Impacts of reducing air pollutants and CO2 emissions in urban road transport through 2035 in Chongqing, China.

The road transport sector in megacities is confronted with pressing local air pollution and carbon dioxide (CO2) control issues. To determine effective policy instruments for saving energy and the co-control of air pollutants and CO2, several mainstream measures were examined and compared in Chongqing's road transport sector from 2017 to 2035. An integration assessment framework was developed by combining the Long-range Energy Alternatives Planning (LEAP) system and a set of quantitative methods for evaluating the co-benefits of emission reductions (including the air pollutant equivalent (APeq), co-control coordinate system, and pollutant reduction cross-elasticity (Elsa/b)). Results showed that the shifting transportation modes scenario presented the most significant potential for energy-saving and emission reductions, reducing energy use by 30.9% and air pollutants and CO2 emissions by approximately 27-32% compared with the business as usual (BAU) scenario in 2035. The improving energy efficiency scenario also provided significant co-benefits for reducing air pollutants and CO2 emissions. Nevertheless, the promoting alternative fuel scenario may increase fine particulate matter (PM2.5) emissions by 2.2% compared to BAU in 2035 under the cleanness of regional electricity in 2017. Our findings suggest that the shifting transportation modes were effective measures to reduce air pollutants and CO2 in the short term synergistically, and highlighted the importance of cleaner electricity generation to develop electric vehicles in the medium and long term.

Introduction

The development of transportation is a key driver for improving the living standards in urban areas and acts as a catalyst for economic development [1]. However, transport is also a critical factor leading to resource depletion and a series of environmental problems [2]. In particular, road transport consumes most of the world's oil products [3] and makes a strong negative contribution to local air quality and global climate change [[4], [5], [6]].

In China, the rapid development of road transport has posed a major challenge to the co-control of air pollutants and carbon dioxide (CO2). The mobile source emissions in densely populated megacities of China such as Shenzhen, Beijing, Shanghai, and Guangzhou have become the leading source of fine particulate matter (PM2.5), accounting for 52.1%, 45.0%, 29.2%, and 21.7% of PM2.5 pollution, respectively [7,8]. Moreover, road transport has joined the power generation and manufacturing industries as one of China's major CO2 emission sectors [[9], [10], [11]]. As of 2017, nearly 7.8% of China's energy-related CO2 emissions are generated by road transport (726.3 million tons), and this contribution is increasing annually [10]. According to forecasts, the vehicle stock in China is expected to increase to 500–600 million in 2050 [12,13] from 285 million in 2017 [14]. Given that the vehicle population will continue to increase in the future, city policymakers have made the co-control of air pollutants and CO2 in the road transport sector a central component of long-term sustainability goals.

Numerous studies have focused on estimating energy demand and greenhouse gas (GHG) emissions in the transport sector [1,[15], [16], [17]]. In particular, many researchers have discussed the peak time and peak path of carbon emissions in the middle-long term [[18], [19], [20]]. Ajanovic and Haas [15] compared possible GHG emission reductions due to different policy measures implemented in passenger car transport in the EU-15. They pointed out that GHG emissions could be reduced by 33% in a selected policy scenario compared to a business-as-usual scenario up to 2030. Hao et al. [19] established a bottom-up accounting framework to analyze the possible future trajectories of GHG emissions from China's freight transport sector, and concluded that GHG emissions would peak by around 2035 with all major mitigation measures. Li and Yu [7] developed a National Energy Technology-Transport (NET-Transport) model to assess CO2 emissions for China's urban passenger transport sector under several mitigation scenarios, and proposed that the CO2 emissions may peak around 2020 with the joint implementation of mitigation measures. In addition, some studies have examined the possible co-benefits of climate change mitigation and air pollutant reduction [11,[21], [22], [23], [24], [25], [26]]. Takeshita [25] used a global energy system model to assess the reduction potential of air pollutants from road vehicles under the scenario of atmospheric CO2 stabilization at 400 ppm by 2100. Mao et al. [11] examined the effects of tax and subsidy policies on reducing CO2 and air pollutant emissions in China's transport sector from 2008 to 2050. Geng et al. [27] analyzed the possibility of reducing CO2 and local air pollution emissions by introducing new emission standard vehicles and alternative fuel vehicles in the public transport sector of Shenyang. Chavez-Baeza and Sheinbaum-Pardo [21] developed a bottom-up model to estimate criteria air pollutants (CO, NO, NMVOC (non-nonmethane volatile organic compounds), and PM10) and greenhouse gas (CH4, N2O, and CO2) emissions caused by passenger vehicles under a baseline scenario and two mitigation scenarios in the Mexico City Metropolitan Area. Alam et al. [28] discussed the co-benefits of using climate change mitigation policies to reduce PM2.5 and CO2 emissions from passenger cars in Ireland. Liu et al. [29] conducted a cost-benefit analysis of the measures in place to reduce GHG and air pollutant emissions from the transport sector in the Pearl River Delta region by 2020. Alimujiang and Jiang [30] analyzed the synergy and co-benefits of reducing air pollutants and CO2 by using electric private cars, taxis, and buses in Shanghai from 2010 to 2016. Jiao et al. [31] evaluated the synergies and cost-effectiveness of emission mitigations on CO2 and air pollutants from the urban transport sector under different control measures in Guangzhou. The studies mentioned above have conducted policy analyses in China and worldwide, and have mostly focused on special vehicle types or urban passenger transport sectors. Studies focused on the entire urban road transport sector (including freight transport, intercity passenger transport, and urban passenger transport) in large cities in China are still limited.

In this analysis, we quantitatively evaluated the energy conservation and co-benefits of reducing air pollutants (including carbon monoxide (CO), sulfur dioxide (SO2), nitrogen oxides (NO), PM2.5), and CO2 emissions through several mainstream mitigation measures that are likely to be implemented in the near future or have already been implemented in Chongqing's road transport sector from 2017 to 2035. Specifically, a LEAP-CQRT (Chongqing's road transport) model based on the Long-range Energy Alternatives Planning (LEAP) system was first proposed to calculate energy consumption, air pollutants and CO2 emissions. Furthermore, the air pollutant equivalent (APeq), co-control coordinate system, and pollutant reduction cross-elasticity (Elsa/b) were applied to identify the co-benefits of reducing air pollutants and CO2 emissions. The results are expected to provide significant insights for energy and environmental policymakers, not only in Chongqing but also in cities with similar topologies.

The remainder of this paper is structured as follows. Section 2 introduces the methodology and data. The Scenario design is set up in Section 3, and the results and discussion are subsequently analyzed in Section 4. Conclusions and policy implications are offered in Section 5.

Methodology and data

LEAP-CQRT model establishment

The LEAP system is an integrated modeling tool that can be used to track energy consumption, air pollutants and GHG emissions in all sectors of an economy [32]. The LEAP model has been widely adopted in the power industry [33,34], iron and steel industry [35], cement industry [36], and transport sector [18,37], as well as in multi-sectoral [38,39] energy demand and pollution emission studies. The modeling framework within LEAP is a bottom-up structure including four activity levels: sector, sub-sector, end-use, and device. Thus, the study divided Chongqing's road transport sector into three sub-sectors according to the characteristics of the road transport sector: highway freight transport, intercity highway passenger transport, and urban passenger transport. Urban passenger transport was further subdivided into public transport and private transport, and each sub-sector was further divided into different traffic types (such as different vehicles and different fuels). The specific classifications are shown in Table 1.

Table 1

The dendritic structure of Chongqing's road transport in the LEAP-CQRT model.

Sector	Subsector		End-use vehicles	Consumed energy
Road transport	Highway freight transport		Light trucks (LT)	Diesel, Gasoline, CNG, Electricity
			Medium trucks (MT)	Diesel, Gasoline, LNG
			Heavy trucks (HT)	Diesel, Gasoline, LNG
	Intercity highway passenger transport		Medium intercity buses (MIB)	Diesel, Gasoline, LNG, Electricity
			Large intercity buses (LIB)	Diesel, Gasoline, LNG, Electricity
	Urban passenger transport	Public transport	Urban buses	Diesel, Gasoline, CNG, Electricity
			Rail transit	Electricity
			Taxis	Gasoline, CNG, Electricity
		Private transport	Private cars	Diesel, Gasoline, CNG, Electricity
			Motorcycles	Gasoline

Notes.

CNG: Compressed Natural Gas; LNG: Liquefied Natural Gas.

LT: Length<6 m, Total mass<4.5 t; MT: Length≧6 m, or 4.5 t ≦ Total mass≦12 t; HT: Total mass≧12 t. MIB: Length<6 m, 10–19 seats; LIB: Length≧6 m, or More than 20 seats.

Energy consumption calculation

The energy consumption in the road transport sector can be calculated based on the volume of traffic turnover and energy consumption per unit traffic turnover, as shown in Eq. (1):where EC is the total energy consumption (million tons of coal equivalents (Mtce)) in the t year; FPT represents traffic turnover (ton-km or person-km) of the type i vehicle with type j fuel in the t year; EI denotes energy consumption per unit traffic turnover of the type i vehicle with type j fuel in the t year; and i, j, and t indicate vehicle type, fuel type and calendar year, respectively.

Calculation of air pollutants and CO2 emissions

For specific types of air pollutants or CO2, the following Eq. (2) is used to calculate the emissions of traditional fuel (Gasoline, Diesel, LNG, and CNG) vehicles.where FE is the type k pollutant emissions of traditional fuel vehicles in the t year; FEF represents type k pollutant emission factor of the type i vehicle with type j fuel in the t year; and k denotes four kinds of air pollutants (CO, SO2, NO, and PM2.5) or CO2.

As electric vehicles release almost zero emissions during driving, it is necessary to consider the pollutants emitted from power generation. Therefore, the air pollutants and CO2 emissions produced by electric vehicles can be expressed by Eq. (3):where EE is the type k pollutant emissions of electric vehicles in the t year; EC is the electric power consumption of the type i electric vehicles in the t year; EEF represents the type k pollutant emission factor of the type i electric vehicles in the t year; and L denotes the electric power transmission loss rate from the output zone to the input zone, calculated at 5.9% [40].

Co-control effect assessment

Reduction effect normalization

Mitigation measures can achieve emission reduction or an increase in different types of pollutants simultaneously. In this study, the air pollutant equivalent (APeq) was used to compare the co-control effects of different mitigation measures on air pollutants and CO2, according to Mao et al. [41]. The APeq combines all of the pollutants (CO, SO2, NO, PM2.5, and CO2 in the present study) into one “integrated” pollutant to reflect the total emission reduction effects, and the normalization formula is as follows:where ΔCO2, ΔCO, ΔSO2, ΔNO, and ΔPM2.5 are the emission reductions of CO, SO2, NO, PM2.5, and CO2 by the implementation of mitigation measures, respectively; and α, β, γ, δ, and ε are the weight factors of corresponding air pollutants or CO2, respectively.

The relative weight factors α, β, γ, δ, and ε are intended to reveal the relative importance of different pollutants in terms of the real externalities. In the weight factor scenario 1 (the primary scenario), the weight factor values were calculated according to the pollutant emission pricing (in RMB yuan based on the 2017 price term throughout this study) obtained from the Environmental Protection Tax Law and the Carbon Emissions Trading Market in the case regions (Table 2). After normalization, the APeq calculation formula can be written as follows:

Table 2

The weight factors of air pollutants and CO2.

Weight factors scenario	Indicator	CO	SO2	NOx	PM2.5	CO2
Scenario 1 (primary scenario)	Pollutant emission prices (yuan/kg)	0.21	3.68	3.68	1.61	0.008
	Weight factors	0.0229	0.4005	0.4005	0.1752	0.0009
Scenario 2	Pollutant trading prices (yuan/kg)	0	0.976	1.20	0	0.008
	Weight factors	0	0.4469	0.5495	0	0.0037
Scenario 3	Damage costs (yuan/kg)	0	85.0	166.2	959.5	0.8
	Weight factors	0	0.0702	0.1372	0.7920	0.0006

The uncertainty of this price weighting method will be addressed by the sensitivity analysis presented in section 4.3.1. The weight factor scenarios 2 and 3 will be compared with scenario 1.

Co-control effects coordinate system

According to Zeng et al. [42], the co-control effects coordinate system was designed and intuitively reflected the emission reduction effect and synergy status of mitigation measures. As shown in Fig. 1, the points located in the first quadrant represent positive reduction effects, suggesting that the measures can reduce both air pollutants and CO2. Conversely, the negative synergistic effect (reducing one pollutant emission but increasing the other, or increasing both air pollutants and CO2 emission) is shown in the other quadrants.

Fig. 1

Schematic of co-control effects coordinate system. Source: Mao et al. [43,44].

Moreover, the angle α or β formed by connecting the point located in the mitigation measures to the origin can indicate the degree of co-benefits. A larger angle indicates that the CO2 reduction effect in the first quadrant is greater when reducing the same air pollutants. For example, measure A and E have the same emission reduction effects for air pollutants, while measure E displays better reduction effects for CO2 than measure A (Fig. 1). Therefore, the co-benefits of measure E are better than those of measure A. If two points share the same angle with the horizontal axis, the point farther away from the origin has more significant co-control effects. For example, measure F has better co-benefits than measure A.

Pollutant reduction cross elasticity

The cross-elasticity of pollutant reduction was used to evaluate the synergy degree of mitigation measures for reducing air pollutants and CO2 based on Mao et al. [43]. Pollutant reduction cross-elasticity is calculated by Eq. (6):where Els is the cross-elastic coefficient of emission reductions; Δa/A represents the rate of change in CO2 emissions; and Δb/B is the rate of change in specific types of air pollutant emissions.

If Elsa/b < 0, the mitigation measures can only reduce emissions of one pollutant emissions, which indicates negative co-control effects. Conversely, the mitigation measures have a co-control effect for air pollutants and CO2 if Elsa/ > 0 (Δa/A > 0 and Δb/B > 0). The emission reduction measure has no synergy if Elsa/b = 0. Furthermore, the degree of reduction is equal for air pollutants and CO2 in the case of Elsa/b = 1. If 0 < Elsa/b < 1 (Δa/A > 0 and Δb/B > 0), the degree of air pollutant emission reduction is higher than that of CO2. Otherwise, the degree of CO2 emission reduction is higher than that of air pollutants if Elsa/b > 1 (Δa/A > 0 and Δb/B > 0).

Data source and processing

The data adopted in the study related to the traffic turnover of highway freight and intercity highway passenger transport were mainly derived from the Chongqing Statistical Yearbook [40]. The urban passenger transport turnover is calculated according to Fan et al. [18] and Liu et al. [23]. The detailed information is presented in Section S1 of the Supplemental Material. Additionally, the traffic turnover of each sub-sector from 2017 to 2035 is forecast in Section S1 of the Supplemental Material based on Han et al. [45] and Liu et al. [23].

The data on the energy consumption per unit traffic turnover of different vehicles were provided by the China Automobile Fuel Consumption Inquiry System [46] and related published literature [19,[47], [48], [49]], and adjusted appropriately combined with the actual situation in Chongqing. The specific details are given in Table S3 of the Supplemental Material.

Referring to the Technical Guidelines for Compilation of Air Pollutant Emission Inventory of Road Motor Vehicles [50], the study designed the emission factors for air pollutants from traditional fuel vehicles by considering the latest emission standards, vehicle composition characteristics, and fuel characteristics of Chongqing in 2017. The CO2 emission factors of traditional fuel vehicles were obtained from the 2006 IPCC Guidelines for National Greenhouse Gas Inventories [51], which have been widely applied to measure the CO2 emissions of the transportation sector in China [18,23,24]. The analysis of air pollutants and CO2 emissions from electric vehicles was simplified to emissions from burning fossil fuels during power generation, and the calculation for emission factors of electric vehicles is shown in Section S2 of the Supplemental Material.

Scenario design

Chongqing is a typical mountainous city in the southwest of China, located at across 105°11′–110°11′ E, 28°10′–32°13′ N with an area of 82,402 km2. In 2018, the population of Chongqing was 31.02 million, and the urbanization ratio reached 65.5%. With the in-depth implementation of the national strategy for developing western regions, Chongqing has experienced rapid economic growth since 2000. Its regional GDP reached 2036.3 billion yuan in 2018, up from 179.1 billion yuan in 2000, with an average annual increase of 12.3%. Chongqing was chosen as a case study area because it provides a typical example of a rapidly developing road transport system in a Chinese megacity. Over the past two decades, the average annual growth rate of the motor vehicle population in Chongqing reached 16.2%, far exceeding the GDP growth. To resolve the challenge of energy conservation and achieve the co-control of air pollutants and CO2 emissions in road transport, the central Chinese government and Chongqing city have issued a series of regulations and plans over recent years. For instance, the “Three-year action plan for promoting the adjustment of transport structure (2018–2020)” requires bulk cargo transportation to shift from highways to railways and waterways. “China's 13th Five-Year Development Plan for Urban Public Transport” focuses on the share rate of urban public transport in cities with different population sizes. In addition, many other development plans, such as the “13th Five Years Plan for Energy Conservation and Environmental Protection of Transportation”, “Technology Roadmap for Energy Saving and New Energy Vehicles”, and “New Energy Automobile Industry Development Plan (2021–2035)” have been issued to establish a green, low-carbon, sustainable transportation system in the future. Among these policy instruments, there is an urgent need to evaluate more effective measures in terms of reducing the rapid energy consumption and air pollutants and CO2 emissions expansion in the road transport sector.

The study employed the business as usual (BAU) scenario as a benchmark for developing a comprehensive policy (CP) scenario from 2017 to 2035. The CP scenario was divided into three sub-scenarios: the shifting transportation mode (STM) scenario, the promoting alternative fuel (PAF) scenario, and the improving energy efficiency (IEE) scenario. The qualitative descriptions of the above five scenarios are defined in the following subsections, and the quantification of key parameters is shown in Table S4 in the supplementary materials.

BAU scenario

As a counterfactual baseline scenario, the BAU scenario assumes that no mitigation measures will impact Chongqing's road transport sector in the future, which means that the transportation structure, fuel consumption patterns, and energy efficiency will be maintained at the base year level. However, the traffic turnover of each sub-sector increases with the development of society and the economy. The main role of this scenario is to serve as a reference for evaluating the potential of energy conservation and emission reduction.

STM scenario

The STM scenario is based on the BAU scenario with additional adjustments to freight and passenger transport structure. For the highway freight transport, an optimistic target was set for greater use of railway and waterway freight transport in both the near and the longer-term future, through which the proportion of highway freight turnover in the total freight turnover (including railway, highway, waterway transport in addition to civil aviation) will gradually decrease. Moreover, intercity highway passenger services are assumed to shrink at different rates from 2017 to 2035, with the rapid development of high-speed intercity railways. In urban passenger transport, the proportion of private transport will gradually decrease, and travel by public transport will be promoted, especially rail transit.

PAF scenario

In recent years, the Chinese government has attached great importance to developing alternative clean fuel vehicle technologies, and many policies have been implemented. The PAF scenario is based on the BAU scenario, and describes a situation where additional transport methods will use more alternative fuels such as CNG, LNG, and electricity in the future to reduce the dependence of road transport on petroleum fuels and realize the optimization of the internal fuel structure. Under this scenario, the market penetration rate of clean energy and new energy vehicles for Chongqing's road transport sector will gradually increase.

IEE scenario

Energy efficiency improvement plays an increasingly important role in energy conservation and emission reduction in the transport sector. Currently, the production of motor vehicles requires light design, ultra-low resistance, and higher fuel efficiency. The IEE scenario is based on the BAU scenario but with a consideration for measures to improve energy efficiency in the future. Under this scenario, the energy consumption per unit turnover of each transport mode will be gradually improved according to varying degrees during the forecast period, with the development of automotive technology.

CP scenario

The CP scenario is based on the BAU scenario but integrates the STM, PAF, and IEE scenarios. In order to quantify and compare the potential impact of each individual strategy, it is assumed that the emission factor of motor vehicles is frozen at the 2017 level during each scenario simulation.

Results and discussion

Energy consumption

The energy consumption simulation of the BAU scenario is shown in Fig. 2. Under this scenario, an average increase from 12.6 to 43.3 Mtce was projected between 2017 and 2035, or 3.4 fold increase in 18 years. Aside from the CP scenario, imposing the STM scenario resulted in the most significant potential to mitigate annual energy consumption, followed by the IEE and PAF scenarios. The STM, IEE, and PAF scenarios would reduce energy consumption in 2035 by 30.9%, 19.1%, and 16.5% compared to the BAU scenario, respectively. As expected, a combination of mitigation measures (CP scenario) resulted in the greatest decrease in energy consumption. The CP scenario reduced energy consumption in 2035 by 49.5% compared with the BAU scenario, and the total cumulated energy saved between 2017 and 2035 was estimated to be 159.1 Mtce.

Fig. 2

Energy consumption of road transport under different scenarios during 2017–2035.

To improve our understanding of the findings, we disaggregated energy consumption by vehicle type (Fig. 3). Much of the energy consumed in the base year came from private transport with a share of 49.7%, followed by highway freight transport (37.3%), public transport (8.6%), and intercity highway passenger transport (4.4%). Private cars and heavy trucks were the dominant sources of energy consumption within the sub-sector, accounting for 46.6% and 28.7%, respectively, and other types of vehicles accounted for a small proportion. The BAU scenario would result in a slight shift in 2035 with the reduction in private cars (46.3%) and the increase in heavy trucks (29.7%). Similar structural changes in energy consumption were found in the STM, PAF, and IEE scenarios. Notably, though the energy consumption of private cars in the above three mitigation scenarios was effectively reduced by 32.9%, 31.9%, and 28.7%, respectively, compared to the BAU scenario in 2035, it still dominated with 44.9%, 37.7%, and 40.8% shares of total energy consumption, respectively. However, under the CP scenario, the joint implementation of mitigation measures reduced energy consumption by 63.1% for private cars compared with the BAU scenario in 2035. As a result, heavy trucks became the largest source of energy consumption by Chongqing's road transport, overtaking private cars.

Fig. 3

Energy consumption by vehicle type under different scenarios during 2017–2035. The columns for each selected year represent the BAU, STM, PAF, IEE, and CP scenarios from left to right, respectively.

The structural change in fuel usage from 2017 to 2035 for the five scenarios is shown in Fig. 4. In 2017, the dominant fuels used in Chongqing's road transport sector were gasoline (53.6%) and diesel (39.0%), followed by CNG (6.2%), LNG (0.8%), and electricity (0.4%). Under the BAU scenario, the fuel usage structure would remain the same as that in 2017. Under the STM and IEE scenarios, gasoline and diesel consumption would continue to dominate the energy mix in the future, and both together would account for 88.7% or more of the energy in 2035. However, under the PAF and CP scenarios, as the market penetration rate of clean energy and new energy vehicles increased, the demand for CNG, LNG, and electricity would rise, and the dependency of Chongqing's road traffic on oil products would substantially decrease. The share of CNG, LNG, and electricity would increase from the BAU scenario (6.2%, 0.7%, and 0.4%, respectively) in the PAF (19.4%, 12.1%, and 5.2%, respectively) and CP scenarios (23.3%, 11.7% and 8.6%, respectively) in 2035. Therefore, the PAF and CP scenarios promise an environmentally friendly future with a clean, diverse energy system in Chongqing's road transport.

Fig. 4

Energy consumption by fuel type under different scenarios during 2017–2035. (a) BAU scenario; (b) STM scenario; (c) PAF scenario; (d) IEE scenario; (e) CP scenario.

Air pollutants and CO2 emissions

Air pollutant emissions

The simulation of air pollutant emissions under the five scenarios is indicated in Fig. 5. The results showed that if no mitigation measures were taken in the BAU scenario, the emissions of CO, SO2, NO, and PM2.5 would grow substantially from 2017 to 2035, with increases of 443.5, 75.3, 224.1, and 6.5 Kt, respectively. Compared with previous studies in other cities, the air pollutant emissions were generally predicted to be higher under the BAU scenario in this study [24,52]. Under the STM, IEE, and CP scenarios, the air pollutant emissions from road transport in Chongqing were improved to various degrees. Specifically, compared with the BAU scenario in 2035, the proportions of emission reduction of CO, SO2, NO, and PM2.5 were 29.8%, 32.3%, 31.9%, and 26.7% in the STM scenario, respectively, 18.9%, 13.0%, 12.2%, and 13.2% in the IEE scenario, respectively, and 57.6%, 50.9%, 43.2%, and 30.9% in the CP scenario, respectively. However, though the PAF scenario contributed to reducing CO, SO2, and NO emissions, it had a negative impact on PM2.5 emissions. In fact, the PM2.5 emissions under the PAF scenario would increase by 2.2% compared to the BAU scenario in 2035, due to the fact that PM2.5 emission intensity during the production of electricity was higher than that of conventional fuel vehicles. Overall, although the STM, IEE, and CP scenarios presented emission reductions compared with the BAU scenario, all mitigation scenarios increased emissions by 2035 compared to 2017. If only these mitigation measures mentioned above were implemented for Chongqing's road traffic in the future, the atmospheric environment of Chongqing would be expected to become worse instead of getting better. Our study showed that in order to achieve a green and sustainable road transport system, more radical policies and measures are necessary in addition to those applied in the model.

Fig. 5

Air pollutant emissions of the road transport under different scenarios during 2017–2035. (a) CO; (b) SO2; (c) NO; (d) PM2.5.

The air pollutant emissions by vehicle type under five scenarios from 2017 to 2035 are shown in Fig. 6. The highway freight transport was the largest contributor to emissions of all substances except CO in the base year. In particular, heavy trucks were the dominant source of SO2, NO, and PM2.5, to which they contributed 63.7%, 50.1%, and 45.5%, respectively, of the total road transport emissions in 2017. Compared with the BAU scenario in 2035, the SO2, NO, and PM2.5 emissions of heavy trucks in all mitigation scenarios were effectively reduced, and 38.0%, 63.3%, and 2.3% reduction would be achieved in the CP scenario, respectively. The STM scenario had a better emission reduction effect than the PAF and IEE scenarios. Although private transport accounting for the highest energy consumption compared with the other sub-sectors in the base year (Fig. 3), private transport was a dominant contributor only to the CO emissions. In Particular, the CO emissions from private cars accounted for 36.3% of the total emissions. This is because private cars mainly use gasoline as fuel, and CO is a major pollutant byproduct of gasoline engines. According to the results of the scenario analysis, the STM, PAF, and IEE scenarios could reduce the CO emissions of private cars by 32.9%, 42.0%, and 28.8%, respectively, compared with the BAU scenario in 2035. Joint implementation of the above three mitigation scenarios could reduce CO emissions by 69.6%. However, private cars were still the primary source of CO by 2035, accounting for 34.6%, 29.2%, 31.8%, and 26.0% under the STM, PAF, IEE, and CP scenarios, respectively. In general, our results demonstrated that heavy trucks and private cars were still dominant to air pollutant emissions for any mitigation scenarios because of the high activity level in the road transport sector.

Fig. 6

Air pollutant emissions by vehicle type under different scenarios during 2017–2035. (a) CO; (b) SO2; (c) NO; (d) PM2.5. The columns for each selected year represent the BAU, STM, PAF, IEE, and CP scenarios from left to right, respectively.

CO2 emissions

Similar to energy consumption, CO2 emissions decreased in the four mitigation scenarios compared to the BAU scenario, as shown in Fig. 7. Under the BAU scenario, CO2 emissions were estimated at 25.6 Mt in 2017 and 88.4 Mt in 2035, increasing by 3.4 times in 18 years with an annual growth rate of 7.1%. Compared with studies in other cities, CO2 emissions in the BAU scenario projected in this paper were significantly higher [18,53]. The STM, PAF, and IEE scenarios reduced the total amount of CO2 emissions in 2035 by 30.6%, 18.6%, and 19.0% compared to the BAU scenario, respectively. It is worth noting that the PAF scenario consumed more energy than the IEE scenario in 2035. However, the two scenarios shared a similar emission reduction effect for CO2 due to the cleaner fuel structure in the PAF scenario. Furthermore, when jointly implementing the above three mitigation scenarios, the CO2 emissions from Chongqing's road transport sector would increase to 43.9 Mt in 2035 with an emission reduction rate of 50.3% in the CP scenario compared with the BAU scenario.

Fig. 7

CO2 emissions of road transport under different scenarios during 2017–2035.

The CO2 emissions by vehicle type under the five scenarios from 2017 to 2035 are shown in Fig. 8. A similar distribution structure to that of the energy consumption in Fig. 4 was observed. Specifically, private transport and highway freight transport were the main sources of CO2 emissions in the base year. Among them, private cars and heavy trucks accounted for 45.8% and 30.2% of total emissions, respectively. Under the BAU scenario, the CO2 emissions of private cars and heavy trucks would reach 40.2 and 27.5 Mt in 2035, increasing by 3.4 and 3.6 times compared with the base year, respectively. Under the STM, PAF, and IEE scenarios, private cars would continue to remain the largest contributor to CO2 emissions, with equivalent shares of 44.0%, 37.7%, and 40.0% of total CO2 emissions in 2035. However, the CO2 emissions of heavy trucks gradually surpassed those of private cars under the CP scenario, and heavy trucks became the main source of CO2 emissions in 2035. Moreover, the CO2 emissions from rail transit increased with the fastest annual growth rate of 16.0% under the CP scenario, while these changes had a minor impact on reducing CO2 emissions from road transport in Chongqing as it contributed only a tiny share of total emissions. Predictably, we found that shifting the transportation mode, promoting alternative fuel, and improving energy efficiency could effectively reduce CO2 emissions in Chongqing's road transport. Each mitigation scenario had distinct characteristics with regard to emission reduction, which was determined by the sub-sectoral distribution of CO2 emissions and the mitigation efforts of each sub-sector. Whether the CO2 emissions in Chongqing's road transport sector peak or not will mainly depend on the CO2 emissions of heavy trucks and private cars.

Fig. 8

CO2 emissions by vehicle type under different scenarios during 2017–2035. The columns for each selected year represent the BAU, STM, PAF, IEE, and CP scenarios from left to right, respectively.

The normalized evaluation of emission reduction effect

The comprehensive emission reduction effects of air pollutants and CO2 under different mitigation scenarios in 2035 were compared using APeq, as shown in Fig. 9(a). The APeq values of the STM, PAF, and IEE scenarios were 82.2, 34.6, and 38.5 Kt, respectively, of which the STM scenario had the most significant emission reduction effect. Moreover, the PAF scenario presented a better emission reduction effect for CO and SO2 compared with the IEE scenario (Fig. 5), while the IEE scenario was superior to the PAF scenario in terms of comprehensive emission reduction. The APeq value of the CP scenario was 123.6 Kt, which was 1.5, 3.6, and 3.2 times those of the STM, PAF, and IEE scenarios, respectively. Overall, all mitigation scenarios provided positive emission reduction benefits, which decreased in the following order: CP > STM > IEE > PAF.

Fig. 9

Normalization of emission reduction effects for different mitigation scenarios in 2035.

The normalized evaluation of the emission reduction effect could be different when using different weights to calculate APeq. Therefore, a sensitivity analysis was conducted according to Chongqing's current pollutant trading market scheme and the pollutant damage costs. The sensitivity analysis scenarios were as follows:

Scenario 1: The weight factor values in the primary scenario were obtained mainly based on the Environmental Protection Tax Law and have been analyzed above.

Scenario 2: The use of pollutant trading prices as an approximate reflection of the weight factor values was also an option. We utilized 976, 1200, and 8 yuan/ton (in the 2017 price term) as the trading prices of SO2, NO, and CO2 emissions, respectively, based on Chongqing's 2017 pollutant and carbon emissions trading market. To date, the CO and PM2.5 emissions have not been part of the pollutant trading market in Chongqing. Therefore, the weights of CO and PM2.5 were set to 0 in scenario 2.

Scenario 3: The weight factors were determined in scenario 3 based on pollutant damage costs from environmental and public health economics. The “Handbook on the External Costs of Transport” by the European Commission presented the damage costs of transport-related key air pollutant species in EU-28 (European Union-28), as well as the climate change costs of CO2 [54]. According to the EU handbook, we employed the EU-28 average damage costs of €11.1, €21.7, €125.1, and €101.7 (converted to 2017 price term) for 1 kg SO2, NO, PM2.5, and CO2, respectively. Because the CO was not included in the EU handbook, the weight of CO was set to 0 in scenario 3.

The APeq under different sensitivity analysis scenarios is shown in Fig. 9. In scenario 2, SO2, NO, and CO2 weights increased slightly, whereas CO and PM2.5 were set to 0 (Table 2). Consequently, the APeq values increased, but the sequence of mitigation scenarios was the same as that obtained in scenario 1. In scenario 3, the weight of CO was set to 0. The weights of SO2, NO, and CO2 decreased, whereas the weight of PM2.5 increased. As a result, the APeq values were reduced, but the mitigation scenario sequence did not change. The comparison indicated that the APeq values to different weights varied widely, suggesting that the comprehensive emission reduction effects of different mitigation measures were related to the control target priorities.

Co-control effect for coordinate system analysis

The co-control effects coordinate system under different mitigation scenarios is shown in Fig. 10. All of the coordinates in the STM, IEE, and CP scenarios were located in the first quadrant. Therefore, the implementation of the above three mitigation scenarios all provided the co-benefits of reducing air pollutants and CO2 emissions. However, the PM2.5-CO2 coordinate of the PAF scenario was in the second quadrant, indicating that the mitigation scenario did not sufficiently control PM2.5 emissions, and exhibited negative synergy (Fig. 10 (d)). This occurred for the same reason described in the above section: the PM2.5 emission intensity during the power generation of power plants was higher than that of conventional fuel vehicles.

Fig. 10

CO, SO2, NO, PM2.5, and CO2 co-control effects coordinate system for different mitigation scenarios in 2035.

Based on the angle between different mitigation scenarios and the horizontal axis, each mitigation scenario had different synergistic control effects for different types of pollutants. Specifically, the STM scenario had a higher emission reduction effect on CO2 than the IEE and CP scenarios when reducing the equivalent CO. However, the emission reduction effects of the IEE and CP scenarios on CO2 were higher than that of the STM scenario when reducing the equivalent SO2, NO and PM2.5. The synergistic reduction effect of the PAF scenario on NO-CO2 emissions was the highest among the four mitigation scenarios, while the STM, IEE, and CP scenarios had higher synergistic reduction effects on CO–CO2, SO2–CO2, and PM2.5-CO2 emissions than the PAF scenario. The synergistic effect of the IEE scenario was better than that of the CP scenario for a more substantial CO2 reduction effect when reducing the equivalent CO, SO2, and NO. However, the emission reduction effect of the CP scenario on CO2 was higher than that of the IEE scenario when reducing the equivalent PM2.5.

Pollutant reduction cross-elasticity analysis

The pollutant reduction cross-elasticity was used to further quantify the degree of co-control effect of the four mitigation scenarios. For the STM scenario, the degree of SO2 and NO emission reduction was higher than that of CO2 (0 < Elsa/b < 1), while the degree of CO2 emission reduction was higher than that of CO and PM2.5 (Elsa/b > 1). For the PAF scenario, the degree of CO and SO2 emission reduction was higher than that of CO2 (0 < Elsa/b < 1), while the degree of CO2 emission reduction was higher than that of NO (Elsa/b > 1). In Particular, the PM2.5 emission reduction elasticity coefficient of the PAF scenario was below 0 (Elsa/b < 0), indicating that promoting clean and new energy vehicles provided negative co-control effects. For the IEE scenario, the degree of CO2 emission reduction was higher than that of SO2, NO and PM2.5 (Elsa/b > 1). The CO2 emission reduction elastic coefficients of CO for the IEE scenario were almost equal to 1 (Elsa/b = 1), showing that this scenario could reduce CO2 and CO simultaneously to the maximum potential. For the CP scenario, the degree of CO and SO2 emission reduction was higher than that of CO2 (0 < Elsa/b < 1), while the degree of CO2 emission reduction was higher than that of NO and PM2.5 (Elsa/b > 1).

Sensitivity analysis

To examine the impact of key parameter settings in scenarios on emissions, a sensitivity analysis was conducted. Specifically, we assumed that each key parameter would increase by 20% of the original value. For each calculation, only one key parameter was changed while keeping all others constant. These results were subdivided into three categories: 1) major effect, in which a key parameter increase of 20% led to a change in emissions of greater than 10%; 2) intermediate effect, in which a key parameter increase of 20% led to emission changes between 5% and 10%; and 3) minor effect, in which a key parameter increase of 20% led to a change in emissions of less than 5%.

The sensitivity analysis of key parameters in the mitigation scenarios showed that the sensitivity of air pollutants and CO2 emissions to these parameters varied widely (Fig. S1). CO, SO2, NO, PM2.5, and CO2 emissions were most sensitive to regional GDP under the BAU scenario. In addition, the population under the BAU scenario had a major effect on CO emissions and played an essential role in PM2.5 and CO2 emissions (intermediate effect). In the STM scenario, the share of road freight in the total freight turnover volume had a major effect on SO2, NO, and PM2.5, and an intermediate effect on CO and CO2. The share of public transport also played an essential role in CO and CO2 emissions. In the IEE sub-scenario, the energy intensity of heavy trucks had a major effect on SO2 and NO, and an intermediate effect on PM2.5 and CO2. The energy intensity of private cars had an intermediate effect on CO and CO2. In the PAF scenario, the share of new or clean energy vehicles had a minor impact on CO, SO2, NO, PM2.5, and CO2 emissions. In summary, the accuracy of GDP forecasts is crucial in the simulation process.

Conclusions and policy implications

Conclusions

This study established the LEAP-CQRT model to assess the impacts of several mainstream mitigation measures on the trend of energy consumption, air pollutants and CO2 emissions change in Chongqing's road transport. Furthermore, the co-benefits of reducing air pollutants and CO2 emissions under different mitigation scenarios were examined and compared by adopting a set of quantitative analysis methods. The main conclusions included four key points. 1) The proposed mitigation scenarios could trigger reductions in energy consumption in Chongqing's road transport of 30.9% (STM scenario), 16.5% (PAF scenario), and 19.1% (IEE scenario), while the joint implementation of the above three sub-scenarios had the potential to reduce energy consumption by 49.5% by 2035, compared with the BAU scenario. 2) The STM scenario presented the most significant potential for reducing emissions relative to the PAF and IEE scenarios. It could reduce CO, SO2, NO, PM2.5, and CO2 emissions by 29.8%, 32.3%, 31.9%, 26.7%, and 30.6% compared with the BAU scenario in 2035, respectively. 3) Heavy trucks and private cars were essential sources of pollution emissions in Chongqing's road transport sector, which together accounted for 45.8%, 72.8%, 60.9%, 56.6%, and 70.0% of CO, SO2, NO, PM2.5, and CO2 emissions under the CP scenario in 2035, respectively. 4) The comprehensive emission reduction effect of each scenario from the best to the worst was CP (123.6 Kt), STM (82.2 Kt), IEE (38.5 Kt), and PAF (34.6 Kt). The STM, IEE, and CP scenarios all provided co-benefits for reducing air pollutants and CO2. However, the PAF scenario may increase PM2.5 emissions by 2.2% compared to BAU in 2035, presenting negative synergistic effects on PM2.5-CO2.

This paper had several limitations that require further improvement and exploration. In the current study, the road transport demand is exogenously given, and the price-demand elasticity is not considered. In the follow-up research, a linkage between the LEAP and top-down models such as CGE will be established to fill the gap. Moreover, we only considered exhaust emissions from fossil fuel combustion, neglecting non-exhaust particulate matter emissions, including abrasion (brake, tire, and road surface wear) and road dust resuspension. In further studies, exhaust and non-exhaust emissions could be integrated to provide deeper insights.

Policy implications

Co-control strategies will be an essential step to achieve air pollution and carbon emission reduction targets in Chongqing's road transport sector, especially because China has committed to reaching peak carbon emissions by 2030 and carbon neutrality by 2060. Based on the above research findings, specific policy recommendations were proposed. 1) In the short term, switching to electricity in Chongqing's road transport sector may not effectively reduce local air pollution or CO2 emissions due to reliance on fossil fuel electricity generation (68% coal-fired power in 2018). Chongqing's road transport sector should consider shifting transportation modes from the perspective of technology, especially promoting freight transport from roads to railways or waterways. Our modeling results demonstrate that the STM scenario has the most significant comprehensive emission reduction effect. 2) In the medium and long term, electrification of mobility is a promising option for the co-control of air pollutants and CO2 in Chongqing's road transport sector. In particular, a significant portion of future electricity generation in Chongqing may come from sustainable energy sources in order to achieve carbon peak and neutrality targets by 2030 and 2060, respectively. 3) Finally, improving the energy efficiency of motor vehicles was found to be necessary and should be considered throughout the design of emissions reduction plans and policies. Furthermore, with the increasing challenge of follow-up to constraining vehicle emissions, integrated policies and early action are the keys to tackling air pollutants and CO2 in Chongqing's road transport sector.

This research validates how an integrated assessment framework designed by combining the LEAP model and a set of quantitative methods for evaluating the co-benefits of reducing emissions could provide a helpful tool to assess road transport policies in large cities. In addition, the analytical framework used in this paper could be extrapolated to evaluate energy trajectories and co-control strategies for other sectors such as the household, agriculture, industry, construction, service, and transformation sectors. Furthermore, policymakers could benefit from this study by using our modeling approach to design and examine the effectiveness of different mitigation measures to achieve the co-benefits of reducing air pollutants and CO2.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.