Spatial–Temporal Evolution and Influencing Factors of Land-Use Carbon Emissions: A Case Study of Jiangxi Province

Abstract

Land-use carbon emissions denote the release or sequestration of greenhouse gases (e.g., CO₂, N₂O) resulting from human land-use activities, with land-use changes exerting a major influence on land-use carbon emissions. Revealing the coupling mechanism between land-use changes and carbon emissions is of crucial theoretical significance for achieving “dual carbon” goals and mitigating global climate change. Based on the land-use change data of Jiangxi Province, this study explored the Spatial–temporal relationship between land-use carbon emissions and land-use changes in Jiangxi Province from 2000 to 2020 using a model of land-use dynamic degrees, a model of land-use transfer matrices, and the IPCC carbon emission accounting model. In this study, the factors influencing changes in land-use carbon emissions were comprehensively analyzed using an LMDI model and the Tapio decoupling model. The results indicated that: (1) Jiangxi Province’s land-use changes show a “two-increase, four-decrease” trend, with construction land and unused land experiencing the most significant shifts, while water, grassland, cropland, and forestland changes stayed near 1%. (2) Net land-use carbon emissions exhibit a rapid then gradual increase, with higher emissions in the north/south and lower levels in central regions. While overall land-use carbon emission intensity is declining, per capita emissions continue to rise. (3) Land-use carbon emission changes are primarily driven by emission intensity, land-use structure, efficiency, and economic level. In Jiangxi, economic growth mainly increases land-use carbon emissions, while land-use efficiency enhancement counters this trend. Jiangxi Province shows weak land-use carbon emission–economic growth decoupling, with land-use carbon emissions rising more slowly than economic growth. This study not only provides a typical case analysis and methodological framework for understanding the carbon emission effects of human–land relationships in rapidly urbanizing regions but also offers a specific scientific basis and policy insights for Jiangxi Province and other similar regions to formulate differentiated territorial spatial planning, promote ecological protection and restoration, and achieve green and low-carbon development pathways under the “dual carbon” goals.

1. Introduction

With the rise in carbon emissions (CEs), global climate change has already begun to pose a threat to human society [1]. Therefore, enhancing urban resilience and effectively reducing and controlling CEs has become an important goal of environmental policies worldwide [2,3]. Meanwhile, LUC is a crucial driving factor in global climate change [4,5]. In 2020, the Chinese government proposed the “dual carbon” goals of CEs reaching their peak and achieving carbon neutrality [6]. Achieving these ambitious goals requires a deep understanding of carbon emission dynamics at regional scales, where land-use decisions have immediate and profound impacts. While enhancing research on the patterns and mechanisms of land-use carbon emissions (LUCEs) can provide a basis for developing differentiated emission reduction policies, significant gaps remain. Existing studies on LUCEs, as reviewed, have extensively covered regional carbon cycles, terrestrial carbon sources/sinks, and the spatiotemporal evolution of emissions. However, many of these studies exhibit limitations: (1) They often focus on single aspects (e.g., either accounting or factor analysis) without integrating spatial evolution with a comprehensive driver decomposition, leading to an incomplete picture of the emission mechanism. (2) Research findings from national or developed regional scales may not be directly applicable to specific, rapidly urbanizing but economically developing regions like Jiangxi Province, due to distinct socioeconomic contexts and land-use transition patterns. (3) Furthermore, the dynamic relationship between economic growth and LUCEs in such contexts is not fully explored, limiting the formulation of targeted decoupling policies. To address these gaps, this study selects Jiangxi Province, a typical developing region in the Yangtze River Basin, as a case study. To construct a comprehensive analytical framework, we employ a suite of methodologies that includes land-use dynamic analysis, LUCE accounting, LMDI decomposition, and the Tapio decoupling (TD) model. This framework aims not only to quantify the spatiotemporal evolution of LUCEs but also to disentangle the complex contributions of various driving factors (e.g., LUCE intensity, land-use structure (LUS), economic level) and explicitly evaluate the decoupling state between economic growth and CEs. Enhancing research on the patterns and mechanisms of LUCEs can provide a basis for developing differentiated emission reduction policies and measures for different regions. As an important province in the Yangtze River Basin, significant efforts are being taken in Jiangxi Province to lower CEs. However, rapid changes in land-use methods, efficiency, and spatial structure have complicated regional research on LUCEs in this province. Therefore, exploring the characteristics of the Spatial–temporal evolution of LUCEs and their influencing factors in Jiangxi Province is highly necessary; indeed, it will prove crucial for high-quality national territorial spatial planning, formulating precise emission reduction policies, and achieving green, low-carbon sustainable development.

Domestic and international research on LUCEs has mainly focused on regional carbon cycling processes [7,8], carbon sources and sinks in terrestrial ecosystems [9,10,11,12], the relationship between land-use changes (LUCs) and carbon cycling [13,14,15], the Spatial–temporal evolution of LUCEs and LUCE efficiency [16,17,18], models for calculating LUCEs [19,20], analyses of carbon ecological footprints [21,22], the factors influencing LUCEs [23,24,25], and studies on the intensity of LUCEs and ecological compensation for emissions [25,26]. Equally, research on LUCEs has mostly been conducted at national, provincial, municipal, and watershed scales [23,27,28,29,30]. Its results indicate there are distinct differences in LUCEs between different regions. Widening gaps between the carbon sources and sinks within regions exist, with CEs from construction land constituting a major source of LUCEs and the carbon absorption capacity of forest land, grasslands, water areas, and unused land continuously decreasing at the same time. Thus, altering the patterns of LUCEs and promoting CE reduction efforts to reach lower levels in key areas have become crucial. Analyzing the factors that affect LUCEs is key to reducing excessive CEs and advancing carbon peak targets, and is a prerequisite for identifying the major contributing factors and implementing corresponding control measures [31,32]. To this end, methods such as factor decomposition [33,34], correlation analysis [35,36], gray relational analysis [37,38], and regression analysis [39,40] have often been employed, among others, including in studying multiple regions of China. Using the Logarithmic Mean Divisia Index (LMDI) [41] is a particularly effective method. In considering the spatial effects between units, the target can be decomposed without leaving unexplained residuals, and relatively simple transformation expressions can be used for both additive and multiplicative decomposition, enhancing the interpretability of the results for studies on the factors influencing LUCEs [42].

The levels of economic development in different regions of Jiangxi Province vary, with significant spatial and temporal differences in their LUCEs. Research on urban LUCEs can provide the basis for optimizing LUS, improving land-use efficiency (LUE), and reducing LUCEs in this region. On this premise, in this study, we analyze the changing patterns of land use in Jiangxi Province from 2000 to 2020 using data on LUC, models on the land-use dynamic degrees (LUDDs), and land-use transfer (LUT) matrices. Furthermore, by integrating the CE factors from the IPCC inventory methodology and its energy consumption and CE model with data on social, economic, and energy consumption variables, the net LUCEs, the intensity of LUCEs, and the intensity of LUCEs per capita in Jiangxi are calculated. The Spatial–temporal evolution of LUCEs is then analyzed further. Finally, an LMDI model and a TD model are applied to calculate and examine the contributions and degrees of the factors influencing LUCEs, elucidating the inherent driving relationships between these factors and net LUCEs, and between changes in LUCEs and economic growth. These findings will provide guidance for the “dual carbon” goals in Jiangxi Province, as well as offering crucial insights for low-carbon development in economically underdeveloped regions more generally.

2. The Study Area and Data Sources

2.1. Overview of the Study Area

Jiangxi Province, as a key region within the Yangtze River Economic Belt, has undergone rapid urbanization and industrialization, leading to significant changes in LUS and CEs. Its diverse ecosystems play a crucial role in carbon sinks. Therefore, selecting Jiangxi as a case study is representative of developing regions in central China and provides valuable insights for achieving national “dual carbon” goals.

There are 11 cities, namely Nanchang, Jingdezhen, Pingxiang, Jiujiang, Xinyu, Yingtan, Ganzhou, Ji’an, Yichun, Fuzhou, and Shangrao, across the entirety of Jiangxi Province. Located in the southeast of China, it spans a total area of 166,900 km² and is characterized mainly by the hills and mountains of Jiangnan in terms of its terrain and landforms. Within a period of rapid urbanization and industrialization, the land-use patterns in Jiangxi Province are changing rapidly, with close economic ties between different regions.

By the end of 2023, Jiangxi Province had a permanent population of 45.15 million and an urbanization rate of 63.13% among this permanent population. Its GDP reached CNY 3.22 trillion, while its per capita regional GDP was CNY 71,216. This study’s scope and geographic location are shown in Figure 1.

2.2. The Data Source

Data on the administrative boundaries of Jiangxi Province were sourced from the Standard Map Service System (https://www.tianditu.gov.cn/ (accessed on 31 January 2024)) under approval number GS(2019)1822. Land-use raster data on Jiangxi Province were derived from the land cover classification dataset developed by the Data Center of the Institute of Resources and Environment of the Chinese Academy of Sciences (http://www.resdc.cn (accessed on 31 January 2024)), at a resolution of 30 m. Although this dataset undergoes rigorous quality control and maintains high overall accuracy, minor errors may occur in small-scale LUC detection. Under the land-use classification system, land is reclassified as construction land, cultivated land, forest land, grassland, water areas, or unused land using ArcGIS 10.8 software. The Digital Elevation Model (DEM) data were sourced from the Geospatial Data Cloud (http://www.gscloud.cn (accessed on 31 January 2024)) at a 30 m resolution, suitable for terrain analysis, though slight elevation deviations may exist in flat areas. Data on population, energy consumption, economic factors, and so on were all sourced from the corresponding annual statistical yearbooks for Jiangxi Province and each city in it, and the “National Economic and Social Development Statistical Bulletin”. While these sources are authoritative, inconsistencies in statistical criteria or missing values in certain years were addressed through interpolation and cross-validation. Despite these potential limitations, the overall data reliability supports the empirical analysis conducted in this study.

3. Methods

In this study, LUCEs in Jiangxi Province are taken as the research object. The research content and influencing factors of LUCEs are qualitatively analyzed, followed by quantitative analysis. Based on the results of the analysis, effective policy recommendations are proposed. The research framework is shown in Figure 2.

3.1. The LUDD Model

This study employs a series of complementary models to systematically analyze the spatiotemporal evolution and influencing factors of LUCEs in Jiangxi Province. The LUDD model is used to quantify the speed and extent of LUCs, providing a more intuitive reflection of trends than single indicators. The LUT matrix clearly displays the transition paths between different land types, revealing structural changes. The LUCE accounting model is based on IPCC standards, ensuring comparability and scientific rigor. The LMDI model decomposes influencing factors without residuals, yielding accurate and reliable results that effectively identify the contribution of each driver. The TD model is a widely used tool for assessing the relationship between economic growth and environmental pressure, helping to determine the decoupling status between CEs and economic development. Together, these models form a comprehensive analytical chain from description and accounting to attribution and evaluation.

LUDD refers to the changes in different land use types within a specific time frame. It can visually reflect the changes in the area and rate of a specific type of land-use during a given research period. It is calculated as follows [43]:

$$K={\displaystyle \frac{A_1-A_0}{A_0}}\times {\displaystyle \frac{1}{T}}\times 100$$

(1)

where K represents the LUDDs; $A_0$ and $A_1$ represent the land-use areas at the beginning and end of the research period; and T is the duration of the research period. A higher value for K indicates more drastic LUCs, while a lower value indicates a more stable situation.

The LUDD model provides an overall trend of LUCs, laying the foundation for subsequent LUT matrix and LUCE accounting.

3.2. A LUT Matrix

A LUT matrix is a tool used to describe and analyze changes and transitions between different land-use types. When studying changes in land use, land is typically divided into various use types. A LUT matrix is a commonly used tool in studies on LUCs that can help decision-makers and researchers understand the LUC process. It records the changes and transitions between various land-use types in different time periods, which are usually presented in the form of a row–column matrix. The calculation involved is expressed as follows [44]:

$$B_{xy}=\left[\begin{array}{cccc}{B}_{11}& B_{12}& \dots & B_{1n}\\ B_{21}& B_{22}& \dots & B_{2n}\\ ⋮& ⋮& \ddots & ⋮\\ B_{n1}& B_{n2}& \dots & B_{nn}\end{array}\right]$$

(2)

where $B_{xy}$ represents the area of land type x transitioning to land type y, and n represents the number of land-use types.

The LUT matrix complements the LUDD model by revealing specific transition paths between land-use types, providing a basis for understanding the spatial sources of CEs.

3.3. The LUCE Accounting Model

LUCEs are divided into direct and indirect LUCEs [45]. Direct LUCEs refer to CEs caused by land use and are applicable to different land types [31,32], while indirect LUCEs refer to the CEs from production and life on various types of land represented by the amount of CO₂ produced by energy consumption and are applicable to accounting for construction land. In considering the findings of existing research [29,30,31,32] and the specific situation in Jiangxi Province, the coefficients for its direct LUCEs are confirmed, as shown in

Table 1. LUCE coefficient per unit area for various land-use types.

Land-Use Types	LUCE Coefficients (Unit: t/(hm2·a))
Cultivated land	0.4595
Forest land	−0.6125
Grassland	−0.0205
Water area	−0.2570
Unused land	−0.0005

. The direct LUCEs can then be obtained.

$$E_d=\sum e_i=\sum A_i\times {\epsilon }_i$$

(3)

where $E_d$ represents the total direct LUCEs, and $e_i$, $A_i$, and ${\epsilon }_i$ represent the LUCEs, area, and LUCE coefficient for land-use type i, respectively.

Fossil fuel consumption is one of the main sources of CEs in China [46]. CEs are mainly related to energy production and consumption. Indirect estimation methods can be used to calculate them. This study calculates LUCEs for eight types of energy sources [47], and the corresponding energy conversion coefficients are shown in

Table 2. Primary energy CE conversion coefficients.

Energy Sources	Standard Coal Conversion Coefficient (t Standard Coal/t)	CE Coefficient (t Carbon/t Standard Coal)
Raw coal	0.7143	0.7559
Coke	0.9714	0.855
Crude oil	1.4286	0.6185
Gasoline	1.4714	0.5538
Kerosene	1.4714	0.5714
Diesel	1.4571	0.4483
Fuel oil	1.4286	0.6185
Liquefied petroleum gas	1.7143	0.5042

Note: CE coefficients based on IPCC 2006.

. The formula for calculating indirect LUCEs is as follows:

$$E_c=\sum e_{ci}=\sum E_{ni}\times {\delta }_i\times {\xi }_i$$

(4)

where $E_c$ represents the LUCEs from construction land; $e_{ci}$ represents the CEs for energy type i; $E_{ni}$ represents the consumption of energy type i in daily life; ${\delta }_i$ is the conversion coefficient for energy type i to standard coal; and ${\xi }_i$ represents the CE coefficient for energy type i.

In summary, the formula for calculating the total LUCEs is as follows:

$$E=E_d+E_c$$

(5)

where E represents the total LUCEs.

CE intensity is defined as the amount of CEs per unit of GDP output, which can reflect the level of low-carbon economic development. It is calculated using the following formula:

$$IC={\displaystyle \frac{C_i}{GD{P}_i}},i=1,2,3\dots \dots ,n$$

(6)

where $IC$ represents the CE intensity of energy consumption; $C_i$ represents the net CEs; $GD{P}_i$ represents the GDP of the research area; and i represents the year in question.

Per capita LUCEs are defined as the total LUCEs divided by the total population, and they are calculated using the following formula:

$$PC={\displaystyle \frac{C_i}{P_i}},i=1,2,3,\dots \dots ,n$$

(7)

where $PC$ represents the LUCEs per capita; $C_i$ represents the net LUCEs; and $P_i$ represents the resident population.

The LUCE accounting results provide direct CE data inputs for the LMDI decomposition and TD analysis.

3.4. The LMDI Model

There are no unexplained residuals in the decomposition results from the LMDI model, allowing relatively simple transformation expressions for additive and multiplicative decomposition to be obtained. Therefore, the driving factors behind LUCEs are commonly modeled this way. Referring to the Kaya identity [48] and starting with five aspects—unit LUCE intensity, LUS, LUE, economic level, and population scale—an LMDI model of the factors affecting LUCEs in Jiangxi Province can be established. The LMDI decomposition process includes additive and multiplicative models. The calculation in the additive model uses the following formula:

$$C=\sum {\displaystyle \frac{C_i}{L_i}}\times {\displaystyle \frac{L_i}{L}}\times {\displaystyle \frac{L}{G}}\times {\displaystyle \frac{G}{P}}\times P$$

(8)

where C represents the total LUCEs; $C_i$ represents the different types of LUCEs; $L_i$ represents the areas of different land-use types; L represents the total land area of the region; G represents the GDP; and P represents the permanent resident population. Let

$$f_i={\displaystyle \frac{G_i}{L_i}};s_i={\displaystyle \frac{L_i}{L}};l={\displaystyle \frac{L}{G}};g={\displaystyle \frac{G}{P}}$$

(9)

The total regional LUCEs can be expressed as

$$C=\sum f_i\times s_i\times l\times g\times p$$

(10)

where $f_i$ represents the intensity of LUCEs; $s_i$ represents the LUS; l represents the LUE (land area per unit GDP); g represents the economic level (per capita GDP); and p represents the population scale.

The LMDI model can be used to decompose the contributions of the factors influencing LUCEs, where $C_0$ is defined as the LUCEs at the beginning of the study period, and $C_t$ is defined as the LUCEs at the end of the t-th period; then, the change $\mathrm{\Delta }C$ in LUCEs can be expressed as

$$\mathrm{\Delta }C=C_t-C_0=\mathrm{\Delta }{C}_{fi}+\mathrm{\Delta }{C}_{si}+\mathrm{\Delta }{C}_l+\mathrm{\Delta }{C}_g+\mathrm{\Delta }{C}_p$$

(11)

where

$$\mathrm{\Delta }{C}_{fi}=\sum W\times ln{\displaystyle \frac{f_i^t}{f_i^0}}$$

(12)

$$\mathrm{\Delta }{C}_{si}=\sum W\times ln{\displaystyle \frac{s_i^t}{s_i^0}}$$

(13)

$$\mathrm{\Delta }{C}_l=\sum W\times ln{\displaystyle \frac{l^t}{l^0}}$$

(14)

$$\mathrm{\Delta }{C}_g=\sum W\times ln{\displaystyle \frac{g^t}{g^0}}$$

(15)

$$\mathrm{\Delta }{C}_p=\sum W\times ln{\displaystyle \frac{p^t}{p^0}}$$

(16)

where $\mathrm{\Delta }{C}_{fi}$, $\mathrm{\Delta }{C}_{si}$, $\mathrm{\Delta }{C}_l$, $\mathrm{\Delta }{C}_g$, and $\mathrm{\Delta }{C}_p$ represent the values for the contributions of the indicators $f_i$, $s_i$, l, g, and p, respectively, and the expression W is given by $W={\displaystyle \frac{C_t-C_0}{ln{C}_t-ln{C}_0}}$.

The formula for the multiplication model calculation is

$$D_{f_i}=exp\left(\sum {\displaystyle \frac{\left(C_i^t-C_i^0\right)/\left(ln{C}_i^t-ln{C}_i^0\right)}{\left(C^t-C^0\right)/\left(ln{C}^t-ln{C}^0\right)}}\times ln\left({\displaystyle \frac{f_i^t}{f_i^0}}\right)\right)$$

(17)

$$D_{{\mathrm{s}}_i}=exp\left(\sum {\displaystyle \frac{\left(C_i^t-C_i^0\right)/\left(ln{C}_i^t-ln{C}_i^0\right)}{\left(C^t-C^0\right)/\left(ln{C}^t-ln{C}^0\right)}}\times ln\left({\displaystyle \frac{{\mathrm{s}}_i^t}{{\mathrm{s}}_i^0}}\right)\right)$$

(18)

$$D_1=exp\left(\sum {\displaystyle \frac{\left(C_i^t-C_i^0\right)/\left(ln{C}_i^t-ln{C}_i^0\right)}{\left(C^t-C^0\right)/\left(ln{C}^t-ln{C}^0\right)}}\times ln\left({\displaystyle \frac{l^t}{l^0}}\right)\right)$$

(19)

$$D_{\mathrm{g}}=exp\left(\sum {\displaystyle \frac{\left(C_i^t-C_i^0\right)/\left(ln{C}_i^t-ln{C}_i^0\right)}{\left(C^t-C^0\right)/\left(ln{C}^t-ln{C}^0\right)}}\times ln\left({\displaystyle \frac{d^t}{d^0}}\right)\right)$$

(20)

$$D_P=exp\left(\sum {\displaystyle \frac{\left(C_i^t-C_i^0\right)/\left(ln{C}_i^t-ln{C}_i^0\right)}{\left(C^t-C^0\right)/\left(ln{C}^t-ln{C}^0\right)}}\times ln\left({\displaystyle \frac{P^t}{P^0}}\right)\right)$$

(21)

$$D={\displaystyle \frac{C_t}{C_0}}=D_{f_i}\times D_{s_i}\times D_l\times D_g\times D_p$$

(22)

where $D_{fi}$, $D_{si}$, $D_l$, $D_g$, and $D_p$ represent the contribution rates of the indicators $f_i$, $s_i$, l, g, and p, respectively, and D represents the total contribution rate.

The LMDI model decomposes the contributions of various factors based on LUCE data and, together with the TD model, reveals the driving mechanisms and decoupling status of CEs.

3.5. The TD Model

The decoupling index is used to represent the relationship between the economic growth rate and the growth rate of resource consumption. When the economic growth rate is higher than the growth rate of resource consumption, this is called “decoupling”, which means that economic growth no longer depends on resource consumption. The decoupling index can be used to evaluate the sustainability of economic development patterns and guide the formulation of policy on environmental protection and resource utilization [49]. This study utilizes the TD model to explore the interrelationship between economic growth and LUCEs in Jiangxi Province, and the calculation involved is expressed as follows:

$$T_i={\displaystyle \frac{M_c}{M_g}}={\displaystyle \frac{\mathrm{\Delta }C/C}{\mathrm{\Delta }G/G}}$$

(23)

where $T_i$ represents the decoupling index for the i-th year in Jiangxi Province; $\mathrm{\Delta }C$ represents the net growth in LUCEs; C represents the initial net LUCEs at the beginning of the study period; $\mathrm{\Delta }G$ represents the increase in GDP; and G represents the initial total GDP at the beginning of the study period.

The decoupling relationship between LUCEs and economic growth can be classified into eight types, as shown in

Table 3. Decoupling types of LUCEs and economic growth.

Decoupling Types		${\mathit{M}}_{\mathit{c}}$	${\mathit{M}}_{\mathit{g}}$	${\mathit{T}}_{\mathit{i}}$
Decoupling	Strong decoupling	<0	>0	T < 0
	Weak decoupling	>0	>0	0 < T < 0.8
	Regressive decoupling	<0	<0	T > 1.2
Linkage	Expansionary linkage	>0	>0	0.8 < T < 1.2
	Regressive linkage	<0	<0	0.8 < T < 1.2
Negative decoupling	Strong negative decoupling	>0	<0	T < 0
	Weak negative decoupling	<0	<0	0 < T < 0.8
	Expansionary negative decoupling	>0	>0	T > 1.2

and

Table 4. Decoupling relationship between LUCEs and economic growth.

Decoupling Types	Decoupling Relationship
Strong decoupling	Optimal: Economic growth, LUCEs significantly reduced
Weak decoupling	Better: Economic growth, LUCEs increase, but the increase is smaller than the economic growth rate
Expansionary linkage	Economic growth, LUCEs increase, with both increasing at a similar rate
Expansionary negative decoupling	Economic growth, LUCEs increase, with the increase greater than the economic growth rate
Regressive decoupling	Economic decline, LUCEs decrease, with the decrease greater than the economic decline
Regressive linkage	Economic decline, LUCEs decrease, with both decreasing at a similar rate
Weak negative decoupling	Worse: Economic decline, LUCEs decrease, but the decrease is smaller than the economic decline
Strong negative decoupling	Worst: Economic decline, LUCEs increase

. Strong decoupling is the optimal solution, indicating that the relationship between economic growth and LUCEs is weaker, and economic growth can be achieved without LUCEs increasing. Conversely, strong negative decoupling represents the worst situation, indicating that economic growth is in decline while LUCEs continue to increase. Neutral decoupling, weak decoupling, and other types reflect varying relationships between economic growth and LUCEs. In the process of pursuing low-carbon economic development, strong decoupling signifies a good level of decoupling between economic growth and LUCEs, which is desirable and holds significant importance for promoting sustainable development.

The TD model uses LUCE and economic growth data to assess the decoupling relationship, corroborating the LMDI results and forming a complete logical chain from description to policy evaluation.

The aforementioned models collectively serve the objectives of this study: to reveal the spatiotemporal evolution of LUCEs, identify key driving factors, and assess their decoupling relationship with economic growth. Through multi-model collaborative analysis, this study provides a scientific basis for Jiangxi Province to formulate differentiated emission reduction strategies, optimize territorial spatial planning, and support the achievement of the “dual carbon” goals.

3.6. Method Limitations and Assumptions

• Data uncertainties: Energy consumption data from statistical yearbooks may have reporting biases or inconsistencies in statistical caliber.
• Model assumptions: The LUCE coefficients are based on existing literature and may not fully reflect local characteristics of Jiangxi Province. The LMDI model assumes that influencing factors are independent, while interactions may exist in reality.
• Unconsidered factors: Factors such as the impact of climate change on carbon sink capacity and the time-lag effects of policy interventions were not incorporated into the models, which may affect the comprehensiveness of the results.

Despite these limitations, this study strives to ensure scientific rigor and comparability by adopting internationally recognized methods and publicly available data, and provides directions for future research improvements.

4. Results and Analysis

4.1. Analysis of LUC in Jiangxi Province

4.1.1. Analysis of LUC in Jiangxi Province

From 2000 to 2020, land-use data for Jiangxi Province were imported into ArcGIS 10.8 software. According to the “Classification of Land-Use Status” (GB/T 21010-2017 [50]) and the land-use classification standards of the Chinese Academy of Sciences, the land-use data for Jiangxi Province in 2000, 2005, 2010, 2015, and 2020 were obtained, as shown in Figure 3. By processing the land-use data, the areas and area ratios of various land-use types in Jiangxi Province from 2000 to 2020 were calculated, as shown in

Table 5. Land-use area and ratio in Jiangxi Province from 2000 to 2020.

Year	Unit	Cultivated Land	Forest Land	Grassland	Water Area	Construction Land	Unused Land	Total
2000	Area (km2)	45,298.14	103,795.86	7282.28	6812.27	2821.38	920.79	166,930.72
	Ratio (%)	27.14	62.18	4.36	4.08	1.69	0.55	100.00
2005	Area (km2)	45,125.32	103,712.11	7116.21	7018.05	3335.44	623.59	166,930.72
	Ratio (%)	27.03	62.13	4.26	4.20	2.00	0.37	100.00
2010	Area (km2)	44,990.95	103,514.48	6800.51	7117.58	3961.08	546.13	166,930.72
	Ratio (%)	26.95	62.01	4.07	4.26	2.37	0.33	100.00
2015	Area (km2)	44,630.69	103,287.93	6760.13	7126.66	4596.16	529.15	166,930.72
	Ratio (%)	26.74	61.87	4.05	4.27	2.75	0.32	100.00
2020	Area (km2)	44,195.84	102,494.12	7125.17	7165.41	5409.61	540.57	166,930.72
	Ratio (%)	26.48	61.40	4.27	4.29	3.24	0.32	100.00

. These data systematically reflect the land-use status and characteristics of changes in Jiangxi Province from 2000 to 2020.

The results from Figure 3 and Table 5 indicate that, in Jiangxi Province, land use is dominated by forest land and cultivated land. Forest land is mainly distributed in the southern, western, and northeastern regions, while cultivated land is primarily distributed in the northern and central regions, with average area proportions of 61.91% and 26.87%, respectively, during the study period. Water areas are mainly distributed in the Poyang Lake region, with an average area proportion of 4.22% during the study period.

In terms of LUCs, between 2000 and 2020, the areas of construction land and water areas in Jiangxi Province grew, while the total area of other land types diminished. The area of construction land increased by 2588.23 km², while the changes in the area of forest land, grassland, and water areas were all less than 5%. The combined area of cultivated land and unused land decreased by 1482.52 km², with most of this reduction corresponding to cultivated and unused land being converted into construction land.

4.1.2. Analysis of LUDDs

The LUDD model was applied to estimate the type-specific LUDDs and overall LUDDs for four time periods in Jiangxi Province—2000–2005, 2005–2010, 2010–2015, and 2015–2020—as shown in

Table 6. LUDD results (%) in Jiangxi Province from 2000 to 2020.

Time Period	Cultivated Land	Forest Land	Grassland	Water Area	Construction Land	Unused Land
2000–2005	−0.08	−0.02	−0.46	0.60	3.64	−6.46
2005–2010	−0.06	−0.04	−0.89	0.28	3.75	−2.48
2010–2015	−0.16	−0.04	−0.12	0.03	3.21	−0.62
2015–2020	−0.19	−0.15	1.08	0.11	3.54	0.43
2000–2020	−0.50	−0.25	−0.44	0.99	9.57	−14.07

.

The results from Table 6 indicate that, from 2000 to 2020, there were significant changes in the areas of construction land and unused land. The area of construction land continued to grow, while the overall trend for unused land was a decrease in area. The changes in area for the other land types remained around 1.00%. Specifically, the area of construction land increased by 9.57%, while the area of unused land decreased by 14.07%. The areas of cultivated land, forest land, and grassland decreased by 0.50%, 0.25%, and 0.44%, respectively, while water areas increased by 0.99%.

In terms of the different time intervals, between 2000 and 2005, the area of construction land increased by 3.64%, and the area of unused land decreased by 6.46%, while the changes in the other land-use types remained at less than 0.60%. Between 2005 and 2010, the area of construction land increased by 3.75%, and unused land decreased by 2.48%, with the changes in the other land-use types not exceeding 0.90%. Between 2010 and 2015, the area of construction land increased by 3.21%, with minimal changes in the other land-use types of no more than 0.20%. Lastly, between 2015 and 2020, the area of construction land increased by 3.51%, with the changes in all other land-use types being less than 1.00%.

4.1.3. Analysis of the Characteristics of LUT

To gain a clearer understanding of the transitions between different land-use types in Jiangxi Province, a LUT matrix was constructed based on LUT data for the periods 2000–2005, 2005–2010, 2010–2015, and 2015–2020, as shown in

Table 7. LUT area (km²) in Jiangxi Province from 2000 to 2020.

Land-Use Types in 2000	Land-Use Types in 2005
	Grassland	Cultivated land	Construction land	Forest land	Water area	Unused land	Total
Grassland	6975.09	81.41	9.09	206.74	8.60	1.32	7282.25
Cultivated land	41.36	44,209.11	438.83	475.18	117.92	15.54	45,297.94
Construction land	1.23	39.19	2767.58	9.03	2.27	2.06	2821.36
Forest land	83.86	591.29	97.42	102,993.67	28.76	0.63	103,795.63
Water area	14.49	199.26	21.24	27.20	6544.49	5.56	6812.24
Unused land	0.18	5.07	1.27	0.30	316.00	598.47	921.30
Total	7116.21	45,125.32	3335.44	103,712.11	7018.05	623.59	166,930.72
Land-Use Types in 2005	Land-Use Types in 2010
	Grassland	Cultivated land	Construction land	Forest land	Water area	Unused land	Total
Grassland	6688.78	69.88	40.42	302.41	13.10	1.61	7116.21
Cultivated land	25.31	44,130.75	474.77	329.12	154.65	10.73	45,125.32
Construction land	5.51	144.70	3150.95	14.18	19.96	0.14	3335.44
Forest land	73.14	490.63	269.43	102,834.77	42.48	1.67	103,712.11
Water area	6.10	142.17	23.56	33.16	6756.99	56.07	7018.05
Unused land	1.62	12.62	1.93	0.62	130.37	476.43	623.59
Total	6800.47	44,990.74	3961.06	103,514.26	7117.55	546.64	166,930.72
Land-Use Types in 2010	Land-Use Types in 2015
	Grassland	Cultivated land	Construction land	Forest land	Water area	Unused land	Total
Grassland	6664.83	34.07	42.47	54.78	4.15	0.10	6800.40
Cultivated land	33.24	44,088.36	383.83	443.91	40.93	0.47	44,990.75
Construction land	1.72	65.13	3879.13	11.85	3.18	0.06	3961.07
Forest land	53.77	412.12	270.91	102,753.35	22.46	0.18	103,512.79
Water area	6.47	30.52	19.16	23.85	7035.31	2.26	7117.58
Unused land	0.10	0.48	0.65	0.18	20.63	526.08	548.13
Total	6760.13	44,630.69	4596.16	103,287.93	7126.66	529.15	166,930.72
Land-Use Types in 2015	Land-Use Types in 2020
	Grassland	Cultivated land	Construction land	Forest land	Water area	Unused land	Total
Grassland	6268.02	159.02	50.68	257.99	23.25	0.38	6759.34
Cultivated land	169.93	41,542.16	751.93	1975.91	187.56	2.62	44,630.11
Construction land	10.93	291.76	4209.31	65.59	18.32	0.23	4596.15
Forest land	661.41	2044.04	364.43	100,089.68	120.44	1.02	103,281.02
Water area	14.28	156.36	32.76	102.31	6796.16	23.96	7125.84
Unused land	0.49	2.32	0.50	0.94	19.67	514.35	538.26
Total	7125.06	44,195.66	5409.61	102,492.42	7165.40	542.57	166,930.72

. ArcGIS 10.8 was used to analyze the patterns of LUCs in Jiangxi Province and further reveal the Spatial–temporal patterns of LUT, as shown in Figure 4.

The results from Table 7 and Figure 4 indicate that, from 2000 to 2005, the area of construction land increased significantly, with a rate of increase of 18.20%. This increase is primarily attributable to the conversion of cultivated land into construction land, with an inflow area of 438.83 km², while the area of unused land decreased by 32.20%, primarily due to the creation of water areas, with an outflow area of 316.00 km². From 2005 to 2010, the ratio of construction land increased by 18.70%, with inflows mainly from forest land and cultivated land, corresponding to areas of 269.43 km² and 474.77 km², respectively. The area of grassland decreased by 4.40%, mainly represented due to its conversion into forest land, with an outflow area of 302.00 km². From 2010 to 2015, the ratio of construction land increased by 17.30%, with inflows mainly from cultivated land and forest land, representing areas of 383.83 km² and 270.91 km², respectively. From 2015 to 2020, the ratio of construction land increased by 17.60%, with inflows mainly from forest land and cultivated land, representing areas of 364.43 km² and 751.93 km², respectively. The area of grassland decreased by 4.40%, which was mainly turned into forest land, with an outflow area of 302.00 km². The area of forest land decreased by 789.00 km², primarily due to its conversion into cultivated land, with an inflow area of 2044.04 km².

Overall, between 2000 and 2020, there were various types of transfers between land areas in Jiangxi Province. The area of construction land continued to increase, with a cumulative net increase of 2588.25 km². The areas of grassland, cultivated land, and forest land decreased slightly, primarily due to the conversion of these types of land into construction land and water areas. Meanwhile, the area of unused land rapidly diminished, with a subsequent trend of a stable decrease in the speed of land transfer.

4.2. Analysis of the Spatial–Temporal Pattern of LUCEs in Jiangxi Province

4.2.1. Analysis of Changes in LUCEs

Using the LUCE accounting model, the net CEs and the carbon sources and sinks for different land-use types in Jiangxi Province from 2000 to 2020 were calculated, as shown in

Table 8. LUCEs (10⁴ t) from different land-use types in Jiangxi Province from 2000 to 2020.

LUCEs		2000	2005	2010	2015	2020
Carbon sources	Cultivated land	208.1450	207.3510	206.7330	205.0780	203.0800
	Construction land	1945.2430	3275.5420	4824.9200	6025.9870	6420.6460
	Total	2153.3880	3482.8920	5031.6530	6231.0650	6623.7260
Carbon sinks	Forest land	−635.7500	−635.2370	−634.0260	−632.6390	−627.7760
	Grassland	−1.4930	−1.4590	−1.3940	−1.3860	−1.4610
	Water area	−17.5080	−18.0360	−18.2920	−18.3160	−18.4150
	Unused land	−0.0050	−0.0030	−0.0030	−0.0030	−0.0030
	Total	−654.7550	−654.7350	−653.7150	−652.3430	−647.6550
Net LUCEs		1498.6330	2828.1570	4377.9380	5578.7230	5976.0710

.

The results from Table 8 indicate that, during the study period, the net LUCEs in Jiangxi Province continuously rose from 14.986330 million tons to 59.760710 million tons, with an annual growth rate of 7.60%. LUCEs originating from construction land accounted for over 90.00% of the total LUCEs and displayed an increasing trend.

Meanwhile, the total carbon sink continuously decreased during the study period, albeit with a small decline. Forest land was the main carbon sink source, accounting for over 95.00% of the total carbon sink, and showed a slow declining trend, decreasing by a total of 79.740 thousand tons during the study period. Carbon sinks from water areas accounted for around 3.00% of the total carbon sink and showed continuous growth, increasing by a total of 9.070 thousand tons. The carbon sinks from grassland and unused land accounted for less than 0.30% of the total carbon sink and exhibited a fluctuating decreasing trend.

4.2.2. Analysis of the Spatial and Temporal Patterns in LUCEs

Based on the data on land use and relevant energy consumption, the spatial differences in LUCEs between various cities in Jiangxi Province were analyzed. The net LUCEs and changes in LUCEs were calculated for each city, as shown in

Table 9. Net LUCEs (10⁴ t) in prefecture-level cities of Jiangxi Province from 2000 to 2020.

	Year	2000	2005	2010	2015	2020
Cities
Fuzhou		74.6757	158.9509	265.9444	338.7817	337.0922
Ganzhou		125.9881	261.2563	423.2471	556.0216	762.2564
Ji’an		93.1622	180.0881	299.5258	405.9788	473.5572
Jingdezhen		83.0199	142.5571	219.5790	259.5874	222.8579
Jiujiang		180.8465	311.2840	486.0071	638.1502	767.7413
Nanchang		461.2274	836.0259	1133.4260	1438.0457	1443.5117
Pingxiang		91.6307	175.4936	254.2555	314.3233	228.8407
Shangrao		121.5123	259.9594	401.9703	531.3981	597.1401
Xinyu		62.7308	140.8908	318.0851	333.7324	245.3899
Yichun		154.8952	267.3452	406.7329	541.3914	659.2441
Yingtan		48.9440	94.3060	169.1646	221.3119	238.4392

and

Table 10. Net LUCE changes (10⁴ t) in prefecture-level cities of Jiangxi Province from 2000 to 2020.

	Year	2000–2005	2005–2010	2010–2015	2015–2020
Cities
Fuzhou		84.2752	106.9935	72.8373	−1.6895
Ganzhou		135.2682	161.9907	132.7746	206.2347
Ji’an		86.9259	119.4377	106.4530	67.5784
Jingdezhen		59.5372	77.0218	40.0085	−36.7295
Jiujiang		130.4374	174.7231	152.1431	129.5910
Nanchang		374.7986	297.4000	304.6198	5.4659
Pingxiang		83.8629	78.7619	60.0678	−85.4826
Shangrao		138.4472	142.0108	129.4278	65.7420
Xinyu		78.1600	177.1943	15.6473	−88.3425
Yichun		112.4500	139.3877	134.6585	117.8527
Yingtan		45.3620	74.8586	52.1473	17.1273

, respectively.

The results from Table 9 and Table 10 indicate significant intra-urban differences in LUCEs in Jiangxi Province. Among the cities under study, Nanchang City had the highest LUCEs, with a net change in LUCEs of 9.822843 million tons during the study period. Between 2015 and 2020, the net change in LUCEs in Nanchang City was only 54.659 thousand tons, indicating a deceleration in the growth rate of its net LUCEs. Jiujiang City had the second-highest LUCEs, with a rapid increase in its net LUCEs during the study period, which reached 7.677413 million tons in 2020. The net LUCEs in Ganzhou City showed fluctuating growth, with net LUCEs of 7.622564 million tons in 2020. From 2015 to 2020, the change in its net LUCEs reached 2.062347 million tons, constituting the greatest change in the province. Although Ganzhou City has implemented various measures to reduce its LUCEs, time is required for these measures to be implemented effectively. Therefore, in the short term, its LUCEs are expected to continue to grow. The net LUCEs of other prefecture-level cities have also steadily increased following a period of rapid growth.

The economic development of various cities in Jiangxi Province varied significantly, leading to variations in their LUCEs. Utilizing ArcGIS 10.8, the spatial patterns of LUCEs in Jiangxi Province were classified into five levels, with the first level being the lowest and the fifth level being the highest, as shown in Figure 5.

The results from Figure 5 indicate that there were significant spatial differences in the net LUCEs between various cities in Jiangxi Province from 2000 to 2020, showing a pattern of spatial distribution in which high levels of emissions were observed in the north and south of the province and low levels were observed in the central region. In 2000, Nanchang City had a net level of LUCEs corresponding to tier three, while the net LUCEs of the other cities fell into tier one and tier two. By 2005, the net LUCEs of the cities in Jiangxi Province gradually increased, with Yingtan City falling into tier one and the net LUCEs of the other cities all moving up one tier. In 2010, the net LUCEs of Nanchang City rose by one tier, with the net LUCEs of all other cities except Yingtan City reaching tier three. In 2015, the net LUCEs of Ganzhou, Yichun, Jiujiang, Shangrao, and Yingtan increased by one tier, while the net LUCEs of the other cities remained unchanged. In 2020, the net levels of LUCEs of the cities in Jiangxi Province remained unchanged, mainly due to the implementation of a green and low-carbon development strategy in the province, which significantly affected LUCE control.

In general, from 2000 to 2020, the net levels of LUCEs in various cities in Jiangxi Province increased. The spatial distribution of the net LUCEs in each region within the province remained relatively stable. Among these cities, Nanchang, as the provincial capital, experienced a rapid increase in population inflow and energy consumption, causing its net LUCEs to reach the highest level. The net levels of LUCEs for the other cities grew steadily during this period.

4.2.3. Analysis of the Spatial and Temporal Patterns in the Intensity of LUCEs

Based on the LUCE intensity model, the intensity of LUCEs $IC$ was calculated, as shown in

Table 11. LUCE intensity (t/10,000 CNY) in Jiangxi Province and its cities from 2000 to 2020.

Cities	GDP (Billion CNY)					IC (t/10,000 CNY)
	2000	2005	2010	2015	2020	2000	2005	2010	2015	2020
Fuzhou	125.14	262.00	630.01	1105.14	1573.00	0.60	0.61	0.42	0.31	0.21
Ganzhou	266.20	500.11	1119.74	1973.87	3645.00	0.48	0.52	0.38	0.28	0.21
Ji’an	155.90	303.14	720.53	1328.52	2168.00	0.61	0.60	0.42	0.31	0.22
Jingdezhen	95.27	193.10	461.50	772.06	957.00	0.87	0.74	0.48	0.34	0.23
Jiujiang	213.07	428.92	1032.06	1902.68	3241.00	0.87	0.74	0.48	0.34	0.24
Nanchang	435.10	1007.70	2200.11	4000.01	5746.00	1.07	0.83	0.52	0.36	0.25
Pingxiang	99.49	228.10	520.39	912.39	963.00	0.92	0.77	0.49	0.34	0.24
Shangrao	173.43	388.11	901.00	1650.81	2624.00	0.72	0.68	0.45	0.32	0.23
Xinyu	64.94	177.32	631.22	946.80	1001.00	0.97	0.80	0.50	0.35	0.25
Yichun	185.58	372.21	870.00	1621.02	2790.00	0.84	0.72	0.47	0.33	0.24
Yingtan	53.95	123.53	344.89	639.26	983.00	0.91	0.77	0.49	0.35	0.24
Jiangxi Province	1868.06	3984.24	9431.45	16,852.55	25,691.00	0.80	0.71	0.46	0.33	0.23

. The calculation of IC utilized nominal GDP, with data directly sourced from the Jiangxi Statistical Yearbook. While this approach may introduce deviations due to price changes, nominal GDP was prioritized based on data consistency and availability. Results were obtained on the intensity of LUCEs for each city, as shown in Figure 6.

The results from Table 11 and Figure 6 indicate that the intensity of LUCEs in Jiangxi Province and its cities declined overall during the study period. Spatially, a pattern of higher emissions in the north and lower emissions in the south emerged, with the intensity of emissions gradually decreasing outward from the provincial capital of Nanchang. Overall, the intensity of LUCEs in Jiangxi Province decreased from 0.80 tons/10,000 CNY to 0.23 tons/10,000 CNY, representing a decrease of 71.20%. Nanchang City exhibited the most significant change in the intensity of its LUCEs, decreasing from 1.07 tons/10,000 CNY in 2000 to 0.25 tons/10,000 CNY in 2020, marking a reduction of 76.60%. Nanchang City significantly reduced the intensity of its LUCEs while simultaneously safeguarding its economic growth, indicating the effectiveness of its low-carbon development model. Ganzhou City experienced the lowest decrease in the intensity of its LUCEs, at 56.20%, while the LUCE intensity of the other cities declined, with reductions exceeding 65.00%. This indicates that these prefecture-level cities have optimized their energy structures well, gradually transitioning from a singular to a diversified energy structure.

4.2.4. Analysis of the Spatial and Temporal Patterns in LUCEs per Capita

Based on the population data from Jiangxi Statistical Yearbooks from 2000 to 2020, the LUCEs per capita were calculated for each city in Jiangxi Province, as shown in

Table 12. Per capita LUCEs (t/person) in each city of Jiangxi Province from 2000 to 2020.

Cities	Total Population (Ten Thousand People)					PC (t/Person)
	2000	2005	2010	2015	2020	2000	2005	2010	2015	2020
Fuzhou	360.49	381.31	403.96	399.28	432.00	0.21	0.42	0.66	0.85	0.78
Ganzhou	794.16	845.69	907.27	960.63	983.50	0.16	0.31	0.47	0.58	0.78
Ji’an	447.59	464.76	495.04	530.36	539.00	0.21	0.39	0.61	0.77	0.88
Jingdezhen	143.78	150.82	163.16	166.73	171.00	0.58	0.95	1.35	1.56	1.30
Jiujiang	446.83	466.20	497.91	516.59	523.50	0.42	0.68	0.99	1.25	1.48
Nanchang	432.55	475.17	502.25	520.38	538.00	1.07	1.77	2.26	2.77	2.69
Pingxiang	173.78	179.67	188.09	198.34	199.50	0.53	0.98	1.35	1.58	1.15
Shangrao	642.72	676.00	740.33	774.40	791.00	0.19	0.39	0.55	0.69	0.76
Xinyu	105.93	110.72	118.01	123.52	125.00	0.59	1.27	2.70	2.70	1.97
Yichun	510.20	528.17	557.93	596.99	602.00	0.31	0.51	0.73	0.91	1.10
Yingtan	106.40	111.15	121.92	127.26	129.00	0.46	0.85	1.39	1.74	1.85
Jiangxi Province	4164.43	4389.66	4695.87	4914.48	5033.50	0.36	0.64	0.93	1.14	1.19

. Further analysis yielded the spatial distribution of LUCEs per capita, as shown in Figure 7.

The results from Table 12 and Figure 7 indicate that, in terms of temporal changes in LUCEs, the LUCEs per capita in various cities in Jiangxi Province increased from 2000 to 2015. On the other hand, from 2015 to 2020, the LUCEs per capita varied, with increases in some cities and decreases in others. In terms of spatial distribution, from 2000 to 2020, the LUCEs per capita in Jiangxi Province generally exhibited a trend of higher levels in the north and lower levels in the south. The levels of LUCEs gradually decreased when moving outward from the provincial capital, Nanchang, and the industrial city of Xinyu. In 2000, the LUCEs per capita in various cities in Jiangxi Province were generally low, with the highest LUCEs per capita in Nanchang at 1.07 tons per person. By 2005, the LUCEs per capita in Jiangxi Province began to gradually increase, with Xinyu City showing the fastest growth rate, reaching 1.27 tons per person. In 2010, the LUCEs per capita in Jiangxi Province increased significantly, with Xinyu City reaching 2.70 tons per person. In 2015, the LUCEs per capita in Jiangxi Province reached 1.14 tons per person, with Nanchang City reaching its peak at 2.77 tons per person and Xinyu City reaching 2.70 tons per person. The LUCEs per capita in other cities in the province also gradually increased. By 2020, the trend of growth in the LUCEs per capita in various cities in Jiangxi Province had decelerated.

Overall, the LUCEs per capita in Jiangxi Province rose, with the growth rate of LUCEs exceeding the growth rate of the total population in most of its cities. Jiangxi Province is actively promoting a low-carbon development strategy, and thus its per capita LUCEs are expected to decrease further, contributing to sustainable social development.

4.3. Analysis Using the LMDI Model

The Kaya identity and the LMDI decomposition method were used to decompose the factors influencing LUCEs in the study area. The relationships between changes in LUCEs and five influencing factors (i.e., LUCE intensity, LUS, LUE, economic level, and population scale) were ascertained. According to Table 8, the net LUCEs in Jiangxi Province increased by 44.774380 million tons from 2000 to 2020. The results of the additive decomposition of the factors influencing LUCEs in Jiangxi Province are outlined in

Table 13. Additive decomposition values of LUCE factors (10⁴ t) in Jiangxi Province.

Year	2000–2005	2005–2010	2010-2015	2015–2020
LUCE intensity	903.0001	861.6737	397.5766	−619.1287
LUS	426.5309	688.0956	803.2698	1016.7204
LUE	−1595.0972	−3061.7672	−2876.8382	−2435.3406
Economic level	1484.1686	2822.1824	2651.2617	2296.8959
Population size	110.9221	239.5957	225.5152	138.2010
Total change	1329.5245	1549.7802	1200.7850	397.3480

, while the trends are shown in Figure 8. These results indicate that economic level was the factor with the greatest impact on LUCEs, contributing to an increase of 92.545086 million tons. LUS was the next most influential factor, promoting an increase of 29.346167 million tons of LUCEs. LUCE intensity was the third most influential factor, leading to an increase of 15.431217 million tons of LUCEs. Population scale contributed to an increase of 7.142340 million tons of LUCEs, representing a lower relative impact. Conversely, LUE suppressed LUCEs, leading to a reduction of 99.690432 million tons.

The results of the multiplicative decomposition of the factors influencing LUCEs in Jiangxi Province are outlined in

Table 14. Multiplicative decomposition results of LUCE factors in Jiangxi Province.

Year	2000–2005	2005–2010	2010–2015	2015–2020
LUCE intensity	1.5393	1.2750	1.0836	0.8983
LUS	1.2260	1.2141	1.1760	1.1925
Land-use efficiency	0.4668	0.4218	0.5595	0.6559
Economic level	2.0318	2.2160	1.7077	1.4884
Population size	1.0544	1.0699	1.0466	1.0242

. These results indicate that from 2000 to 2020, the economic level, LUS, LUCE intensity, and population scale in Jiangxi Province drove an increase in its net LUCEs. Among these factors, the facilitating effect of economic level was the most pronounced, with an average contribution rate of 1.8610. On the other hand, LUE mitigated the net LUCEs, but at an average contribution rate of only 0.5260. Due to adjustments to Jiangxi Province’s energy structure across the overall study period, LUCE intensity played a driving role in increasing its net LUCEs from 2000 to 2015 but a mitigating role from 2015 to 2020, with an average contribution rate of 1.1991. LUS and population scale had average contribution rates to the growth of net LUCEs of 1.2022 and 1.0488, respectively.

In summary, improving LUE can lower LUCEs but only to a limited extent. In the future, LUCEs in Jiangxi Province will continue to increase. To reduce CEs to lower levels, developing a green economy, optimizing LUS, strongly promoting new sources of energy, and pushing for the clean and efficient utilization of fossil fuels will be integral.

4.4. Analysis of the TD Model

Using the LUCE accounting results for Jiangxi Province from 2000 to 2020 and Equation (23), the decoupling index between LUCEs and economic growth during the study period was obtained, as shown in

Table 15. Decoupling status of LUCEs and economic growth in Jiangxi Province from 2000 to 2020.

Year	${\mathit{M}}_{\mathit{c}}$	${\mathit{M}}_{\mathit{g}}$	${\mathit{T}}_{\mathit{i}}$	Decoupling Status
2000–2005	0.89	1.13	0.78	Weak decoupling
2005–2010	0.55	1.37	0.4	Weak decoupling
2010–2015	0.27	0.79	0.35	Weak decoupling
2015–2020	0.07	0.52	0.14	Weak decoupling
2000–2020	2.99	12.75	0.23	Weak decoupling

.

The results from Table 15 indicate that from 2000 to 2020, there was weak decoupling between LUCEs and economic growth in Jiangxi Province. This shows that the rate of growth of LUCEs was lower than the rate of growth of the economy. The increase in net LUCEs decreased from 89% in the 2000–2005 period to 7% in the 2015–2020 period, while the rate of economic growth decreased from 113% in the 2000–2005 period to 52% in the 2015–2020 period. Overall, from 2000 to 2020, the pace of economic growth in Jiangxi Province gradually shifted toward high-quality development, showing a stable and positive trend. Simultaneously, the rate of growth in LUCEs also exhibited a declining trend over time.

5. Discussion

This section primarily summarizes and analyzes the research findings, revealing the mechanisms driving the growth of LUCEs and their relationship with economic growth, while also elucidating the practical implications of these findings for policy formulation and territorial spatial planning in Jiangxi Province.

5.1. Summary of Findings

Empirical analysis reveals the trajectory of LUC and associated LUCEs in Jiangxi Province from 2000 to 2020. The expansion of construction land, primarily at the expense of cultivated land and forest land (Table 5, Figure 4), directly led to a substantial increase in net LUCEs, which grew from 14.99 million tons to 59.76 million tons (Table 8). Spatially, net LUCEs exhibited a pattern characterized by “higher levels in the northern and southern regions and lower levels in the central region” (Figure 5). Further analysis using the LMDI and Tapio decoupling models elucidated the influencing factors and their dynamic mechanisms: economic development was the dominant driver behind the growth in LUCEs, while improvements in LUE served as the most significant mitigating factor (Table 13). A consistent “weak decoupling” state between economic growth and LUCEs was maintained throughout the study period (Table 15), indicating that Jiangxi Province remains in a stage of carbon-intensive economic growth, though the growth rate of LUCEs has been slower than that of the economy.

5.2. Mechanistic Analysis

The LMDI decomposition results reveal the driving factors behind the growth of LUCEs. Economic development level was identified as the primary positive contributing factor, reflecting the development patterns characteristic of rapidly industrializing and urbanizing regions. Both the decomposition results and LUT analyses indicate massive conversions of cultivated land, forest land, and grassland to construction land (Table 7). However, LUE demonstrated a significant negative contribution effect, suggesting a substantial increase in economic output per unit of land area in Jiangxi Province. This reflects a transition toward more intensive and efficient land-use patterns, which has partially offset LUCE growth resulting from economic expansion.

The spatial pattern of LUCEs aligns closely with regional economic development levels. Northern economic centers (e.g., Nanchang and Jiujiang) exhibit high LUCE characteristics, though Nanchang achieved a deceleration in LUCE growth during 2015–2020 through its transition to new industrialization. Southern cities (e.g., Ganzhou) demonstrated accelerated emission growth driven by rapid industrialization. The study confirms a “weak decoupling” relationship between economic growth and LUCEs in Jiangxi Province, attributable to policy measures including energy efficiency improvements and industrial restructuring (Table 11), though achieving “strong decoupling” remains an ongoing challenge.

5.3. Policy Implications

First, optimizing LUS is of paramount importance. Strict control over the scale of newly added construction land, coupled with the revitalization of existing stock, is critical to reducing LUCEs. Second, studies on LUE indicate that enhancing LUE should serve as a central pillar of Jiangxi’s low-carbon strategy, thereby fostering compact urban development and industrial intensification. Third, given the persistent dominance of economic development as a key driver, cities should implement policies to accelerate the energy transition—promoting the clean and efficient use of fossil fuels while advancing the development of new energy sources. Finally, cases such as Nanchang, which have successfully slowed CE growth through industrial upgrading, offer valuable, replicable references for other cities in the province.

6. Conclusions

This study utilized land-use grid and socioeconomic data on Jiangxi Province spanning 2000 to 2020 to estimate its LUCEs based on LUCE coefficients. Subsequently, decomposition analyses were conducted on the primary factors influencing changes in its LUCEs during this period. The main conclusions are as follows:

• During the period 2000–2020, cultivated land and forest land remained the dominant land-use types in Jiangxi Province. While the area of construction land and water area showed an increasing trend, other land-use types experienced a decline. Frequent land-use conversions were observed, primarily characterized by transformations into construction land and water area, indicating an accelerated urbanization process.
• LUCEs in Jiangxi Province demonstrated a growth trend characterized by an initial rapid increase followed by a slower pace, rising from 14.99 million tons in 2000 to 59.76 million tons in 2020. Construction land constituted the primary source of LUCEs, accounting for over 90% of total LUCEs. While LUCE intensity showed a continuous decline, per capita LUCEs exhibited dynamic growth throughout the period.
• LUS, economic development level, and population scale were the primary drivers of LUCE growth in Jiangxi Province, while LUE served as a mitigating factor. Over time, the role of LUCE intensity shifted from positive to negative. A long-term state of weak decoupling was maintained between LUCEs and economic growth, indicating that the growth rate of LUCEs consistently lagged behind that of the economy.
• During the study period, a state of weak decoupling was observed between LUCEs and economic growth in Jiangxi Province, characterized by economic growth outpacing the increase in CEs. This indicates that Jiangxi Province is transitioning toward a green, low-carbon economic model.

Using land-use data and established coefficients, this study first estimated Jiangxi’s LUCEs from 2000 to 2020, and then analyzed the main factors driving their dynamics. The findings have significant theoretical and practical implications. However, the LMDI model employed in this study is fundamentally a decomposition technique and cannot directly account for spatial spillover effects between different cities or regions. Future research could adopt spatial econometric models to delve into the spatial spillover effects of LUCEs, thereby offering more targeted guidance for regional collaborative emissions reduction.

7. Recommendations

Based on our analysis of the Spatial–temporal characteristics of the LUCEs in Jiangxi Province and the factors influencing these emissions, this study proposes corresponding strategies for reducing CEs to lower levels, providing a reference for carbon reduction efforts in underdeveloped regions.

• It is vital to optimize LUS rationally to reduce LUCEs. Jiangxi Province should strengthen its spatial planning and land-use control; strictly adhere to ecological protection boundaries; control the use of its ecological spaces; prohibit unauthorized changes to the purpose and nature of ecosystems like forests, wetlands, and grasslands; strictly control the scale of new construction land; and revitalize its existing urban and rural construction land. To enhance the carbon sequestration capacity of its ecosystems, efforts should be focus targeted at governance of its mountain, water, forest, cultivated land, lake, grassland, and sand systems, both overall and at the watershed level. Foundational support for accounting for carbon sequestration in these ecosystems can be strengthened by relying on and expanding competent natural resource surveying and monitoring systems.
• Actively engaging in the green and low-carbon transformation of energy is urgent. Jiangxi Province should promote the clean and efficient use of fossil fuel energy; reasonably control the growth of coal and petroleum consumption; and pursue reductions in fossil fuel energy consumption through substitution. Development efforts for new energy sources should be intensified, guided by adequate planning and with increases in their utilization. Accelerating the construction of a new type of power system is essential to driving a sustainable transformation of energy infrastructure.
• It is crucial to advocate for a green and low-carbon lifestyle and translate green concepts into conscious actions among the entire population. Jiangxi Province should enhance public education and awareness of low-carbon living. Encouraging enterprises to fulfill their social responsibilities, proactively adapt to the requirements of green and low-carbon development, and enhance their level of green innovation is also an important step.
• Carbon reductions can be achieved by promoting a circular economy. Jiangxi Province should advocate a circular economy in its development zones, with the aim of enhancing their resource output and recycling rates, optimizing the industrial layouts in industrial parks, and thoroughly pursuing circular-economy-related transformations in these areas. Efforts should be made to improve the comprehensive utilization of bulk solid waste, implement major projects for the efficient utilization of mineral resources, and reasonably optimize levels of mineral resource extraction. Enhancing resource recycling, establishing a sound network for recycling waste materials, and promoting the collection of recyclables will be key in this regard.