Gaseous and Particulate Pollution in the Wu-Chang-Shi Urban Agglomeration on the Northern Slope of Tianshan Mountains from 2017 to 2021

: Rapid social development has led to serious air pollution problems in cities, and air pollutants, including gaseous pollutants and particulate matter, have an important impact on climate, the environment, and human health. This study analyzed the characteristics, potential sources, and causes of air pollution in the Wu-Chang-Shi urban cluster. The results showed that NO 2 , CO, SO 2 , PM 10 , and PM 2.5 had a tendency to decrease, while O 3 showed an increasing trend. The concentrations of SO 2 , NO 2 , CO, PM 2.5 , and PM 10 showed the highest values in winter and the lowest values in summer, with similar seasonal variations. However, the concentration of O 3 was highest in the summer and lowest in the winter. Compared with the pollutant concentrations in other Chinese cities, PM 2.5 , PM 10 , and NO 2 are more polluted in the Wu-Chang-Shi urban. Meteorological factors have a greater impact on pollutant concentrations, with higher concentrations of major pollutants observed when wind speeds are low and speciﬁc wind directions are observed, and higher secondary pollutant O 3 concentrations observed when wind speeds are low and speciﬁc wind directions are observed. The backward trajectory and concentration weighting analysis show that the particulate pollutants in the Wu-Chang-Shi urban in winter mainly come from Central Asia and surrounding cities. O 3 showed an increasing trend before and after the novel coronavirus outbreak, which may be related to changes in NO X , volatile organic compounds, and solar radiation intensity, and the concentrations of SO 2 , NO 2 , CO, PM 10 , and PM 2.5 showed an overall decreasing trend after the outbreak and was smaller than before the outbreak, which is related to the reduction of industrial and anthropogenic source emissions during the outbreak.


Introduction
Global urbanization and industrialization are accelerating, resulting in increased energy consumption and air pollutant emissions.Air pollution events are frequent and expanding in scope, seriously affecting regional air quality and the global climate [1,2].Many countries are facing serious environmental problems, and air pollution is causing great harm to the health of the inhabitants, which has received widespread attention from all sectors of society and the public [3][4][5].The World Health Organization (WHO) air quality guidelines show that 92% of the global population lives in urbans that do not meet the guidelines [6].The Asian Development Bank study shows that 7 of the top 10 polluted cities in the world were in China, and only 5 of the 7 cities met WHO air quality standards [7].In the whole of China, air pollution is characterized by a combination of local and regional factors and the cross-coupling of various pollutants [8,9].Air pollution has caused serious damage to the health of Chinese residents.In 2013, a large area of moderate to severe haze occurred in east-central China, covering 2.07 million square kilometers and affecting 11 provinces (municipalities and districts), and In 2015, air pollution has caused 1.6 million deaths in China, and the effects are widespread and persistent [10,11].Air pollutants mainly come from human activities, such as industrial production, transportation, etc, and some come from natural emissions, such as crustal movement, forest fires, etc [12,13].At the moment, the atmospheric environment of China's urbans is primarily polluted by SO 2 , NO x , O 3 , and PM [14].
Air pollution research in China began about 20 years later than in Europe and the U. S. [15].In 1972, China held its first national conference on environmental protection after participating in the United Nations Conference on the Human Environment, which pioneered the control of air pollution in China by eliminating smoke and dust and controlling point source pollution in the atmosphere.In 1974, a serious photochemical pollution incident occurred in Xigu District, Lanzhou City, Gansu Province, and this photochemical pollution incident sounded an alarm to us.As a result, fixed-point monitoring of nitrogen oxides, ozone, and atmospheric particulate matter was carried out for the first time in Lanzhou City.China's rapid industrialization and urbanization have been accompanied by high-intensity air pollution emissions since the reform and opening up [16].Large-scale regional heavy haze pollution events have occurred in economically developed areas of east-central and southern China.Particulate matter prevention and control efforts are planned in key urbans [17].Therefore, in China, air pollution is of increasingly widespread concern to the government, the public, and researchers [18].In particular, the sudden outbreak of new coronary pneumonia (COVID-19) at the end of 2019 has caused widespread global concern [19,20].While we took strict preventive and control measures after the outbreak, air quality conditions during the outbreak became a major concern [21].In general, air quality in China improved after the emergence of the COVID-19 outbreak, but air pollution events still occurred due to harsh climatic conditions in northwest China [22,23].Many scholars have studied the spatial and temporal distribution characteristics, source analysis and causes of pollutants, and health effects of air pollution in major cities in China (Beijing-Tianjin-Hebei region, Fenwei Plain, Yangtze River Delta region, Pearl River Delta region, Sichuan Basin, etc.), and many research results have been obtained [24].However, these studies have mainly focused on the economically developed and densely populated southeast region of China, and fewer studies have been conducted on the northwest region.
The Wu-Chang-Shi urban is located on the northern slope of the Tianshan Mountains.The core urban for the construction of the "Silk Road Economic Belt" is home to 31% of Xinjiang's industrial enterprises.In 2016, coal consumption was nearly 65 million tons, and soot emissions, sulfur dioxide emissions, and nitrogen oxide emissions accounted for 25%, 38.4%, and 45.3% of the total in Xinjiang, respectively.For this reason, Xinjiang has made the Wu-Chang-Shi urban a key urban for air quality monitoring in Xinjiang.The monitoring results show that the ambient air quality in the Wu-Chang-Shi urban is not optimistic.From November 2018 to March 2019 alone, the Xinjiang Uygur Autonomous Region Heavy Pollution Weather Emergency Response Command Office launched five heavy pollution weather attacks in the Wu-Chang-Shi urban, a yellow weather warning.It can be seen that the air quality changes in the Wu-Chang-Shi urban need to be studied in depth.
As the core area of the economic zone on the northern slope of Tianshan Mountain and an important node of the "Belt and Road", it is significant to study and control the air pollution in the Wu-Chang-Shi urban.At present, the studies conducted for the Wu-Chang-Shi urban are limited to single pollutants and lack continuity; there is a lack of studies on multiple pollutants, and the regional air quality is still unsatisfactory, especially in winter when particulate matter pollution is the most serious, and heavy pollution weather often occurs [25,26].Therefore, in this study, we analyzed the pollution characteristics of six criteria air pollutants (SO 2 , NO 2 , CO, O 3 , PM 10 , and PM 2.5 ) in Wu-Chang-Shi urban from Atmosphere 2023, 14, 91 3 of 22 2017 to 2021 and compared them with those in other Chinese cities, as well as investigating the characteristics of changes before and after the COVID-19 outbreak.Based on the latest air pollution dataset, we also investigated potential sources of particulate pollutants in winter.This study provides some scientific basis for the prevention and control of atmospheric compound pollution in the Wu-Chang-Shi urban.

Studying urban
The Wu-Changji-Shi region is located at the northern edge of the Tianshan Mountains and the southern edge of the Junggar Basin (42 • 52 -45 • 28 N, 88 • 40 -91 • 33 E), with a long and narrow topography and the city located in the pre-mountain basin.Its geographical location is shown in Figure 1.The population is 5.46 million (3.51 million in Urumqi, 1.61 million in Changji, and 346,400 in Shihezi).As an important part and center of the economic belt on the northern slope of Tianshan Mountain, there are about 3200 industrial enterprises in Wu-Chang-Shi urban.In 2016, coal consumption in the region was 65 million tons; emissions were 263,000 tons and NO x emissions were 323,000 tons.Due to the low average temperature and long heating period, the annual heating period is up to 6 months, from October of the previous year to April of the following year.With socioeconomic development, emissions in urbans have continued to rise in recent years.The Wu-Chang-Shi urban has a typical temperate continental climate with noticeable seasonal temperature changes and an average temperature ranging from −13.2 • C to 26.5 • C. The average annual precipitation is about 256 mm, mainly concentrated in spring and summer, followed by autumn, and the average annual wind speed is 1.5 m s −1 .

Data Collection
In this paper, we selected 24-h data and daily average data of air pollutants concentrations of six air pollutants (SO 2 , NO 2 , CO, O 3 , PM 2.5 , and PM 10 ) and data from AQI in Wu-Chang-Shi urban from 2017 to 2021, which were obtained from China General Environmental Monitoring (http://113.108.142.147:20035/emcpublish/accessed on 1 March 2022).We also compared the air pollution concentration data of the Wu-Chang-Shi urban in 2017 with the air pollution concentration monitoring data of 168 key cities in China.The air pollutants concentration data of 168 key cities were obtained from the National Bureau of Statistics of China (http://www.stats.gov.cn/accessed on 1 March 2022).The information used in the backward trajectory model is the Global Data Assimilation System (GDAS) data for 2017-2021 provided by the National Centers for Environmental Prediction (NCEP).The meteorological information was obtained from the National Meteorological Science Data Sharing Service Platform ( http://data.Cma.cn/accessed on 5 March 2022).The statistical summary of the six criteria pollutants and air quality index for 2017-2021 is shown in Table 1.Previously, we developed a force plate for humans using multi-eddy current displace ment sensor elements in the same way [27].By using multiple sensor elements, multi-axial forces and moments were detected.On the contrary, the design guideline of this study is to accurately measure the uniaxial force using a single sensor element.Planar springs are suitable for measuring vertical force because a horizontal deformation does not tend to occur compared to that in 3D spring structures.
As shown in Figure 2, eddy currents are generated in the target metal when it moves due to the electromagnetic induction effect.This phenomenon changes the impedance of the planar coil.The sensor measures the distance based on the oscillation changes caused by this phenomenon.The insufficient size of the target metal or planar coil of the eddy current displacement sensor can cause a reduction in sensitivity.Therefore, the relationship between the size of the planar coil and the target metal should be appropriate.Aluminum is used as the target metal because of its high conductivity and ease of processing.First, we checked the data to control the quantity and quality of pollutant data according to the National Environmental Protection Standards of the People's Republic of China (HJ630-2011) and the National Ambient Air Quality Standards of the People's Republic of China (GB3095-2012) to ensure that no data were missing and that the data were complete.In addition, we also processed the published pollution data as a way to assess the pollution status [27].We analyzed the data for independence, normality, and reliability using the run test and the Kolmogorov-Smirnov test, which showed a Cronbach's alpha of 0.866 (>0.6), which confirmed the quality of the data and that the data characteristics were favorable.Statistical methods have also been used in previous studies to confirm the reliability of the data [28][29][30].

Data Analyses
We use MeteoInfo software and its Trajstat plug-in to implement the calculation of the backward trajectory model, HYSPLIT is the model for calculating trajectories in TrajStat.The HYSPLIT model is relatively mature and has been commonly used in studies related to the sources and transport paths of atmospheric pollutants [31][32][33][34].Meteoinfo software has powerful data processing capabilities, developed by Chinese meteorologists, and the HYSPLIT model is an integrated model system jointly developed by the United States and Australia [35].Compared to HYSPLIT, the Trajstat plug-in has the advantage of an intuitive display of air mass trajectories and facilitates subsequent work (trajectory clustering, PSCF, CWT, etc.) to be performed [36].As a result, we use the HYSPLIT and NCEP/NCARGDAS datasets in this paper to calculate the backward air mass trajectories of winter air masses arriving at the Wu-Chang-Shi urban to determine potential long-range transport paths.We determined the potential source areas of the transported airflow by potential source contribution function (PSCF) and concentration-weighted trajectory (CWT) analysis calculations.

HYSPLIT Model and Backward Trajectory
The HYSPLIT model is a hybrid Eulerian and Lagrangian diffusion model with more complete motion, diffusion, and deposition processes [37,38].It is now commonly used for the analysis of air pollutants' transport pathways and sources [31,39].In this paper, the Wu-Chang-Shi urban is used as the simulated reception point, the starting altitude of the trajectory is 500 m above the ground, and the GDAS meteorological data is used to calculate the backward airflow trajectory for 72 h (4-time points of 00:00, 06:00, 12:00, 18:00 (Beijing time) per day) when arriving at the Wu-Chang-Shi urban.For the calculation time of the backward trajectory, we choose a backward trajectory length of 72 h considering that the airflow transport is usually within 1 km at the boundary layer and the particulate matter lasts about 2-3 days at the regional scale or mesoscale [40].

Cluster Analysis
One method of multivariate statistical techniques is cluster analysis, which mathematically classifies them into progressive typological categories.HYSPLIT is to group and cluster all trajectories according to the spatial similarity, transmission speed, and direction of air mass trajectories, calculate the Spatial heterogeneity (SPVAR) and total spatial heterogeneity (TSV) of each trajectory combination, and determine the number of clusters by the relationship between TSV and n.There are 2 clustering methods in TrajStat software, namely Euclidean distance, and angular distance, and this paper focuses on the airflow trajectory direction that reaches the receiving point.In this paper, the latter method is used.The study urban is gridded into a horizontal grid of 0.25 • × 0.25 • , and the TrajStat software is used to cluster the PSCF and CWT analysis of the Wu-Chang-Shi urban to obtain the different types of transport airflow and potential source urbans in winter.

Potential Source Contribution Function
The PSCF method is a conditional probability function that indicates the pollution contribution of a grid to a recipient point by the ratio of pollution trajectories to the residence time of all trajectories passing through a grid, and it defines that the value of an element corresponding to the arrival of an air mass passing through a region at a recipient point exceeds a set threshold (the thresholds for PM 2.5 and PM 10 in this paper are 75 µg m −3 and 150 µg m −3 , respectively), and calculates the element value of each trajectory within the grid [41,42].The element value of each trajectory within the grid range is calculated, and if the element value is higher than the set threshold, the trajectory is considered a polluted trajectory, where m ij is the number of polluted trajectories passing through the grid (i, j) in the study urban, n ij is the number of all trajectories in the grid (i, j), and w ij is the weighting factor, and the weighting function is determined by referring to the research results of Zeng et al. and Tian et al. [43][44][45][46].namely: W ij is defined as follows:

Concentration Weighted Trajectory Analysis
The PSCF algorithm cannot distinguish the magnitude of the contribution of grids with the same PSCF value to the pollutant concentration at the recipient point, i. e., the range of the extent to which the elemental value of the trajectory within the grid is above the set threshold, so we introduce the CWT method to reflect the degree of pollution contribution of different grid urbans to the study urban [47].The specific method is as follows: where C ij is the average weight concentration of grid (i, j), l is the trajectory, M is the number of trajectories in the grid (i, j), C l is the mass concentration of the pollutant at the receiving point corresponding to trajectory l when it passes through the grid (i, j), and τ ijl is the time trajectory l stays in the grid (i, j) [48], adopts the same weight factor W ij as PSCF to reduce the uncertainty of C ij [49], namely:

Urban Gaseous and Particulate Pollutants Profiles
Based on all measurements, the average concentrations of air pollutants in the three cities over the entire study period (2017-2021) are shown in Table 2 below.Through comparative analysis, we found that the concentration values of AQI and O 3 were Shihezi > Changji > Urumqi, the highest concentration values of SO 2 and CO were found in Changji, the highest concentration values of NO 2 were found in Urumqi, followed by Changji and Shihezi, the highest value of PM 10 was found in Changji among the three cities, and the highest value of PM 2.5 was found in Shihezi.Compared to the other two cities, Shihezi needs to pay more attention to air quality changes.Figure 2 shows the inter-annual concentration change trends of six major atmospheric pollutants and AQI in three cities from 2017 to 2021 in the past five years.Its changing trends are different.SO 2 , NO 2 , and CO in Changji City had a tendency of first decreasing and then increasing, but an overall trend of decrease.SO 2 , NO 2 , and CO all showed their lowest values in 2020 (8.48 µg m −3 , 32.61 µg m −3 , and 0.93 µg m −3 ), PM 2.5 and PM 10 are on a downward trend with the lowest values in 2021 (51.75 µg m −3 , 114.52 µg m −3 ).Among the six air pollutants in Shihezi, SO 2 , PM 10 and PM 2.5 showed a decreasing trend, and CO and NO 2 showed a decreasing and then increasing trend, all reaching their lowest values in 2020 (0.85 µg m −3 , 32.87 µg m −3 ).Among the 6 air pollutants in Urumqi, SO 2 , CO, PM 10 , and PM 2.5 trended to decrease, and the values reached the lowest in 2021 (7.21 µg m −3 , 0.81 µg m −3 , 71.87 µg m −3 , 39.74 µg m −3 ), NO 2 showed it first decreased and then increased, and the value reached the lowest value in 2020 (35.52 µg m −3 ), but an overall trend of decrease.O 3 showed an upward trend in all three cities.From Figure 2 and Table 2, we can see that the five-year AQI averages of the three cities are 99.60,102.62, and 93.08, respectively.The AQI value in Shihezi exceeded the national secondary standard, but did not exceed the national tertiary standard, being the largest among the three cities.This indicates that the air quality in Shihezi is in a lightly polluted state, while the air quality in the other two cities is in a good state.In terms of changing trends, the AQI in all three cities showed a decreasing trend, indicating that air quality is gradually improving.

Seasonal Variation
The seasonal changes of air pollutants in the Wu-Chang-Shi region from 2017 to 2021 are shown in Figure 3 (Figure 3 is for Changji City, Supplementary Figures S1 and S2 are for Shihezi and Urumqi).The characteristics of the six air pollutants in each of the four seasons are shown below.
The concentrations of SO 2 , CO, and NO 2 in the cities of Changji, Shihezi, and Urumqi all reached their maximum in winter.In Changji, the lowest concentrations of SO 2 and CO were in summer, at 8.14 µg m −3 and 0.63 µg m −3 , respectively; spring and summer were the seasons with the lowest concentrations of NO 2 (30.95 µg m −3 ).In Shihezi, SO 2 and NO 2 concentrations were lowest in summer with 8.14 µg m −3 and 0.63 µg m −3 , respectively, while CO (30.95 µg m −3 ) was lowest in spring and summer.In Urumqi, the concentrations of SO 2 (8.14 µg m −3 ), CO (30.95 µg m −3 ), and NO 2 (0.63 µg m −3 ) were the lowest in summer.The seasonal variations of SO 2 , NO 2 , and CO in all three cities showed a "U" shape, with concentrations in winter (December to February) > autumn (September to November) > spring (March to May) > summer (June to August).PM 2.5 and PM 10 concentrations in Changji, Shihezi, and Urumqi all show a winter > spring > autumn > summer pattern, and all showed "U"-shaped changes.
Atmosphere 2023, 14, x FOR PEER REVIEW 8 of 24 air quality in Shihezi is in a lightly polluted state, while the air quality in the other two cities is in a good state.In terms of changing trends, the AQI in all three cities showed a decreasing trend, indicating that air quality is gradually improving.

Seasonal Variation
The seasonal changes of air pollutants in the Wu-Chang-Shi region from 2017 to 2021 are shown in Figure 3 (Figure 3 is for Changji City, Supplementary Figures S1 and S2 are for Shihezi and Urumqi).The characteristics of the six air pollutants in each of the four seasons are shown below.The above findings are related to primary emissions from household heating and unfavorable dispersion conditions in the Wu-Chang-Shi urban.Household heating in the Wu-Chang-Shi urban usually starts in October this year and lasts until April of the following year, which requires the burning of large amounts of fossil fuels during the heating period, while a large number of family visits and tourists increases during the Spring Festival, leading to a sharp increase in traffic emissions, which, together with unfavorable diffusion conditions in winter, leads to higher air pollutant concentrations in the urban [50].However, the cessation of heating in summer reduces the burning of fossil fuels and reduces anthropogenic emission sources, which, together with abundant precipitation and enhanced atmospheric convective activity in summer, results in faster diffusion and lower concentrations of pollutants.Frequent sandstorms and dust in spring also increase PM 2.5 and PM 10 concentrations [4].The concentrations of SO2, CO, and NO2 in the cities of Changji, Shihezi, and Urumq all reached their maximum in winter.In Changji, the lowest concentrations of SO2 and CO were in summer, at 8.14 μg m −3 and 0.63 μg m −3 , respectively; spring and summer wer the seasons with the lowest concentrations of NO2 (30.95 μg m −3 ).In Shihezi, SO2 and NO concentrations were lowest in summer with 8.14 μg m −3 and 0.63 μg m −3 , respectively while CO (30.95 μg m −3 ) was lowest in spring and summer.In Urumqi, the concentration of SO2 (8.14 μg m −3 ), CO (30.95 μg m −3 ), and NO2 (0.63 μg m −3 ) were the lowest in summe The seasonal variations of SO2, NO2, and CO in all three cities showed a "U" shape, wit concentrations in winter (December to February) > autumn (September to November) spring (March to May) > summer (June to August).PM2.5 and PM10 concentrations i Changji, Shihezi, and Urumqi all show a winter > spring > autumn > summer pattern, an all showed "U"-shaped changes.
NO2 is photochemically active and reacts chemically with other precursors to pro duce O3 in the presence of sunlight, so the concentration of NO2 is lower in spring an summer.On the contrary, the maximum O3 concentration appeared in summer, and th average O3 concentrations in Changji, Shihezi, and Urumqi were 75.47 μg m −3 , 108.79 μ m −3 , and 74.91 μg m −3 , respectively.The seasonal variation of O3, in contrast to other pr mary pollutants, is due to the high temperatures and intense solar radiation in summe that favor the formation of O3.
The above findings are related to primary emissions from household heating an unfavorable dispersion conditions in the Wu-Chang-Shi urban.Household heating in th Wu-Chang-Shi urban usually starts in October this year and lasts until April of the fo lowing year, which requires the burning of large amounts of fossil fuels during the heatin period, while a large number of family visits and tourists increases during the Spring Fes tival, leading to a sharp increase in traffic emissions, which, together with unfavorabl

Influence of Meteorological Factor on Pollutants Concentration
A growing number of studies show that pollutants are closely related to meteorological elements and that local emissions and meteorological factors affect pollutants concentrations [51,52].Previous research has shown that meteorological factors can cause changes in atmospheric pollutant concentrations even when local emissions are relatively stable.To further study the meteorological factors that affect air pollutants, we have studied wind speed and direction, which affect the speed and direction of atmospheric transport.
This paper presents a comprehensive and comparative analysis of wind speed, wind direction, and concentrations of six atmospheric pollutants, to visualize the relationship between atmospheric pollutants and meteorological factors.The results are shown in Figure 4 below (Figure 4 below shows the city of Changji, Supplementary Figures S3 and S4 is for Shihezi and Urumqi).The results of the study showed that the highest concentrations of SO 2 , NO 2 , and CO were found in Changji when the wind direction was southwest and the wind speed was less than 1.5 m s −1 .This may be related to the fact that the wind brings emissions from industrial areas and also increases volatiles, causing O 3 photochemical pollution.This was followed by the southeast wind when the wind speed was greater than 1.5 m s −1 .The high concentrations of SO 2 , NO 2 , and CO were always associated with the southwest and southeast winds.This result also indicates that winds from specific wind directions have a greater influence on atmospheric pollutants.The consistency of PM 2.5 and PM 10 concentrations is highest with the southwest wind, when the wind speed is less than 1.5 m s −1 and PM 2.5 and PM 10 concentrations will increase.When the wind speed is greater than 2.5 m s −1 , the southwesterly wind causes the maximum O 3 concentration, and the O 3 concentration increases from east to west until it reaches the maximum value.The trend of higher O 3 concentration in the west and lower in the east may be related to the fact that more cities are clustered in the west of Changji and fewer cities in the east of Changji.In Shihezi, the elevated SO2, NO2, and CO concentrations are caused by southwesterly winds and wind speeds less than 1.0 m s −1 .lower wind speeds allow pollutants from industrial urbans and other cities to be deposited in the city, resulting in elevated pollutant concentrations.elevated SO2, NO2, and CO concentrations are also associated with southeasterly winds.The southwest and southeast winds cause higher concentrations of PM2.5 and PM10 and show maximum concentrations when the wind speed is less than 1.5 m s −1 , which may be related to the fact that there are more industrial urbans and farmland to the west and south of Shihezi city.However, O3 concentrations reach a maximum in southeast and southwest winds with wind speeds greater than 2.0 m s −1 .High values of SO2, NO2, and CO occur in Urumqi during southeasterly winds, and wind speeds are less than 2.0 m s −1 .This may be due to the fact that a large number of industrial and commercial urbans are distributed in the south and east of Urumqi, and winds bring more pollutants from this area.Secondly, the southwest wind increases the concentration of SO2, NO2, and CO.Southeast winds increase PM2.5 and PM10 concentrations, with maximum concentrations occurring at wind speeds less than 2.0 m s −1 , which may be related to the more open topography and more wasteland and desert in eastern Urumqi.O3 concentrations are mainly influenced by southwest and southeast winds, with the highest O3 concentrations occurring at wind speeds greater than 2.0 m s −1 .
The above results suggest that pollutant concentrations may be mainly influenced by local emissions and transport.On the one hand, when the wind speed is less than 2.0 m s −1 , local and local emissions have the greatest influence on pollutant concentrations.Increased human activity and consequently increased emissions lead to higher pollutant concentrations.Air pollutants are affected more by certain wind directions than others for several reasons.On the one hand, the migration and transport of pollutants are higher in In Shihezi, the elevated SO 2 , NO 2 , and CO concentrations are caused by southwesterly winds and wind speeds less than 1.0 m s −1 .lower wind speeds allow pollutants from industrial urbans and other cities to be deposited in the city, resulting in elevated pollutant concentrations.elevated SO 2 , NO 2 , and CO concentrations are also associated with southeasterly winds.The southwest and southeast winds cause higher concentrations of PM 2.5 and PM 10 and show maximum concentrations when the wind speed is less than 1.5 m s −1 , which may be related to the fact that there are more industrial urbans and farmland to the west and south of Shihezi city.However, O 3 concentrations reach a maximum in southeast and southwest winds with wind speeds greater than 2.0 m s −1 .High values of SO 2 , NO 2 , and CO occur in Urumqi during southeasterly winds, and wind speeds are less than 2.0 m s −1 .This may be due to the fact that a large number of industrial and commercial urbans are distributed in the south and east of Urumqi, and winds bring more pollutants from this area.Secondly, the southwest wind increases the concentration of SO 2 , NO 2 , and CO.Southeast winds increase PM 2.5 and PM 10 concentrations, with maximum concentrations occurring at wind speeds less than 2.0 m s −1 , which may be related to the more open topography and more wasteland and desert in eastern Urumqi.O 3 concentrations are mainly influenced by southwest and southeast winds, with the highest O 3 concentrations occurring at wind speeds greater than 2.0 m s −1 .
The above results suggest that pollutant concentrations may be mainly influenced by local emissions and transport.On the one hand, when the wind speed is less than 2.0 m s −1 , local and local emissions have the greatest influence on pollutant concentrations.Increased human activity and consequently increased emissions lead to higher pollutant concentrations.Air pollutants are affected more by certain wind directions than others for several reasons.On the one hand, the migration and transport of pollutants are higher in the upwind urban in specific directions.On the other hand, the wind speed in a particular direction is low, causing the accumulation of air pollutants, and the long-term transport of dust aerosols may be related to the wind speed in specific directions being higher than that in other directions.

Backward Trajectory Clustering Analysis during Winter
Combining the literature research and the results of the above study, trajectories in winter were clustered into 4 or 5 classes using TrajStat software (as shown in Figure 5), and the effects of various trajectories on atmospheric particulate matter in the Wu-Chang-Shi urban were quantified by combining backward trajectory analysis and particulate matter concentration data from the Wu-Chang-Shi urban (as shown in Table 3).
tmosphere 2023, 14, x FOR PEER REVIEW 12 of 24 direction is low, causing the accumulation of air pollutants, and the long-term transport of dust aerosols may be related to the wind speed in specific directions being higher than that in other directions.

Backward Trajectory Clustering Analysis during Winter
Combining the literature research and the results of the above study, trajectories in winter were clustered into 4 or 5 classes using TrajStat software (as shown in Figure 5), and the effects of various trajectories on atmospheric particulate matter in the Wu-Chang-Shi urban were quantified by combining backward trajectory analysis and particulate matter concentration data from the Wu-Chang-Shi urban (as shown in Table 3).In winter in Changji, airflow is predominantly west and northwest, followed by southwest airflow.The west and northwest airflow (tracks 1, 2, and 3 from southeastern Kazakhstan, north-central Bortala Mongol and Yili Kazakh Autonomous Prefecture, Kuitun and Shihezi urbans) account for the highest proportion of the total winter airflow tracks, 98.88%.The south-westerly flow from the eastern part of the Ili Kazakh Autonomous Prefecture, the border of the Ili Kazakh Autonomous Prefecture and Changji City, the southern part of Changji City, and the western part of Turpan (track 4) accounted for 1.12% of the season.The highest concentrations of particulate matter were carried by the western and southwestern airflow in winter, with trajectories 1 and 4 corresponding to PM 10 concentrations of 226.06 µg m −3 and 304.90 µg m −3 , respectively, and PM 2.5 concentrations of 170.32 µg m −3 and 176.44 µg m −3 , respectively.It is indicated that the west and southwest airflow from the north-central and east of Ili Kazakh Autonomous Prefecture, the border between Ili Kazakh Autonomous Prefecture and Changji City, south of Changji City, and west of Turpan City are the main transport paths affecting the atmospheric particulate matter concentration in winter in Changji City.
Shihezi is mainly influenced by airflow from westward and northwestward directions in winter, with track 2 accounting for the highest percentage of 36.01%, which shows the characteristics of long transport distance and strong force, originating from southeastern Kazakhstan, passing through the northeastern part of Ili Kazakh Autonomous Prefecture and eastern Shihezi, and arriving at Shihezi; track 1 is second with 21.08%, passing through southeastern Ili Kazakh Autonomous Prefecture, and is Trajectory 3, 4 and 5 are smaller and relatively uniform, mainly influenced by southeastern Kazakhstan and Tianshan Mountains, passing through the northern part of Bortala Mongol Autonomous Prefecture, Tacheng area, southern Karamay, central Tianshan Mountains and southern Shawan to reach Shihezi.The highest concentrations of particulate matter were carried by the westward airflow in winter, with trajectories 2 and 4 corresponding to PM 10 concentrations of 227.07 µg m −3 and 319.90 µg m −3 , respectively, and PM 2.5 concentrations of 194.81 µg m −3 and 140.50 µg m −3 , respectively, with trajectory 3 corresponding to the next highest PM 2.5 concentration of 180.11 µg m −3 .It can be seen that the most important transport path affecting the winter particulate concentration in Shihezi is the westward airflow trajectory.
Compared with Shihezi, Urumqi is mainly influenced by a westerly flow originating in the southeast of neighboring Kazakhstan and northeast of Uzbekistan in winter, with track 1 accounting for the highest share (38.62%) and tracks 2 and 4 accounting for a smaller share, which passes through the southeast of Kazakhstan and the north-central part of Ili Kazakh Autonomous Prefecture to reach Urumqi.The second is influenced by a provincial westward airflow (track 3, accounting for 33.40%), which originates in the south of Ili Kazakh Autonomous Prefecture and reaches Urumqi via the southwest of Urumqi.The highest concentrations of particulate matter were carried by the westward airflow in winter, with trajectories 1 and 3 corresponding to PM 10 concentrations of 157.40 µg m −3 and 157.33 µg m −3 , respectively, and PM 2.5 concentrations of 38.62 µg m −3 and 33.40 µg m −3 .This suggests that there may be regional transport of particulate matter from southeastern and southern Kazakhstan, southern Ili Kazakh Autonomous Prefecture, and southwestern Urumqi to the city of Urumqi.
In summary, the west and north-west flow originates from the south and south-east Kazakhstan, north-east Uzbekistan, the Turpan Basin, north-central Ili Kazakh Autonomous Prefecture, and other regions, while the south-west flows originate from north-east Ili Kazakh Autonomous Prefecture, north-central Tianshan and the junction of Ili Kazakh Autonomous Prefecture and Changji City and south of Shihezi and Changji, making them the most important transport pathways affecting atmospheric particulate matter concentrations in winter in the Wu-Chang-Shi region.

Potential Source Area Analysis
The source direction and spatial distribution of air mass trajectories affecting the Wu-Changji-Shi urban can be determined by HYSPLIT, but the potential sources and contributions affecting PM cannot be determined.Therefore, we determined the potential sources and contributions of PM in the Wu-Changji-Shi urban in winter by PSCF and CWT analysis.

PSCF Analysis
The potential winter WPSCF results for particulate matter in the Wu-Chang-Shi urban from 2017 to 2021 are shown in Figure 6 below.The larger the WPSCF value, the higher the percentage of pollution trajectories in the grid area.The distribution of PM 10 potential contributing source areas in winter in Changji is similar to that of PM 2.5 , with a large and concentrated range of potential contributing source areas, but the range of PM 10 high-value areas is larger than that of PM 2.5 , and the high-value areas of WPSCF (WPSCF > 0.9) are mainly concentrated in Changji and the neighboring cities of Urumqi, Wujiaqu, Shihezi, and Turpan.Under the influence of westerly winds, a potential contributing source area with wide coverage from west to east (0.6 < WPSCF < 0.8) exists from southeastern Kazakhstan via north-central Ili Kazakh Autonomous Prefecture to Changji, which contributes significantly to the mass concentration of PM 10 .The high WPSCF values (WPSCF > 0.8) of PM 2.5 in winter are mainly distributed in Changji city and adjacent areas, and the areas with high WPSCF values also have a trend of distribution from west to east, similar to PM 10 overall.
The distribution of potential contributing source areas for winter particulate matter in Shihezi is similar to that in Changji, but the overall extent is smaller.The distribution of high-value areas of potential contributing source areas for PM 10 and PM 2.5 in Shihezi is almost the same, with both WPSCF values reaching above 0.8, which are mainly influenced by the western part of Ili Kazakh Autonomous Prefecture and surrounding areas.Also, the areas with high WPSCF values are influenced by westward and northwestward airflow, and the overall distribution range is similar to that of Changji.The distribution of the potential contributing source areas of PM in winter in Urumqi is more obvious compared with Changji and Shihezi. the areas with high values of PM 10 and PM 2.5 show a zonal distribution from the western region of Yili Kazakh Autonomous Prefecture through Shihezi and other places to Urumqi, and the WPSCF values also both reach above 0.8.Under the influence of westerly winds, there is a southwest-northeast trending potential contributing source area zone (0.2 < WPSCF < 0.7) covering a wide area from southeastern Uzbekistan through southeastern Kazakhstan and areas such as Aksu and Bortala Mongol Autonomous Prefecture to the receiving point Urumqi, which contributes more to the mass concentrations of PM 10 and PM 2.5 .The distribution of potential contributing source areas for winter particulate matter in Shihezi is similar to that in Changji, but the overall extent is smaller.The distribution of high-value areas of potential contributing source areas for PM10 and PM2.5 in Shihezi is almost the same, with both WPSCF values reaching above 0.8, which are mainly influenced by the western part of Ili Kazakh Autonomous Prefecture and surrounding areas.Also, the areas with high WPSCF values are influenced by westward and northwestward airflow, and the overall distribution range is similar to that of Changji.The distribution of the potential contributing source areas of PM in winter in Urumqi is more obvious compared with Changji and Shihezi. the areas with high values of PM10 and PM2.5 show a zonal distribution from the western region of Yili Kazakh Autonomous Prefecture through Shihezi and other places to Urumqi, and the WPSCF values also both reach above 0.8.Under the influence of westerly winds, there is a southwest-northeast trending potential contributing source area zone (0.2 < WPSCF < 0.7) covering a wide area from southeastern Uzbekistan through southeastern Kazakhstan and areas such as Aksu and Bortala Mongol Autonomous Prefecture to the receiving point Urumqi, which contributes more to the mass concentrations of PM10 and PM2.5.
In summary, the WPSCF winter distribution characteristics of PM2.5 and PM10 in the Wu-Chang-Shi region are relatively similar, with a relatively concentrated range of potential contributing source urbans in the surrounding cities.The range of high WPSCF urbans is mostly greater for PM2.5 than PM10, with PM10 greater than PM2.5 in Changji City, and there are 1, 2, and 3 potential source urban contributing bands, whose WPSCF values Part of the range of the urban covered is the main source of sand and dust storms in China, In summary, the WPSCF winter distribution characteristics of PM 2.5 and PM 10 in the Wu-Chang-Shi region are relatively similar, with a relatively concentrated range of potential contributing source urbans in the surrounding cities.The range of high WPSCF urbans is mostly greater for PM 2.5 than PM 10 , with PM 10 greater than PM 2.5 in Changji City, and there are 1, 2, and 3 potential source urban contributing bands, whose WPSCF values Part of the range of the urban covered is the main source of sand and dust storms in China, and some of the urbans (e. g.Aksu region, Karamay city, and Turpan) are relatively polluted urbans in China, and therefore they can be considered the main potential source urbans in winter.

CWT Analyse
The WPSCF analysis method can help us determine the potential source urban, but it has limitations and does not accurately reflect the pollution level of the pollution trajectory.Therefore, according to the CWT analysis method, weighting the concentration of pollutants on the source grid to reflect the pollution level of the pollution trajectory (As shown in Figure 7 below).From the resulting map, it can be seen that the WCWT in the Wu-Chang-Shi urban has a good agreement relative to the WPSCF results, but the potential source area is more widely distributed, except for Shihezi.
it has limitations and does not accurately reflect the pollution level of the pollution trajectory.Therefore, according to the CWT analysis method, weighting the concentration of pollutants on the source grid to reflect the pollution level of the pollution trajectory (As shown in Figure 7 below).From the resulting map, it can be seen that the WCWT in the Wu-Chang-Shi urban has a good agreement relative to the WPSCF results, but the potential source area is more widely distributed, except for Shihezi.The high-value area (WCWT > 80 μg m −3 ) of the potential contributing source area of PM2.5 in winter in Changji is much more widely distributed than PM10, mainly in the local and surrounding areas of Changji, such as Urumqi, Turpan, Shihezi, Kuitun, Ili Kazakh Autonomous Prefecture, and northern Aksu region, with some scattered distribution in southeastern Kazakhstan, northeastern Kyrgyzstan, eastern Uzbekistan, Karamay City and the southern part of Tacheng region also have some sporadic distribution.Compared with PM2.5, the high WCWT of PM10 (WCWT > 170 μg m −3 ) is mainly scattered in the areas around Changji city and the central and western parts of Ili Kazakh Autonomous Prefecture.This is related to the winter heating and the intensity of air pollutant emissions in The high-value area (WCWT > 80 µg m −3 ) of the potential contributing source area of PM 2.5 in winter in Changji is much more widely distributed than PM 10 , mainly in the local and surrounding areas of Changji, such as Urumqi, Turpan, Shihezi, Kuitun, Ili Kazakh Autonomous Prefecture, and northern Aksu region, with some scattered distribution in southeastern Kazakhstan, northeastern Kyrgyzstan, eastern Uzbekistan, Karamay City and the southern part of Tacheng region also have some sporadic distribution.Compared with PM 2.5 , the high WCWT of PM 10 (WCWT > 170 µg m −3 ) is mainly scattered in the areas around Changji city and the central and western parts of Ili Kazakh Autonomous Prefecture.This is related to the winter heating and the intensity of air pollutant emissions in the region.In addition, cold air is frequently active in the west, northwest, and southwest directions in winter, forming three corridors of transport source areas from west to east, northwest, and southwest.
The spatial distribution of potential contributing source areas with low WCWT values for PM 2.5 and PM 10 in winter in Shihezi is similar, and overall the WCWT values for PM 2.5 are lower than those for PM 10 , with significant differences in distribution ranges compared to the WPSCF analysis.This may be related to the locally adopted industrial emission limits.high WCWT values of PM The high values of WCWT of PM 2.5 and PM 10 in winter in Urumqi are both greater than 90 µg m −3 , and the distribution of the potential contributing source areas of both is very similar, but the coverage of the latter is slightly larger than that of the former, and the high WCWT areas of PM 2.5 and PM 10 show a band distribution from southeastern Kazakhstan through the central and western regions of Ili Kazakh Autonomous Prefecture to the receiving point Urumqi.This is related to winter heating and the intensity of air pollutant emissions in the region.In addition, similar to Changji city, a source pathway is formed from southwest to northeast transport due to the frequent activity of cold air in the west in winter, which is also consistent with the above analysis of WPSCF.
In summary, the distribution characteristics of WCWT high-value areas of PM 2.5 and PM 10 in the Wu-Chang-Shi urban in winter are relatively similar.The urbans with high WCWT values for PM 2.5 and PM 10 are relatively concentrated, including Wu-Chang-Shi locality and neighboring cities and counties, Ili Kazakh autonomous prefecture, Bortala Mongolian Autonomous Prefecture, southeastern Kazakhstan, northern Aksu, and the junction of Kazakhstan and Kyrgyzstan.This is the most important potential source of particulate matter.

Comparisons with Other Chinese Cities
Since studying the concentration of air pollutants in only one region and not comparing it with other cities does not reflect the level of pollution in the region, we compared the Wu-Chang-Shi region with 168 cities monitored by the Chinese in 2017.(As shown in Figure 8 below).Nationally, pollution is more severe in cities in north-central China and less severe in cities in southern China [24].The main pollutants commonly found in China are PM2.5, followed by PM10 and O3 [39]  Nationally, pollution is more severe in cities in north-central China and less severe in cities in southern China [24].The main pollutants commonly found in China are PM 2.5 , followed by PM 10 and O 3 [39].The annual average concentrations of PM 10 (67.29 µg m −3 , 61.87 µg m −3 , 70.31 µg m −3 ) and PM 2.5 (98.49µg m −3 , 102.14 µg m −3 , 115.05 µg m −3 ) in Changji, Shihezi, and Urumqi were higher, but in moderate pollution.91.07% and 81.55% of the cities for PM 2.5 and PM 10 exceeded the CAAQS secondary standard, respectively.The annual average concentration of O 3 ranges from 72.39 µg m −3 (Changji) to 124.47 µg m −3 (Tangshan).Among the 168 cities, no city exceeded the O 3 secondary standard, but 54.17% of the cities exceeded the primary standard.The SO 2 concentrations in Changji City, Shihezi City, and Urumqi City were 72.39 µg m −3 , 73.13 µg m −3, and 86.90 µg m −3 respectively.No city exceeded the Class I standard and was within the range of moderate pollution levels.For SO 2 , only three cities in China, Linfen, Lvliang, and Jinzhong exceeded the secondary standard, with the lowest (6.45 µg m −3 ) and highest (83.44 µgm −3 ) annual average SO 2 concentrations being in Fuzhou and Jinzhong respectively.The concentrations of SO 2 in Changji City, Shihezi City, and Urumqi City were 18.09 µg m −3 , 15.25 µg m −3 , and 13.52 µg m −3 respectively, which belonged to low pollution levels.None of the cities exceeded the first-grade carbon monoxide standard, and the annual average CO concentrations (1.33 µg m −3 , 1.15 µg m −3 , 1.40 µg m −3 ) in Changji, Shihezi, and Urumqi were in the lightest pollution level.In terms of NO 2 , 40% of cities exceed secondary standards.The NO 2 concentrations in Changji City, Shihezi City, and Urumqi City were 45.24 µg m −3 , 42.80 µg m −3 , and 49.64 µg m −3 respectively, which belonged to the moderate pollution level.By comparing Changji, Shihezi, and Urumqi with the 168 cities mentioned above, we can see that the pollution of PM 2.5 and PM 10 is more serious in Wu-Chang-Shi, while NO 2 and O 3 are relatively light, and SO 2 and CO are the least polluted.As a result, improving air quality in Changji, Shihezi, and Urumqi should be based on PM 2.5 and PM 10 control and management, addressing traffic sources, reducing primary pollutants and NO 2 pollutants emissions, and reducing dust entrainment in the desert's surrounding urbans.Volatile organic compound emissions should be controlled by prioritizing NO x control to prevent increases in O 3 pollution.

Characteristics of Changes in Air Pollutant Concentrations before and after the New Coronary Pneumonia Outbreak
As the novel coronavirus became a pandemic in China in late 2019 or early 2020, the Chinese government then adopted a series of containment policies to control its rapid spread [53].It is due to these containment policies that the atmospheric pollutant concentrations have changed [54], we compared pollutant concentration data for the year before and for the entire year following the national outbreak of New Crown pneumonia in early 2020 and the adoption of containment.The time period for the comparative analysis is taken from January to July before the dotted line in the chart below (Months in which pollutant concentrations were affected by the outbreak).
CO is mainly from incomplete combustion of fossil fuels and biomass (burning of crop residues in open urbans).As can be seen in Figure 9 below, CO in the Wu-Chang-Shi urban showed a decreasing trend from January to July after the outbreak, and the concentration values after the outbreak were all smaller than those before the outbreak, decreasing by 0.3 µg m −3 (Urumqi), 0.14 µg m −3 (Changji), and 0.15 µg m −3 (Shihezi).The variation over a whole year shows that the concentration values after the outbreak are also smaller than those before the outbreak.This indicates that the national closure policy adopted after the outbreak has to some extent reduced CO emissions from industrial and domestic fuels and domestic coal combustion to a certain extent, and the effect is relatively obvious.The combustion of sulfur-containing fuel combustion (oil, coal) increases the concentration of SO 2 in the atmosphere, indicating that it is the main source [55,56].Overall SO 2 emissions in the Wu-Chang-Shi region improved after the epidemic; average SO 2 concentrations in Changji, Shihezi, and Urumqi decreased by 3.97 µg m −3 , 2.95 µg m −3 , and 2.17 µg m −3 , respectively, from January to July, compared to the same period before the outbreak; however, emissions after the epidemic were smaller than those before the epidemic in spring and summer and larger than those before the epidemic in autumn and winter.On the one hand, because of the strong ecological protection policies implemented by the government in recent years, the upgrading of key industries (electricity and iron and steel), the extensive use of new energy sources, and the obvious effect of energy conservation and emission reduction, the concentration of air pollutants were reduced.On the other hand, many coal-fired industries could not be shut down after the epidemic, especially in the autumn and winter seasons when a large amount of coal is burned for urban heating, leading to an increase in SO 2 emissions.The main sources of gaseous NO2 are fossil fuel combustion, coal and oil combustion, etc.After the outbreak, NO2 in the Uchang region showed an overall decreasing trend from January to July, decreasing by 6.69 μg m −3 (Urumqi), 7.44 μg m −3 (Changji), and 1.44 μg m −3 (Shihezi), respectively, year-on-year.The lowest NO2 concentrations before and after the epidemic occur in summer, which is due to the strong photochemical reactions in summer that consume NOx However, the post-epidemic concentrations were greater than the pre-epidemic values in winter, which were mainly influenced by winter heating in the Northwest.Gaseous O3 is mainly emitted from the combustion of fossil fuels and exhaust gases from industrial production, and vehicle exhaust from transport.We found that the average O3 concentrations in Changji, Shihezi, and Urumqi were greater from January to July after the outbreak than in the same period before the outbreak, and increased by 12.47 μg m −3 and 0.16 μg m −3 and 2.56 μg m −3 , respectively, showing an increasing trend after the outbreak.As both NO2 and NO emissions decreased during the epi- The main sources of gaseous NO 2 are fossil fuel combustion, coal and oil combustion, etc.After the outbreak, NO 2 in the Uchang region showed an overall decreasing trend from January to July, decreasing by 6.69 µg m −3 (Urumqi), 7.44 µg m −3 (Changji), and 1.44 µg m −3 (Shihezi), respectively, year-on-year.The lowest NO 2 concentrations before and after the epidemic occur in summer, which is due to the strong photochemical reactions in summer that consume NO x However, the post-epidemic concentrations were greater than the pre-epidemic values in winter, which were mainly influenced by winter heating in the Northwest.Gaseous O 3 is mainly emitted from the combustion of fossil fuels and exhaust gases from industrial production, and vehicle exhaust from transport.We found that the average O 3 concentrations in Changji, Shihezi, and Urumqi were greater from January to July after the outbreak than in the same period before the outbreak, and increased by 12.47 µg m −3 and 0.16 µg m −3 and 2.56 µg m −3 , respectively, showing an increasing trend after the outbreak.As both NO 2 and NO emissions decreased during the epidemic, which suppressed O 3 production and reduced O 3 consumption, the compensating effect of NO 2 on O 3 exceeded the depleting effect of NO, and O 3 concentrations increased.The change in the ratio of NO x to VOCs may also lead to an increase in O 3 production, while the decrease in particulate matter concentration during the epidemic may lead to an increase in light radiation, which may also lead to an increase in O 3 content.
The sources of PM 2.5 and PM 10 can be divided into natural and anthropogenic sources, such as ultrafine particulate matter emitted directly from various industrial processes, such as coal combustion, metallurgy, chemicals, and internal combustion engines, and ultrafine particulate matter formed secondarily in the atmosphere with aerosols.Through the study, we found that PM 2.5 and PM 10 concentrations in the Uchang Shi area were lower than the pre-epidemic values from January to July of the outbreak, and showed a decreasing trend due to the reduction of industrial and domestic emissions after the outbreak, Comparing other periods of the year, we found that the values after the outbreak were also lower than those before the outbreak.However, in individual places in spring, PM 10 concentration values after the outbreak were greater than those before the outbreak, which may be related to the frequent dusty weather in northwest China in spring.As a whole, the concentrations of all air pollutants except O 3 are lower after the epidemic than before it, due to the national policy of closure and control after the epidemic, which banned people's living activities and thus reduced some industrial and domestic emissions, as well as the significant effectiveness of pollution control in recent years.

Conclusions
The main conclusions from the comprehensive analysis of the pollution characteristics and potential sources of air pollutants in the Wu-Chang-Shi urban from 2017 to 2021 are as follows: NO 2 , CO, SO 2 , PM 10 , and PM 2.5 had a tendency to decrease, while O 3 trended to increase during this period.The concentrations of pollutants were significantly higher in winter than in other seasons, but O 3 instead had the highest concentrations in summer.Due to the influence of seasonal and domestic heating emissions, the concentrations of PM 10 and PM 2.5 showed characteristics of high winter and low summer.Compared with the pollutant concentrations in other cities in China, the pollution of PM 2.5 , PM 10 , and NO 2 in the Wu-Chang-Shi urban was relatively serious.By studying the relationship between atmospheric pollutants and meteorological factors, we could conclude that the wind speed and direction of the north, northwest, and southwest winds were associated with lower concentrations of primary pollutants, while temperature increases were associated with higher concentrations of secondary pollutants (O 3 ).
Backward trajectory cluster analysis shows that in winter, the Wu-Chang-Shi urban is dominated by west and northwest airflow (98.88% in Changji, 78.92% in Shihezi, and 66.60% in Urumqi), followed by southwest airflow (1.12% in Changji, 21.08% in Shihezi, and 33.40% in Urumqi), the west and northwest airflows come from the south and southeast of Kazakhstan, the northeast of Uzbekistan, the Turpan Basin, the central and northern parts of Ili Kazakh Autonomous Prefecture, and the southwest airflow comes from the northeast of Ili Kazakh Autonomous Prefecture, the central and northern Tianshan Mountains, Ili Kazakh Autonomous Prefecture and Changji City The long-distance transport of dust in the junction and the south of Shihezi and Changji, Central Asia and adjacent urbans makes them the most important transport paths affecting the concentration of atmospheric particulate matter in the Wu-Chang-Shi urban in winter.WPSCF and WCWT analysis showed that the distribution characteristics of PM 2.5 and PM 10 in the Wu-Chang-Shi urban were similar in winter.The urbans with high values for PM 2.5 and PM 10 were relatively concentrated, including the Wu-Chang-Shi locality and neighboring cities and counties, Ili Kazakh Autonomous Prefecture, Bortala Mongolian Autonomous Prefecture, southeastern Kazakhstan, northern Aksu, and the junction of Kazakhstan and Kyrgyzstan.

Figure 1 .
Figure 1.(a) Concept of the proposed force plate.This force plate consists of a planar spring and an eddy current displacement sensor.(b) Side view of the proposed force plate.

Figure 2 .
Figure 2. Comparison of the AQI and annual average concentrations of air pollutants in the Wu-Chang-Shi urban for 2017-2021.(μg m −3 ).

Figure 2 .
Figure 2. Comparison of the AQI and annual average concentrations of air pollutants in the Wu-Chang-Shi urban for 2017-2021.(µg m −3 ).
NO 2 is photochemically active and reacts chemically with other precursors to produce O 3 in the presence of sunlight, so the concentration of NO 2 is lower in spring and summer.On the contrary, the maximum O 3 concentration appeared in summer, and the average O 3 concentrations in Changji, Shihezi, and Urumqi were 75.47 µg m −3 , 108.79 µg m −3 , and 74.91 µg m −3 , respectively.The seasonal variation of O 3 , in contrast to other primary pollutants, is due to the high temperatures and intense solar radiation in summer that favor the formation of O 3 .

Figure 5 .
Figure 5. Clustering of backward trajectories of airflow in the Wu-Chang-Shi region during winter 2017-2021.

Figure 5 .
Figure 5. Clustering of backward trajectories of airflow in the Wu-Chang-Shi region during winter 2017-2021.

Figure 6 .
Figure 6.PSCF distribution of winter particulate matter in the Wu-Chang-Shi urban.

Figure 6 .
Figure 6.PSCF distribution of winter particulate matter in the Wu-Chang-Shi urban.
2.5 are sporadically distributed in local areas of Shihezi city, with WCWT values reaching more than 180 µg m −3 .PM 10 is mainly influenced by the areas around Shihezi, such as Changji, Urumqi, Shawan, and Kuitun, as well as the northern part of Aksu, the central and western part of Ili Kazakh Autonomous Prefecture, and the sporadic area in southeastern Kazakhstan of the city, the WCWT values reached above 160 µg m −3 , indicating that these areas contributed more to the PM 10 pollution concentration in Shihezi.In addition, lower WCWT values of PM 10 (WCWT < 60 µg m −3 ) are sporadically distributed in southeastern Kazakhstan and northern Xinjiang.

Atmosphere 2023 , 24 Figure 8 .
Figure 8.Comparison of annual average concentrations of six air pollutants in 168 major cities in 2017 (μg m −3 ).The red circles represent the Wu-Chang-Shi urban.

Figure 8 .
Figure 8.Comparison of annual average concentrations of six air pollutants in 168 major cities in 2017 (µg m −3 ).The red circles represent the Wu-Chang-Shi urban.

Table 1 .
Summary statistics for the six criteria pollutants and AQI in the Wu-chang-shi urban for 2017-2021.(µg m −3

Table 3 .
Statistical results of the mass concentration of all kinds of airflow in the four seasons of the Wu-Chang-Shi urban.
* An asterisk indicates the concentration of pollutants in the table.