Characterization of the Air Quality Index for Urumqi and Turfan Cities, China

Received: November 13, 2018
Revised: January 3, 2019
Accepted: January 4, 2019

Download Citation: RIS|BibTeX|https://doi.org/10.4209/aaqr.2018.11.0410  

Cite this article:

Yin, Z., Cui, K., Chen, S., Zhao, Y., Chao, H.R. and Chang-Chien, G.P. (2019). Characterization of the Air Quality Index for Urumqi and Turfan Cities, China. Aerosol Air Qual. Res. 19: 282-306. https://doi.org/10.4209/aaqr.2018.11.0410

HIGHLIGHTS

• Air Quality Index for two Cities.
• Atmospheric PM_2.5 Concentration in two Cities.
• Atmospheric SO₂ Concentration in two Cities.
• Atmospheric NO_x Concentration in two Cities.
• Atmospheric O₃ Concentration in two Cities.

Keywords: AQI; PM10; PM2.5; SO2; NOx; CO; O3.

INTRODUCTION

With economic growth and urban construction, and with industrialization, urbanization is further expanded, so environmental pollution has become more and more serious (Heal et al., 2012; Chen et al., 2013). Industrial activities, the burning of chemical fuels, the emission of smoke from domestic stoves, and the emission of automobile exhaust caused by an increase in the number of automobiles have caused serious air pollution. At present, air pollution is a major risk affecting human health.

The harmful effects of atmospheric pollutants on the human body vary. PM_2.5 causes cardiovascular disease and lung cancer. PM_2.5 also affects air visibility and contributes to global climate change (Matawle et al., 2015; Liang et al., 2016; Du et al., 2018). The harm of PM₁₀ cannot be ignored. It can affect visibility and temperature. Nitrogen dioxide in the atmosphere has a variety of toxicities and damages the bronchus and lungs after entering the human body, which can induce various types of respiratory inflammation (Ezzati and Kammen, 2002; Tong et al., 2018). SO₂ and NO_xreleased from the combustion of fossil fuels or some types of industrial production are discharged into the atmosphere and undergo a chemical reaction to form sulfuric acid or nitric acid. After rainwater falls onto the ground, they cause acidification of groundwater and surface water, contaminate the soil, and affect crop yields (Crutzen, 1979; Garg et al., 2001). Carbon monoxide (CO) is a very toxic pollutant. Carbon monoxide (CO) in the air passes through the respiratory system and enters the body's bloodstream (Peng et al., 2005; Brauer et al., 2016), resulting in oxygen deprivation of the body tissues, with the most significant effects on the heart and brain (Amato et al., 2011; Wei et al., 2012). After entering the human body, ozone stimulates the respiratory tract, triggers bronchitis and emphysema, and affects the nervous system of humans. The Air Quality Index (AQI) is a conceptual value that evaluates air quality, representing the air quality status and trends in a given area. It includes a measurement of nitrogen dioxide (NO₂), sulfur dioxide (SO₂), carbon monoxide (CO), ozone (O₃), fine particulate matter (PM_2.5), and coarse particulate matter (PM₁₀), which determine the air quality index on a given day. At present, the number of deaths caused by air pollution continues to increase. The harm caused by air pollution seriously affects sustainable development. Therefore, air pollution deserves serious attention.

As air pollution becomes more serious, China has begun to implement corresponding measures to gradually improve air quality. For example, in 2013, China began to implement the "Air Pollution Prevention Action Plan," which also improved its air quality. Based on this, the current study focused on the air pollution characteristics of Urumqi and Turpan through the use of AQI values. Both Urumqi and Turpan are located in the northwest part of China. Urumqi is one of the last ten cities in China with poor air quality, which is of great importance for the study of air quality in northwestern China (Pope and Dockery, 2013). The local government provided an important basis for the implementation of an air pollution control strategy.

This study discussed the content of six standard pollutants PM_2.5, PM₁₀, SO₂, NO₂, CO, and O₃ in air over a three-year period in Urumqi and Turpan. In addition, the study provides an Air Quality Index (AQI) analysis section illustrating changes in air quality over the study period.

METHODS

Urumqi (43°45′N, 87°36′E) is located in the middle of Xinjiang, bordering the city of Turpan. Turpan (42°25'N, 89°36'E) is located in the central part of the Xinjiang Uygur Autonomous Region and is located in the Tianshan mountain basin. Xinjiang is located in the northwestern part of China. It is far from the sea, surrounded by high mountains that block the flow of the oceans, and is difficult to reach, thus forming a clear temperate continental climate (Jin et al, 2017; Wu et al., 2017). There is a large temperature difference depending on hours of sunshine, with annual hours of sunshine ranging from 2500–3500. There is little rainfall, so the climate is dry. The two cities have been developing rapidly in recent years. According to past surveys, the industrialization and urbanization of a city are directly proportional to environmental pollution (Che et al., 2009). This study is of great significance to the environmental protection and human health of these two cities in the Xinjiang region.

Data were obtained for three years from January 2015 to December 2017 in both Urumqi and Turpan cities. The PM mass concentration (including daily PM_2.5 and PM₁₀) and gaseous pollutants (including daily SO₂, NO₂, CO, and 8 hr-averaged O₃) were obtained from the China air quality online monitoring and analysis platform (http://www.aqistudy.cn).

Air Quality Index (AQI)

The sub-AQIs for the six standard contaminants were calculated by observing the concentrations, as shown in Eq. (1) (She et al., 2017; Shen et al., 2017). The overall AQI represents the maximum of the sub-AQI of all pollutants, where when the AQI is higher than 50, the highest sub-AQI contributor is defined as the primary pollutant on that day, as shown in Eq. (2) (She et al., 2017; Shen et al., 2017) 

where

IAQI_P: the air quality sub index for air pollutant P;
C_P: the concentration of pollutant P;
C_low: the concentration breakpoint that is ≤ C_P;
C_high: the concentration breakpoint that is ≥ C_P;
I_low: the index breakpoint corresponding to C_low;
I_high: the index breakpoint corresponding to C_high.

AQI refers to the air pollution index. According to the ambient air quality standards and the impact on the human social environment, the AQI simplifies the concentration of pollutants into a single numerical value to reflect the current air quality status. Six standard air pollutants have a dramatic impact on health. The daily AQI value is calculated from the 24-hour average concentrations of SO₂, NO₂, PM_2.5, PM₁₀, CO, and the daily maximum 8-hour concentration of O₃. According to some studies, AQI in the air is generally divided into six categories: Level I: 0–50 (good, green); Level II: 51–100 (medium, yellow); Level III: 101–150 (the sensitive group is not healthy; Orange); IV: 151–200 (unhealthy; red); V: 201–300 (very unhealthy; purple), VI: 300–500 (dangerous; brown) (Hu et al., 2015; She et al., 2017; Zhao et al., 2018).

RESULTS AND DISCUSSION

PM_2.5 Concentration

PM_2.5 refers to particulate matter less than 2.5 microns in diameter. It is suspended in the air for a long time and has a significant impact on the human body. PM tends to accumulate in human respiratory tract and thus is classified as a severe health hazard (Tao et al., 2009; Bilal et al., 2017; Xu et al., 2017). PM_2.5 enters the body's respiratory tract and cardiovascular system, leading to respiratory diseases. The monthly average PM_2.5concentration in the ambient air of Urumqi and Turfan for the period 2015–2017 are shown in Tables 1(a), 1(b) and 1(c), respectively.

In Urumqi, the average monthly PM_2.5 concentration ranged from 24–147 µg m^–3 in 2015, with an average of 65 µg m^–3; in 2016, it was 24–201 µg m^–3, with an average of 73 µg m^–3; in 2017, it was 19–228 µg m^–3, with an average of 71 µg m^–3. According to the average, the PM_2.5 concentration was the lowest in 2015. From 2015 to 2016, the PM_2.5 concentration increased by about 10.6%. From 2016 to 2017, it decreased by about 2.70%. In the three years under investigation, the average concentration of PM_2.5 was 70 µg m^–3, which was 7 times higher than the WHO air quality regulatory standard (10 µg m^–3). Therefore, it is important to decrease the levels of PM_2.5 in the ambient air. During the three years examined, the maximum daily average concentration occurred in January 2016 (397.0 µg m^–3), and the minimum occurred in August 2015 (9.20 µg m^–3). This is because the burning of coal during the heating period in winter causes an increase in the amount of pollutants in the air.

In the case of Turpan, the average concentration of PM_2.5 in 2015 ranged from 23–115 µg m^–3, with an average of 65 µg m^–3; in 2016, it was 32–133 µg m^–3, with an average of 70 µg m^–3; in 2017, it was 35–136 µg m^–3, with an average of 68 µg m^–3. From 2015 to 2017, the PM_2.5 content in the air increased slightly, an increase of about 4.60%. The average daily maximum occurred in May 2016 (359 µg m^–3), and the minimum occurred in July 2015 (8.0 µg m^–3). The Turpan region is located in the northwestern part of China, where sandstorms often occur, so in May 15, 2016, the content of PM_2.5 and PM₁₀ in the air increased abnormally. Overall, the average annual values for these three years were 65 µg m^–3, 70 µg m^–3 and 68 µg m^–3, respectively, with a three-year average of 68 µg m^–3, which is 6.8 times higher (10 µg m^–3) than the WHO air quality regulatory standard.

Regarding seasonal changes, the average concentrations of PM_2.5 in spring, summer, autumn, and winter in Urumqi in 2015 were 80, 34, 33 and 111 µg m^–3, respectively; and in 2016, they were 101, 25, 32 and 134 µg m^–3, respectively. In 2017, the numbers were 101, 23, 29 and 130 µg m^–3, respectively. On the basis of the three years, it can be seen that the PM_2.5 content in the air changes due to seasonal changes. Generally, the PM_2.5 content in the winter air is greater than the PM_2.5 content in the spring air when the content of PM_2.5 in the air is greater than the PM_2.5 content in summer and autumn. Among the seasons, the PM_2.5 content in the air in winter was the highest in 2016, and the PM_2.5 content in the air in the summer of 2017 was the lowest. The highest value was in the winter of 2016 (134 µg m^–3), which was 5.8 times the lowest value in the summer of 2017 (23 µg m^–3).

In the case of Turpan, in 2015, the average PM_2.5 concentrations in spring, summer, fall, and winter were 71, 34, 39, and 115 µg m^–3, respectively, and those in 2016 were 74, 49, 46, and 113 µg m^–3, respectively. Those in 2017 were 68, 47, 45, and 112 µg m^–3, respectively. The PM_2.5 concentration in the winter air was higher than the PM_2.5 concentration in the spring air, and PM_2.5 concentration was the lowest in the summer and autumn air. Among the seasons, the PM_2.5 content in the air in the winter was the highest in 2015, and the PM_2.5 content in the air in the summer was the lowest in the summer of 2015. The highest value was in the winter of 2015 (115 µg m^–3), which was about 5.8 times the lowest value in the summer of 2015 (34 µg m^–3).

Previous studies have reported that urban transport has become a major source of atmospheric particulates in China's megacities due to rapid urbanization and industrialization (Song et al., 2012; Zhao et al., 2013; Lanzafame et al., 2015; Xie et al., 2016; Tao et al., 2017). In addition to transportation, fuel combustion and industrial production are also important sources of PM_2.5. PM_2.5 has a small particle size and a long residence time in air and therefore has a great impact on air quality and human health. 

PM₁₀ Concentration

PM₁₀ is a particulate with a particle size of less than 10 microns. Increases in PM₁₀ pollution has important effects on air quality and human health. The sources of PM₁₀ can be divided into natural factors and human factors. Natural factors include sandstorms and resuspension of local soils. The human factors generally include coal combustion (Evagelopoulos et al., 2006; Matawle et al., 2015), various industrial activities, and the cement ground car grinding process (Kong et al, 2011; Yang et al., 2016). The content of PM₁₀ pollution plays a crucial role in the study of air quality. The monthly mean PM₁₀ concentrations in the ambient air of Turpan and Urumqi are listed in Tables 2(a), 2(b) and 2(c), respectively, for the period from 2015 to 2017.

In the case of Urumqi, the average PM₁₀ concentration in 2015 was between 77 and 220 µg m^–3, with an average of 131 µg m^–3, and in 2016, it was 65 to 287 µg m^–3, with an average of 123 µg m^–3. In 2017, it was 61 to 271 µg m^–3, with an average of 115 µg m^–3. For three years, the concentration of PM₁₀ in the area gradually decreased. From 2015 to 2016, it decreased by about 6.11%, and from 2016 to 2017, it decreased by about 6.50%.

In the case of Turfan, the average concentration of PM₁₀ in 2015 was ranged from 61–282 µg m^–3, with an average of 140 µg m^–3, and in 2016, it was 105–277 µg m^–3, with an average of 171 µg m^–3. In 2017, the concentration ranged from 95–219 µg m^–3, and the average value was 160 µg m^–3. According to the average annual PM₁₀ concentration, the annual average value of PM₁₀ in the three years from 2015 to 2017 gradually increased. The PM₁₀ concentration was the highest in 2016 and increased by approximately 22.1% from 2015 to 2016. From 2016 to 2017, thePM₁₀ decreased by approximately 6.43%. The above results also show that the concentration of PM₁₀ in Turpan was about eight times higher than the World Health Organization's air quality supervision standard (20 µg m^–3) during the study period. Therefore, efforts to implement more effective strategies to improve air quality in this region must be increased.

The three-year PM₁₀ concentration in Urumqi for the three years under investigation ranged from 61 to 287 µg m^–3 with an average of 123 µg m^–3, and the PM₁₀ concentration in Turpan was between 61 and 282 µg m^–3, with an average of 157 µg m^–3. Therefore, the PM₁₀ concentration in Turpan was higher than that in the Urumqi during the three years under study. It is possible that urbanization and industrialization in Turpan had already preceded that occurring in Urumqi. With economic development, it is also necessary to strengthen the protection of the environment.

Regarding the seasonal changes in the Urumqi region, the average PM₁₀ concentrations in spring 2015, summer, autumn, and winter were 123, 102, 100, and 199 µg m^–3, respectively. The average values in 2016 were 93, 70, 84, and 246 µg m^–3, respectively, and the mean values in 2017 were 91, 67, 101, and 203 µg m^–3. The PM₁₀ concentration in the winter was higher than that in the spring in the three year study period, which is far greater than the concentration of PM₁₀ in the summer. For Turpan, in the spring of 2015, the average concentrations of PM₁₀ in the summer, autumn, and winter were 126, 81, 95, and 136 µg m^–3, respectively. In the spring of 2016, the average concentrations of PM₁₀ in the summer, autumn, and winter were 117, 78, 90, and 112 µg m^–3, respectively. In the spring of 2017, the average concentrations of PM₁₀ in the summer, autumn, and winter were 93, 67, 90, and 90 µg m^–3, respectively. In the three year period under observation, the average concentration of PM₁₀ in the spring and winter was higher than that in the summer and autumn, and the PM₁₀ concentration was the lowest in summer. Generally speaking, the air quality in winter is poor. This is because the winter temperature is low, and a lot of coal is used for heating in the winter and spring (Shang et al., 2015). Exhaust gas from fuel combustion contributes greatly to the accumulation of elements such as V, Cr, Cu, and Fe (Lelieveld et al., 2015; Matawle et al., 2015), and low-temperature air may hinder the diffusion of pollutants (Tang et al., 2017; Xing et al., 2017; Wang et al., 2018). In the summer, air circulation is smooth and conducive to pollution, so the PM₁₀ concentration in the summer is lower than the PM₁₀ concentration in the winter.

Within the three year study period, the concentration of PM₁₀ in Turpan was very high and fluctuated greatly with the change of seasons, mainly because sandstorms occur frequently in the area, and the area is surrounded by mountains, which is not conducive to air circulation. Coal combustion and power plant production are also sources of PM_2.5 and PM₁₀ pollution. A recent report showed that cars also exacerbate air pollution. PM does great harm to the human body, affecting the human respiratory and nervous systems (Lee et al., 2008; Wu et al., 2013; Wang et al., 2016). Previous studies have shown that air quality can be assessed by the ratio of PM_2.5 to PM₁₀ (Zaveri et al., 2008). Fossil fuel combustion, industrial activities, and vehicle exhaust emissions are important sources of PM₁₀ (Kong et al., 2011). Therefore, the concentration of PM₁₀ must be appropriately lowered to improve the air quality.

PM_2.5/PM₁₀ Ratio

PM_2.5 and PM₁₀ are particulates in the air. We often use PM_2.5/PM₁₀ to reflect air pollution. The monthly averages of the PM_2.5/PM₁₀ ratios in the ambient air of Urumqi and Turpan are shown in Tables 3(a), 3(b), and 3(c).

In Urumqi, the monthly PM_2.5/PM₁₀ ratio ranged from 0.28 to 0.71 in 2015, with an average of 0.47. The monthly PM_2.5/PM₁₀ ratio for 2016 ranged from 0.33 to 0.78, with an average of 0.52. The monthly PM_2.5/PM₁₀ ratio for 2017 ranged from 0.29 to 0.88, with an average of 0.53. In Turpan, the monthly PM_2.5/PM₁₀ ratio for 2015 was 0.32 to 0.78, with an average of 0.47. In 2016, the monthly PM_2.5/PM₁₀ ratio was 0.31 to 0.67, with an average of 0.43. In 2017, the monthly PM_2.5/PM₁₀ ratio was 0.30 to 0.71, with an average of 0.43. These results show that the annual average PM_2.5/PM₁₀ ratios in both Urumqi and Turpan were all lower than those reported in the Beijing-Tianjin-Hebei region (0.83), the Yangtze River Delta region (0.76), and the Pearl River Delta region (0.74) (Alghamdi et al., 2015). The PM_2.5/PM₁₀ ratio refers to the ratio of PM_2.5 to PM₁₀ and reflects the PM_2.5 and PM₁₀ pollution proportion of PM pollution in the air. It has a certain guiding significance for PM pollution control.

In the case of Urumqi, in 2015, the three highest monthly averages for the PM_2.5/PM₁₀ ratio were 0.71 in January, 0.62 in February, and 0.66 in December. The three lowest monthly averages for the PM_2.5/PM₁₀ ratio were 0.33 in June, 0.28 in July, and 0.33 in August. In 2016, the three highest monthly averages of the PM_2.5/PM₁₀ ratio were 0.72 in January, 0.68 in February, and 0.78 in December. The three lowest monthly averages were 0.33 in April, 0.38 in July, and 0.35 in August. In 2017, the three highest monthly averages of PM_2.5/PM₁₀ ratio were 0.88 in January, 0.87 in February, and 0.71 in December. The three lowest monthly averages were 0.32 in May, 0.29 in July, and 0.29 in September.

In the case of Turpan, in 2015, the three highest monthly averages for the PM_2.5/PM₁₀ ratio were 0.75 in January, 0.63 in November, and 0.78 in December. The three lowest monthly averages for the PM_2.5/PM₁₀ ratio were 0.32 in July, 0.36 in August, and 0.37 in September. In 2016, the three highest monthly averages of the PM_2.5/PM₁₀ ratio were 0.67 in January, 0.44 in November, and 0.65 in December. The three lowest monthly averages were 0.35 in August, 0.31 in September, and 0.35 in October. In 2017, the three highest monthly averages of the PM_2.5/PM₁₀ ratio were 0.71 in January, 0.51 in November, and 0.67 in December. The three lowest monthly averages were 0.34 in March, 0.31 in August, and 0.30 in September.

In general, the PM_2.5/PM₁₀ value is higher when the winter temperature is low. When the summer temperature is high, the PM_2.5/PM₁₀ value is smaller. PM₁₀ contains PM_2.5 based on the particle size. Compared with PM₁₀, PM_2.5 is a comprehensive pollutant; its range is relatively wide, and considerable parts of PM_2.5 are nitrogen oxides, sulfur dioxide and VOC derivative compounds that are produced by chemical conversion in the air. Therefore, to control PM_2.5, it is necessary not only to control particulate matter, but also to control sulfur dioxide, nitrogen oxides, volatile organic compounds, and so on. Because PM_2.5 has a small particle size, long residence time in the atmosphere, long transport distance, easy inhalation into the lungs, and a significant influence on human health and air quality, it should be of great concern. 

SO₂ Concentration

SO₂ is a simple sulfur oxide, which is mainly derived from volcanic eruptions, coal combustion, and the combustion of petroleum and chemical fuels (Kato et al., 2016). SO₂ can also undergo a series of reactions to form acid rain, which has negative effects on the ecosystem and the social environment. According to a US Environmental Protection Agency report, SO₃ and H₂SO₄ aerosols can have a range of adverse effects on the human body, including respiratory irritation and dyspnea (Li et al., 2017a). Therefore, SO₂ pollution deserves scientific study. Figs. 1(a), 1(b), and 1(c) provide the average concentrations of SO₂ in Urumqi and Turpan, respectively, from 2015 to 2017.

Fig. 1(a). Monthly average atmospheric SO₂ concentrations in Urumchi and Turpan in 2015.

Fig. 1(b). Monthly average atmospheric SO₂ concentrations in Urumchi and Turpan in 2016.

Fig. 1(c). Monthly average atmospheric SO₂ concentrations in Urumchi and Turpan in 2017.

In Urumqi, the monthly average concentration of SO₂ in 2015 was ranged from 0.700–27.7 ppb, with an average of 5.53 ppb. The average concentration in 2016 ranged from 1.10–20.3 ppb, with an average of 5.06 ppb. The average concentration in 2017 ranged from 2.10–16.5 ppb, with an average of 4.74 ppb. It was observed that the SO₂ concentration gradually decreased by approximately 8.50% from 2015 to 2016 and decreased by 11.7% from 2016 to 2017.

In Turpan, the average concentration of SO₂ in 2015 ranged from 0.700 to 47.6 ppb, with an average of 7.00 ppb; the average SO₂ concentration in 2016 ranged from 0.700 to 26.6 ppb, with an average of 5.13 ppb; the average concentration of SO₂ in 2017 ranged from 0.700 to 174 ppb, with an average is 4.93 ppb. The annual average concentration of SO₂ decreased by approximately 26.7% from 2015 to 2016 and decreased by approximately 3.90% from 2016 to 2017.

In the three year study period, the SO₂ concentration averaged 5.11 ppb in Urumqi and 5.69 ppb in Turpan, and the results showed that the concentrations of SO₂ in Urumqi and Turpan were lower than the World Health Organization air quality standard of 7.00 ppb, and that of Turpan was slightly higher than the concentration of SO₂ in Urumqi. The combustion of coal produces SO₂, which is easily oxidized into SO₃ in the air and has a significant impact on the environment (Kato et al., 2016). Most of the SO₃ entering the air will form sulfuric acid vapor, which corrodes buildings. SO₂ has also attracted widespread attention, and China has also carried out corresponding control measures, such as gas desulfurization and modification of electrostatic precipitation and wet electrostatic precipitation (Schecter et al., 2006). These measures have played a certain role in the control of SO₂. Urumqi adopted desulfurization construction of coal-fired plants as well as measures such as coal-fired heating control during the warm season. Accordingly, seasonal changes are further discussed.

In Urumqi, the average concentrations of sulfur dioxide in the spring, summer, autumn, and winter in 2015 were 4.83, 3.10, 3.87, and 10.33 ppb, respectively. In 2016, they were 4.07, 2.77, 4.60, and 8.80 ppb, respectively. In 2017, they were 3.67, 2.97, 4.43, and 7.90 ppb, respectively. In Turpan, the average SO₂ concentrations in the spring, summer, autumn, and winter of 2015 were 4.43, 2.80, 4.55, and 16.7 ppb, respectively. In 2016, they were 2.80, 2.57, 4.43, and 10.7 ppb, respectively. In 2017, they were 4.20, 2.33, 3.50, and 9.68 ppb, respectively. In the spring, summer, autumn and winter in Turpan, the average SO₂ concentrations were 3.81, 2.57, 4.16, and 12.4 ppb, and in the spring, summer, autumn and winter of Urumqi, the average SO₂ concentrations were 4.19, 2.95, 4.30, and 9.01 ppb. It is clear that the concentration of SO₂ in winter in these two regions is significantly higher than in other seasons. In general, the lowest SO₂ content in the air occurs in the summer because the burning of coal in the area in the winter is higher than in the other three seasons. The area around Turpan is surrounded by mountains, which lowers air circulation. Therefore, the SO₂ concentration changes with the season. Both cities are located in the northwestern part of China. They are cold in the winter and hot in the summer, especially in the case of Turpan. Extreme temperatures and low rainfall also cause differences in the levels of air pollutants in the winter and summer.

NO₂ Concentration

NO₂ is a toxic gas that can undergo a series of complex reactions in the air (Thompson, 1992; Olivier et al., 1998; Khokhar et al, 2016). Nitrogen dioxide is also one of the causes of acid rain and poses a significant environmental hazard. Generally speaking, NO₂ is mainly derived from human activities, including industrial NO₂ emissions, vehicle exhaust emissions, and fuel combustion (Cheng et al., 2018). Figs. 2(a), 2(b), and 2(c) show the average concentrations of nitrogen dioxide in Urumqi and Turpan from 2015 to 2017, respectively.

Fig. 2(a). Monthly average atmospheric NO₂ concentrations in Urumchi and Turpan in 2015.

Fig. 2(b). Monthly average atmospheric NO₂ concentrations in Urumchi and Turpan in 2016.

Fig. 2(c). Monthly average atmospheric NO₂ concentrations in Urumchi and Turpan in 2017.

In the case of Urumqi, the monthly NO₂ concentration was between 8.80 and 61.4 ppb, with an average of 24.8 ppb in 2015. In 2016, these values ranged from 10.2 to 68.7 ppb, with an average of 18.5 ppb, and from 8.80 to 66.2 ppb in 2017, with an average of 24.3 ppb. The average NO₂ concentration decreased by approximately 25.4% from 2015 to 2016 and increased by 31.4% from 2016 to 2017.

In Turpan, the monthly average NO₂ concentration in 2015 ranged between 3.41 and 36.5 ppb, with an annual average of 17.3 ppb. The monthly average NO₂ concentration in 2016 ranged between 1.90 and 43.3 ppb, with an annual average of 19.0 ppb. In 2017, these values ranged from 4.87 to 43.3 ppb, with an annual average of 21.0 ppb. The average NO₂ concentration increased slowly during the three-year period, rising by approximately 9.83% from 2015 to 2016 and increasing by 19.5% from 2016 to 2017.

Overall, the average NO₂ concentration range for Urumqi and Turpan during the three-year period was 8.80–68.7 ppb and 1.90–43.3 ppb, respectively. The corresponding average concentrations are 22.5 and 19.1 ppb, respectively. The annual average NO₂ level in Urumqi (22.5 ppb) was slightly higher than WHO's air quality supervision standard (19.5 ppb). The average annual NO₂ level in Turpan (19.1 ppb) was close to the World Health Organization's air quality supervision standard (19.5 ppb).

Urumqi is the capital of Xinjiang and, according to official statistics, the population of Urumqi (2.67 million) is more than that of Turpan (650,000) (http://www.xjtj.gov.cn). In recent years, along with economic development, urbanization and industrialization has accelerated. There are more vehicles and factories in Urumqi than in Turpan. Studies have shown that motor vehicles are the main source of nitrogen dioxide emissions (Sillman, 1999; Pudasainee et al., 2006). In 2015, the number of motor vehicles in Urumqi was 800,000. As of the end of 2016, the number of motor vehicles in Urumqi was 943,100, an increase of 14.2% over the previous year. By 2017, the number of motor vehicles reached 1.09 million. In Turpan, in 2015, the number of motor vehicles was 145,000. As of the end of 2016, the number of motor vehicles was 111,000, a decrease of 23.3% from the previous year. In 2017, the number of motor vehicles was 126,800 (http://www.tjcn.org). As a result, The number of cars in Urumqi is higher than that in Turpan, the amount of nitrogen oxides (NO_x) emitted by automobile exhaust gases in Urumqi is greater, resulting in higher concentrations of NO₂ in the atmosphere.

In terms of seasonal variations, in Urumqi, the NO₂ concentrations in the spring, summer, autumn, and winter in 2015 were 21.5, 18.9, 21.8, and 37.0 ppb, respectively, and the NO₂ concentrations in 2016 were 22.0, 19.4, 23.9, and 38.4 ppb respectively; the NO₂ concentrations in 2017 were 20.2, 16.5, 24.0, and 36.4 ppb. In the Turpan region, the NO₂ concentrations in the spring, summer, autumn and winter in 2015 were 13.1, 11.9, 19.6, and 24.5 ppb respectively, and the NO₂ concentrations in 2016 were 13.6, 12.5, 22.1, and 27.6 ppb respectively; the NO₂ concentrations in 2017 were 15.9, 13.8, 24.7, and 29.7 ppb. In Urumqi, the maximum value of NO₂ occurred in the winter of 2016 (38.4 ppb), and the minimum occurred in the summer of 2017 (16.5 ppb). In Turpan, the maximum value of NO₂ occurred in the winter of 2017 (29.7 ppb), and the minimum occurred in the summer of 2015 (11.9 ppb). It can be seen from the data that in Urumqi, the concentration of NO₂ in the autumn and winter was higher than that in the spring and summer. In Turpan, the concentration of NO₂ was the highest in the winter and the lowest in NO₂ in the summer. This phenomenon may be related to different meteorological conditions in different seasons. Both cities are located in northwestern China, with a dry climate and large temperature differences. The increase in the use of coal and vehicles during winter results in an increase in NO₂ concentration; therefore, corresponding measures should be taken to reduce NO₂ emissions.

CO Concentration

Carbon monoxide (CO) is a common air pollutant that poses a significant threat to human health. Carbon monoxide (CO) is mainly derived from the combustion of fossil fuels and the presence of CO in domestic gas or coal stoves and automobile exhaust. Once CO enters the body, it immediately binds to hemoglobin, which forms carboxyhemoglobin (COHb), which causes tissue hypoxia (Dary et al., 1981; Li et al., 2017b), causing symptoms such as vomiting, dizziness, and even death (Jarvis et al, 1986; Scharte et al, 2000). Carbon monoxide (CO) is an important indicator of air pollutants, and thus, research on carbon monoxide has important implications. The monthly mean concentrations of CO in Urumqi and Turpan from 2015 to 2017 are shown in Tables 4(a), 4(b), and 4(c), respectively.

In Urumqi, the monthly average CO concentration ranged from 0.516 to 2.28 ppm in 2015, from 0.421 to 2.76 ppm in 2016, and from 0.605 to 2.60 ppm in 2017, corresponding to an annual average of 1.13, 1.17, and 1.13 ppm, respectively. The carbon monoxide concentration did not fluctuate in the three year period. The annual average concentration of carbon dioxide in 2016 increased by approximately 3.54% from 2015 and decreased by approximately 3.42% from 2016 to 2017.

In Turpan, the monthly average CO concentrations ranged from 0.480 to 3.07 ppm in 2015, 0.598 to 4.06 ppm in 2016, and 0.555 to 2.27 ppm in 2017, for which the corresponding annual averages were 1.33, 1.32, and 1.18 ppm, respectively. The average CO concentration in 2015 and 2016 did not change much, and it decreased by 0.75% from 2015 to 2016. From 2016 to 2017, the average CO concentration decreased by 10.6%, so carbon monoxide in the air has decreased annually.

The annual average CO concentrations in Urumqi and Turpan were 1.14 and 1.28 ppm, respectively. The concentrations of carbon monoxide in Urumqi and Turpan were both lower than the 8-hour average of the WHO Air Quality Standard (8.00 ppm), indicating that carbon monoxide has no serious impact on the air quality of the two cities.

As for seasonal changes in CO concentration, in Urumqi, the CO concentrations in the spring, summer, autumn, and winter of 2015 were 0.861, 0.558, 0.897, and 2.21 ppm, respectively; in 2016, they were 0.821, 0.479, 0.972, and 2.43 ppm, respectively. In 2017, they were 0.849, 0.629, 0.958, and 2.08 ppm, respectively. As for Turpan, in 2015, the concentrations of CO in spring, summer, autumn, and winter were 0.512, 0.578, 0.875, and 1.144 ppm, respectively, and in 2016, they were 0.666, 0.758, 1.17, and 1.26 ppm, respectively. In 2017, they were 0.933, 0.696, 1.17, and 1.54 ppm, respectively. In Urumqi, the CO concentration was the highest (2.43 ppm) in the winter of 2016, and the CO concentration was the lowest (0.479 ppm) in the summer of 2016. The highest value was five times higher than the lowest value. In Turpan, the CO concentration was the highest (1.54 ppm) in the winter of 2017, and the CO concentration was the lowest (0.512 ppm) in the spring of 2015. The highest value was three times higher than the lowest value. From the data, it can be seen that the CO concentration in the two cities varies greatly between the seasons, and in the winter, the concentration of CO was the highest in Urumqi, whereas it was at its lowest level in summer. In the spring and summer, both values were at the middle level. Turpan, however, had the lowest CO concentration in the spring and summer of 2015 and 2016, with the CO concentration in autumn at a moderate level and the highest concentration in winter. Seasonal variations in CO concentrations are related to temperature and are also related to the increase in CO concentration caused by coal combustion in winter.

O₃ Concentration

O₃ is an oxygen allotrope with poor stability and is a source of greenhouse gases. High levels of O₃ have strong oxidative properties that adversely affect human health and vegetation (Monks et al., 2015). O₃ is a greenhouse gas that plays an important role in global warming (Stocker et al., 2013). Ozone concentrations have been rising over the past decade (Akimoto, 2003; Vingarzan, 2004; Xu et al., 2011; Feng et al., 2015). With the development of urbanization and industrialization in China, O₃ has become a growing public concern (Ou et al., 2016; Gong et al., 2018). In contaminated air, the largest contributors to O₃ forming precursors are NO_x and VOCs, especially unsaturated VOCs. The simplified general equation for the regulation of atmospheric photochemistry is summarized in Wang et al. (2018).

Figs. 3(a), 3(b), and 3(c) show the changes in monthly average O₃ concentrations in Urumqi and Turpan in 2015, 2016, and 2017, respectively.

Fig. 3(a). Monthly average atmospheric O₃ concentrations in Urumchi and Turpan in 2015.

Fig. 3(b). Monthly average atmospheric O₃ concentrations in Urumchi and Turpan and in 2016.

Fig. 3(c). Monthly average atmospheric O₃ concentrations in Urumchi and Turpan in 2017.

In Urumqi, the monthly average concentration of O₃ in 2015 was between 2.30 and 77.4 ppb, and the average in 2015 was 28.2 ppb. The monthly average concentration of O₃ in 2016 was between 4.20 and 75.5 ppb, and the average in 2016 was 27.7 ppb. In 2017, the monthly average concentration was between 6.10 and 72.2 ppb, with an average of 34.0 ppb. We can see that the O₃ level in Urumqi dropped by about 1.77% from 2015 to 2016, and the O₃ level increased by about 22.7% from 2016 to 2017.

In the Turpan region, the monthly average concentration of O₃ in 2015 was between 4.70 and 67.6 ppb; the average concentration of O₃ in 2015 was 37.5 ppb; the monthly average concentration of O₃ in 2016 was between 6.99 and 95.1 ppb, and the average concentration of O₃ in 2016 was 45.9 ppb. The monthly average concentration of O₃ in 2017 was between 8.39 and 91.3 ppb, and the average concentration of O₃ in 2017 was 41.0 ppb. It increased by approximately 22.4% from 2015 to 2016 and decreased by approximately 10.7% from 2016 to 2017. The results show that the Turpan O₃ concentration is lower than the WHO air quality supervision standard, but we still need to implement appropriate control measures to reduce the O₃ concentration.

In Urumqi, seasonal average O₃ concentrations in the spring, summer, autumn, and winter of 2015 were 29.3, 51.5, 21.4, and 10.7 ppb, respectively. In the spring, summer, autumn and winter of 2016, they were 28.3, 47.5, 24.4, and 10.5 ppb, respectively, and in 2017, they were 35.6, 52.2, 30.8, and 17.5 ppb for spring, summer, autumn, and winter, respectively. In Turpan, the seasonal concentrations of ozone in the spring, summer, autumn, and winter of 2015 were 46.4, 51.3, 31.1, and 21.1 ppb, respectively, while those in 2016 were 61.7, 61.2, 37.9, and 23.0 ppb, respectively. In 2017, these values were 44.4, 62.4, 35.6, and 21.4 ppb, respectively. According to the data, in Urumqi, in the summer of 2015, 2016, and 2017, the ozone concentrations were 51.5, 47.5, 61.2, and 52.2 ppb, respectively, which exceeded the WHO's air quality supervision standard of 46.6 ppb. In Turpan, for the summer of 2015, the spring of 2016, and the summer of 2017, the ozone concentrations were 55.3, 61.7, 61.2, and 62.4 ppb, respectively, which exceeded the WHO's air quality regulated standard of 46.6 ppb.

Although the average O₃ concentration in both cities was below the World Health Organization's 46.6 ppb air quality regulated standards, in some seasons, the air O₃ concentration was higher than the air quality monitoring standard. According to the data, in Urumqi and Turpan, O₃ concentration in the air was in the following order: summer > spring > autumn > winter. This shows that summer is the season with the highest O₃ pollution. Previous studies have shown that in humid environments, the presence of water vapor facilitates the removal of O₃ (Fiore et al., 2002),and the movement of the gas stream also has an effect on the concentration of O₃. Urumqi has a large temperature difference between day and night, with severe changes in temperature and heat and less precipitation. Turpan has typical drought and desert characteristics, where the annual amount of radiation is large, especially in spring and summer. This leads to high temperatures in the summer and strong radiation in these two cities, causing NO_x to form O₃ under ultraviolet light. The higher the temperature is, the longer the illumination time is, and the easier it is for ozone to exceed the standard. The summer is a season with severe ozone pollution. After entering autumn, the ozone concentration gradually decreases. According to a survey conducted by the Environmental Monitoring Station in Urumqi, ozone in the air is not emitted directly, but is a secondary pollutant generated under the effect of sunlight. Motor vehicles are also one of the sources of O₃. Pollutants emitted by powerplants and the petrochemical industry are emitted under specific meteorological conditions. Therefore, it is necessary to control plant pollution emissions, reduce vehicle exhaust, improve gasoline quality, etc. to reduce the formation of O₃. 

AQI Analysis

AQI is the ambient air quality index. It is a universal global air quality assessment system used to evaluate air quality and assess health risks. Pollutants monitored include sulfur dioxide, nitrogen dioxide, PM₁₀, PM_2.5, carbon monoxide, and ozone. The air quality index (AQI) range and corresponding air quality categories can be divided into six categories: excellent air quality, good air quality, mild pollution, moderate pollution, heavy pollution, and severe pollution. Based on the different air quality categories, appropriate air pollution measures are implemented. Figs. 4(a)–4(f) shows a small portion of the six AQI categories in different seasons in Urumqi and Turpan from 2015 to 2017, where the accumulated days with the major pollutants are shown in the table. Figs. 3(a)–3(b). In Urumqi, in 2015, the daily AQI range was 30–357, and the annual average was 105. In 2016, the daily AQI range was 27–432, and the annual average was 110. In 2017, the daily AQI range was 35–402. The annual average was 110. In Turpan, in 2015, the daily AQI range was 37–500, and the annual average was 110. In 2016, the daily AQI range was 53–500, and the annual average was 130. In 2017, the daily AQI range was 54–500. The annual average was 121. As can be seen from the data, the AQI remained basically unchanged in Urumqi during the three years under investigation. There was no significant improvement in Turpan’s AQI during this period, with the highest AQI value in 2016. Among them, the maximum AQI in Turpan reached 500, which indicates that there was severe pollution in Turpan.

Fig. 4 (a). The fractions of the six AQI categories for Urumchi in (A) Spring, (B) Summer, (C) Fall, and (D) Winter in 2015.

Fig. 4(b). The fractions of the six AQI categories for Turpan in (A) Spring, (B) Summer, (C) Fall, and (D) Winter in 2015.

Fig. 4(c). The number fractions of the six AQI categories for Urumchi in (A) Spring, (B) Summer, (C) Fall, and (D) Winter in 2016.

Fig. 4(d). The fractions of the six AQI categories for Turpan in (A) Spring, (B) Summer, (C) Fall, and (D) Winter in 2016.

Fig. 4(e). The number fractions of the six AQI categories for Urumchi in (A) Spring, (B) Summer, (C) Fall, and (D) Winter in 2017.

Fig. 4(f). The fractions of the six AQI categories for Turpan in (A) Spring, (B) Summer, (C) Fall, and (D) Winter in 2017.

In Urumqi, Figs. 4(b)-(A), 4(d)-(A), and 4(f)-(A) show that in the spring of 2015, the I, II, III, IV, V, and VI ratios were 9%, 63%, 22%, 4%, 2%, and 0%, respectively. In 2016, the ratios of I, II, III, IV, V, and VI were 21%, 62%, 9%, 5%, 3%, and 0%, respectively. In 2017, the ratios of I, II, III, IV, V, and VI were 19%, 54%, 17%, 7%, 3%, and 0%, respectively. The proportion of Class I increased, rising by 10% from 2015 to 2017, but there was no significant fluctuation in the proportion of the remaining categories. As can be seen from Table 5(a), PM₁₀ was the most common major pollutant in the spring, followed by PM_2.5, and NO_x pollutants also were found. This shows that in addition to controlling PM pollution in Urumqi in the spring, controlling NO_x pollution is also critical.

Figs. 4(b)-(B), 4(d)-(B), and 4(f)-(B) show that in the summer of 2015, the ratios of I, II, III, IV, V, and VI were 9%, 78%, 13%, 0%, 0%, and 0%, respectively. In 2016, the ratios of I, II, III, IV, V, and VI were 16%, 83%, 1%, 0%, 0%, and 0%, respectively. In 2017, the ratios of I, II, III, IV, V, and VI were 11%, 87%, 2%, 0%, 0%, and 0%, respectively. From 2015 to 2017, Grades I and II increased by 2% and 8%, respectively, and Grade III decreased by 11%. This shows that Urumqi's summer air quality is gradually improving. As can be seen from Table 5(a), in the summer of Urumqi, PM₁₀ was the primary air pollutant, followed by O₃. High temperatures in the summer and intense radiation accelerate the formation of O₃. From 2015 to 2017, the number of days with O₃ pollution in the air increased. Therefore, it is urgent to take effective measures to prevent this increase of O₃ in the summer.

Figs. 4(b)-(C), 4(d)-(C), and 4(f)-(C) show that in the fall of 2015, the I, II, III, IV, V and VI scales were 13%, 66%, 18%, 2%, 1%, and 0%, respectively. In 2016, the ratios of I, II, III, IV, V, and VI were 17%, 66%, 10%, 4%, 3%, and 0%, respectively. In 2017, the ratios of I, II, III, IV, V, and VI were 10%, 65%, 21%, 4%, 0%, and 0%, respectively. From 2015 to 2017, Grades I and II decreased by 3% and 1%, respectively, and Grades III and IV increased by 3% and 2%, respectively. In 2015 and 2016, the proportions of V grades were 1% and 3%, respectively. In 2017, the proportion of V grades was 0%. Table 5(a) shows that PM₁₀ was the main air pollutant, followed by PM_2.5 and NO_x. NO_x has increased significantly relative to the other two seasons because of the burning of coal during the autumn heating period, which increases NO_x emissions.

Figs. 4(b)-(D), 4(d)-(D), and 4(f)-(D) show that in the winter of 2015, I, II, III, IV, V, and VI proportions were 2%, 22%, 21%, 21%, 25%, and 9%, respectively. In the winter of 2016, I, II, III, IV, V, and VI were 0%, 6%, 23%, 14%, 42%, and 15%, respectively. In the winter of 2017, I, II, III, IV, V and VI were 0%, 15%, 23%, 11%, 34%, and 17%, respectively. It can be seen that the winter air quality in Urumqi was relatively poor. Comparing the proportions of different AQI categories from 2015 to 2017, I, II, and IV fell by 2%, 7%, and 10%, respectively, while Grades V and VI increased by 9% and 8%, respectively. In the three years under investigation, the air quality in Urumqi deteriorated. Table 5(a) shows that PM_2.5was the main air pollutant, followed by PM₁₀ and NO_x. This was also because the combustion of coal in winter boilers leads to an increase in PM_2.5 emissions, so PM_2.5 should be controlled during the heating period.

Figs. 4(a)-(A), 4(c)-(A), and 4(e)-(A) show that in 2015, the proportions of I, II, III, IV, V and VI were 0%, 70%, 13%, 8%, 0%, and 9%, respectively. In 2016, these proportions were 0%, 53%, 28%, 3%, 6%, and 10%, respectively. In 2017, these proportions were 0%, 60%, 23%, 13%, 1%, and 3%, respectively. Comparing the proportion of different AQI categories from 2015 to 2016, the I level was 0%, while the II and IV levels were reduced by 24.3% and 62.5%, respectively. Level III rose from 13% to 28%. The V level also rose by 6%. The decrease in Grade II and the increase in Grade III and Grade V indicate that Turpan’s air quality gradually deteriorated from 2015 to 2016. Grade VI occurred frequently in 2015 and 2016, at 6% in 2015 and 10% in 2016. This indicates that air pollution incidents frequently occurred in the spring of 2015 and 2016 in Turpan and that air pollution incidents from 2015 to 2016 increased. Grade II increased by 13.2% from 2016 to 2017, but the III, V and VI grades decreased. This may be due to changes in meteorological conditions and increased vegetation coverage, which is conducive to the dilution of pollutants. Table 5(b) shows that PM₁₀ was the most common major atmospheric pollutant, followed by PM_2.5 and O₃. Especially in 2016, the frequency of O₃ as the main pollutant was significantly higher than other years. This was also due to the emission of NO_x and VOC, as well as the occurrence of high temperature solar radiation. This result shows that, in addition to reducing PM pollution in the spring, controlling ozone pollution is essential for improving air quality.

Figs. 4(a)-(B), 4(c)-(B), and 4(e)-(B) show the distribution of six AQI species in Turpan in summer. In 2015, the proportions of I, II, III, IV, V, and VI were 2%, 87%, 8%, 1%, 1%, and 1%, respectively; in 2016, they were 0%, 61%, 33%, 3%, 0%, and 3% respectively. In 2017, they were 0%, 76%, 19%, 2%, 1%, and 2%, respectively. During the three years, the number of days with Level II accounted for the majority, followed by Level III. There was a significant increase in Grade III weather from 2015 to 2016, from 8% to 33%. In 2017, Level III accounted for 19%. Compared with spring, the proportion of days of IV, V, and VI were relatively small, which indicates that the air in summer was relatively good. As can be seen from Table 5(b), in the three year period, PM₁₀ and O₃ were the most common pollutants in Turpan in the summer. This was also due to the increase in O₃ concentration due to the increase in VOC emissions in recent years. High temperature weather and solar radiation can promote the formation of O₃. This result is consistent with previous research results (Atkinson and Arey, 2003; Zhang and Ying et al., 2011; Li et al., 2012; He et al., 2017; Shen et al., 2017).

Figs. 4(a)-(C), 4(c)-(C), and 4(e)-(C) show that in the fall of 2015, the proportions of I, II, III, IV, V, and VI were 9%, 81%, 10%, 0%, 0%, and 0%, respectively. In 2016, the I, II, III, IV, V and VI proportions were 8%, 87%, 5%, 0%, 0%, and 0%, respectively, and in 2017, they were 4%, 89%, 7%, 0%, 0%, and 0%, respectively. Grade II accounted for the majority. From 2015 to 2017, Grade II gradually increased from 81% to 89%, and the ratio of I and III decreased. At the same time, the ratios of IV, V, and VI were 0%. This shows that the air quality in Turpan was good in the autumn and that there were no serious pollution incidents. As can be seen from Table 5(b), PM₁₀ was the main pollutant in the autumn in Turpan, followed by PM_2.5, and the concentration of O₃ was significantly reduced in autumn.

Figs. 4(a)-(D), 4(c)-(D), and 4(e)-(D) show that in the winter of 2015, the proportions of I, II, III, IV, V, and VI were 8%, 58%, 27%, 4%, 3%,and 0%, respectively. In 2016, the ratios of I, II, III, IV, V, and VI were 2%, 84%, 8%, 4%, 2%, and 0%, respectively. In 2017, the ratios of I, II, III, IV, V, and VI were 4%, 88%, 8%, 0%, 0%, and 0%, respectively. Grade II was still the majority. From 2015 to 2017, the proportion gradually increased. From 2015 to 2016, it increased by 26%. From 2016 to 2017, it increased by 4%, while the proportion of Grade III decreased from 27% in 2015 to 8% in 2016. There were grades IV and V in 2015 and 2016, with 4% in 2015 and 2016, and 3% in 2015 and 2016. In 2017, the proportion of Grade IV and V was 0%. This shows that the air quality in Turpan gradually improved in the winter during the three year period. As can be seen from Table 5(b), PM_2.5 was the main pollutant in winter in Turpan, followed by PM₁₀. This is mainly related to large quantities of coal burning and adverse meteorological conditions in winter (Sun et al., 2014; Zhou et al., 2017). Compared with summer, due to the low temperatures in winter, the working temperature of car engines is not high, causing incomplete combustion that results in PM_2.5.

In general, the air quality in Turpan is better than that in Urumqi. Overall, during the three years of the study, in Urumqi, in spring, the average ratios of I, II, III, IV, V, and VI were 16.3%, 60.0%, 16.0%, 5.33%, 2.67%, and 0%, respectively; were 12.0%, 82.7%, 5.33%, 0%, 0%, and 0% in summer, were 13.3%, 65.7%, 16.3%, 3.33%, 1.33%, and 0% in fall, respectively, and were 0.67%, 14.3%, 22.3%, 15.3%, 33.7%, and 13.7% in winter, respectively. The average ratios of I, II, III, IV, V, and VI in the spring of Turpan were 0%, 61.0%, 21.3%, 8.00%, 2.33%, and 7.33%, respectively. In the summer, they were 0.67%, 72.7%, 20.0%, 2.00%, 0.67%, and 2%, respectively, and were 7.00%, 85.7%, 7.33%, 0%, 0%, and 0%, respectively, in fall. In winter, they were 4.67%, 76.7%, 14.3%, 2.67%, 1.67%, and 0%, respectively. In general, the summer air quality was the best, and the spring and winter air quality were the worst. Due the seasonal changes, the main pollutants in the air also change accordingly. In Urumqi, in the spring, PM₁₀ was the main air pollutant, followed by PM_2.5, and NO_x pollution also occurred. In summer, PM₁₀ and O₃ were the main air pollutants. In the fall, PM₁₀ was the main air pollutant, followed by PM_2.5 and NO_x. Especially in the autumn in Urumqi, the number of days of NO_x pollution was high. In the winter, PM_2.5 was the main air pollutant. In Turpan, in the spring and autumn, PM₁₀ was the main air pollutant, where the majority of pollution were from PM₁₀. In the summer, in addition to PM₁₀, O₃ was also a main air pollutant. In the winter, PM_2.5 replaced PM₁₀ as the primary pollutant. Therefore, as the seasons change, it is necessary to control PM_2.5, PM₁₀, and O₃. In Urumqi, in addition to controlling PM_2.5, PM₁₀, and O₃, it is also important to take appropriate measures to control NO_x.

CONCLUSION

The results of this study on atmospheric deposition in Urumqi and Turpan can be summarized as follows:

1. The average concentration of PM₅ in Urumqi for the 3 years under investigation was 19–228 µg m^–3; the annual average concentration was 69 µg m^–3; the average concentration of PM_2.5 in Turpan for the 3 years was 23–136 µg m^–3, and the average annual concentration was 69 µg m^–3. The average concentration of PM_2.5 in the two cities was very similar, and the concentration of PM_2.5 in the winter was higher than that in the summer.
2. The 3-year PM₁₀ concentration in Urumqi ranged between 61 and 287 µg m^–3, with an average of 123 µg m^–3. The Turpan data ranged from 61 to 282 µg m^–3, with an average of 157 µg m^–3. The PM₁₀ levels in Turpan were higher than those in Urumqi. In general, PM₁₀ concentration in the summer was lower than that in the winter. The PM₁₀ concentration in the summer in Urumqi (80.2 µg m^–3) was 63.0% lower than that in the winter (217 µg m^–3). In Turpan, the concentration in the summer (75.0 µg m^–3) was 34.2% lower than that in the winter (114 µg m^–3).
3. The three-year average SO₂ concentration fluctuated from700 to 27.7 ppb in Urumqi and from 0.700 to 47.6 ppb in Turpan, with corresponding averages of 5.11 and 5.69 ppb, respectively. The results show that the SO₂ concentrations in the spring and winter of the two cities fluctuated significantly, but the SO₂ concentration had little impact on the air quality of the two cities.
4. During the three-year period, the average NO₂ concentration in Urumqi ranged from 15.9 to 43.9 ppb, with an average of 22.3 ppb. The average NO₂ concentration in Turpan ranged from 9.74 to 34.1 ppb, with an average of 19.1 ppb. In general, NO₂ concentrations in the autumn and winter were higher than those in the spring and summer. The concentration of NO₂ in Urumqi was higher than that in Turpan, and NO₂ was an important pollutant. Therefore, it is necessary to control the emission of NO₂ from automobile exhaust and winter boilers in order to meet the environmental protection requirements.
5. During the three-year period, the average CO in Urumqi ranged from 0.421 to 2.76 ppm, and the average value in Turpan ranged from 0.480 to 4.06 ppm, corresponding to an average of 1.14 and 1.28 ppm, respectively. The concentration of carbon monoxide in the two cities was lower than the 8:00 ppm required by the WHO's air quality supervision standard, which indicates that carbon monoxide has never had a serious impact on the air quality in the two cities. Under normal circumstances, the highest concentration of carbon monoxide was in the winter, and the lowest concentration of carbon monoxide occurred in the summer.
6. During the three-year period, the average O₃ concentration in Urumqi ranged from 7.20 to 57.4 ppb, and the average concentration of Urumqi was 30.0 ppb. The average O₃ concentration in Turpan ranged from 13.0 to 73.2 ppb, with an average concentration of 41.5 ppb. O₃ pollution generally occurred during the summer months. The highest concentration exceeded the WHO Air Quality Regulation Standard (46.6 ppb). Ozone pollution has a significant impact on the environment, and controlling ozone pollution is a serious challenge.
7. The air quality of Urumqi and Turpan fluctuates seasonally. In the spring in Urumqi, air quality was mainly concentrated in Grade II and Grade III, with mildly polluted weather. PM₅ and PM₁₀ were the main pollutants. In the spring in Turpan, unlike the spring in Urumqi, air quality was mainly concentrated in Class II, Class III, and Class IV, where PM₁₀ and O₃ were the main pollutants.
8. In the summer in Urumqi, air quality was mainly concentrated in Class II, and PM₁₀ and O₃ were the main pollutants. In the summer in Turpan, air quality was mainly concentrated in Class II and Class III, and PM₁₀ and O₃ were the main pollutants.
9. In the autumn in Urumqi, air quality was mainly concentrated in Class II and Class III, and PM₅, PM₁₀, and NO_x were the main pollutants. In the fall in Turpan, air quality was mainly concentrated in grades II and III, and PM_2.5 and PM₁₀ were the main pollutants.
10. In the winter in Urumqi, air pollution was serious; PM₅ was the main pollutant, and occasionally NO_x pollution occurred. In the winter in Turpan, air quality was mainly concentrated in Class III, Class IV, and Class V, and PM_2.5 and PM₁₀ were the main pollutants. The winter air pollution levels of the two cities were much higher than in the other three seasons.
11. The results of this study provide useful information for the establishment of air pollution control strategies and for the future studies made by the scientific community on this issue.

Aerosol Air Qual. Res. 19 :282 -306 . https://doi.org/10.4209/aaqr.2018.11.0410