Emission Characteristics of Air Pollutants and CO2 from 11 Cities with Different Economic Development around the Bohai Sea in China from 2008–2017

Cities around the Bohai Sea are one of the main population cluster areas in China, which are characterized by high levels of sustainability performance and human capital, as well as resource-intensive industries. In this study, levels of economic development metrics and emissions of air pollutants (BC, CO, NH3, NOx, OC, PM2.5, PM10, and SO2) and CO2 across eleven cities around the Bohai Sea from 2008 to 2017 were compared to illustrate the potential relationships between air pollutants/carbon emissions and socioeconomic developments. Meanwhile, the associations between the levels of economic development metrics (GDP per capita), emissions, and energy use per GDP have also been examined. Large differences across these 11 cities presenting different economic development levels and energy consumption characteristics have been observed. Cities with development dependable on the consumption of fossil fuels and the development of resource-intensive industries have emitted large amounts of air pollutants and CO2. Furthermore, the emissions and energy use per GDP for all the cities follow environmental Kuznets curves. The comparison results suggested that the developing cities dependable on resource-intensive industries around the Bohai Sea would obtain greater socioeconomic benefits owing to the interregional cooperation policies under top-down socioeconomic development plans and bottom-up technology development, accompanied by reduced emissions of air pollutants and CO2.


Introduction
Sustainable development of society faces steep challenges regarding climate change and atmospheric pollution with several pollutants accounting for both issues [1][2][3]. Climate change resulting from the emission of greenhouse gases (e.g., CO 2 , CH 4 , N 2 O, O 3 ) could reduce food production and pose health burdens to residents through extreme heat and weather events (i.e., floods and droughts) in vulnerable areas where countries are located [1,2,4]. Meanwhile, Low-and Middle-Income Countries (LMICs) pursue economic development with high dependence on the consumption of natural sources (e.g., forests, minerals) and fossil fuels (e.g., coal and crude oil) [5]. A study estimated that LMICs countries were projected to consume primary energy (i.e., fossil fuel) as much as double by 2040 from 2016, while the demand for primary energy in high-income countries may decrease by 8% from 2016 to 2040 [5,6]. High-income countries contributed to high fractions (50-70%) of global greenhouse gas emissions in the atmosphere [3]. Since CO 2 emissions per capita are found to be correlated with socioeconomic development, large amounts of air pollutants (e.g., PM 2.5 , black carbon (BC), SO 2 , NO x ) and CO 2 are projected to be emitted with the development of LMICs countries, which could change regional weather patterns and increase health risks due to exposure to air pollution.
Nearly 95% of cities in LMICs did not achieve the air quality guideline of 10 µg m −3 for yearly average PM 2.5 in 2016, whereas half of the cities in high-income countries met the Toxics 2022, 10, 547 2 of 17 guideline [7][8][9]. Cities experiencing environmental degradation and climate change have limited capacities to manage and adapt to these risks because of the challenges associated with developments [10].
Prior studies have documented the long-term relationships between GDP per capita and CO 2 emissions per capita in cities of representative countries worldwide [3,11]. This relationship follows the environmental Kuznets curves (EKC), where emissions first increase, then decrease with income growth [12,13]. Liu et al. [3] demonstrated that CO 2 emissions per capita were highly associated with GDP per capita and the human development index (HDI) in 184 countries worldwide. Awan et al. [11] examined the validity of the EKC hypothesis using CO 2 per capita and economic developments in five countries, including China, France, the Russian Federation, the UK, and the USA from 1993 to 2017. A study by Ru et al. [14] indicated that long-term associations between income growth and emissions of SO 2 and CO 2 were dependent on energy sectors globally using the country-level emission inventory. However, the emission of BC did not follow the EKC pattern. The emissions of CO 2 from the industrial and residential sectors, as well as the emissions of SO 2 from the power and industrial sectors, obeyed the EKC curve, while emissions of SO 2 and CO 2 from other sectors did not.
The United Nations 2030 Agenda appeals for the 17 sustainable development goals (SDG) of cities (e.g., no poverty, good health and well-being, affordable and clean energy, economic growth, climate action) in the inclusive, safe, resilient, and sustainable pathway [15][16][17]. Many countries including China pledge sustainable development by incorporating the 2030 Sustainable Development Goal (SDG) agenda into their national development plan [16]. It was estimated that China may have been responsible for~30% of global CO 2 emissions in 2019 [18]. The Chinese central government is committed to reducing the CO 2 emissions per GDP by 13.5% in 2025 relative to that in 2020, as well as 65% in 2030 compared to 2005, to fulfill the responsibility for the support of global sustainability [18]. In addition, northern citizens in China have been choked with severe air pollution with PM 2.5 higher than 75 µg m −3 over the last few decades [9,19,20]. To achieve the 2030 SDG goals, a top-down scheme of macro-level governance for the priority regions accounting for air pollution and CO 2 in China is essential and should be recognized for effective local governance and regional coherence of national SDG implementation [17].
The Bohai Sea is the largest inland sea in northern China, and its continental shelf is a mining area for natural gas, crude oil, and coal [21][22][23]. The littoral zone of the Bohai Sea is an economically developed region with a high population density in China [24,25]. The socioeconomic development of some cities in this region relies on the large consumption of specific energy resources originating from fossil fuels [26]. Large-scale industries including iron and steel smelting are densely distributed in this region [27][28][29]. For instance, Tangshan in Hebei Province along the littoral zone of the Bohai Sea is the largest producer of iron and steel among cities in the world [30]. It produced approximately 80 million tons of iron and steel in 2012, which is almost equal to the national steel production in the United States [31,32]. Therefore, high-intensity emissions of greenhouse gases and air pollutants from the combustion of fossil fuels have been observed in this region [18,33]. Several field studies have documented extreme air pollution and severe haze events over the coastal region of the Bohai Sea where large-scale industries exist [33][34][35].
Aiming to contribute to co-benefit strategies for city-level sustainable development as well as reductions in emissions of air pollution and CO 2 in the littoral zone of the Bohai Sea, our study is performed by assessing the relationships between emissions of air pollution/CO 2 and GDP in 11 representative cities with different development levels and capital structures. The metrics of air pollution include BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 from different sectors (industrial, power, residential, agriculture, and transportation), as well as GDP from the primary sector (agriculture), the secondary sector (mining and manufacturing), and tertiary sector from 2008 to 2017, are adopted in the comprehensive analysis. This study can provide policy guidance for the potential Toxics 2022, 10, 547 3 of 17 of diverse sustainable development pathways in the context of air pollution and climate mitigation in the short term in those respective cities.

Description of Studied Cities
In this study, 11 cities, namely, Beijing, Tianjin, Dalian, Yingkou, Panjin, Jinzhou, Qinhuangdao, Tangshan, Dongying, Weifang, and Yantai, located along the littoral zone of the Bohai Sea were selected to evaluate the impact of greenhouse gas emissions and air pollution on socioeconomic development from 2008 to 2017 (Table 1). As illustrated in Figure S1, four cities, Dalian, Yingkou, Panjin, and Jinzhou, are in Liaoning Province and are distributed in the northeast of the Bohai Sea. The cities of Qinhuangdao and Tangshan are located in the north of the Bohai Sea and are subordinated to Heibei Province. Beijing and Tianjin are municipalities directly under the Chinese government and are situated in the central location of the Bohai Sea. Three cities, Dongying, Weifang, and Yantai, are affiliated with Shandong Province and are located south of the Bohai Sea.

Emission Dataset
The Multi-resolution Emission Inventory for China (MEIC) provides gridded total emissions and emissions by sector of greenhouse gases (CO 2 ) and air pollutants (i.e., BC, CO, NH 3 , NO x , organic carbon (OC), PM 2.5 , PM 10 , and SO 2 ) at 0.5 • × 0.5 • over land by combining the bottom-up technical method with the latest emission inventories in China, respectively [36,37]. The emission inventories in the MEIC include unit-based emission inventories for power plants and cement plants, a high-resolution country-level vehicle emission inventory, a residential combustion emission inventory based on national survey data, and an explicit profile-based non-methanevolatile organic compound (NMVOC) speciation framework [38,39]. The data area available from 2008 to 2017 at http://www.meicmodel.org (accessed on 1 November 2021). We selected the MEIC model because it provides sufficiently high spatial-temporal resolution outputs compared with other alternative models [36]. Furthermore, the accuracies of the results from the MEIC model are proven to be consistent with several outputs estimated from other alternative models (e.g., Transport and Chemical Evolution over the Pacific, TRACE-P; Intercontinental Chemical Transport Experiment-Phase, INTEX-B; Model Inter-Comparison Study for Asia, Hemispheric Transport of Air Pollution, HTAP) [40][41][42][43]. The monthly total emissions and emissions by sector of greenhouse gas (i.e., CO 2 ) and air pollutants (i.e., BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 ) for the 11 studied cities from 2008 to 2017 were estimated based on the sum of emissions of all the gridded cells (0.5 • × 0.5 • ) over the city. The dataset of greenhouse gas and air pollutants for Jinzhou City during the periods from 2014 to 2017 was absent. The annual emissions of each city are the sum of monthly emissions in one year. It is speculated that city-level estimates are accurate within 20% using gridded emissions, whereas comparisons across cities tend to be more accurate (<10% error) ( Table 2).

Socioeconomic Data
Economic predictors include GDP, GDP of the primary sector (agriculture), GDP of the secondary sector (industry), and GDP of the tertiary sector (excluding agriculture and industry) and are summarized from the China City Statistical Yearbook.
Population data from 2008 to 2017 for 11 cities were obtained from the China Statistical Yearbook and are used to estimate the GDP per capita and GDP per capita of the industry sector using Equations (1) and (2), respectively. GDP per capita = city-level GDP/city-level population (1) GDP per capita of the industry sector = GDP of the industry sector/city-level population (2) The energy use for each city in the form of standard coal consumption was collected from the City Statistical Yearbook and were used to calculate the energy use per GDP with Equation (3). In addition, air pollutants and CO 2 per GDP in each city were estimated using Equations (4) and (5).
Energy use per GDP = city-level energy use/GDP Air pollutants per GDP = city-level air pollutant/GDP (4) CO 2 per GDP = city-level CO 2 /GDP (5)

Statistics Analysis
We assessed the correlation between energy use per GDP and air pollutants (BC, CO, CO 2 , NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 ) per GDP as well as CO 2 per GDP using the Pearson correlation. Moreover, the correlation between economic development metrics including GDP, GDP per capita, GDP of the industry sector, and GDP per capita of the industry sector with air pollution as well as CO 2 emissions were performed with the Pearson correlation. The statistical analysis was conducted using the Statistical Package for the Social Sciences (IBM SPSS Version 20.0, Armonk, NY, United States).

Monthly Trends in Emissions
The monthly trends in emissions of BC, CO, OC, PM 2.5 PM 10 , and SO 2 in the eleven cities from 2008 to 2017 presented clear cycles with greater emissions in cold seasons (November, December, and January) and lower emissions in warm seasons (June, July, and August). In contrast, higher emissions of NH 3 were observed in warm seasons (June, July, and August) relative to those of cold seasons (November, December, and January) in all the cities from 2008 to 2017 ( Figures S2-S12). The emissions of CO 2 and NO x from 2008 to 2017 had relatively constant levels with no significant monthly trends ( Figures S2-S12).
These cities presented large differences in the orders of magnitude in emissions of air pollutants (i.e., BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 ) and CO 2 . For BC, the monthly levels were comparable within the same orders of magnitude, with the highest level in Tianjin (0.8-1.8 metric kilotonnes) and the lowest (0.15-0.25 metric kilotonnes) in Panjin ( Table 3). The monthly emissions of NH 3 and OC were within the orders of magnitude across 11 cities, which varied from 0.8-6.0 metric kilotonnes and 0.2-2.8 metric kilotonnes, respectively. The highest monthly emission of CO 2 was found in Tangshan (12-19 metric million tonnes), followed by Beijing (8.0-16 metric million tonnes), Tianjin (6.0-16 metric million tonnes), Dongying (2.5-6.0 metric million tonnes), Yantai (4.5-5.6 metric million tonnes), Dalian    (Table 3).

Annual Trends in Emissions
The annual levels of air pollutants, including BC, CO, NO x , OC, PM 2.5 , and PM 10 ,for the eleven cities decreased gradually from 2008 to 2017, though some peak emissions occurred at the periods of 2011 and 2013 for some cities (Figure 1). The annual levels of BC for the 11 cities varied from 2 to 14 metric kilotonnes (Figure 1a,b). The large differences in annual CO emissions were seen across the eleven cities from 2008 to 2017. The levels of annual CO in three cities (Beijing, Tianjin, and Tangshan) ranged from 2000 to 3000 The annual levels of air pollutants, including BC, CO, NOx, OC, PM2.5, and PM10,for the eleven cities decreased gradually from 2008 to 2017, though some peak emissions occurred at the periods of 2011 and 2013 for some cities (Figure 1). The annual levels of BC for the 11 cities varied from 2 to 14 metric kilotonnes (Figure 1a,b). The large differences in annual CO emissions were seen across the eleven cities from 2008 to 2017. The levels of annual CO in three cities (Beijing, Tianjin, and Tangshan) ranged from 2000 to 3000 metric kilotonnes, and the levels of annual CO in four cities (Dalian, Dongying, Yantai, and Weifang) were in the range of 200-1500 metric kilotonnes. In comparison, three cities (Panjin, Jinzhou, and Yinkou) emitted an annual CO lower than 200 metric kilotonnes (Figure 1c,d). The levels of NOx in two cities (Tangshan and Tianjin) were higher than 250 metric kilotonnes from 2008 to 2017, and the annual NOx emissions in the range of 80-200 metric kilotonnes were Beijing, Dalian, Yantai, Jinzhou, Qinhuangdao, and Weifang. The annual emissions of NOx in Dongying, Yingkou, and Panjing were in the range of 20-60 metric kilotonnes from 2008 to 2017. Higher annual OC emissions in ten cities (6-25 metric kilotonnes) were seen relative to that (1-3 metric kilotonnes) of Panjin from 2008 to 2017 (Figure 1k,l). Very high annual emissions (60-100 metric kilotonnes) of PM2.5 and PM10 were observed in Tangshan from 2008 to 2017. The annual emissions of PM10 and PM2.5 were within the range of 20-60 metric kilotonnes in the other nine cities, while the annual emissions of PM10 and PM2.5 in Panjin were lower than 10 metric kilotonnes (Figure 1m-p).  The levels of annual CO 2 emissions observed were found to increase from 2008 to 2013, while the levels of annual CO 2 emissions kept at the peak level of 2013 from 2014 to 2017 for all eleven cities. The greatest annual emissions of CO 2 were seen in Tangshan, followed by Beijing, Tianjin, Dongying, Yantai, Dalian, Weifang, Qinhuangdao, Yingkou, Jinzhou, and Panjin. The largest discrepancies in annual CO 2 emissions were observed in Tangshan (150-200 metric million tonnes) and Yingkou (5-10 metric million tonnes).
The contributions of the industry sector accounted for about 20-40% of annual emissions of air pollutants (BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10, and SO 2 ) as well as CO 2 for all the eleven cities ( Figure S13). The trends in annual emissions of air pollutants (BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 ), and CO 2 from 2008 to 2017 were similar to those in total emissions for the eleven cities. Similar trends as those of total annual emissions were also observed in per capita air pollutants and CO 2 , as well as per capita air pollutants and CO 2 of the industry sector across the eleven cities (Tables S1-S3).

Economic Metrics
Beijing is the capital of China, which had the highest GDP value (1180-2990 billion RMB) during the period of 2008 to 2017 among the eleven cities. Tianjin is a municipality directly under the central government, which has the second-highest GDP value  Table 1). The industry sector contributed about~20% of the total GDP, which was found to be 300-600 billion RMB during the period of 2008 to 2017. The contributions of the industry sector were responsible for 40-60% of GDP for the other ten cities. Tianjin had the GDP of the industry sector in the range of 300-700 billion RMB, which was higher than those of nine cities, including Dongying (100-300 billion RMB), Yantai (100-300 billion RMB), Dalian (100-300 billion RMB), Tangshan (100-300 billion RMB), Weifang (100-300 billion RMB), Qinhuangdao (50-100 billion RMB), Yingkou (50-100 billion RMB), Jinzhou (50-100 billion RMB), and Panjin (50-100 billion RMB) ( Figure S14). The highest GDP per capita during the period of 2008 to 2017 among the eleven cities was Dongying, followed by Beijing, Dalian, Yantai, Tianjin, Panjin, Tangshan, Weifang, Yingkou, Qinhuangdao, and Jinzhou ( Table 1). The ranking of GDP of the industry sector per capita for these eleven cities was consistent with that of GDP per capita.

Associations between Economic Metrics and Emissions of Ambient Species
We evaluated the associations between annual economic development metrics including GDP, GDP per capita, GDP of the industry sector, GDP per capita of the industry sector, and annual emissions of ambient species (BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 ) in 11 cities. GDP was observed to be correlated well with annual emissions of ambient species (BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 ) (Pearson's r range, 0.22-0.87, p < 0.01). GDP per capita correlated with annual emissions of ambient species (BC, CO, NH 3 , OC, PM 2.5 , PM 10 , and SO 2 ) (Pearson's r range, 0.33-0.51, p < 0.05), while GDP per capita was not associated with NO x per capita (p > 0.05). Annual emissions of ambient species (BC, CO, NO x , OC, PM 2.5 , PM 10 , and SO 2 ) exhibited a correlation with GDP of the industry sector, as well as GDP per capita of the industry sector (Pearson's r range, 0.22-0.86, p < 0.05) ( Table S4). As shown in Figure 2, the energy use per GDP was significantly associated with air pollutants (BC, CO, CO 2 , NH 3 , NO x , OC, PM 2.5 , PM 10

Energy Use and Emission per GDP for Cities with Different Development Levels
As shown in Figure 3, CO2 emissions per GDP versus cities with different deve opment levels, the energy use per GDP versus cities with different development leve and PM2.5 emissions per GDP versus cities with different development levels are pr sented. Since the three metrics, including CO2 emissions per GDP, energy use per GD and PM2.5 emissions per GDP versus cities, were highly correlated with each other, t trends of the three metrics against cities with different development levels were found be comparable. For most cities, including Beijing, Tianjin, Dalian, Jinzhou, Panji Qinhuangfao, Dongying, Weifang, and Yantai, the emission and energy use per GD decreased with the enhancements of economic growth, following the environment Kuznets curves. The emission and energy use per GDP in Tangshan and Dongying al obeyed the environmental Kuznets curves, but levels of emission and energy use p GDP were found to be higher than those of other cities.

Energy Use and Emission per GDP for Cities with Different Development Levels
As shown in Figure 3, CO 2 emissions per GDP versus cities with different development levels, the energy use per GDP versus cities with different development levels, and PM 2.5 emissions per GDP versus cities with different development levels are presented. Since the three metrics, including CO 2 emissions per GDP, energy use per GDP, and PM 2.5 emissions per GDP versus cities, were highly correlated with each other, the trends of the three metrics against cities with different development levels were found to be comparable. For most cities, including Beijing, Tianjin, Dalian, Jinzhou, Panjin, Qinhuangfao, Dongying, Weifang, and Yantai, the emission and energy use per GDP decreased with the enhancements of economic growth, following the environmental Kuznets curves. The emission and energy use per GDP in Tangshan and Dongying also obeyed the environmental Kuznets curves, but levels of emission and energy use per GDP were found to be higher than those of other cities.

Disparities in Emissions and Economic Development across Cities
During the periods from 2008 to 2017, GDP per capita in China increased steadily from~25,000 RMB (~3700 US dollars) to~65,000 RMB (~9500 US dollars) with an average annual growth of~7%, which reached an average level in middle-income countries [44] ( Figure S15a). However, large differences in GDP per capita across the studied 11 cities were observed. Eight cities (Beijing, Tianjin, Dalian, Panjin, Tangshan, Jinzhou, Donging, Weifang, and Tianjin) had GDP per capita higher than the national mean level in China, while the values of GDP per capita in three cities (Yingkou, Jinzhou, and Qinhuangdao) were lower than the national mean level in China ( Figure S15b). Since emissions of air pollution and CO 2 per capita were highly associated with GDP per capita at the initial stage of economic development following the EKC hypothesis [13,45], the large different emissions across the 11 cities were ascribed to the regional disparities in economic development. Our result is consistent with several studies that emissions of air pollution and CO 2 per capita comply with the EKC hypothesis in cities in Asian countries [7,13,14]. The sum of emissions of air pollution and CO 2 accounted for 10-30% of total emissions in China. Since city-level emissions of air pollution and CO 2 across the world are rarely limited, the level comparisons in the emissions of air pollution and CO 2 across cities around the world could provide a comprehensive evaluation of emissions of air pollution and CO 2 at a city level in the future. In general, CO 2 emissions per GDP and emissions of air pollutants per GDP in each city decreased by 30-50% from 2008 to 2017, which illustrated the replacement of old technology with advanced technology driven by economic development could lessen the emissions of CO 2 and air pollutants [3,46,47].
These 11 cities belonging to five regions (Beijing, Tianjin, Liaoning Province, Hebei Province, and Shandong Province) can be simply divided into different city types by characteristics of economic developments, industrial structures, as well as emission levels of air pollutants and CO 2 [26,36]. Cheng et al. [26] classified 210 cities in China into seven groups based on their sustainability performance in 2016. Beijing is characterized by a high human capital structure with sustainable development driven by high quality and quantity of human capital. Tianjin, four cities (Dalian, Yingkou, Jinzhou, and Panjin) in Liaoning Province, and two cities (Weifang and Yantai) in Shandong Province are featured with high produced capital structure and are highly dependable on the development of the industry. Dongying in Shandong Province is an energy-producing city for crude oil ( Table 4). The cities of Tangshan and Qinhuangdao were not grouped in Cheng et al. [26] due to the absences of the data. Tangshan pursues economic development based on abundant coal reserves and steel production [32], while Qinhuangdao is a produced capital city, with development driven by the manufacturing industry [48,49]. Table 4. The classification of 11 cities according to the development types.

Types City a Reference
Human capital-dominated development Beijing [26] Produced capital-dominated development Tianjin, Dalian, Yingkou, Jinzhou, and Panjin in Liaoning Province, as well as Weifang and Yantai in Shandong Province [26] Energy-producing city Dongying in Shandong Province [26] a The cities of Tangshan and Qinhuangdao were not grouped due to the absences of the data.
For a human capital city with high values of GDP per capita, Beijing canexpand its dependency on human capital gradually and abandon the economic development of the industry with high emissions of air pollutants and CO 2 . For those cities associated and reliant on the development of produced capital structure, cities can rely on the industry with low emissions of air pollutants/CO 2 and improve the mitigation technology for their emission reductions. Proper strategies for reducing ambient species, including PM 2.5 , BC, SO 2 , and CO 2 , from specific sectors could alleviate climate change and air pollution, as well as simul- taneously improve socioeconomic development [1,50]. There are several co-benefit policies for the reduction of air pollutants (e.g., PM) and CO 2 when they are from the same sources. Shindell et al. [1] presented 14 measures that could reduce greenhouse gas emissions in the short term with improvements in human health and food security. Annenberg et al. [50] illustrated that a low-carbon development policy could lead to a decrease in mortality associated with exposure to air pollution and events resulting from climate change, with an increase in the gross domestic product (GDP). However, improper mitigation policies for energy use can result in trade-offs between air pollution and greenhouse gas emissions. For example, switching from coal heating to natural gas was found to reduce PM and BC emissions by 30-80% but increased CO 2 emissions slightly in some cities, according to prior studies [51,52]. While in a study conducted by Wilson and Stafell, displacing coal with natural gas could lead to reducing per-capita annual emissions by 400 kg CO 2 between 2015 and 2016 in the UK [53]. For those cities with development dependent on fossil fuels, including coal and crude oil, the increased dependence on the development of resource-intensive industries following ordinary development routes previously can lead to low levels of sustainable development performance. Residents should consider the low levels of sustainability performance in resource-exhausted cities and thus abandon their resource dependency gradually. These cities can pave a development pathway for industries with high-value-added products and low emissions of air pollutants and CO 2 . The shift to low-carbon pathways for resource-dependent cities can be achieved by increasing the capacities to attract high quality and quantity of human capital [53][54][55][56][57].

Mitigation of Air Pollution and CO 2 Emission at a City Level
For the cities (Yingkou, Jinzhou, and Qinhuangdao) with GDP per capita lower than the national mean level in China, the priorities for the citizens are to seek socioeconomic development in low-carbon development pathways. For the cities (Tangshan and Dongying) with energy use and emission per GDP higher than those other cities, the priorities for the citizens are to pave the low-carbon development with developed technologies. Recent studies indicate that interregional cooperation policies could aid the city to move forward in an economically optimal pathway for carbon emissions reductions and socioeconomic advancement [58][59][60][61]. The pathway for moving to low-carbon development with developed technologies is expected to prioritize PM and GHG reductions, as well as share greater co-benefit in protecting public health. Driscoll et al. [46] found that annual estimated CO 2 and PM 2.5 can be reduced by 10-40% in the USA after implementing conventional energy with after-treatment technology and renewable energy in 2417 power plants, avoiding 21-33% of premature deaths associated with air pollution. A study conducted in China suggested that the reductions of 1469 million tonnes of CO 2 and decreases in 15-22% of all-cause mortality associated with air pollution could be achieved in China around 2030 if low-carbon policies on conventional energy with advanced after-treatment technology were performed [47]. Key steps toward the targets for carbon emissions reductions and socioeconomic advancement in the cities with low levels of development can be supported by the city's appropriate mitigation action under 14th China's Five-Year Plans. These development plans can aid the citizens to expand their industrial productivity and promote economic development locally [62].

Conclusions
Our study presented the differences in emissions of air pollutants and CO 2 across 11 cities around the Bohai Sea. Beijing, presenting high human capital structures, has emitted total amounts of CO 2 in the range of 100-150 metric million tonnes and air pollutants (BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 ) in the range of 1500-2500 metric kilotonnes during the period of 2008-2017. Another eightcities, including Tianjin, Dalian, Panjin, Tangshan, Jinzhou, Donging, Weifang, and Tianjin, characterized by high produced capital structures have produced total amounts of CO 2 in the range of 40-200 metric million tonnes and air pollutants (BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 ) in the range of 500-3000 metric kilotonnes during the period of 2008-2017. The three cities (Panjin, Jinzhou, and Qinhuangdao) with GDP per capita lower than the national mean level in China discharged total amounts of CO 2 in the range of 20-50 metric million tonnes and air pollutants (BC, CO, NH 3 , NO x , OC, PM 2.5 , PM 10 , and SO 2 ) in the range of 200-500 metric kilotonnes during the period of 2008-2017. The high energy use cities of Tangshan and Dongying corresponded to higher emissions of air pollutants and CO 2 per GDP than other cities. Cities with high air pollutants and CO 2 per GDP should pursue short-term policies to reduce air pollution and increase human development. Short-term policies should focus on technological development in carbon reductions and socioeconomic development simultaneously. Three Cities (Yingkou, Jinzhou, and Qinhuangdao) with lower GDP per capita lower than the national mean level in China are essential for promoting the development of cities in low-carbon pathways.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/toxics10090547/s1, Figure S1: The sketch map for the locations of the eleven cities. Figure Figure S13: Contribution of the different sectors to the yearly air pollutants including BC (a), CO (b), NH 3 (d), NO x (e), OC (f), PM 2.5 (g), PM 10 (h), SO 2 (i), as well as CO 2 (c) in the eleven cities from the years of 2008-2017. Figure S14: Contributions of the primary sector (agriculture), the secondary sector (industry), and the tertiary sector (excluding agriculture and industry) to GDP from 2008 to 2017 in the11 cities. Figure S15: (a) Changes in GDP per capita from 2008 to 2017 in China. (b) Changes in ratio of GDP per capita and national mean level from 2008 to 2017 in 11 cities. Ratio of GDP per capita and national mean level was estimated using GDP per capita for each city divided GDP per capita in China. Table S1: The summary of the annual level of air pollutants (Metric kilotonnes) of the industry sector and CO 2 of the industry sector. Table S2: The summary of the annual level of air pollutants (kg per capita) per capita and CO 2 (tonnes per capita) per capita. Table S3: The summary of the annual concentration of air pollutants (kg per capita) of the industry sector per capita and CO 2 (tonnes per capita) of the industry sector per capita. Table S4. Correlation (Pearson's r) between economic development metrics including GDP, GDP per capita, GDP of industry sector, and GDP per capita of industry sector with air pollution, as well as CO 2 emission.
Author Contributions: Z.Z., Q.L., J.L. and Y.L. designed the experiment and collected the data. Z.Z. and Q.L. wrote the draft. All authors have read and agreed to the published version of the manuscript.

Funding:
The study was supported by Taishan Scholarship.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: The data and materials could be available at http://www.meicmodel.org (accessed on 1 November 2021) and also be acquired upon the request from the corresponding author.