Research on Spatiotemporal Changes in Carbon Footprint and Vegetation Carbon Carrying Capacity in Shanxi Province

Abstract

The climate and ecological problems caused by excessive carbon dioxide emissions are attracting more and more attention, and the need for carbon reduction has reached a consensus. Carbon peak and carbon neutrality is a solemn commitment made by China to the world and a strategic arrangement to achieve economic and social transformation. This study focused on Shanxi Province, a base of power source and heavy chemical industry in China. Based on energy consumption data, carbon emission data set at the county level, land use data and socioeconomic data, we built a carbon ecological pressure index in order to analyze the spatiotemporal evolution characteristics of the carbon footprint, carbon carrying capacity and carbon ecological security status of each county in Shanxi Province from 2000 to 2020. The results showed that: (1) the total carbon footprint of Shanxi Province increased, and the number of high carbon emission counties showed an increasing trend. The largest part of carbon footprint was coal. (2) The vegetation carbon-carrying capacity showed an increasing trend in general, and forest land was the main contributor to the carbon-carrying capacity. (3) The carbon deficit of Shanxi Province was greater than 0 and behaved as a carbon source. The carbon ecological security decreased from a relatively safe level to a general safe level. (4) The carbon ecological pressure index gradually increased. It was predicted that the carbon ecological security level of each county will remain basically unchanged by 2025 and some will still be at a carbon ecological insecurity level. In general, the carbon ecological pressure index of some counties was still large. It is necessary to strengthen the use of clean energy, optimize the industrial structure and increase the carbon sink of forest land in order to reduce carbon emissions and increase the carbon sink, so as to ensure carbon ecological security and realize the goal of a low-carbon economy.

1. Introduction

Global warming caused by emissions of carbon dioxide and other greenhouse gases has attracted wide attention. The global carbon dioxide concentration in 2020 was 413.2 ppm, 149% of preindustrial levels [1]. Global warming will cause sea levels to rise, ocean acidification, glacier melting and other issues that will have irreversible consequences for the ecosystem. Food security will also be jeopardized by high temperatures and drought. According to State of the Climate in 2021, the global average temperature was about 1.11 (±0.13) °C above preindustrial levels. China is a climate-change-sensitive region, with a significant upward trend in annual average surface temperature from 1951 to 2020, and the warming rate is significantly higher than the global average during the same period, with a warming rate of 0.26 °C/10 years [2]. Controlling carbon dioxide emissions is now widely accepted as a way to reverse the trend of global warming. In 2015, the Paris Agreement set the long-term goal of limiting the increase in global temperature to 2 °C and aiming for 1.5 °C [3]. In 2005, China became the world’s largest carbon emitter, and its total carbon emissions reached 10 billion tons in 2012 [4]. China has actively assumed responsibility and has made emission reduction commitments on many occasions. At the United Nations Climate Change Conference in 2009, China established the target of reducing carbon emission intensity by 40 to 45% by 2020 compared to 2005, which has already been accomplished ahead of schedule. In the National Independent Contribution to Address Climate Change paper submitted in 2015, a target was proposed of peaking carbon dioxide emissions around 2030, endeavoring to reach the peak as early as possible and reducing carbon emission intensity by 60%–65% compared to 2005. At the 75th General Assembly of the United Nations, China made a solemn pledge to achieve peak carbon dioxide emissions by 2030 and carbon neutrality by 2060. A total of 125 countries proposed carbon neutrality targets by June 2020 [5].

Correspondingly, research on carbon footprint and carbon carrying capacity has become increasingly abundant in this context. There is no standard definition of a carbon footprint; however, it can be summed up as the total amount of greenhouse gases released during the course of a product or service’s life cycle [6], or through human activity, measured in carbon dioxide equivalent [7]. The methods used to measure carbon footprint mainly include the bottom-up input–output method [8], top-down life-cycle method [9], IPCC calculation method [10] and carbon footprint calculator [11]. Most people use the first three. The first three methods were widely used. Research on carbon footprint spans several scales, including country [12], city [13], household [8], product [14], etc. In terms of sectors and industries, carbon footprint research involves the industrial sector [15], the agricultural and forestry sector [16], the transportation industry [17], the medical industry [18], etc. Additionally, studies on the spatiotemporal evolution and influencing factors of carbon footprints have been carried out by scholars. Pan et al. [19] analyzed the characteristics of spatial distribution and spatiotemporal evolution of the carbon footprint of energy consumption in China, as well as studied the decoupling effect between environmental carbon load and economic growth based on an improved Tapio model. Chen et al. [20] used the STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology) model and PLS (partial least squares regression) model to analyze the main influencing factors of the carbon footprint of energy consumption in Beijing. The results showed that population, per capita income and urbanization level were all positively correlated with carbon emissions, with urbanization level being the most significant influence on Beijing’s carbon footprint of energy consumption.

Carrying capacity is a physical concept, and carbon carrying capacity is defined as improving the quality of human life within the carbon carrying capacity of the supporting ecosystem [21]. It refers to the amount of carbon dioxide that vegetation can fix through photosynthesis [22]. Being the greatest carbon reservoir in terrestrial ecosystems, forests have extremely high carbon carrying capacity and carbon sequestration capacity [23]. In addition to forests, the carbon carrying ability of crops and grasslands is primarily taken into account. Huang et al. calculated the carbon carrying capacity of 30 provinces in China and analyzed regional differences [24]. Tu et al. measured the tourism carbon carrying capacity in East China and studied the connection between tourism-related carbon emissions and carrying capacity. The findings indicated that the region is under enormous ecological pressure because of tourism [25].

The concept of ecological security was introduced as early as the 1970s [26]. One component of ecological security is carbon security, which refers to the degree of assurance that carbon emissions will not have an impact on the climate, environment or our life and production activity in a region [27]. In recent years, low-carbon economies, energy efficiency and emission reduction have been popular; research on carbon security has attracted the attention of scholars. A carbon safety assessment was conducted using the principle of carbon balance. To measure carbon ecological security, the primary indicators are as follows: carbon pressure index, also known as carbon bearing intensity index; carbon footprint index; and net carbon footprint. Research on carbon ecological security is mostly focused on a large scale, with more research on the national and provincial levels. Wei et al. [27] assessed the carbon ecological security level of Xinjiang Province in China, and the findings revealed that its carbon ecological security showed a decreasing trend, with the rise in fossil energy consumption being the primary factor. Huang et al. [28] explored carbon security and its influencing factors in Zhejiang Province, and found that the level of carbon security was mainly influenced by carbon footprint, and that the level of carbon security was relatively low. Wu et al. [29] analyzed China’s carbon ecological security and pointed out that, overall, the level of carbon security and eco-economic coordination saw a decreasing trend, and carbon deficit appeared in most provinces.

China is one of the countries in the world with a fragile ecological environment, over 60% of the total land area is ecologically sensitive [30], particularly in regions like the Loess Plateau [31]. Shanxi Province is situated in the Loess plateau, which is an ecologically fragile area. Shanxi Province is also a resource-based province, rich in mineral resources. The economy has grown rapidly in recent years, but because of its extensive reliance on resources, the economic structure is out of balance. And its coal-based energy structure has led to high carbon emissions, resulting in greater pressure on its environment and limiting sustainable development. To realize energy saving and carbon emission reduction, socio-economic transformation is an absolute necessity to ensure the sustainable development of Shanxi Province. Based on the calculation of the carbon footprint and vegetation carbon carrying capacity of counties in Shanxi Province, this study established the carbon ecological pressure index to evaluate the carbon ecological security status. The study aims to provide suggestions for energy saving, emission reduction and risk prevention in Shanxi Province, as well as to assist in realizing the goal of carbon neutrality, which is of great significance for advancing the socio-economic transformation and ecological protection in Shanxi Province. The study of the carbon ecological security of such a large resource-based province also has significance for the world’s attempt to curb climate change. The specific objectives of this study are: (1) to analyze the spatiotemporal change characteristics of carbon footprint and vegetation carbon carrying capacity from 2000 to 2020; and (2) to evaluate the carbon ecological security level in the study area from 2000 to 2020 and predict it for 2025.

2. Date and Methods

2.1. Study Area

Shanxi Province, lying between 34°34′ N and 40°44′ and 110°14′ E and 114°33′ E, is located in North China. It consists of 117 districts and counties in 11 prefecture-level cities, with a total area of 156,700 km² (Figure 1). The terrain is high in the northeast and low in the southwest, with an altitude above 1000 m in most areas. The landform types are complex and varied, mainly mountainous and hilly, accounting for 80% of the total area. It has a temperate continental monsoon climate, with an average annual temperature of 4.2~14.2 °C, and annual precipitation between 358 and 621 mm. Of the rivers within the region, the Yellow River and Haihe River are its two major water systems. Shanxi Province is rich in mineral resources, with five kinds of resource reserves ranking top in China, and it is a typical resource-based area. Forest land, grassland and cultivated land are the main land use types, with forest land area making up about 39% of the total area. The forest coverage rate is about 23.6%. Permanent residents numbered 34.80 million in 2021 and the regional gross domestic product increased by 9.1% to CNY 2259.016 billion.

2.2. Data Collection

The energy consumption data of Shanxi Province used in this study (2000, 2005, 2010, 2015, 2020) were provided by China Energy Statistical Yearbook. County-level carbon emission data were obtained from the CEADs database (https://www.ceads.net.cn/ (accessed on 18 January 2023)), and the county-level carbon emission data, in China during the period of 1997 to 2017, were collected by Chen et al. using the particle swarm optimization-back propagation (PSO-BP) algorithm to invert the night light data; the fitting effect reached 0.988 [32]. And the county-level carbon emissions in 2020 were predicted by the exponential smoothing method [33], an important time-series forecasting method that takes a weighted average of historical data as a forecast for future moments. The land use data were from the 30 m resolution China Land cover data collection, from 1985 to 2021, conducted by Wuhan University (http://doi.org/10.5281/zenodo.4417809 (accessed on 18 January 2023)), which is one of the few publicly available 30 m resolution long-term year-by-year land cover databases in China [34]. The area of forest land and grassland in each district and county was obtained by processing land use data using ArcGIS10.2. Socio-economic data such as crop yield, population and GDP were derived from Shanxi Statistical Yearbook.

2.3. Research Methods

2.3.1. Calculation of Carbon Footprint

Carbon footprint in this study is defined as the total amount of carbon dioxide released into nature by human beings during social and economic activities. As a typical resource-based province, carbon dioxide generated by energy consumption is the main source of carbon emissions in Shanxi Province; as such, this study only calculates energy consumption carbon emissions. We calculated carbon emissions from the end consumption of the following six fossil fuels: coal, coke, natural gas, gasoline, diesel and kerosene, as well as the carbon emissions from the secondary energy consumption of electricity. Electricity consumption does not directly result in carbon emissions, which are calculated according to the energy input of thermal power generation. The formula is as follows:

$$F={{\displaystyle \sum }}_{i=1}^n{E}_i\times e_i\times P_i\times 44/12$$

(1)

where F is the total carbon footprint (unit Mt); E_i is the actual consumption of energy i; e_i is the standard coal coefficient of energy i (refer to appendix of China Energy Statistical Yearbook for values); and P_i is the carbon emission coefficient of energy i, which is obtained by converting units from the default carbon content value in the IPCC Guidelines for National Greenhouse Gas Inventories.

The carbon footprint of Shanxi Province was calculated using the above formula. Since it is difficult to obtain county-level energy consumption data, and it is often inconsistent and incomplete. Thus, we used Chen et al.’s county-level carbon emission data on China to obtain the carbon footprint of counties in Shanxi Province.

2.3.2. Calculation of Vegetation Carbon Carrying Capacity

Carbon carrying capacity refers to the amount of carbon dioxide fixed by photosynthesis of various vegetation. The carbon sequestration capacity of forest land, grassland and crops was mainly taken into account in Shanxi Province. The formula is as follows:

$$C=C_f+C_g+C_p$$

(2)

where C is the regional carbon carrying capacity (unit Mt); C_f is the carbon sequestration in forest land (unit Mt); C_g is the carbon sequestration in grassland (unit Mt); and C_p is the carbon sequestration of crops (unit Mt).

(1)

Calculation of carbon sequestration in forest land and grassland

$$C_f=S_f\times NE{P}_f\times 44/12$$

(3)

$$C_g=S_g\times NE{P}_g\times 44/12$$

(4)

Here, C_f is the carbon sequestration amount of forest land; C_g is the carbon sequestration amount of grassland; S_f is the area of forests land; and S_g is the area of grassland. NEP is the net ecosystem productivity, which is the carbon sequestration amount per unit area of vegetation, i, in one year. When we calculated carbon sequestration in forest land and grassland, considering the classification of land use types in the land use data, we measured the carbon sequestration amount of forest, shrub and grassland. Shanxi Province has a temperate climate, referring to the research of Xie et al. [35], and the NEP of temperate forest is 4.5, the NEP of grassland is 0.9483, the shrub NEP takes the average of forest and grassland, and is 2.725.

(2)

Calculation of carbon sequestration of crops

$$C_p=\beta \times z\times \sum \frac{P_i}{Y_i}\times 44/12$$

(5)

Here, C_p is the regional carbon sequestration of crops; β is the correction coefficient, calculated according to the share of straw in the industry, which is 0.05 [36]; z is the conversion coefficient of biomass and carbon sequestration, which is 0.5 [36]; P_i is the economic yield of crop, i, and this study mainly considers seven crops: wheat, corn, sorghum, millet, soybean, cotton and oil; and Y_i is the economic coefficient of crop, i, which is the ratio of crop economic yield to biological yield. The economic coefficient of 7 crops is 0.37, 0.49, 0.39, 0.4, 0.25, 0.35 and 0.39, respectively. Due to the yield of wheat, corn, sorghum and millet, these were not counted separately in each district and county in 2000, only the total yield of food crops was counted, so the average economic coefficient of the four crops, 0.35, was taken to calculate the crops’ carbon sequestration.

2.3.3. Carbon Ecological Security Model

The balanced relationship between carbon emission and carbon sequestration can reflect regional carbon ecological security. Carbon footprint represents the amount of carbon released to nature during economic and social activities, and carbon carrying capacity represents the carbon sequestration capacity of vegetation. There will be a carbon ecological deficit if the carbon footprint exceeding the carbon carrying capacity, that is, carbon emissions cannot be fully absorbed by vegetation carbon sequestration, and regional ecological security will deteriorate. There will be a carbon surplus if the carbon carrying capacity is greater than the carbon footprint, which is conducive to curbing climate warming and improving the regional ecological security level. The carbon ecological pressure index was chosen to assess the balanced between carbon emission and carbon sequestration, and to indicate the regional carbon ecological security level. The formula is as follows:

$$CTI=F/C$$

(6)

where CTI is carbon ecological pressure index; F is carbon footprint; and C is the carbon carrying capacity.

2.3.4. Spatial Autocorrelation Analysis

The global Moran’s index (Global Moran’I) is a commonly used index for global spatial autocorrelation analysis, which can describe the average degree of correlation between all spatial units and their surroundings from a global perspective. This study used the Global Moran’I to measure the degree of spatial autocorrelation between carbon ecological security level and the global perspective on a county-level, and to explore whether the spatial distribution of carbon ecological security level was clustered. The formula is as follows:

$$I=\frac{n\times {{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1}^n{w}_{ij}\left(x_i-\overline{x}\right)\left(x_j-\overline{x}\right)}{{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1}^n{w}_{ij}{{\displaystyle \sum }}_{i=1}^n{\left(x_i-\overline{x}\right)}^2}$$

(7)

where x_i, x_j is the carbon ecological pressure index of the county i and county j, respectively; n is the total number of county units; and W_ij is a spatial weight matrix representing the proximity of the county i and county j. The range of I is from −1 to 1. The more it approaches 1, the spatial agglomeration is more significant; and the more it approaches −1, the spatial divergence is more significant.

The local Moran’s index (Local Moran’I) is used for local spatial autocorrelation analysis, analyzing the spatial aggregation characteristics of a specific location and identifying the clustering pattern of this location and surrounding ones on a certain attribute. It was used to identify the carbon ecological security level of the position at which the aggregation phenomenon occurred, and the clustering pattern to which it belonged to in this study. The formula is as follows:

$$I_i=\frac{n\left(x_i-\overline{x}\right){{\displaystyle \sum }}_{j=1}^n{w}_{ij}\left(x_j-\overline{x}\right)}{{{\displaystyle \sum }}_{i=1}^n{\left(x_i-\overline{x}\right)}^2}$$

(8)

where n is the number of counties adjacent to county, i, and the meaning of other variables is the same as Formula (7).

2.3.5. Gray Forecast GM (1, 1) Model

Gray forecast is used to predict the gray process which changes within a certain range and is related to time series. GM (1, 1) model is a widely used gray forecast model with only one variable and first-order differential. It is a method used to predict the future development trend of things by identifying the degree of difference between the development trend of system factors, generating and processing the original data, finding out the law of system change and then establishing the corresponding differential equation model. GM (1, 1) model can be used to predict data with small sample size and incomplete sequence, and it does not consider the distribution rule or change trend. It is appropriate for the short-term and medium-term prediction of exponential growth. The establishment of the model includes three steps: the first is data inspection and processing in order to calculate the level ratio of the sequence. It is necessary to perform translation transformation of the sequence to make the level ratio fall in the range if the range is not passed. The second is to establish a model and, through the solution of albinism equation, obtain the predicted value. Thirdly, it is necessary to carry on the model error test. The model accuracy is good when the posterior difference ratio c < 0.35 and the small error probability p > 0.95. Taking the carbon footprint and carbon carrying capacity from 2000–2020 as the original series, we used the GM (1, 1) model to predict the carbon footprint and carbon carrying capacity in 2025, and to obtain carbon ecological pressure index of each county in Shanxi Province in 2025 in order to analyze the future carbon ecological security level.

3. Results

3.1. Spatiotemporal Change Characteristics of Carbon Footprint

3.1.1. Analysis of Temporal Evolution of Carbon Footprint

At the five nodes selected in this study, the total carbon footprint of Shanxi Province exhibited an increased trend (Figure 2). The total carbon footprint increased from 160.22 Mt in 2000 to 434.21 Mt in 2020, with an increase of 171%. With an average annual growth rate of 14% from 2000 to 2010, the carbon footprint increased more rapidly compared to the increase that occurred from 2010 to 2020. This is explained by the fact that Shanxi Province, as a typical resource-based province, has formed a coal-dominated high-carbon energy consumption structure. There are many energy-intensive industries, and extensive energy consumption and carbon emissions are generated in the process of rapid economic development. During this period, the GDP and total energy consumption of Shanxi Province showed a similar trend to that of the carbon footprint. It can be seen that energy consumption benefits economic growth but also leads to high carbon emissions and pollution problems. The growth rate of the carbon footprint from 2010 to 2020 slowed down significantly, with an annual growth rate of 1.3%, which was related to economic transformation, industrial restructuring and clean energy use. For example, opinions on supporting Shanxi Province to further deepen reform, and promote the transformation and development of a resource-based economy, was issued in 2017, which proposed requirements for promoting industrial transformation and building a clean energy supply system. The scale of wind power generation and photovoltaic power generation continued to increase, reducing the consumption of coal resources in the power system, and promoting the reduction of carbon emissions.

The total carbon footprint was summarized and analyzed according to three categories: coal, oil and natural gas. It can be found that the carbon footprint of coal and natural gas both showed an increasing trend during the study period, while only oil’s carbon footprint decreased slightly from 2015 to 2020. Coal had the largest carbon footprint of the three types of energy, accounting for more than 90% of the total carbon footprint, but the proportion has been declining. The carbon footprint of oil ranked next, with a slight downward trend between 2015 and 2020. Natural gas had a smaller footprint, accounting for about 1% to 3%, but it began to grow obviously after 2010. This is mainly due to the 20-year natural gas supply contract signed by Shanxi Natural Gas Corporation and petroChina in 2004, and the improvement in the natural gas pipeline network construction in Shanxi.

At a county-level, the carbon footprint of all districts and counties increased in 2020 compared to 2000. During 2000–2005 and 2005–2010, the carbon footprint of all districts and counties increased, and the average annual growth rate of all districts and counties was 14.2%, which showed a rapid growth rate. The growth rate of the carbon footprint of all districts and counties slowed down significantly after 2010, with an average annual growth rate of 1.1% from 2010 to 2020. The carbon footprint of all districts and counties was maintained in a relatively stable range. Only a few counties had a growth rate of more than 10%, and the carbon footprint of some counties showed a downward trend (Figure 3). In terms of total carbon footprint, there were 3 counties with a total carbon footprint greater than 5 Mt in 2000, and 9, 20, 20 and 24 counties in 2005, 2010, 2015 and 2020, respectively. The number of counties with high carbon footprint increased, especially from 2000 to 2010.

3.1.2. Analysis of Spatial Evolution of Carbon Footprint

The carbon footprint of each county in Shanxi Province during the study period was divided into six levels using the natural breakpoint method (Figure 4). The carbon footprint presented as high in the middle region, and low in the east and west sides of the province. The areas with high carbon footprint were distributed in five regions: the Yungang District and the Pingcheng District of Datong City, the urban district, the mining district and suburbs of Yangquan City, Zezhou County and the Cheng District of Jincheng City, Hongdong County of Linfen City and Xiaoyi City, and basically did not change over time. Only small scale expansion occurred, on the basis of these five regions, in 2020. The high total carbon footprint was related to the rapid economic development, industrial agglomeration and high population density. For example, the Yungang District of Datong City always had the highest carbon footprint, which was related to its rich resource reserves and industrial layout dominated by coal production, leading to high carbon emissions. In addition, its population density was also high. The low total carbon footprint was related to slow economic development, small population density and backward industries. Daning County and Yonghe County of Linfen City always had small carbon footprints, and their GDP and population remained low within the province. To reduce the carbon footprint and carbon emissions, we should focus on these high-value regions and the regions with expansion in 2020.

3.2. Spatiotemporal Change Characteristics of Vegetation Carbon Carrying Capacity

3.2.1. Analysis of Temporal Evolution of Carbon Carrying Capacity

From 2000 to 2020, the carbon carrying capacity of Shanxi Province showed a slight upward trend in general (Figure 5), from 85.30 Mt in 2000 to 87.79 Mt in 2020, with an increase of 2.92%. The carbon carrying capacity decreased by 5.11% in 2005 that was mainly related to the expansion of construction land and the decrease in forest land. In other periods, the carbon carrying capacity showed varying degrees of growth, which was due to the implementation of the “Grain for Green” project, the treatment of coal mining subsidence and the development of other ecological restoration projects; carbon sequestration capacity has been improved.

According to the survey, cultivated land, forest land and grassland were the dominant land use types in Shanxi Province, and these areas were 3.8695 million ha, 6.0957 million ha and 3.1051 million ha, accounting for about 24.7%, 39% and 19.8%, respectively. Therefore, we mainly calculated the carbon sequestration capacity of forest land (including forest and shrub), grassland and crops as the total carbon carrying capacity. We can see that forest land carbon carrying capacity was the main contributor of carbon carrying capacity in Shanxi Province, accounting for more than 70%. The change trend of forest and carbon carrying capacity was consistent with the change trend of total carbon carrying capacity during our study period, with a slight increase in other years except for a decline of 11.7% during the period of 2000–2005. The crops carbon carrying capacity accounted for a relatively small proportion, but it has been constantly improving, from 0.30 Mt in 2000 to 2.78 Mt in 2020, with an average annual growth rate of 41%. It indicated that the productivity of cultivated land and the agricultural technology level have improved, and the output of crops has increased. The grassland carbon carrying capacity showed an increase and then a decrease, with a slight downward trend after 2010.

At the county-level, the carbon carrying capacity of most counties in 2020 had improved compared with that in 2000, and only 7 districts and counties had a slight decrease. The average degree of change from 2000 to 2020 was 14.2%. Among them, the carbon carrying capacity of Linyi County of Yuncheng City and Quwo County of Linfen City had a significant increase, while in the Xiaodian District of Taiyuan City it decreased greatly. The change trend of carbon carrying capacity in each district and county had a high consistency with that in Shanxi Province. The carbon carrying capacity of 83 districts and counties decreased from 2000 to 2005, accounting for 71% of the number of districts and counties. However, only four districts and counties had a decreasing trend from 2005 to 2010, and 35% of the districts and counties had a decreasing trend from 2010 to 2015 and 2015 to 2020.

3.2.2. Analysis of Spatial Evolution of Carbon Carrying Capacity

The carbon carrying capacity of counties in Shanxi Province during the study period was divided into six levels by the natural breakpoint method (Figure 6). The areas with low carbon carrying capacity were mainly distributed in the north and southwest. There were five low-value agglomeration areas of carbon carrying capacity: the Xinrong District; Pingcheng District–Yungang District–Huairen City; Jiancaoping District–Wanbolin District–Xiaodian District–Jinyuan District–Qingxu County; Lucheng District–Luzhou District–Shangdang District; and Xiangfen County–Quwo County–Yanhu District and other southwestern districts and counties. The mining district of Yangquan City has always been the county with the lowest carbon carrying capacity during the study period. Four aggregation areas, including Wutai County-Yu County, Jiaocheng County, Heshun County-Zuoquan County, and Qinyuan County-Anze County-Qinshui County, had high carbon carrying capacity. Both vegetation coverage and carbon sink were high in these areas. Qinyuan County and Qinshui County were the counties with the highest carbon carrying capacity during the study period. The high-value and low-value areas of carbon carrying capacity basically did not change significantly over time, and the low-value aggregation areas narrowed in 2010.

3.3. Analysis of Carbon Ecological Security

3.3.1. Carbon Balance Analysis

In order to investigate the carbon balance in Shanxi Province, we calculated the carbon deficit based on the carbon footprint and carbon carrying capacity. In Shanxi Province, the carbon deficit was greater than 0 from 2000 to 2020, meaning that the carbon footprint was greater than the carbon carrying capacity. This indicated that the carbon emissions could not be completely absorbed, and the region served as a carbon source on the whole, which is not conducive to curbing climate warming. There was a rising demand for energy with the rapid development of the economy, and the growth rate of the carbon footprint was higher than that of the carbon carrying capacity. The carbon deficit showed an increasing trend, with an annual growth rate of 18.1%, and the growth rate dropped slightly after 2010.

In terms of county-level carbon deficit (Figure 7), this increased gradually on the whole. The counties with large carbon deficits and large carbon footprints had higher overlap. The number of counties with carbon deficit greater than 0 exhibited an upward trend; that is, more counties served as carbon source on the whole. There were 76 counties with carbon deficit greater than 0 in 2000, by 2005, there were 93 counties and by 2020, there were 99 counties; however, the increase in the rate of the number of counties with carbon deficit greater than 0 slowed down in the later period. During the study period, there were three main types of carbon deficit changes in each county: firstly, some, including 76 counties such as Xiaodian District and Yingze District, have always had a carbon deficit greater than 0. A total of 53 of them showed an upward trend in their carbon deficit from 2000 to 2020, indicating the need to pay attention to carbon ecological security. Secondly, counties with a carbon deficit that has always been less than 0, including Pingshun County, Qinyuan County and other 17 counties, served as carbon sinks on the whole. Thirdly, the carbon deficit from less than 0 to greater than 0, including Yangqu County, Tianzhen County, etc., meant that the carbon footprint gradually increased beyond the carbon carrying capacity. And the counties gradually became a carbon source, on the whole, since the vegetation could not totally offset the carbon emissions.

3.3.2. The Carbon Ecological Pressure Index Analysis

From 2000 to 2020, the carbon ecological pressure index of Shanxi Province increased, from 1.87 in 2000 to 4.56 in 2010 and then 4.95 in 2020, while the rate of change lowered in the last decade. At the county-level, the carbon ecological pressure index of most counties (110 and 116, respectively) increased during 2000–2005 and 2005–2010. After 2010, the carbon ecological pressure index of some districts and counties showed a decreasing trend. The carbon ecological pressure index of 47 districts and counties increased from 2000 to 2020. In Shanxi Province, the carbon ecological security level of counties was divided into five levels according to the carbon ecological pressure index [29], as shown in

Table 1. Carbon ecological security classification.

carbon ecological pressure index	≤1	1–3.46	3.46–7.45	7.45–60	>60
carbon ecological security level	very safe	relatively safe	general	unsafe	very unsafe

.

From 2000 to 2005, the overall carbon ecological security of Shanxi Province was at a relatively safe level, and at a general level from 2010 to 2020. According to statistics on the number of districts and counties in each carbon ecological security level during the study periods (Figure 8), it can be found that each period occupied five carbon ecological security levels. The number of districts and counties in the very safe level was the highest in 2000, and then showed a downward trend, with a significant decline in 2005. The number of districts and counties in the very unsafe level was small, but rose in volatility, suggesting that the ecological security situation of some districts and counties needs to be monitored.

The carbon ecological security level in Shanxi Province presented a distribution pattern of high in the east and west and low in the middle region (Figure 9). From 2000 to 2010, the carbon ecological security level of each district and county changed significantly, while the carbon ecological level was stable and changed little during the period of 2010 to 2020. During the study period, Lingqiu County, Pingshun County and other 17 districts and counties were always at a very safe level, with low carbon ecological pressure and good environmental quality. However, the GDP of most of them ranked low in the whole province, so they can enhance their economic vitality on the basis of ecological protection in the future. The Xiaodian District, Pingcheng District and five other districts and counties were always in a very unsafe state, carbon ecological pressure index was high, which was the key area of focus. There were also areas such as Pinglu District and Yingxian County where the carbon ecological pressure index was at an unsafe threshold in 2020, and there is a need to remain aware of these areas turning from normal to unsafe.

We conducted spatial autocorrelation analysis on the carbon ecological pressure indexes with the help of ArcGIS10.2, and explored the spatial agglomeration characteristics of carbon ecological security. The global Moran’s I index in 2000, 2010, 2015 and 2020 was 0.064, 0.073, 0.065, 0.061 and 0.048, respectively; the Z-values were all less than 2.58 and failed to pass the 99% significance test. The only year that passed the 95% significance test was 2005. The results show that the spatial distribution of the carbon ecological pressure index in each county in Shanxi Province was random, that is to say, the carbon ecological security level in a county was basically not affected by the surrounding region. However, global uncorrelation does not mean local uncorrelation. We further conducted local autocorrelation analysis, and cold and hot spot analysis, taking 2005 as an example (Figure 10). The local autocorrelation analysis revealed that the carbon ecological security in Shanxi Province had a “High–High” type of aggregation in the areas of the Yungang District and Pingcheng District. The hot spot analysis showed that Shanxi Province had high carbon ecological pressure index value clustering in the areas of the Yungang District, Pingcheng District, Yunzhou District and Xinrong District.

The GM (1, 1) model was used to predict the carbon footprint and carbon carrying capacity of counties and districts in Shanxi Province in 2025, and data with unqualified prediction accuracy of the model were not included. Then, we calculated the carbon ecological pressure index and divided the carbon ecological security level (Figure 11). As observed, the carbon ecological security level of counties and districts will change little and basically remain unchanged by 2025, and the number of counties in carbon ecological insecurity level will remain quite large. Compared with 2020, the carbon ecological security level of Quwo County and Yushe County will improve. However, it is worth noting that the carbon ecological security level of Loufu County, Xinfu District, Wutai County and Tianzhen County will reduce, as these areas may be at risk of ecological security deterioration.

4. Discussion

4.1. Analysis of Carbon Footprint

The problem of global warming caused by carbon emissions has attracted increasing attention, and many countries have made commitments to reduce carbon emissions. Achieving peak carbon dioxide emissions and carbon neutrality is a solemn commitment to build a community of a shared future for mankind. It is also a strategic arrangement to promote economic and social transformation and realize a harmonious coexistence between man and nature. What is the level of regional carbon ecological security? This is the question that should be answered at the beginning of economic and social transition [29]. As a province with rich resources, the structure of coal-based energy consumption is difficult to change in the short term in Shanxi Province. Due to the large energy consumption and high carbon emissions in the process of economic development, causing environmental degradation, air pollution, and other ecological problems, the task of reducing emissions and increasing sinks is arduous.

The five aggregation areas with high carbon footprint are the focus of attention for emission reduction, such as the Yungang District of Datong City, which consistently had the highest carbon footprint values. A high economic output value, high industrial agglomeration, high population density and a strong circulation of people and goods are common characteristics of high carbon emission agglomeration areas [37]. It is found that population size is the dominant factor affecting carbon emissions in the counties of Shanxi Province, and carbon emissions can be effectively controlled by controlling population size and improving agglomeration [38,39]. In order to promote emission reduction, one must gradually eliminate dependence on coal resources and enhance the use of clean energy. It is necessary to promote the construction of green mines and smart mines, change the methods of coal mining and rely on technology to reduce carbon emissions in production. The regulations on the promotion of the efficient and clean utilization of coal in Shanxi Province came into effect in 2023, aiming to promote the efficient and clean utilization of coal throughout the process and ensure the realization of the goal of peak carbon dioxide emissions and carbon neutrality. Secondly, it is important to optimize the industrial structure. The industrial structure of Shanxi Province, which is dominated by coal, coking, metallurgy and electric power, exerted greater pressure on the ecological environment [40], and the high dependence on resources also led to a lack of risk resistance. The industrial structure dominated by coal is changing in recent years, and the share of the tertiary industry in GDP exceeded that of the secondary industry in GDP for the first time in 2015. It is necessary to promote the optimization of industrial structures, upgrade and transform traditional industries, extend the industrial chain and improve the efficiency of resource utilization. Furthermore, it is important to encourage the development of industries that transform coal into value-added industries, non-coal industries and high-tech industries, to accelerate the construction of strategic emerging industry clusters and to promote the rapid development of new industries. The added value of industrial strategic emerging industries increased at an average annual rate of 7.8% from 2016 to 2020, gradually becoming a key support for the transformation and development of Shanxi Province. Thirdly, it is necessary to cultivate residents’ awareness of low-carbon living, increase publicity and promote low-carbon green actions for all. In addition, Daning County and Yonghe County of Linfen City consistently had a small carbon footprint, and their GDP was low; it is important to enhance economic vitality on the basis of the protection of ecology.

4.2. Analysis of Carbon Carrying Capacity

The low-value areas of carbon carrying capacity in Shanxi Province are mainly distributed in the north and southwest, which are the key areas that need to increase their carbon sink. Forest land is the main contributor to the carbon carrying capacity of Shanxi Province, and it is important to improve forest coverage and increase forest carbon sink. On the one hand, afforestation work is carried out to increase the area of forest land and improve the forest stock. Forest types such as broadleaved mixed forest and larch forest have high carbon stocks [41], which can increase the proportion of tree species, such as broadleaved mixed forest, to structurally enhance the carbon sink function. On the other hand, we should pay attention to strengthening the management and monitoring of forest land resources, and, in time, take engineering measures and biological measures to restore the vegetation according to the actual situation of damaged forest land. And wetland is also a very important carbon reservoir, and it has a continuous carbon sequestration capacity, which plays an important role in mitigating climate change [42]. Wetland reclamation, overgrazing and the construction of large-scale water conservancy projects may lead to wetland degradation and reduce the carbon sequestration capacity of the wetland. Wetland in Shanxi Province covers 54,400 ha, mainly distributed in Xinzhou City and Yuncheng City. The enterprise of wastewater is the main cause of wetland degradation in Shanxi Province, and strengthening the environmental protection awareness of enterprises and enhancing the supervision of enterprise discharge are the primary measures necessary [43]. Wetland protection and restoration should receive attention. We must restore the hydrology and vegetation of the original wetland through engineering means, biological means and hydrological restoration, etc., and gradually restore its carbon sequestration capacity and other ecosystem service functions.

4.3. Analysis of Carbon Ecological Security

Although the carbon deficit of Shanxi Province showed an increasing trend, the increase rate slowed down after 2010. According to the carbon ecological pressure index, Shanxi Province was at a relatively safe and general safe level. For example, in 2020, the districts and counties with higher carbon ecological pressure index, such as the Xiaodian District, Jianchao District and Luzhou District, all had high regional GDP, high population density and high energy consumption in the process of economic development, resulting in a high carbon footprint. And Xiaodian District was the district with the highest GDP in Shanxi Province in 2020, and the only one with more than 100 billion. Most of these districts and counties are the main urban areas with large construction land areas, or large arable land areas, with less ecological land area, such as forest land and grassland, and vegetation carbon sequestration is low.

We need to focus on the 53 counties with increasing carbon deficit, and the 5 counties always at the very unsafe level, such as the Xiaodian District and Pingcheng District. For areas such as the Pinglu District and Ying County, where the carbon ecological pressure index is at the critical value of insecurity in 2020, as well as areas where the ecological security level may decline in 2025 according to the forecasted results, these should be controlled to prevent them from turning to unsafe levels.

4.4. Study Shortcomings and Limitations

Currently, most studies on carbon footprint, carbon emission or carbon carrying capacity are concentrated on larger scales such as national and provincial. In this paper, we studied the carbon footprint, carbon carrying capacity and carbon ecological security level of counties in Shanxi Province, which enriches the case studies on county-level carbon ecological security and provides a reference for Shanxi Province to achieve emission reduction and ensure ecological security.

Different scholars take different values regarding the NEP of various vegetation, which leads to a difference in carbon carrying capacity calculation results. Wetland is one of the largest carbon pools in the world and has a continuous carbon sequestration capacity. Our study mainly considered the carbon carrying capacity of vegetation such as forest land, grassland and crop, and did not calculate the carbon sequestration of the wetland. The energy consumption data of each county are missing, and we chose to use the 1997–2017 county-level carbon emission data to carry out the study, and the data for 2020 are made up by the moving smoothing method, which may have a small error.

Shanxi Province is the main coal deposit in China, including three major coal bases, Jinbei Coal Base, Jinzhong Coal Base and Jinbei Coal Base, which are used partly for exporting, in addition to meeting the province requirements and playing an important role in the national coal supply. Shanxi Province exported coal to 28 provinces and autonomous regions, and exported it to Asia, Europe and other countries. Shanxi Province produced 1.079 billion tons of raw coal and transferred 621 million tons of coal out of the province in 2020. In addition to coal, the outward transfer of other energy sources is also growing rapidly. This part of the carbon emissions in the process of energy production is accounted for in Shanxi’s carbon emissions, but not actually consumed by the province. For example, thermal power generation is one of the main sources of carbon emissions in Shanxi, but a large amount of electricity is exported, not all of which is consumed by Shanxi Province, and the calculation and allocation scheme of this part of carbon emissions should be strengthened in the future. The process of coal mining is accompanied by a large amount of fossil energy consumption and the escape of carbon dioxide and methane; the carbon emission caused by this escape was not considered.

We chose the carbon ecological pressure index to characterize the carbon ecological security level, which is also influenced by other factors. The criteria for the classification of carbon ecological safe level has also not been standardized, but this does not affect the comparison of carbon ecological security level in the study area. In future studies, we can build a richer carbon ecological security evaluation system to reflect the carbon ecological security level more comprehensively.

5. Conclusions

In this study, we explored the spatiotemporal change in carbon footprint, carbon carrying capacity and carbon ecological security in the counties and districts of Shanxi Province, and predicted the carbon ecological security level in 2025. The results indicated that:

(1)

The carbon footprint of Shanxi Province showed the characteristics of high in the middle region and low in the east and west sides of province. Coal had the largest carbon footprint of the three types of energy, accounting for more than 90% of the total carbon footprint. The total carbon footprint from 2000 to 2020 showed an increasing trend, and the number of high carbon emission counties increased.

(2)

From 2000 to 2020, the carbon carrying capacity of Shanxi Province showed a slight upward trend in general, and forest land carbon carrying capacity was the main contributor to carbon carrying capacity. The carbon carrying capacity of most counties in 2020 had improved compared with that in 2000. The high and low carbon carrying capacity areas did not change significantly over time.

(3)

The carbon deficit in Shanxi Province was greater than 0 from 2000 to 2020, meaning that the region served as a carbon source on the whole. Counties with large carbon deficit and large carbon footprint had higher overlap. The carbon ecological pressure index of Shanxi Province increased and the province carbon ecological security decreased from a relatively safe level to a general safe level during the study period. The spatial distribution of the carbon ecological pressure index in each county was random. The carbon ecological security level showed a distribution pattern of high in the east and west and low in the middle region.

(4)

According to the prediction, the carbon ecological security level of counties and districts will show little change by 2025, and will basically remain unchanged compared with 2020, while there will still be districts and counties at carbon ecological insecurity levels.