CO₂ Emissions and Scenario Analysis of Transportation Sector Based on STIRPAT Model: A Case Study of Xuzhou in Northern Jiangsu

Abstract

To support carbon peaking and neutrality goals in the city transportation sector, this paper accounts for CO₂ emissions from the transport sector in Xuzhou City, North Jiangsu Province, from 1995 to 2023. This study explores the relationship between transport-related carbon emissions and economic growth, using the TAPIO decoupling index. Meanwhile, a carbon emission prediction model based on the STIRPAT framework is constructed, with scenario analysis applied to forecast future emissions. Results show three decoupling stages: the first, dominated by weak and expansive negative decoupling, reflects extensive economic growth; the second features weak decoupling with expansive coupling, indicating a more environmentally coordinated phase; the third transitions from expansive negative decoupling and weak decoupling to strong decoupling and expansive coupling, suggesting a shift in development patterns. Under the baseline, low-carbon, and enhanced low-carbon scenarios, by 2030, the CO₂ emissions of the transportation industry in Xuzhou will be 10,154,700 tons, 9,072,500 tons, and 8,835,000 tons, respectively, and the CO₂ emissions under the low-carbon scenario and the enhanced low-carbon scenario will be reduced by 10.66% and 13.00%, respectively. Based on these findings, the study proposes carbon reduction strategies for Xuzhou’s transport sector, focusing on policy regulation, energy use, and structural adjustments.

The transportation sector, as a pillar industry underpinning economic and social development, also plays a crucial role in the domains of energy consumption and environmental protection [1,2,3]. According to the International Energy Agency’s (IEA) 2019 annual report, carbon emissions from China’s transportation sector reached 901 million tons, accounting for 9.12% of the country’s total greenhouse gas emissions [4]. Faced with the dual pressures of tight carbon reduction targets and a fixed timeline for the national “dual carbon” strategy, in-depth examinations of the sector’s dynamic evolution and systematic explorations of its driving mechanisms have become core issues for advancing the green transformation of transportation. These efforts not only contribute to theoretical innovation in environmental economics but also provide a scientific basis for policymaking in emission reduction [5,6].

Current research on carbon emissions in the transportation sector primarily focuses on the influences of economic, social, industrial, and technological factors, as well as associated predictive-modeling approaches. Lin et al. [7], using a quantile regression model, analyzed the mechanisms through which energy structure, economic development level, carbon emission coefficients, and population size affect transport-related carbon emissions. Talbi [8] applied a vector autoregressive model and found that carbon emissions from the transportation sector are influenced by factors such as the energy intensity of road transport, total energy consumption, economic growth, urbanization, and fuel efficiency. Xu et al. [9] introduced a non-parametric additive regression model to demonstrate the nonlinear coupling between private car ownership, freight turnover, economic scale, and energy efficiency. Gao et al. [10], through an improved STIRPAT extended model, deconstructed the regulatory effects of variables such as population base, economic level, energy consumption intensity, infrastructure investment, urbanization rate, and private vehicle penetration on carbon emissions in the transportation sector. Wang et al. [11] employed the IPAT model to study the relationships between carbon emissions and factors such as economic level, population size, and industrial structure. Wang et al. [3], based on a double-leveled econometric model, identified the main influencing factors of carbon emissions in the transportation sector to be transport structure, economic development level, energy efficiency of transport equipment, quality of transport organization, and infrastructure density.

A review of the existing literature reveals that the influencing factors of carbon emissions in the transportation sector can generally be categorized into four dimensions: economic and social development, transport system characteristics, energy structure and consumption, and environmental regulation [12]. However, current studies largely focus on macro-level indicators, such as economic development, population, freight volume, and urbanization rate, with relatively limited quantitative analysis of the transport sector in resource-depleted cities [13]. Moreover, there remains a gap in research that effectively integrates factor analysis with modeling processes tailored to the transportation sector.

To provide a reliable and robust assessment and projection of CO₂ emissions in the regional transportation sector, this paper accounts for data on CO₂ emissions from Xuzhou City’s transport industry in Northern Jiangsu from 1995 to 2023 and applies a multidimensional modeling framework to reveal viable emission reduction pathways. First, the TAPIO decoupling index is employed to quantitatively assess the dynamic relationship between transport carbon emissions and economic growth in the region. Building on this, an extended STIRPAT model is constructed, incorporating eight key influencing variables across four dimensions: population (year-end total population, urbanization rate), economic development (GDP per capita, share of tertiary industry), technological level (energy structure, carbon intensity), and transport activity (passenger turnover, freight turnover). Finally, we simulated carbon emission trends from 2024 to 2030 under baseline, low-carbon, and enhanced low-carbon scenarios based on models. This study aims to provide tailored regulatory strategies and peaking programs for Xuzhou City’s transport industry, while also offering a theoretical reference for similar cities conducting carbon emission research.

1. Research Methodology and Data Sources

1.1. Research Methodology

1.1.1. Measurement of Carbon Dioxide Emissions

Carbon emissions from the transportation sector: CO₂ emissions from the transportation sector refer to the total amount of carbon dioxide generated during the operation of various modes of transport [14]. In Xuzhou, transportation can be divided into external and internal categories. New energy vehicles are not included. External transportation mainly includes long-distance buses, railways, and air travel, while internal transportation primarily involves private cars, public buses, motorcycles, and similar vehicles. The CO₂ emissions from external transportation are calculated using Equation (1):

$${TN}_{\mathrm{CO}2}=\sum _i^n{TV}_i\times {EF}_i$$

(1)

where TN_CO2 represents the CO₂ emissions from external transportation (in 10,000 tons); TV_i denotes the passenger turnover of the type-i external transport mode (in 10,000 person-kilometers); EF_i refers to the CO₂ emission conversion factor for the type-i external transport, in kg/(km·person). Specifically, the conversion factors are 0.028 (kg/(km·person)) for long-distance buses, 0.049 (kg/(km·person)) for railways, and 0.300 (kg/(km·person)) for air transport [15].

The CO₂ emissions from internal transportation are calculated using Equation (2):

$${TW}_{\mathrm{CO}2}=\sum _i^n{L_i\times EF}_i\times {DV}_i$$

(2)

where TW_CO2 represents the CO₂ emissions from internal transportation (in 10,000 tons); EF_i refers to the CO₂ emissions generated by the operation of the type-i vehicle for 1 km. That is, the CO₂ emission conversion factor (kg/km); DV_i is the annual mileage of the type-i vehicle (in 10,000 km); L_i is the number of type-i vehicles. Specifically, the conversion factors are 0.820 (kg/km) for public buses, 0.200 (kg/km) for private cars, 0.070 (kg/km) for motorcycles, and 0.460 (kg/km) for light trucks [16].

1.1.2. Scope and Definition of Decoupling

In this study, “decoupling” is strictly defined at the prefecture-city scale (Xuzhou City’s transport sector, 1995–2023) and aims to quantify the annual decoupling status between CO₂ emissions from the municipal transport sector and economic growth [17]. Economic growth is proxied by real GDP, while CO₂ emissions refer solely to direct combustion-related emissions from fossil-energy use within the transport sector. Emissions attributable to electricity consumption by new-energy vehicles are explicitly excluded. The temporal dimension is framed by China’s sequential Five-Year Plan periods from the Ninth to the Fourteenth FYP, thereby enabling an assessment of how policy interventions shape decoupling trajectories. Following the decoupling model developed by Sun et al. [18], the specific formulation is presented as Equation (3):

$$W_{n+1}={\displaystyle \frac{\left(T_{n+1}-T_n\right)/T_n}{\left({\mathrm{GDP}}_{n+1}-{\mathrm{GDP}}_n\right)/{\mathrm{GDP}}_n}}={\displaystyle \frac{\%∆TC}{\%∆\mathrm{GDP}}}$$

(3)

where n represents the year; W_n+1 is the decoupling elasticity in year n + 1; T_n and T_n+1 are the CO₂ emissions from the transportation sector in years n and n + 1, respectively; and GDP_n and GDP_n+1 are the city’s GDP values in years n and n+1, respectively.

Based on Equation (3), and following the decoupling classification standards proposed by Jie et al. [19], eight decoupling states are identified:

• Strong decoupling (ΔTC < 0, ΔGDP > 0, W < 0);
• Weak decoupling (ΔTC > 0, ΔGDP > 0, 0 ≤ W < 0.8);
• Expansive coupling (ΔTC > 0, ΔGDP > 0, 0.8 ≤ W < 1.2);
• Expansive negative decoupling (ΔTC > 0, ΔGDP > 0, W ≥ 1.2);
• Strong negative decoupling (ΔTC > 0, ΔGDP < 0, W < 0);
• Weak negative decoupling (ΔTC < 0, ΔGDP < 0, 0 ≤ W <0.8);
• Recessive coupling (ΔTC < 0, ΔGDP<0, 0.8 ≤ W < 1.2);
• Recessive decoupling (ΔTC < 0, ΔGDP < 0, W ≥ 1.2).

1.1.3. Establishment of STIRPAT Model

The IPAT model was initially proposed by Ehrlich et al. [20] in 1971 to analyze the relationships among population (P), affluence (A), technology (T), and environmental impact (I). To overcome the linear constraints of the original model, York et al. [21] introduced the stochastic STIRPAT model, with its multiplicative form as Equation (4) and log-linear form as Equation (5). This improved model introduces random error terms and elasticity coefficients, significantly enhancing the flexibility of empirical analysis.

I = a × P^b × A^c × T^d × e

(4)

lnI = lna + b × lnP + c × lnA + d × lnT + lne

(5)

where a, b, c, and d are evaluation parameters; I, P, A, and T represent environmental impact, population, affluence, and technology factors, respectively; and e is a stochastic error term.

To enhance the explanatory power, this study extends the classical STIRPAT model using a multi-dimensional indicator decomposition method. Specifically, eight variables are selected across four dimensions:

• Population dimension: Urbanization rate (C) and year-end total population (P) to represent population structure and scale;
• Economic dimension: GDP per capita (R) for economic development and tertiary industry share (D) for industrial structure;
• Technology dimension: Energy structure (Q) and carbon intensity (Z) to capture the impact of technological progress;
• Transportation dimension: Passenger turnover (K) and freight turnover (H) to quantify transportation demand.

Accordingly, Equation (5) is extended to Equation (6):

TC = EXP (lna₀ + a₁ × lnC + a₂ × lnP + a₃ × lnR + a₄ × lnD + a₅ × lnQ + a₆ × lnI + a₇ × lnK + a₈ × lnH)

(6)

1.1.4. Methodological Independence

The Tapio elasticity index and the STIRPAT specification are logically independent: the former functions as a diagnostic measure of past decoupling between CO₂ emissions from the urban transport sector and economic development, whereas the latter operates as a causal-inference framework aimed at forecasting future sectoral CO₂ emissions on the basis of their driving indicators. Crucially, the two models share no common parameters in their mathematical construction. Consequently, their parameter remain entirely unrelated.

1.2. Data Sources

Relevant parameters were extracted from the Xuzhou Statistical Yearbook (1995–2023). The carbon intensity (Z) is defined as the ratio of CO₂ emissions from the transportation sector to the total economic output of the transportation sector, measured in tons per CNY 10,000. The energy structure (Q) is represented by the proportion of fossil energy consumption within the total energy consumption of the transportation sector, expressed as a percentage. GDP per capita (R) is calculated as the ratio of regional GDP to the year-end total population, measured in yuan per person. Population size (P) refers to the total permanent resident population at year-end, in ten thousand persons. The urbanization rate (C) is defined as the proportion of the non-agricultural registered population in the total population, expressed as a percentage. Passenger turnover (K) is measured as the total product of the number of passengers transported by various means of transport and the distance they are carried, in ten thousand person–kilometers. Freight turnover (H) refers to the total product of the weight of goods transported and the transport distance by all modes, in 100 million ton–kilometers. The share of the tertiary industry (D) is measured as the proportion of the value added by the tertiary sector in the regional GDP, expressed as a percentage. Gross domestic product (G) is directly taken from the annual GDP figures published in the statistical yearbooks, in CNY 100 million.

This paper focuses on calculating CO₂ emissions from the perspective of transportation and does not account for emissions from water transport or the impact of clean energy sources (such as wind, hydro, solar, and nuclear power) on CO₂ emissions.

2. Results and Discussion

2.1. Carbon Emissions from the Transportation Sector

From 1995 to 2023, the CO₂ emissions from the transportation industry in Xuzhou increased from 755,100 tons in 1995 to 7,711,900 tons in 2023, exhibiting an overall upward trend. The annual growth rate ranged from −0.92% to 23.64%, with an average value of 8.79%. Specifically, during the period from 1995 to 2016 (the fluctuating emission increase stage), the growth rate exhibited significant volatility. In contrast, during the period from 2017 to 2023 (the stable emission increase stage), the fluctuations were relatively stable. This indicates that the transportation system has become more mature and stable (Figure 1).

2.2. Decoupling Status from the Transportation Sector

Based on series data on transport-related carbon emissions and GDP in Xuzhou from 1995 to 2023, a quantitative analysis using the TAPIO decoupling index reveals a four-phase dynamic evolution in the decoupling relationship between the two (

Table 1. Decoupling relationship between CO₂ emissions from Xuzhou’s transportation sector and economic growth (1995–2021).

Year	ΔTC/%	ΔGDP	w	Decoupling Relationship
1995–1996	3.699	20.964	0.176	Weak decoupling
1996–1997	4.751	3.316	1.433	Expansionary negative decoupling
1997–1998	2.976	6.453	0.461	Weak decoupling
1998–1999	23.643	7.514	3.147	Expansionary negative decoupling
1999–2000	12.401	6.793	1.826	Expansionary negative decoupling
2000–2001	6.561	10.579	0.620	Weak decoupling
2001–2002	11.292	9.956	1.134	Expansionary connectivity
2002–2003	5.987	13.735	0.436	Weak decoupling
2003–2004	14.440	21.971	0.657	Weak decoupling
2004–2005	22.515	18.966	1.187	Expansionary connectivity
2005–2006	2.382	19.365	0.123	Weak decoupling
2006–2007	13.981	19.417	0.720	Weak decoupling
2007–2008	8.506	21.044	0.404	Weak decoupling
2008–2009	14.417	13.068	1.103	Expansionary connectivity
2009–2010	12.192	22.339	0.546	Weak decoupling
2010–2011	2.883	20.716	0.139	Weak decoupling
2011–2012	3.087	13.090	0.236	Weak decoupling
2012–2013	7.400	12.529	0.591	Weak decoupling
2013–2014	6.518	9.825	0.663	Weak decoupling
2014–2015	4.261	7.171	0.594	Weak decoupling
2015–2016	−0.918	8.864	−0.104	Strong decoupling
2016–2017	9.355	9.014	1.038	Expansionary connectivity
2017–2018	10.207	5.950	1.715	Expansionary negative decoupling
2018–2019	9.237	5.111	1.807	Expansionary negative decoupling
2019–2020	8.288	3.777	2.194	Expansionary negative decoupling
2020–2021	7.233	10.897	0.664	Weak decoupling
2021–2022	10.206	3.633	2.809	Expansionary negative decoupling
2022–2023	8.619	5.802	1.486	Expansionary negative decoupling

). Specifically, the relationship can be categorized as strong decoupling (1 period, 3.57%); weak decoupling (15 periods, 53.57%); expansive coupling (4 periods, 14.28%); and expansive negative decoupling (8 periods, 28.57%). Among these, weak decoupling and expansive negative decoupling together account for 82.14%, representing the dominant modes during the study period. This indicates that Xuzhou’s transportation sector is still in a developmental stage where economic growth has not yet been fully decoupled from carbon emissions.

The decoupling relationship between Xuzhou’s transport-related carbon emissions and economic growth shows clearly staged characteristics:

• Stage I (1995–2000): Dominated by weak decoupling and expansive negative decoupling, this stage reflects an extensive mode of economic development. Economic growth was driven by the expansion of high-energy-consuming industries, which led to a continuous surge in transportation-related carbon emissions. In particular, from 1996–1997 and 1998–2000, the growth rate of transportation-related emissions far overtook GDP expansion, manifesting atypical environmental pressure dynamics;
• Stage II (2001–2012): Characterized by a mixed pattern of predominantly weak decoupling with some expansive coupling. Benefiting from the guidance of the “11th Five-Year Plan”, the industrial structure began shifting toward the tertiary sector. This stage achieved a model of relative decoupling, reflecting a more coordinated relationship between the environment and the economy;
• Stage III (2013–2023): Mainly featured expansive negative decoupling and weak decoupling, followed by strong decoupling and expansive coupling. This stage represents a transitional development mode, marked by diverse patterns in the interaction between the transportation sector and economic growth, shifting from coordinated development to extensive development, then to high-quality growth, and eventually reverting again to extensive development. The phase-wise progression highlights ongoing reforms in the urban economic system. These trends align with China’s national “dual carbon” strategy and echo Jiangsu Province’s and Xuzhou City’s policies on building a green, low-carbon, and circular economy.

It is worth noting that, as a traditional resource-based hub city, Xuzhou is in a critical stage of transitioning between old and new growth drivers. Due to the cyclical nature of industrial restructuring, some high-emission and high-energy-consuming industries have yet to be fully phased out, resulting in fluctuations in the decoupling relationship between economic growth and environmental/resource pressures [22,23].

2.3. Carbon Emissions Forecasting Model for the Transportation Sector

Based on the STIRPAT model, the carbon emissions from the transportation sector in Xuzhou have eight key influencing factors, including urbanization rate (C), year-end population (P), GDP per capita (R), the share of the tertiary industry (D), energy structure (Q), carbon intensity (Z), passenger turnover (K), and freight turnover (H). In order to determine the relevant coefficients by ridge regression, the logarithmic data for each of the above coefficients are shown in

Table 2. The observations of variables used in the STIRPAT model.

Year	Q	Z	R	P	C	K	H	D	TC
1995	4.459	0.938	8.468	6.747	2.986	4.730	6.116	3.381	4.324
1996	4.454	0.737	8.649	6.756	3.020	4.833	6.076	3.420	4.361
1997	4.467	0.701	8.673	6.765	3.045	4.762	6.044	3.473	4.407
1998	4.465	0.617	8.726	6.775	3.068	4.722	6.075	3.492	4.436
1999	4.491	0.725	8.792	6.777	3.091	4.687	6.119	3.512	4.649
2000	4.442	0.744	8.846	6.798	3.250	4.824	6.243	3.548	4.765
2001	4.473	0.701	8.933	6.804	3.285	4.849	6.278	3.566	4.829
2002	4.496	0.731	9.024	6.807	3.321	4.883	6.299	3.562	4.936
2003	4.492	0.729	9.149	6.812	3.444	4.809	6.340	3.554	4.994
2004	4.493	0.563	9.367	6.821	3.497	4.966	6.444	3.543	5.129
2005	4.511	0.520	9.545	6.830	3.529	4.511	6.305	3.558	5.332
2006	4.496	0.351	9.729	6.840	3.535	4.588	6.670	3.562	5.356
2007	4.498	0.303	9.912	6.847	3.538	4.732	6.824	3.582	5.486
2008	4.496	0.196	10.107	6.853	3.572	5.244	7.001	3.606	5.568
2009	4.505	0.256	10.232	6.864	3.676	5.333	7.030	3.613	5.703
2010	4.501	0.843	10.440	6.880	3.824	5.531	7.923	3.698	5.818
2011	4.467	0.726	10.633	6.884	4.159	5.367	8.074	3.718	5.846
2012	4.437	0.643	10.755	6.898	4.206	5.372	7.318	3.716	5.877
2013	4.437	0.586	10.870	6.915	4.271	5.191	7.387	3.761	5.948
2014	4.520	1.165	10.956	6.931	4.277	5.731	7.387	3.799	6.011
2015	4.450	0.620	11.017	6.936	4.059	5.286	7.173	3.821	6.053
2016	4.413	0.474	11.094	6.948	4.072	5.272	7.587	3.858	6.044
2017	4.410	0.528	11.169	6.946	4.084	5.214	7.640	3.884	6.133
2018	4.441	0.687	11.218	6.952	4.104	5.181	7.771	3.913	6.230
2019	4.444	0.785	11.264	6.949	4.114	5.168	7.887	3.914	6.319
2020	4.441	0.804	11.298	6.945	4.132	4.970	7.903	3.915	6.398
2021	4.433	0.901	11.401	6.943	4.137	4.670	7.902	3.898	6.468
2022	4.538	0.781	11.309	6.939	4.140	4.505	7.941	3.893	6.565
2023	4.442	0.716	11.367	6.937	4.146	5.024	8.048	3.925	6.648

.

The multicollinearity problem of the logarithmic values of the eight influencing factors can be solved by the ridge regression analysis method. The ridge regression analysis method is a kind of supplement to the least squares regression, which can effectively solve the multicollinearity problem and improve the accuracy, stability, and reliability of the model’s calculation. Therefore, this paper chooses the ridge regression analysis method to solve the multicollinearity problem and uses the ridge regression function of SPSS software (version 24.0) to fit Equation (5). The ridge regression coefficient K is in the interval of (0, 1) and takes the numeric value with a step size of 0.001. When K = 0.17, the change of the ridge plot was gradually smooth, in Figure 2. The specific ridge regression estimation results are shown in

Table 3. The estimation results of STIRPAT model for transportation industry.

Factors	Unstandardized Coefficient	Standard Error	t-Statistic	p-Value	VIF
Constant term	−23.546	2.865	−8.218	0.000 ***	-
Q	1.765	0.535	3.297	0.004 ***	0.080
Z	−0.202	0.083	−2.441	0.024 **	−0.058
R	0.161	0.01	16.005	0.000 ***	0.232
P	2.136	0.167	12.772	0.000 ***	0.204
C	0.299	0.041	7.243	0.000 ***	0.184
K	−0.128	0.059	−2.186	0.041 **	−0.056
H	0.198	0.028	6.991	0.000 ***	0.202
D	0.863	0.093	9.301	0.000 ***	0.204
R2	0.985
F	F(8, 20) = 160.703, p = 0.000

Note: *** represents p < 0.01, ** represents p < 0.05.

.

The coefficients of the ridge regression passed the 1% significance level test for all variables except for carbon emission intensity (Z) and social passenger turnover (K). The variance inflation factor of each variable is below 1.0, indicating that the multicollinearity is not serious. R² is 0.985, which is a good overall fit, and the F-statistic also passes the 1% significance level test, implying that the eight factors of energy structure, carbon emission intensity, GDP per capita, population size, urbanization rate, social passenger turnover, freight turnover, and tertiary industry share can explain the change of 98.5% of CO₂ emissions in the transport industry. This indicates the rationality of choosing the above eight factors.

Based on the results of the ridge regression analysis, the specific form of the STIRPAT extended model, considering eight influencing factors, is constructed as Equation (7):

lnTC = 1.765lnQ − 0.202lnZ + 0.161lnR + 2.136lnP + 0.299lnC − 0.128lnK + 0.198lnH + 0.863lnD − 23.546

(7)

Convert Equation (7) into exponential form as Equation (8):

TC = 5.84 × 10⁻¹¹Q^1.765Z^−0.202R^0.161P^2.136C^0.299K^−0.128H^0.198D^0.863

(8)

Equations (7) and (8) show that energy structure, carbon intensity, GDP per capita, population size, urbanization rate, social passenger turnover, freight turnover, and the share of tertiary industry will directly affect the CO₂ emissions of the transport industry. Through the fitting of the model to the historical data, the fitting effect is relatively significant, and the comparison between the fitted value and the real value of the model on carbon emissions is shown in Figure 3, which shows that the model can better simulate the growth and changing trend of carbon emissions in the transport industry in Xuzhou. In 1995~2008, the fitted values are basically close to the real value. In 2009~2023, the real value fluctuates up and down around the fitted value, and the error is large.

From the positive and negative effects reflected by the elasticity coefficients, the elasticity coefficients of energy structure, GDP per capita, population size, urbanization rate, freight turnover, and the share of tertiary industry are positive, indicating that there is a significant positive correlation between the above factors and CO₂ emissions, and the changes in these six factors have driven the growth of carbon emissions in the Xuzhou transport industry. Meanwhile, the elasticity coefficients of carbon emission intensity and the social passenger turnover are negative, indicating that these two factors have a negative correlation with CO₂ emissions, and the decline in energy structure and social passenger turnover can relatively inhibit the growth of carbon emissions. The results show that economic and social development drives the growth of demand in the transport industry, which leads to an increase in carbon emissions. However, the decline in energy structure and the proportion of fossil energy will reduce carbon emissions. If the transport sector accelerates the process of transforming its energy mix from fossil to non-fossil, it will strongly contribute to the reduction in carbon emissions. The peak of carbon emissions and the time to reach the peak will depend on the combined effect of these factors.

2.4. Scenario Parameter Settings

This study uses scenario analysis to construct the carbon emissions forecasting model for Xuzhou’s transportation sector. Based on national and Jiangsu Province’s social and economic development plans, the current status of Xuzhou’s transportation sector, the “dual carbon goals” policy, and the domestic and international literature, eight key influencing factors were identified [22,23,24,25,26,27,28]. Three forecasting scenarios, namely the baseline scenario (BEN), low-carbon scenario (LN), and enhanced low-carbon scenario (SL), were developed, with differentiated parameter settings for each scenario in 2024–2025 and 2026–2030 (

Table 4. Growth rate settings for various influencing factors under different scenarios.

Year	Scenario	Rate Setting of Various Influencing Factors/%
		Urbanization Rate	Total Population at the End of Each Year	Per Capita GDP	Proportion of Tertiary Industry	Energy Consumption Structure	Carbon Emission Intensity	Passenger Turnover	Cargo Turnover
2024–2025	BEN	2.0	4.0	6.0	2.0	−0.2	−4.0	13.5	2.4
	LN	1.8	3.0	4.0	2.5	−0.4	−5.0	10.0	2.0
	SL	1.5	2.5	2.5	3.0	−0.6	−8.0	8.0	1.8
2026–2030	BEN	1.5	2.0	5.5	1.5	−0.4	−3.5	4.0	1.5
	LN	1.2	1.5	3.5	2.0	−0.8	−4.0	3.0	1.4
	SL	1.0	1.2	2.0	2.5	−1.0	−6.0	2.0	1.2

).

(1)

Urbanization rate. In April 2021, the Xuzhou Municipal Government issued the Outline of the 14th Five-Year Plan for National Economic and Social Development of Xuzhou City and the Long-Range Objectives Through the Year 2035, which pointed out that, by 2025, the urbanization rate of Xuzhou will reach 73%, and the average annual growth rate of Xuzhou from 2022 to 2025 and 2026 to 2030 will be 2.0% and 1.5%, respectively [27]. Based on this, the growth rate of the urbanization rate in different periods is set at 2.0% in the period 2024–2025 and 1.5% in the period 2026–2030 under the baseline scenario, respectively, and the low-carbon and enhanced low-carbon scenarios are adjusted accordingly;

(2)

Population size. Li et al. found that the registered population of Xuzhou City in 2025, 2030, and 2035 will be 11.271 million, 11.843 million, and 12.4463 million, respectively [29]. Therefore, the growth rate of population size in the baseline scenario for different periods is set at 4.0% for the period 2024–2025 and 2.0% for the period 2026–2030, respectively;

(3)

GDP per capita. According to the “Xuzhou Sustainable Development Plan (2022–2030)”, the annual disposable income of urban residents in Xuzhou in 2030 will be CNY80,000. During the 14th Five-Year Plan, the 15th Five-Year Plan, and the 16th Five-Year Plan periods, the average annual GDP growth rate of key industries was about 5.5%, 5.0%, and 4.2%, respectively [30], and Wu Meng et al. pointed out that the average annual growth rate of Xuzhou’s per capita GDP in the two stages of 2022–2025 and 2026–2030 is 6.0% and 5.5% [27]. Therefore, the growth rate of per capita GDP in different periods is set at 6.0% in 2024–2025 and 5.5% in 2026–2030 under the baseline scenario, respectively, and the low-carbon and enhanced low-carbon scenarios are adjusted accordingly;

(4)

The proportion of the tertiary industry. According to the Outline of the 14th Five-Year Plan for National Economic and Social Development of Xuzhou City and the Long-Range Objectives Through the Year 2035, the output value of high-tech industries will account for 50% of the output value of industries above a designated size in 2025. Therefore, the growth rate of the proportion of the tertiary industry in different periods of the baseline scenario is set at 2.0% in the period 2024–2025 and 1.5% in the period 2026–2030, respectively, and the low-carbon and intensified low-carbon scenarios are adjusted accordingly;

(5)

Energy structure. The “14th Five-Year Plan” Comprehensive Work Plan for Energy Conservation and Emission Reduction and the Action Plan for Carbon Peaking Before 2030 put forward the goal of reducing energy consumption per unit of GDP by 13.5% by 2025 compared to 2020. At the same time, the Xuzhou Sustainable Development Plan (2022–2030) states that renewable energy will account for 25% of Xuzhou’s energy consumption by 2030. Therefore, the growth rate of the energy structure under the baseline scenario in different periods is set at −0.2% in the period 2024–2025 and −0.4% in the period 2026–2030, respectively, and the low-carbon and enhanced low-carbon scenarios are adjusted accordingly;

(6)

Carbon emission intensity. According to the white paper “China’s Policies and Actions to Address Climate Change”, China’s carbon dioxide emissions per unit of GDP will be reduced by more than 65% by 2030 compared with 2005. The “Xuzhou Sustainable Development Plan (2022–2030)” points out that carbon emissions per unit of GDP (t/CNY 10,000) will drop from 1.06 t/CNY 10,000 in 2021 to 0.5 t/CNY 10,000 in 2030. Therefore, the growth rate of the energy structure under the baseline scenario in different periods is set at −1.0% in the period 2024–2025 and −1.5% in the period 2026–2030, respectively, and the low-carbon and enhanced low-carbon scenarios are adjusted accordingly;

(7)

Passenger turnover and freight turnover. According to the 2023 Statistical Annual Report on National Economic and Social Development of Xuzhou City and the 2024 Statistical Annual Report on National Economic and Social Development of Xuzhou City, passenger turnover and freight turnover in 2024 will increase by 20.2% and 3.9%, respectively, compared to 2023. Therefore, the growth rates of passenger turnover and freight turnover in different periods are set at 13.5% and 2.4% in 2024–2025 and 4.0% and 1.5% in 2026–2030, respectively.

2.5. Scenario Analysis

Based on the STIRPAT model and the differentiated parameter settings for the baseline, low-carbon, and enhanced low-carbon scenarios, dynamic simulations were conducted to forecast the carbon emissions from Xuzhou’s transportation sector from 2024 to 2030. The results are shown in Figure 4.

From the forecast results, there are obvious differences in the peak years of carbon emissions under the three scenarios: the CO₂ emissions of Xuzhou’s transportation industry will be 7,290,600 tons, 7,029,200 tons and 6,965,900 tons in 2025, and 10,154,700 tons, 9,072,500 tons and 8,835,000 tons in 2030, respectively, and the CO₂ emissions under the low-carbon scenario and the enhanced low-carbon scenario will be reduced by 10.66% and 13.00%, respectively.

Overall, the Xuzhou transportation sector’s carbon emissions peak value varies significantly under different scenarios. It is important to note that, even when other variables remain stable, the carbon emissions peak will still show a marked upward trend as the population expands. According to Equation (7), the energy structure coefficient (1.765) is larger than the tertiary industry coefficient (0.863), indicating that the emissions reduction effect brought about by technological progress is greater than that resulting from adjustments in the industrial structure.

Currently, carbon emissions in Xuzhou’s transportation sector are a prominent issue, mainly due to the insufficient share of clean energy, low energy efficiency, and high empty load rates during transportation [24,25]. To address these issues, this paper proposes optimization suggestions from three dimensions:

• Policy regulation: It is recommended that government departments formulate a phased low-carbon development roadmap, set net-zero emission targets for the transportation sector, and promote green travel modes. Additionally, energy consumption intensity and carbon emission indicators should be incorporated into the performance evaluation system of transportation management departments and enterprises [26]. It is also necessary to establish and improve carbon tax policies and carbon emissions trading mechanisms;
• Energy consumption optimization: It is suggested to accelerate the application of clean energy, such as electricity and natural gas, in transportation vehicles, gradually replacing traditional oil-fueled vehicles. At the same time, increased investment in the research and development of new energy technologies in the transportation sector should be encouraged, along with the commercialization of research achievements [27,28];
• Optimization of the freight transport structure: It is recommended to develop high-tech solutions such as the Internet of Things (IoT) and artificial intelligence (AI) in transportation [31], promote green freight transport, integrate multiple channels, and push forward the “road-to-rail” strategy for bulk goods.

3. Conclusions

(1)

Xuzhou’s transportation sector’s CO₂ emissions increased from 755,100 tons in 1995 to 7,711,900 tons in 2023, with an average annual growth rate of 8.39%. Among these, aviation and private cars are the primary sources of carbon emissions in Xuzhou’s transportation sector;

(2)

Factors such as population size, urbanization rate, GDP per capita, the share of the tertiary industry, energy structure, and carbon intensity have a significant positive effect on the carbon emissions of Xuzhou’s transportation sector. In contrast, passenger turnover and freight turnover show significant negative moderating effects;

(3)

The decoupling relationship between transport-related carbon emissions and economic development in Xuzhou can be divided into three distinct stages: the first stage is characterized by weak decoupling and expansive negative decoupling, reflecting an extensive mode of economic development; the second stage is dominated by weak decoupling, with expansive coupling as a secondary feature, representing a phase of environmentally-coordinated economic development; the third stage is marked primarily by expansive negative decoupling and weak decoupling, followed by strong decoupling and expansive coupling, indicating a transitional economic development model;

(4)

Under the baseline, low-carbon, and enhanced low-carbon scenarios, by 2030, the CO₂ emissions of the transportation industry in Xuzhou will be 10,154,700 tons, 9,072,500 tons, and 8,835,000 tons, respectively, and the CO₂ emissions under the low-carbon scenario and the enhanced low-carbon scenario will be reduced by 10.66% and 13.00%, respectively. Carbon emission reductions in the transportation sector can be effectively achieved through a combination of policy tools, structural optimization projects, and enhanced technological innovation, thereby advancing the dual strategic goals of high-quality economic development and improved eco-environmental quality.