3.1 Temporal characteristics of carbon emissions from road transportation in China
Car ownership in China from 2000–2020 grew from 16.1 million to 278 million, with an average annual growth rate of 15.3% (Fig. 2). China has become the country with the largest car ownership in the world, accounting for 20.0% of the global total (1590 million) (OICA 2022). In terms of vehicle type, small passenger cars have experienced the fastest annual growth rate, growing from a proportion of 34.9% (5.68 million) in 2000 to 85.4% (238 million) in 2020. The new energy cars include fuel cell electric vehicles, hybrid electric vehicles, and battery-powered electric vehicles. All emerged in 2010, with the highest growth rate of 557% in 2011. With the implementation of the dual carbon policy, the battery-powered electric vehicle increased from 181 in 2010 to 4.03 million by 2020, representing an annual growth rate of 172%; higher ownership levels are expected for different car companies (He et al. 2022). In addition, the light truck also experienced a high growth rate of 9.88%, accounting for 7.52% of total ownership in 2020. The increased purchasing power and the improvements in foreign trade conditions in China are the major reasons for this growth (Niu et al. 2016).
Based on car ownership values, Fig. 3 shows the carbon emissions calculated for different vehicles. The total carbon emissions from road transportation in China grew from 11.9 to 22.4 Mt CO2e, from 2000 to 2009, representing an annual growth rate of 7.28%. During the 2000–2009 period, increased in the carbon emissions in China’s transportation sector were driven by the per capita growth in GDP, and a transition in the transportation model from a lower to a higher energy consumption-intensive mode (Wang et al. 2011). After 2010, the carbon emissions from car transportation experienced a slower rate of growth, growing from 11.9 to 33.8 Mt CO2e in 2020, representing an annual growth rate of 5.37%. Meanwhile, as showed in Table 1, China’s total carbon emissions from road transportation are much lower than USA and European Union, while comparable with Malaysia and higher than some European countries. However, in terms of the emission proportion accounting for national total, it was estimated to represent only 0.47% (11500 Mt CO2e) (Table 1), as reported by BP Statistical Review of World Energy Review (2022) (BP 2022). This proportion was much lower in China compared to European and American countries, where the transport industry represents more than 27.0% of total carbon emissions (Marrero et al. 2021).
Carbon emissions for different vehicle types also increase each year. The small passenger car contributed the most to the growth of carbon emissions in the passenger car category, with an increase in proportion from 7.12% (0.85 Mt CO2e) in 2000 to 20.6% (6.98 Mt CO2e) in 2020. Once China officially became a member of the WTO in 2001, the WTO increased the pace at which international car manufacturers entered the Chinese market, and promoted the development of China’s car industry (Gallagher 2003). As a result, from 2000 to 2020, the gap between the carbon emissions of passenger cars and trucks significantly narrowed, and the proportion increased annually from 18.9–31.0% for passenger cars, and decreased from 77.3–63.6% for trucks. However, the largest carbon emissions in 2020 were still from heavy trucks, accounting for 28.3% of the total emissions; while the lowest carbon emissions were from the mini passenger car, accounting for 1.41% of the total emissions. In addition, the proportion of carbon emissions from medium and heavy-duty vehicles (including medium and heavy-duty trucks and medium and large passenger cars) in China is 47.5%, which exceeded that of light-duty vehicles (including small passenger cars and light trucks) (14.6 Mt CO2e, 43.1%). By comparison, light-duty vehicles are always generate the most pollutants in the transportation sector in developed countries, such as the USA, United Kingdom and Italy (Liu and Cirillo 2016; Transport & Environment 2019). It can be seen that the carbon emissions of China's automobile transportation industry come more from economic production than from daily life. In general, trucks consistently contribute more carbon emissions than passenger cars when considering national road transportation. Today, carbon mitigation paths are similar for truck and passenger car transport, and include converting to low-carbon fuels, greater vehicle economy, and mode transfer (from trucks to rail or sea) (Axsen et al. 2020). Based on the present carbon emission management system, more road transport companies, particularly those with many medium and heavy-duty vehicles, should be a focus for the future prevention and control of carbon emissions from road transportation (Ai et al. 2022). More targeted measures and regulations are needed to meet the cargo transportation attributes of trucks, while also reducing carbon in China.
Table 1
Carbon emissions from road transportation in different countries
Country | Year | Carbon emissions / (Mt CO2e) | Proportions accounting total national carbon emissions / (%) | Reference |
Malaysia | 2007 | 59.4 | 26.9 | (EI 2023; Ong et al. 2011) |
USA | 2013 | 1515 | 27.0 | (EI 2023; Liu et al. 2016) |
European union | 2014 | 792 | 25.6 | (EI 2023; Marrero et al. 2021) |
Russia | 2015 | 14.5 | 0.71 | (EI 2023; Eder et al. 2019) |
Italy | 2016 | 96.3 | 28.0 | (EI 2023; Transport & Environment 2019) |
Denmark | 2020 | 6.90 | 24.7 | (EI 2023; Eder et al. 2019) |
Germany | 2020 | 94.8 | 15.2 | (EI 2023; Eder et al. 2019) |
Greece | 2020 | 8.60 | 14.4 | (EI 2023; Eder et al. 2019) |
France | 2020 | 69.8 | 26.9 | (EI 2023; Eder et al. 2019) |
Spain | 2020 | 48.7 | 21.0 | (EI 2023; Eder et al. 2019) |
Turkey | 2020 | 83.6 | 19.7 | (EI 2023; Eder et al. 2019) |
China | 2020 | 53.6 | 0.47 | this study |
3.2 Spatial characteristics of carbon emissions in China
Figure 4 shows the carbon emissions from road transportation at prefecture-level for 2020, with the data for other years shown in Fig. S2-S3. Figure 4A shows the overall regional perspective: carbon emissions from the road transportation system are generally higher in East China and lower in West China. In East China, emissions are mostly concentrated in areas with higher economic development. The distribution of carbon emissions in maps is similar to the distribution shown in Local Indicators of Spatial Association (LISA) agglomeration maps (Cheng et al. 2018) and land urbanization maps (Cao et al. 2021). Based on the PCA conducted for this study, carbon emissions (driven by car ownership) are highly correlated with the urbanization rates.
According to province and prefecture level car ownership levels, and the corresponding coefficient calculations for 2020, the carbon emission from road transportation was 53.6 Mt CO2e. Regional differences are not generally considered in development; therefore, the national total car ownership level and corresponding coefficient are used to calculate the carbon emission from road transportation, which is estimated at 33.8 Mt CO2e. This estimate is significantly lower than the results calculated using provincial and prefecture data. Data released by China’s Ministry of Ecology and Environment indicate that the total amount of carbon emissions from road transportation in China was around 40.2 Mt CO2e in 2018 (MEEPRC 2019). This verifies the rationality of this study’s results (based on provincial and prefecture data in 2020).
From a local perspective, the carbon emissions of prefecture-level cities show an agglomeration pattern, concentrated in Beijing-Tianjin-Hebei (BTH) of North China (ID 0, 1 and 2), the Sichuan Chongqing Urban Agglomeration of Southwest China (ID 27 and 29), the Yangtze River Delta Urban Agglomerations (ID 7, 8, 9, and 10) of East China, and the Pearl River Delta Urban Agglomeration (ID 17) of South China. The carbon emissions of prefecture-level cities in Northwest, Southwest, and Northeast China are all concentrated in provincial capitals. When considering the carbon emission differences between cities, the city with the highest emission level is Guangzhou, with 2.57 Mt CO2e in 2020; this is 800 times more than the levels in the Hainan Tibetan Autonomous Prefecture, which has the lowest emission level of 0.0031 Mt CO2e. Shenzhen, Shanghai, Dongguan, and Beijing all have carbon emission levels of more than 1.40 Mt CO2e, and these four cities and Guangzhou consist of more than 16.3% of national total. Henan province (ID 14) has the longest road length in China; the province had 2.09 Mt CO2e in 2020, ranking sixth among all provinces. It is generally clear that the urbanization rate has a major influence on city carbon emissions, due to the road transportation system. In addition, unsynchronized control methods have resulted in significant disparities in vehicle growth patterns and emissions among cities (Yang et al. 2022). The cumulative carbon emissions in prefecture-level cities in this study are comparable with those published corresponding provincial emissions (Sun et al. 2016; Gu et al. 2019; Liu et al. 2022), indicating this study provides a reliable estimate. For example, Fan et al. calculated the carbon emission of Beijing’s road transportation in 2009 based on COPERT 4 to be 0.88 Mt CO2e (Fan, 2011); this study calculated a value of 0.96 Mt CO2e. The total quantity of carbon emissions in the BTH area has been reported to be around 5.12 Mt CO2e in 2015 (Xing et al. 2016); while this study calculated a value of 6.05 Mt CO2e in 2020.
The carbon emissions density in China follows a pattern of decreasing from the eastern coastal cities to central China; the cities with more carbon emissions in Northeast China show agglomeration, as seen in Fig. 4B. The distribution seen on the carbon emission density map is similar to the GDP density map for the primary sectors (Chen et al. 2021). For each city, the carbon emission densities of Beijing, Shanghai, Guangzhou, and Shenzhen are significantly higher compared to other cities, with emissions of 85.1, 242, 354, and 902 tons CO2e/km2, respectively. In comparison, the lowest emission density is only 0.02 tons CO2e/km2 in the Ali region of Tibet Autonomous Region. There is a 45,147 times difference between the first and the last emission density value in China. In addition, transportation’s overall carbon emissions and growth rate were both comparatively low in undeveloped and sparsely populated areas (Yang et al. 2015), For example, Lhasa’s density of car transportation carbon emissions is 1.94 tons CO2e/km2, and Urumqi’s density of car carbon emissions is 12.4 tons CO2e/km2 in 2020.
Generally, both the total carbon emissions and the emission density are concentrated in city centers, which have better economic development levels. Research has also found that carbon emissions appear in high value clusters in the center of Beijing (Zheng et al. 2022); this value relates to the total carbon emissions and economic aggregation (CASTJ et al. 2022). Chai et al. applied ARIMA and ETS models to forecast China’s road traffic energy consumption and carbon emissions in 2016. These findings indicate that the increased road traffic, energy consumption, GDP, and urbanization are all interconnected (Chai et al. 2016). Similarly, China’s cities are all experiencing urban expansion, leading to more transportation routes, vehicles, and carbon emissions.
3.3 Future forecast
3.3.1 Future car ownership in China
Figure 5 shows the projected future car ownership in China. The PCA-BP combined model predicts an expected rapid annual growth rate in car ownership in the early stage (2021–2049), with an annual 3.62% growth rate, given the influences of total national population, GDP, per capita GDP, and urbanization rate. From 2050 to 2060, the annual growth rate is expected to gradually slow to 1.15%. More specifically, the growth rate of car ownership was expected to reach a maximum of 8.97% in 2022, and is predicted to reach a minimum of 1.10% in 2060. As China’s economic development and population growth stagnate, the growth rate in the vehicle population in each city is expected to steadily decrease (Zeng et al. 2016). For example, the growth rate is expected to decrease significantly in 2023; however, car ownership could still reach 472 million, which is about 28 times higher than in 2000. China’s car ownership counts are expected to be 472, 808, and 906 million by 2030, 2050, and 2060, respectively, which would be 29.4 times, 50.2 times, and 56.3 times higher than in 2000, respectively. It is expected that China’s per capita car ownership would be 0.33, 0.62, and 0.76 by 2030, 2050, and 2060, respectively, which would be 26.3 times, 49.1 times, and 59.7 times higher than in 2000.
China’s car ownership in 2060 is predicted to be mainly achieved through the rapid growth of small passenger cars, accounting for 87.0% of all vehicles. As the Chinese citizen income level continues to improve, per capita GDP has significantly impacted small passenger car ownership. China’s car ownership levels would be higher when there is a high per capita GDP level (Wu et al. 2014). These predicted results are comparable with published reference values and government reports. For example, if there are no restrictions on car sales, car ownership should reach 384 (343–432) million by 2030 (Gan et al. 2020), which is close to this study’s prediction results. Hsieh et al. cited inadequate national car ownership statistics in assessing the unpredictability of China’s car ownership projections, which range from 200 to 700 million by 2040 (Hsieh et al. 2018). This is consistent with this study’s estimated car ownership level of 659 million in 2040.
It is predicted that the car ownership in China in 2050 may reach a total of 498 million (Ou et al. 2010), ranging between 530 and 623 million (Huo et al. 2012a). A previous research (Peng et al. 2018) predicted China’s car ownership would reach 543 million by 2050. In addition, the National Development and Reform Commission’s Energy Research Institute (ERI) in China estimated this number to be 501 million cars (excluding motorbikes) by 2050 (Kejun et al. 2010). These research results are used to conservatively predict China’s car ownership in 2050. Based on the discussion in section 3.1, China’s average annual growth rate from 2000 to 2020 is estimated to have reached 15.3%; in 2020, car ownership was close to 280 million. Therefore, this study’s predicted car ownership for 2021–2060 is aligned with China’s actual situation.
3.3.2 Future carbon emissions for different scenarios
Future carbon emissions from road transportation in China relate closely to the car type and emission coefficient, as shown in Fig. 6. Scenario 1 keeps the car ownership growth rate based on the historical value for all car types; as such, it may be a conservative carbon emission prediction, yielding 33.7 Mt CO2e by 2030 and 39.1 Mt CO2e by 2060. Under Scenario 1, the growth rate of carbon emissions from China’s road transportation decreased from 8.77% in 2001 to -4.58% in 2021, with the growth rate decreasing each year. For the overal growth trend, based on the historical growth rate of car ownership, and the most stringent emission standard reflected in China VIb (Liu et al. 2023), three growth stages for carbon emission of road transportation are seen. There is a sharp decline between 2021 and 2024, significant growth between 2025 and 2049, and moderate growth between 2050 to 2060. Correspondingly, the annual growth rates for these three time spans are projected to be -0.79%, 0.79%, and 0.14%, respectively, resulting in 1.16 times more carbon emissions from 2021 to 2060. The transportation industry is expected to have a large effect on whether China can reach its carbon dioxide emissions peak by 2030 as planned, and on future trends after the peak (Wang et al. 2022).
After 2040, the carbon emissions corresponding to Scenarios 1 and 2 are symmetrically distributed on the straight line, at 32.0 Mt CO2e. Scenario 2 involves the upgrade of fuel vehicle emission standards to improve their performance and achieve the goal of reducing emissions. Based on China’s complete implementation of the China VIb emission standard in 2024, automotive carbon emissions are expected to fall significantly in 2024, climb from 2025 to 2026, and then fluctuate every three years. As a result, in Scenario 2, future carbon emissions are expected to be much lower than those predicted in Scenario 1. The trajectory of the change curves of Scenarios 1 and 2 from 2033 to 2060 indicates an approximately symmetrical distribution. The increase in fuel vehicles can somewhat increase the national carbon emissions, but the increase in emission standards reduces carbon emissions to a greater extent. Based on lessons learned from industrialized nations, implementing progressively stricter car emission regulations may be the most successful of all pollution control measures implemented in China (Wu et al. 2017). Therefore, scenario 2 of carbon emissions coefficient reduction is an effective and reasonable way to reduce carbon, and may help achieve China’s “double carbon” goal. Under this assumption, carbon emissions are expected to fall at an annual rate of -0.46% until 2060, decreasing to 27.4 Mt CO2e. The results indicate that applying new emission standards increases fuel economy while decreasing carbon emissions (Sun et al. 2019). Therefore, it is critical to investigate the costs and potential carbon emissions reductions associated with adopting a variety of low-carbon solutions in fuel and car exhaust (Gambhir et al. 2015).
Scenario 3 is an idealized upgrade based on Scenario 1 as it directly replaces all fuel vehicles with new energy vehicles, reducing the carbon emissions at the source. Leading the future growth of car ownership through new energy vehicles is critical for reducing carbon emissions. The new energy vehicles should ideally be completely battery-driven electric vehicles, based on the application of the China IV standard requirements. Battery-driven electric vehicles do not emit carbon emissions during road transportation; therefore, Scenario 3 shows a noticeable downward trend in 2021; the carbon emissions from China’s road transportation are expected to be reduced to 27.2 Mt CO2e and maintained after that. Thus, China is expected to reduce end-user carbon emissions by 19.7% from 2020 levels by 2030. This highlights the need to stimulate the manufacture and purchase of new energy vehicles to reduce carbon emissions created by all vehicle types (Iwan et al. 2019).
Comparisons among Scenarios 1, 2, and 3 show that increasing the number of battery electric vehicles may significantly reduce carbon emissions. Pan et al. also found that a “full conversion” scenario could cut vehicle activity emissions by 95.0%; they modeled four scenarios based on varying battery electric vehicles adoption rates (Pan et al. 2019). Another study found that the widespread future use of battery electric vehicles could reduce carbon emissions by approximately 75.0% (Pan et al. 2021a).Therefore, the forecast for carbon emissions from road transportation indicates that battery-driven electric vehicles may help decarbonize the car industry (Rinawati et al. 2023), and also lower road transportation emissions. Nevertheless, new energy vehicles still face practical problems related to battery energy storage and acquisition costs (Arfeen et al. 2020). Technological advancements and stricter emission regulations are essential considerations for reducing excess emissions in the petroleum and energy sectors generated by fuel vehicles and new energy vehicles (Hofmann et al. 2016). Overall, Scenario 2 is the most effective and executable approach. Scientifically reducing emissions and innovatively reducing the carbon emission coefficient is expected to become a mainstream trend in the car industry to work toward the “double carbon” goal. It is expected that the effect and efficiency of carbon reduction in the car industry should also be significantly improved.