tons of CO₂, respectively. The factors that help mitigate the growth of carbon emissions are mainly population size (P) and economic intensity (CG), with cumulative contributions of −0.2486 and −0.1248 million tons, respectively. This indicates that while improvements in energy efficiency and controlling the scale of the mining workforce have had some positive effects, they are still insufficient to offset the emission increase driven by the main factors.

Coal mining and dressing (CMD) and ferrous metal ore mining and dressing (FMMD) (see Figure 10b,d) exhibit similar trends, where the primary driving factors for carbon emissions are CP and CE, while the main contributing factors for carbon emission reduction are energy efficiency output (EG) and P. In CMD, CP and CE have cumulative contribution values of 0.4339 and 0.2921 million tons, respectively, whereas EG and P contribute −0.1968 and −0.1737 million tons, respectively. In FMMD, CP and CE have cumulative contribution values of 1.1159 and 0.7333 million tons, while EG and P contribute −0.3039 and −0.3824 million tons, respectively. Notably, the driving effect of CP has increased rapidly since 2018. This trend is closely related to the continued rise in demand for coal and ferrous metals as fundamental resources, leading to an increase in carbon emissions per capita in the sector. At the same time, the negative contribution of EG highlights the necessity of improving energy efficiency and optimizing industrial structures in the pursuit of low-carbon development.

Petroleum and natural gas extraction (PNGE) (see Figure 10c) is primarily driven by energy consumption scale (E), with a cumulative contribution of 0.0206 million tons. The key suppressing factor is EG, with a cumulative contribution of −0.0252 million tons. Although energy consumption in the PNGE industry directly contributes to a certain amount of carbon emissions, its small production scale in Guangxi allows for emission reductions through improved energy efficiency.

Non-metallic mineral mining and dressing (NMMD) (see Figure 10f) is primarily driven by G, technological level (T), and E, with cumulative contributions of 0.2067, 0.1705, and 0.1281 million tons, respectively. The key driving factor is CG, with a cumulative contribution of −0.1402 million tons. As NMMD’s production scale expands, the driving effect of G on carbon emissions becomes more evident. While T initially played a significant role in driving carbon emissions, it shifted to suppressing emissions after 2015. This shift reflects the industry’s improving technological innovation, which not only improves production efficiency but also enhances CG’s role in reducing carbon emissions.

The influence of each factor is subject to instability and vulnerability to external factors, which may result in a change in direction. For example, in MI (see Figure 10a), the contribution of E fluctuates between −0.0829 and 0.1707 million tons, while CG ranges from −0.1970 to 0.3450 million tons. Considering this observation, the above analysis focuses more on the cumulative changes in the obtained results.

Overall, the driving factors of carbon emissions in Guangxi’s MI and its five sub-sectors exhibit distinct differences. The primary drivers for MI and NFMD are G and CP, while CMD and FMMD are mainly influenced by CP and CE. For PNGE, the key driver is E, whereas NMMD is driven by G and T. Despite the differences in the contribution of driving factors across the sub-sectors, the overall impact of driving factors is stronger than that of inhibiting factors. This suggests that the increase in carbon emissions driven by economic development outweighs the positive effects of policy measures aimed at reducing emissions. Consequently, more targeted and differentiated carbon reduction strategies should be implemented to effectively control overall carbon emissions and promote the sustainable development of the industry.

4.3. Scenario Simulation and Decoupling Analysis

4.3.1. Robustness Check

To verify whether the model results effectively reflect the actual development patterns of the system, a model verification was performed. The model was first run to obtain prediction results while reducing variable instability across different time steps. Given the numerous variables in the carbon emission model, the “Total output value of MI” variable from the economic development subsystem was selected for testing. Three time steps, 0.25 years, 0.5 years, and 1 year, were used for robustness testing, with the results shown in Figure 11. The analysis reveals that the impact of different time steps on the total carbon emissions of Guangxi’s MI is minimal, indicating the model’s stability and successful verification [82].

4.3.2. Historical Examination

By comparing the simulation results with the actual results, this study selects four variables: “MI carbon emissions”, “Total output value of MI”, “Resource extraction volume”, and “Carbon emission intensity” to verify the model’s effectiveness based on the error magnitude, as shown in Figure 12. Through comparative analysis, it is found that the error values of the selected variables are mostly within 5%, and all variables are within the 10% error range required by the SD model [82], indicating that the established model passes the historical verification and can reflect the actual situation.

4.3.3. Scenario Simulation Analysis

The carbon emission trends of Guangxi’s MI for the period 2022–2035 are simulated under the five development scenarios, as shown in Figure 13. It is observed that only the three scenarios under the ERS achieve the carbon peak target by 2030, while two scenarios under the ESS reach their peak in 2032. In the remaining scenarios, carbon emissions continue to rise, albeit at varying rates. The carbon emission levels, ranked from highest to lowest, are as follows: RDS > BS > GDS.

In the BS, if the historical development trend continues, carbon emissions will keep rising from 4.5061 million tons in 2021 to 11.8459 million tons by 2035. In the RDS, with priority given to economic development, its high energy consumption characteristics lead to an expansion in resource extraction scale, significantly accelerating carbon emission growth. By 2035, carbon emissions in the RDS2 will reach 17.6965 million tons, a 50.70% increase compared with BS. Both RDS1 and RDS3 slow down the carbon emission growth rate by reducing population growth. However, since both scenarios shift resource extraction and energy consumption from medium to high, their carbon emissions increase by 26.38% and 23.49%, respectively, compared with BS, indicating that increasing resource extraction has a more significant driving effect on carbon emissions.

In the GDS (shown in Figure 13), although carbon emissions continue to show an upward trend, the growth rate slows down compared with the BS. Specifically, in the GDS3, carbon emissions will reach 11.779 million tons by 2035. This suggests that a moderate reduction in the proportion of the secondary industry can effectively mitigate carbon emissions. Moreover, by optimizing the industrial structure and collaboratively adjusting variables such as energy consumption intensity and the scale of resource extraction, carbon emissions in GDS1 and GDS2 are expected to decrease to 8.2018 and 7.1654 million tons by 2035, respectively. These adjustments successfully moderate the growth rate of carbon emissions. To enhance these outcomes, the government should further refine the industrial structure, optimize energy consumption patterns, and control the scale of resource extraction. Implementing these multidimensional strategies will establish a robust collaborative emission reduction mechanism and enhance the efficacy of carbon reduction efforts.

In the ESS (shown in Figure 13), to achieve carbon emission peak control, policies of low GDP growth and low energy consumption result in carbon emissions peaking in 2032. Specifically, the carbon emission peaks for the ESS1 and ESS2 are 5.8666 and 5.9015 million tons, respectively. These findings indicate that slowing economic development and controlling energy consumption significantly reduce carbon emissions. Compared with the ESS1, however, the ESS2, adjusts the population and industrial structure from medium to low, yet carbon emissions still increase. This suggests that simply controlling population growth without accelerating industrial restructuring will not effectively curb carbon emissions; in fact, it may even lead to an increase due to the slower pace of industrial transformation. Overall, it is essential to strengthen adjustments to the economic development model and accelerate the transition to clean energy, along with other coordinated measures, to achieve the carbon peak target before 2030.

In the ERS (shown in Figure 13), which adopts a carbon reduction-oriented approach, the carbon peak is achieved ahead of schedule in 2028 across all three scenarios due to decreases in energy consumption and resource extraction. Specifically, the peak values for the ERS1, ERS2, and ERS3 are 4.8475, 4.8627, and 4.7972 million tons, respectively, which are approximately one million tons lower than the peak values in the ESS. These findings suggest that directly regulating sources of carbon emissions through the control of energy consumption and resource extraction can significantly mitigate carbon emissions. Consequently, it is recommended that the Guangxi government actively optimize the energy consumption structure and reduce dependence on traditional high-carbon energy sources [83]. Simultaneously, stringent control over the scale of resource extraction and the promotion of green production technology innovations in the extraction process can not only alleviate environmental pressures but also effectively reduce energy consumption and carbon emissions during the extraction phase, thereby further optimizing the industrial structure.

Figure 14 presents the simulated carbon emission intensity of Guangxi’s MI from 2022 to 2035. Across the scenarios, carbon emission intensity peaks before 2035 in all cases except the BS and GDS3. Specifically, the GDS3 adjusts only the industrial structure and population compared with the BS. However, the carbon emission reduction effects of these two adjustments are weaker than the driving forces of economic growth, causing the carbon emission intensity to continue rising in this scenario. This underscores the need for multidimensional policy interventions to achieve substantial decarbonization.

Carbon emission intensity in most scenarios initially rises before declining, reflecting a transitional phase where economic expansion temporarily elevates emissions. However, with the implementation of carbon reduction policies targeting energy consumption and industrial structure, carbon emission intensity eventually stabilizes and diminishes (shown in Figure 14). Notably, scenarios such as the RDS3, RDS2, GDS1, and ESS1 demonstrate this trend. Specifically, the RDS3 reaches its peak carbon emission intensity in 2034, with a maximum value of 1.56 tCO₂/104 CNY. This peak illustrates that carbon emission demand is closely linked to the scale and developmental level of economic growth; larger scales of economic development correspond to increased carbon emission demands, resulting in elevated overall carbon emission intensities.

In the ERS (shown in Figure 14), carbon emission intensity shows a continuous downward trend. The ERS1, ERS2, and ERS3 all reach their carbon peak in 2021 with a value of 0.78 tCO₂/104 CNY. This outcome suggests that Guangxi has effectively mitigated the increase in carbon emission intensity amidst ongoing economic expansion. This has been accomplished through the judicious management of resource extraction in the construction of green mines, the advancement of energy-efficient and low-carbon technologies, and the enhancement of energy infrastructure. The ERS exemplifies a decrease in carbon emission intensity amid economic growth, offering valuable insights for policymakers striving to meet carbon peak objectives.

4.4. Carbon Emission Decoupling Analysis

4.4.1. Decoupling Effect Analysis of Total Output Value and Carbon Emissions

Figure 15 illustrates the decoupling relationship between carbon emissions and total output value in Guangxi’s MI, derived from Equation (10). Weak decoupling dominated during 2006–2008, 2010–2013, and 2020, reflecting sluggish progress in decoupling economic growth from emissions. In contrast, expansive negative decoupling occurred in 2009 and 2021, while 2015 and 2017 saw expansive connection, and 2016 exhibited recessive connection. A phase of strong negative decoupling emerged in 2018–2019. These observations indicate that the decoupling status between carbon emissions and total output value in Guangxi’s MI experienced significant fluctuations, largely driven by economic expansion and delays in technological advancements. This underscores the necessity for enhanced alignment between carbon reduction efforts and economic growth. Furthermore, a notable increase in carbon emissions in 2021 led to a higher baseline value, exacerbating the observed effect. The variation in carbon emissions in 2022 had a relatively minor impact on the overall decoupling effect, thereby introducing some distortion in the decoupling status. This accentuates the need for improving the coordination mechanisms between carbon emissions and economic growth.

The BS and RDS3 exhibited weak decoupling in 2022, while the RDS1 and RDS2 displayed expansive connection and expansive negative decoupling, respectively (shown in Figure 15). The RDS3 is projected to demonstrate expansive connection during 2033–2035, whereas the remaining scenarios are expected to exhibit expansive negative decoupling throughout 2023–2035. These dynamics are closely tied to the economic trajectory of the Guangxi region, which is currently experiencing rapid industrialization, urbanization, and economic transformation. The region’s development continues to be driven by a resource-dependent economic model. This suggests that future economic development patterns will remain resource-intensive, characterized by a high energy consumption that supports rapid economic growth. Consequently, the regional ecological environment is deteriorating, complicating the effective control of carbon emissions. This exacerbates the conflict between economic development and carbon emission management.

The GDS1 and GDS2 exhibited strong decoupling in 2022, followed by expansive negative decoupling from 2023 to 2028 (shown in Figure 15). The GDS1 demonstrated expansive connection from 2029 to 2030 and transitioned to weak decoupling from 2031 to 2035, while the GDS2 exhibited weak decoupling from 2029 to 2035. The GDS3 maintained expansive negative decoupling from 2023 to 2035. These results suggest some progress in green development, yet a deep decoupling of economic growth from carbon emissions remains unachieved. Despite Guangxi’s implementation of robust policies such as the Guangxi Zhuang Autonomous Region Carbon Peaking Implementation Plan [84], aimed at enhancing energy efficiency and reducing carbon emissions, several challenges persist. The region’s relatively low economic development, limited technological innovation, complex industrial restructuring, and the high costs of green development hinder its transition to a low-carbon economy [85]. As a result, Guangxi has experienced a weak decoupling status, characterized by economic growth that surpasses the reduction in carbon emissions.

The ESS1 and ESS2 exhibited strong decoupling in 2022 and from 2033 to 2035, with intermittent fluctuations in decoupling status observed in the intervening years (shown in Figure 15). These results suggest that energy-saving initiatives are gradually proving effective, though some instability remains. As economic growth slows, businesses may face temporary financial constraints, limiting investments in energy-efficient infrastructure and technological innovation. This could reduce energy efficiency, increase carbon emissions, and weaken decoupling progress. However, over the long term, these adjustments are expected to reduce corporate energy dependence, minimize waste, and curb excessive energy consumption at its source, ultimately helping to lower carbon emissions. As industrial development drives economic expansion, strong decoupling is expected to follow.

The ERS1, ERS2, and ERS3 exhibited strong decoupling in 2022 and from 2029 to 2035, while displaying weak decoupling from 2023 to 2028 (shown in Figure 15). This suggests that carbon emissions are gradually being brought under control, achieving a decoupling of economic growth from carbon emissions. This outcome aligns with the scenario’s emphasis on reducing energy consumption through improved energy efficiency. Guangxi’s “Six Key Actions for High-Quality Development” emphasize improving economic quality while maintaining stable growth. To support this, Guangxi enterprises have expanded clean energy development, improved energy efficiency, and optimized the energy consumption structure. These efforts have ensured a sustained reduction in carbon emissions, promoting balanced economic and environmental development [86].

4.4.2. Decoupling Effect Analysis of Resource Extraction Volume and Carbon Emissions

Using Equation (11), the decoupling effect between carbon emissions and resource extraction in Guangxi’s MI was calculated, as shown in Figure 16. The results indicate that expansive negative decoupling occurred in 2006–2007 and 2015, while weak decoupling was observed in 2008–2009, 2012–2013, and 2017–2021. Strong negative decoupling was present in 2010, expansive connection in 2011, recessive connection in 2014, and weak negative decoupling in 2016. These fluctuations highlight Guangxi’s MI continued reliance on resource extraction. There is a pressing need to enhance the optimization of resource extraction processes and to promote industrial restructuring.

Under the three scenarios of BS (i.e., BS1, BS2, BS3), RDS (i.e., RD1, RD2, RD3), and GDS (i.e., GDS1, GDS2, GDS3), the decoupling status exhibited expansive connection from 2023 to 2035 (shown in Figure 16). This persistent linkage highlights both the region’s entrenched reliance on extractive industries and the formidable challenges in achieving emission reductions. As the economy grows, the demand for resources increases, driving continuous expansion in extraction activities. Although industrial restructuring is implemented in the GDS, the region’s transition pace proves insufficient to counteract escalating energy consumption. This structural inertia maintains a parallel growth trajectory for both resource extraction and carbon emissions, with emission rates closely tracking extraction activities. Consequently, effective decoupling has not been achieved. Unless Guangxi undertakes deeper structural reforms in its industry and energy consumption, the strong link between resource extraction and carbon emissions will remain difficult to break.

Under the ESS, the coordination between resource extraction and carbon emissions was unsatisfactory (shown in Figure 16). Specifically, the ESS1 and ESS2 exhibited recessive connection in 2022, expansive connection during 2023–2030, and predominantly recessive connection or recessive decoupling during 2031–2035. The overall trajectory indicates that carbon emissions initially rise in tandem with resource extraction before subsequently declining. In particular, ESS1 reaches its peak for both resource extraction and carbon emissions in 2032, while ESS2 sees its maximum resource extraction in 2033. This results in ESS2 achieving a strong decoupling in 2033, with resource extraction increasing while carbon emissions decrease. However, the unsustainable nature of this trend leads to a decline in both variables thereafter, shifting to a phase of recessionary decoupling. This pattern reflects the slow progress in Guangxi’s energy-saving technologies and limited innovation momentum. These findings suggest that Guangxi should eschew the pursuit of rapid economic expansion in favor of prioritizing the enhancement of research and the application of energy-saving technologies, improving the efficiency of resource conversion, and striving for comprehensive and coordinated development [87].

Under the ERS (i.e., ERS1, ERS2, ERS3), recessive connection was observed in 2022 and from 2029 to 2035, while expansive connection occurred from 2023 to 2028 (shown in Figure 16). This progression from intensive to attenuated linkage demonstrates a poor decoupling performance between resource extraction and carbon emissions. This implies that, driven by the demand for resource extraction, Guangxi cannot solely rely on adjustments in resource extraction volumes to reduce carbon emissions. It is essential to move away from the previous model of small-scale or fragmented mining and focus on large-scale mineral resource extraction [88]. Furthermore, optimizing resource extraction processes and reducing energy consumption through improved mining techniques are critical for achieving a balance between resource extraction and carbon reduction objectives in the MI.

5. Conclusions and Suggestions

5.1. Conclusions

(1) An analysis of carbon emissions from Guangxi’s MI from 2005 to 2021 reveals a consistent upward trajectory, reaching 4.63 million tons in 2021 with an average annual growth rate of 7.92%. This trend underscores the region’s continued reliance on resource-intensive industries, highlighting significant challenges in decoupling economic growth from carbon emissions. Among its five sub-sectors, the emissions ranked from highest to lowest are follows: NFMD, FMMD, CMD, NMMD, and PNGE. NFMD has the highest carbon emission share at an average of 40%, primarily from electricity consumption. However, NFMD’s carbon emission intensity has dropped by 64% from 2005 to 2021. This trend demonstrates that despite overall emission growth, carbon intensity can still be effectively controlled through decarbonization measures.

(2) This study applied the GDIM factor decomposition model to identify the key drivers of carbon emissions in Guangxi’s MI. The analysis identified G, CP, and CG as primary drivers, collectively accounting for 58.45%, 44.29%, and 26.49% of carbon emissions, respectively. Conversely, P and EG, as critical inhibitory factors, offset emissions by 8.57% and 3.20%, respectively. While the inhibitory effects were smaller in magnitude compared with the driving forces, their impact remains significant and warrants attention. Furthermore, the influence of these factors exhibited substantial variability across different sub-sectors and over time. The varying economic development of different sub-industries leads to differences in resource consumption and energy demand. As policies change, the volatility and disparities in carbon emission drivers become increasingly evident.

(3) By constructing the SD model integrating economic, demographic, energy, environmental, and policy dimensions, this study simulated carbon emission trends for Guangxi’s MI from 2022 to 2035 under multiple scenarios. The results show that under the ERS, emissions are projected to peak in 2028 at 4.80 to 4.86 million tons, achieved primarily by optimizing energy consumption and limiting resource extraction, and this represents the most optimal carbon reduction pathway. In contrast, the ESS delays the peak to 2032, with emissions ranging between 5.81 and 5.86 million tons. Meanwhile, under the BS, RDS, and GDS, average emissions are forecasted to escalate to 11.85, 15.77, and 9.05 million tons by 2035, respectively. These results underscore that a unilateral reliance on population control or industrial restructuring is inadequate for achieving carbon peaking. Therefore, a multidimensional and comprehensive approach to control measures is necessary.

(4) The analysis of decoupling trends in Guangxi’s MI reveals distinct patterns between carbon emissions and economic output. From 2005 to 2021, carbon emissions were primarily characterized by expansive negative decoupling and weak decoupling relative to total output value. From 2022 to 2035, scenarios under the ESS and ERS show gradually improved decoupling effects, with strong decoupling becoming more common. This suggests that energy-saving and carbon reduction goals, coupled with reasonable economic growth and effective carbon constraint policies, can facilitate strong decoupling between economic growth and carbon emissions. In contrast, the GDS1 and GDS2 under the GDS generally exhibited weak decoupling, while the BS, RDS, and GDS3 still show expansive negative decoupling. Meanwhile, from 2005 to 2021, the decoupling of carbon emissions from resource extraction in Guangxi’s MI was mainly weak, with significant fluctuations. In 2022–2035, scenarios under the BS, RDS, and GDS primarily presented expansive connection decoupling, whereas the ESS and ERS were relatively poor, mostly exhibiting recessive decoupling or connection. These findings suggest that the high dependence on resources and the inherent characteristics of the MI in mineral-rich regions closely link carbon emissions to energy consumption and resource extraction. Thus, achieving significant decoupling from resource extraction in the short term remains a considerable challenge.

5.2. Policy Suggestions

These findings underscore the formidable challenges faced by mineral-rich regions like the MI in achieving the “dual carbon” goals. To effectively decouple carbon emissions from economic growth and resource extraction while advancing actionable decarbonization strategies, policymakers should prioritize the following critical areas:

Firstly, governments should strengthen institutional decarbonization leadership by adopting a dual strategy that combines economy-wide regulatory frameworks with prioritized sector-specific interventions. In Guangxi, where mining enterprises operate across diverse geographical and industrial contexts, carbon emission reduction policies must account for significant variations in energy consumption patterns and sector-specific drivers of carbon intensity. A granular analysis of structural disparities and the underlying mechanisms driving emission differences across mining sub-sectors is critical to identifying key bottlenecks in energy efficiency. Building on this, governments should implement precision policies that address both universal challenges and sub-sector-specific barriers. This approach ensures policy coherence, maximizes impact, and accelerates progress toward region-wide carbon neutrality.

Secondly, the strategic management of resource extraction and energy mix optimization are pivotal. Mineral-rich regions often exhibit a heavy reliance on resource extraction and fossil fuels due to their natural endowments, exacerbating tensions between carbon emissions and energy demand. Therefore, optimizing the energy consumption structure is critical to addressing this issue. Addressing this requires dual actions: regulating extraction intensity to alleviate carbon pressure at the source and accelerating clean energy adoption through technological innovation and efficiency improvements to reduce fossil fuel dependency. These measures collectively curb energy consumption while advancing sustainable, low-carbon regional development.

Thirdly, achieving carbon peaking in mineral-rich regions demands a systematic alignment of low-carbon development with economic and resource priorities. While economic growth remains foundational, pursuing decarbonization through the outright suppression of industrial activity or resource extraction is neither feasible nor sustainable. Instead, the approach to attaining a carbon peak may draw upon the ERS frameworks, positioning the low-carbon development objective as a strategic focus. This entails integrating carbon emission metrics into economic planning, a gradual modification of the economic structure, and the promotion of a transition in the mineral industry towards energy efficiency and low-carbon development. Such measures will provide robust institutional backing for reaching carbon reduction goals and for the decoupling of carbon emissions in mineral-rich regions.

This study provides valuable insights that can serve as a reference for carbon reduction efforts in the MI. The findings are particularly relevant to the low-carbon development of other mineral-rich regions. However, the scenario simulations are limited to the year 2035 and do not cover longer-term development trends. Future work will simulate over a longer time horizon and extend the analysis to resource-rich cities nationwide, aiming to develop a more comprehensive analytical framework that supports carbon reduction and sustainable development in the MI from a broader, global perspective.