Multi-Year Variation of Ozone and Particulate Matter in Northeast China Based on the Tracking Air Pollution in China (TAP) Data

With the rapid development of economy and urbanization acceleration, ozone (O3) pollution has become the main factor of urban air pollution in China after particulate matter. In this study, 90th percentile of maximum daily average (MDA) 8 h O3 (O3-8h-90per) and PM2.5 data from the Tracking Air Pollution in China (TAP) dataset were used to determine the mean annual, seasonal, monthly, and interannual distribution of O3-8h-90per and PM2.5 concentrations in Northeast China (NEC). The O3-8h-90per concentration was highest in Liaoning (>100 μg/m3), whereas the highest PM2.5 concentration was observed mainly in urban areas of central Liaoning and the Harbin–Changchun urban agglomeration (approximately 60 μg/m3). The O3-8h-90per concentrations were highest in spring and summer due to more intense solar radiation. On the contrary, the PM2.5 concentration increased considerably in winter influenced by anthropogenic activities. In May and June, the highest monthly mean O3-8h-90per concentrations were observed in central and western Liaoning, about 170–180 μg/m3, while the PM2.5 concentrations were the highest in January, February, and December, approximately 100 μg/m3. The annual mean O3-8h-90per concentration in NEC showed an increasing trend, while the PM2.5 concentration exhibited an annual decline. By 2020, the annual mean O3-8h-90per concentration in southern Liaoning had increased considerably, reaching 120–130 μg/m3. From the perspective of city levels, PM2.5 and O3-8h-90per also showed an opposite variation trend in the 35 cities of NEC. The reduced tropospheric NO2 column is consistent with the decreasing trend of the interannual PM2.5, while the increased surface temperature could be the main meteorological factor affecting the O3-8h-90per concentration in NEC. The results of this study enable a comprehensive understanding of the regional and climatological O3-8h-90per and PM2.5 distribution at distinct spatial and temporal scales in NEC.


Introduction
Surface ozone (O 3 ) is a critical gaseous pollutant that affects air quality and contributes to climate change [1][2][3][4]. O 3 and its interactions with precursor gaseous emissions have been widely researched in recent decades [5,6]. Surface O 3 is formed by photochemical reactions of volatile organic compounds (VOCs) and nitrogen oxides; these reactions are exacerbated by sunlight [7][8][9]. The major contributions to O 3 formation and accumulation emanate from natural emissions and anthropogenic sources such as long-range transport [10][11][12][13].
Worldwide, numerous studies have analyzed the relationship between surface O 3 and temperature, pressure, relative humidity, wind, precipitation, cloud cover, and solar radiation, as well as the meteorological factors related to synoptic circulation patterns [14][15][16][17].
Due to several decades of rapid economic growth and urbanization, China has experienced severe fine particulate matter (PM 2.5 ) pollution originating from anthropogenic emissions [18][19][20]. However, in recent years, O 3 has become a more severe air pollutant than PM 2.5 is in some Chinese cities, resulting from a substantial increase of O 3 precursor emissions such as VOCs [21][22][23][24]. Numerous studies have reported the coordinated pollution of O 3 and particulate matter as the primary urban pollutant in densely populated areas, such as Beijing-Tianjin-Hebei (BTH), the Yangtze River Delta (YRD), the Pearl River Delta (PRD), and the Sichuan Basin [25][26][27]. Zeng et al. (2020) [28] implied that the concentrations of PM 2.5 and O 3 showed different seasonal trends in eastern China. Liu et al. (2021) [29] used the ground-based hyperspectral stereoscopic remote sensing network to provide a promising strategy to support management of PM 2. 5 and O 3 and their precursors and conduct attribution of sources. In the formation of surface O 3 , meteorological factors play a crucial role [30][31][32][33][34]. Yang et al. (2021) [35] indicated that PM 2.5 was primarily affected by wind, temperature, and rainfall, while O 3 was mostly influenced by temperature, relative humidity, and sunshine duration in 284 major cities in China. Dai et al. (2020) [36] identified that the number of days with co-pollution of O 3 and PM 2.5 was mainly dependent on relative humidity, surface air temperature, and wind speed in the Yangtze River Delta region.
Northeast China (NEC) is a key region in Asia, bordering Siberia in the north, the Bohai and Yellow Seas and the Central Plains in the south, Mongolia to the west, and the Korean Peninsula and the Sea of Japan to the east (Figure 1a). Air quality in NEC has been improving because efforts to control PM 2.5 have been intensified. However, O 3 has become a new environmental problem. Due to the unique geographical location and meteorological conditions of NEC, O 3 pollution originates not only from local emissions but also from those of surrounding areas. Figure 1b shows the spatial distribution of population density in NEC. The NEC has an uneven distribution of population, while southwest and central NEC are densely populated with a maximum population density of nearly 50 (×1000 persons), followed by eastern NEC with a population density of less than 15 (×1000 persons). The northern mountain of NEC is sparsely populated, with a population density of less than 0.5 (×1000 persons). The population density in NEC is mainly affected by topography, climate, and economic development, and the natural factors and social factors simultaneously affect the change of population density. From the spatial distribution of population density in the three provinces of NEC, population density is greater in Liaoning than in Jilin province, and lower in Heilongjiang province. The provincial capital cities mainly present a spatial pattern of polycentric concentrated distribution of population. These results indicate that the level of social and economic development is the main factor influencing the distribution of population density in NEC. Therefore, studies in NEC have revealed the importance of considering both local emission sources as well as the long-range transport of pollutants from regions such as the North China Plain [37,38]. Zhu and Liao (2016) [39] reported that surface O 3 levels in NEC were comparable to those of BTH and the YRD and even higher than those in the PRD during spring 2000. NEC is a crucial industrial and agricultural base in China. More research on O 3 pollution and its impact on ecosystem security and agriculture is needed. Understanding the spatiotemporal variation of O 3 and PM 2.5 can aid in providing technical support for the prevention and control of O 3 and PM 2.5 pollution in NEC.
This study analyzed the mean annual, seasonal, monthly, and interannual variation and distribution of O 3 -8h-90per and PM 2.5 in NEC. The innovation in this study was its focus on the multi-year variation and distribution of O 3 -8h-90per and PM 2.5 in NEC. The objective was to investigate the historical variation in surface O 3 -8h-90per (from 2013 to 2020) and PM 2.5 (from 2001 to 2020) in NEC on the basis of the Tracking Air Pollution in China (TAP) dataset. The remainder of this paper is structured as follows: Section 2 describes the study area and data sources. Section 3 presents the multi-year variation in O 3 -8h-90per and PM 2.5 at temporal (annual, seasonal, monthly, and interannual) and spatial (regional and city level) scales. Sections 4 and 5 discusses and presents the conclusions of this study.

Materials and Methods
NEC (120 • -135 • E, 40 • -53 • N) mainly encompasses Heilongjiang, Jilin, and Liaoning provinces. Liaoning is the most populous province and lies in the south of NEC. To the north lie Jilin and the northernmost province of Heilongjiang, which are the major agricultural provinces in the region. Precipitation is highest in summer; winter is characterized by snowfall, and surface snow usually remains for prolonged periods. NEC is the snowiest region in China. In recent years, the intensification of anthropogenic pollutants and adverse meteorological conditions have resulted in numerous processes causing air pollution. In cities with severe pollution and low air quality, particulate matter and O 3 have become the primary pollutants [40][41][42].
The TAP dataset was developed at Tsinghua University as a cooperative effort among several institutions and teams [43]. The aim was to build a multiscale, near-real-time aerosol and gaseous pollutant concentration database in China and provide essential support for pollution characteristics analysis. The TAP database was generated using state-of-the-art technology involving machine learning algorithms [44,45]. The TAP data are determined based on the combination of multisource data including ground measurements, satellite aerosol optical parameter retrievals, model simulations, and meteorology field, land use information as well as population, and elevation data by multilayer machine learning models. It integrates real-time ground observations, near-real-time satellite remote sensing information, and air quality model simulation with multisource big data and provides near-real-time data while ensuring complete spatiotemporal coverage. Surface O 3 -8h-90per concentration data (2013-2020) and PM 2.5 data (2001-2020) for China with a 10 km resolution can be downloaded from http://tapdata.org (last access: 1 July 2021). To our knowledge, this is the first study to use the pollutant concentration data from the newly released TAP dataset for investigating pollution in NEC. In this paper, we define O 3 -8h-90per as the average daily maximum O 3 -8h-90 per at the 90th percentile level. The annual, monthly, and seasonal average of O 3 -8h-90per data was used to analyze the temporal scale and spatial distribution of O 3 (µg/m 3 ), with the minimum value of 1.00 µg/m 3 .
Population data were obtained from the combined datasets of the GPWv3 (Gridded Population of the World available at 5-year intervals from 1990 to 2000) and GPWv4 (available at 5-year intervals from 2000 to 2020), from the NASA Socioeconomic Data and Applications Center (http://sedac.ciesin.columbia.edu/, last accessed on 1 July 2021), and the resolution was 2.5 arcminute.

Annual Distribution of O 3 -8h-90per and PM 2.5 in NEC
The spatial distribution of the mean annual O 3 -8h-90per concentrations in NEC from 2013 to 2020 is presented in Figure 2a [46] reported that PM 2.5 was concentrated mostly in central Liaoning, western Jilin, and Heilongjiang. The highest PM 2.5 concentration was approximately 60 µg/m 3 and occurred in Shenyang, Changchun, Harbin, and other provincial capitals of NEC. With approximately 50 µg/m 3 , the PM 2.5 concentration in western Liaoning was higher than that in other areas. The distribution range of high PM 2.5 concentrations in western Liaoning was not as large as that of O 3 -8h-90per concentrations. Except for the central cities, the PM 2.5 concentration in Jilin was approximately 30-40 µg/m 3 , and the lowest PM 2.5 concentration was approximately 20 µg/m 3 in the eastern and western marginal areas of Jilin. At approximately 10-20 µg/m 3 , the PM 2.5 concentration in northern Heilongjiang was relatively low.
The spatial distribution of O 3 -8h-90per (from 2013 to 2020) and PM 2.5 (from 2001 to 2020) concentrations in NEC was closely related to the environmental background, population density, and meteorological condition. The O 3 -8h-90per pollution in NEC was concentrated mainly in central and western Liaoning, followed by central Jilin; the overall O 3 -8h-90per concentration in Heilongjiang was low. However, high levels of PM 2.5 were observed throughout the densely populated areas of the three provinces. Due to rapid and geographically distinct urbanization patterns, numerous anthropogenic pollution sources are likely to have led to a regional imbalance in the distribution of PM 2.5 concentrations. However, the spatial distribution of high PM 2.5 concentrations was inconsistent with that of high O 3 -8h-90per concentrations. This indicates that although pollutants can increase O 3 precursors to a certain extent, they are not the main factors affecting the distribution of O 3 -8h-90per concentrations in NEC. In particular, important air pollutants such as volatile organic compounds (VOCs) are also important precursors of PM 2.5 , and various factors affect the correlation between O 3 -8h-90per and PM 2.5 . The existence of a high concentration of O 3 in the atmosphere enhances oxidation of the atmosphere, which is conducive to the formation of secondary particulate matter and thus an increase in PM 2.5 pollution. The presence of a large amount of PM 2.5 weakens the solar radiation reaching the near ground, reducing the photodecomposition reaction rate of O 3 .

Seasonal Distribution of O 3 -8h-90per and PM 2.5 in NEC
In this study, the four seasons were spring (March, April, and May), summer (June, July, and August), autumn (September, October, and November) and winter (December, January, and February). The O 3 -8h-90per concentrations were higher in spring and summer, followed by autumn and winter ( Figure 3a  Unlike the distribution of O 3 -8h-90per concentrations, that of PM 2.5 concentrations increased considerably in winter, followed by that in spring and autumn. The lowest PM 2.5 concentrations were observed in summer ( Figure 3b). In spring, the highest PM 2.5 concentrations were observed in central Liaoning and central Jilin (approximately 60 µg/m 3 ). The PM 2.5 concentration in a typical city reaches 80 µg/m 3 . By contrast, the PM 2.5 concentration in Heilongjiang was relatively low, approximately 20-30 µg/m 3 . In summer, the PM 2.5 concentration in NEC decreased. Except for the PM 2.5 concentrations in some central cities of Liaoning, which exhibited values of approximately 50-60 µg/m 3 , the PM 2.5 concentration in NEC was lower than 30 µg/m 3 [48] reported that low concentrations of O 3 were observed during winter (when heating is used), whereas relatively higher concentrations were observed in spring, when longer and stronger solar radiation drives the photochemical processes that lead to the formation of O 3 and heating is not used. Xia et al. (2021) [49] also found that high values of daytime maximum, nighttime minimum, and diurnal difference of summer ozone concentration occurred in several city agglomerations of northern and southern China.    We analyzed the interannual variation in PM 2.5 concentrations in NEC from 2001 to 2020 (Figure 5b).
From 2001 to 2014, the spatial distribution and regional variation of PM 2.5 concentrations in NEC were consistent. The high PM 2.5 concentrations were distributed in a zonal pattern, with the highest value reaching 80 µg/m 3 , observed in the central Liaoning city cluster, central Jilin, and areas adjacent to southern Heilongjiang. In 2015, the areas in NEC with high PM 2.5 exhibited a decreasing trend. These areas were concentrated in Liaoning, Jilin, Heilongjiang, and several typical cities, and the higher PM 2.5 concentration exhibited a point-like distribution with values reaching 80 µg/m 3 . In 2016, the annual mean PM 2.5 concentration in NEC decreased considerably, with the highest PM 2.5 concentration, 50-60 µg/m 3 , distributed among Liaoning, Jilin, and central Heilongjiang. In 2017, the area exhibiting high PM 2.5 concentrations decreased, and the highest PM 2.5 concentrations (approximately 50 µg/m 3 ) were observed in northern Liaoning as well as in central Jilin and Heilongjiang. In particular, from 2018 to 2020, the PM 2.5 concentration in NEC decreased considerably and reached approximately 40 µg/m 3 in typical cities such as Shenyang, Changchun, and Harbin. According to the analyzed data, the interannual characteristics of the spatial variation of O 3 -8h-90per and PM 2.5 reveal that the O 3 -8h-90per concentration in NEC changed mainly regionally, exhibiting an increasing annual trend. By contrast, the PM 2.5 concentration exhibited a zonal distribution characteristic from the southwest to the northeast, and the variation of PM 2.5 concentrations decreased annually. To determine the interannual variations of O 3 -8h-90per and PM 2.5 concentrations in NEC, we investigated the monthly and annual variations of O 3 -8h-90per and PM 2.5 concentrations in the three studied provinces.
As depicted in Figure 6A, the interannual variation of the monthly mean O 3 -8h-90per concentration in NEC exhibited obvious periodic variation during 2013-2020. The highest and lowest O 3 -8h-90per concentrations were observed from May to July and December to January, respectively. The monthly mean O 3 -8h-90per concentrations exhibited an increasing trend in Liaoning, Jilin, and Heilongjiang, respectively. By contrast, the interannual variation of the monthly mean PM 2.5 concentration during 2001-2020 was not as substantial as the periodic variation of O 3 -8h-90per ( Figure 6B). The highest PM 2.5 concentration was observed in January and February, and the lowest was observed in August and September. The PM 2.5 concentrations in Liaoning, Jilin, and Heilongjiang exhibited a small monthly decreasing trend, respectively.
We also investigated the trend in the annual mean variation of O 3 -8h-90per from 2013 to 2020 and PM 2.5 concentrations from 2001 to 2020 in NEC. As depicted in Figure 7A We also investigated the interannual trend of PM 2.5 concentrations in NEC from 2001 to 2020. As depicted in Figure 7B [51] found that the significance of ozone enhancement due to PM 2.5 dropping depends on both the PM 2.5 levels and optical properties of particles in many mega-cities in China.

Interannual Variation and Trends of O 3 -8h-90per at the City Level in NEC
In this section, we present a comparison of city-level interannual variation in O 3 -8h-90per and PM 2.5 concentrations in 35 cities. In Liaoning, the following 14 cities were included: Dandong, Fushun, and Benxi in the east, Jinzhou, Huludao, Fuxin, and Chaoyang in the west, Yingkou, Dalian, and Panjin in the south, Tieling in the north, and Shenyang, Liaoyang, and Anshan in central Liaoning. In Jilin, the following nine cities were included: Yanbian and Baishan in the east, Baicheng, Songyuan, and Siping in the west, Tonghua in the south, and Changchun, Jilin, and Liaoyuan in central Jilin. In Heilongjiang, the following 12 cities were included: Yichun, Qitaihe, Mudanjiang, Jiamusi, Shuangyashan, Hegang, Jixi, Qiqihar, and Daqing in the west, Harbin in the south, Heihe in the north, and Suihua in central Heilongjiang.
As depicted in Figure 8a, the interannual variations in O 3 -8h-90per concentrations in different cities exhibited distinct trends in increases and increase ranges. These results reveal that the distribution of high city-level O 3 -8h-90per concentrations in NEC was consistent with the spatial distribution of ground temperature. Under the combined influence of high temperatures and solar radiation, the O 3 formation is enhanced. In NEC, the mean temperature in various cities was closely related to distinct meteorological factors, such as the location of land and sea, latitude, topography, and altitude. Studies have reported that rising temperatures could be related to high O 3 pollution in NEC [52,53]. The temperature in Liaoning increased from east to west and from north to south. Therefore, the O 3 -8h-90per concentrations in Huludao, Panjin, and Dalian (southwestern Liaoning) were high, whereas those in Fushun and Benxi (eastern Liaoning) were low. A similar temperature difference between east and west was observed in Jilin. The temperature was relatively high in western Jilin, especially in Baicheng and Siping (southwest Jilin); this high temperature was conducive to the formation of O 3 -8h-90per. The temperature in Baishan and Yanbian (eastern Jilin) was relatively low; therefore, the O 3 -8h-90per concentration was low. The temperature was relatively high in southern Heilongjiang; therefore, the cities with high O 3 -8h-90per concentration were distributed mostly in southern Heilongjiang. The northern cities exhibited low O 3 -8h-90per concentrations.
The city-level trends of PM 2.5 in NEC were opposite to those of O 3 -8h-90per, and several cities exhibited a considerable decrease in PM 2.5 (Figure 8b). Compared with the changes in O 3 -8h-90per concentration, the interannual variation of PM 2.5 in NEC revealed PM 2.5 pollution at the city level. The cities with high PM 2.5 concentrations were characterized by industrial activity and large populations. Therefore, the reduction of environmental pollution caused a reduction in the contribution of PM 2.5 to air pollution at the city level in NEC.
In order to discuss the potential impact of important meteorological factors and emission sources on O 3 -8h-90per and PM 2.5 , we further analyzed the annual variation of tropospheric NO 2 column from 2005 to 2020, precipitation (PPT), wind speed (WS), boundary layer height (BLH), and temperature at 2 m (T2) from 2001 to 2020 in NEC. Figure 9 shows that the tropospheric NO 2 column in NEC has been decreasing since 2012, which is consistent with the decreasing trend of the interannual variation of PM 2.5 , indicating the impact of the decreasing intensity of anthropogenic emission sources on PM 2.5 concentration. From the interannual variation of meteorological elements in NEC, the higher wind speed and the increase of BLH are conducive to the diffusion of PM 2.5 to a certain extent. It is worth noting that temperature at 2 m has shown a significant increasing trend since 2012. In addition to the increase of precursors, the rise of surface temperature may be the main meteorological driving factor affecting the O 3 -8h-90per concentration. In addition to emission and meteorological factors, the chemical mechanism affecting O 3 and PM 2.5 is also important and needs further study.

Discussion
With increasing urbanization, environmental air quality has become an increasingly critical public concern. The pollution caused by PM 2.5 and O 3 -8h-90per has considerably affected people and residential environments. In almost ten years, surface O 3 -8h-90per and PM 2.5 pollution has become an environmental concern. Therefore, studying the distribution and variation of surface O 3 and PM 2.5 concentrations has high theoretical and practical value and aids in further investigating the meteorological and chemical factors that determine regional pollution. From the perspective of formation mechanism and meteorology, many previous studies have conducted important research on the variation characteristics of O 3 and PM 2.5 concentration in different regions of China.   [54] pointed out that the PM 2.5 could directly transport from one city to another city, while Tibetan Plateau may be an important source region of high ozone in Sichuan Basin of southwest China. Ma et al. (2021b) [55] illustrated that the increase in volatile organic compounds (VOCs) along with depletions in NO 2 and CO significantly boosted the ozone photochemical production in North China plain. Wang et al. (2021) [56] showed that the annual summertime ozone over Central China was significantly correlated with the springtime thermal forcing, indicated by total atmospheric energy over Tibetan Plateau in an interdecadal timescale. During the COVID-19 lockdown period, the results of Yin et al. (2021) [57] suggest that conventional emission reduction of NO x could not be sufficient to reduce surface O 3 concentration, and ozone pollution needs to be controlled by a variety of pollutants in central China. The significance of this study is to improve the scientific understanding of multi-year changes of O 3 -8h-90per and PM 2.5 concentrations due to climatological characteristics in NEC. This paper focuses on the O 3 -8h-90per and PM 2.5 pollution at different temporal and spatial scales, especially at city level in NEC. These results could provide reference on the formation mechanism of ozone and PM 2.5 in NEC for further study. Moreover, the correlation between O 3 and PM 2.5 is related to the reactions with VOCs. Ozone could oxidize VOCs to less volatile products that likely partition to the particle phase. Therefore, collaborative control of O 3 and PM 2.5 must be accelerated to reduce regional haze pollution and photochemical smog events.

Conclusions
In this study, TAP O 3 -8h-90per and PM 2.5 concentration data were used to analyze the multi-year variation of O 3 -8h-90per and PM 2.5 concentrations at distinct temporal and spatial scales in NEC.
The concentrations of O 3 -8h-90per were highest in northern, western, and southern Liaoning and the Bohai Rim (up to 120 µg/m 3 ). The highest O 3 -8h-90per concentrations (approximately 90-100 µg/m 3 ) in Jilin were distributed in the central and western regions. The O 3 -8h-90per concentration in Heilongjiang was relatively small (approximately 80 µg/m 3 ). The highest concentrations of PM 2.5 were observed in Shenyang, Changchun, Harbin, and other major provincial cities in NEC (approximately 60 µg/m 3 ). The spatial distribution of high PM 2.5 concentrations was inconsistent with that of high O 3 -8h-90per concentrations. The spatial distribution of O 3 -8h-90per and PM 2.5 was closely related to the meteorological factors, population density, and environmental impact. In addition, O 3 could also oxidize pollutants in the air and increase PM 2.5 , while PM 2.5 can reduce the solar radiation reaching the ground and decrease the rate of O 3 photochemical formation.
The O 3 -8h-90per concentrations were highest in spring and summer, followed by autumn and winter. In spring, the highest O 3 -8h-90per concentrations (approximately 120-130 µg/m 3 ) were observed in central and western Liaoning. Then, in summer, the O 3 -8h-90per concentration increased considerably in the whole province (to approximately 150 µg/m 3 ). In autumn, the O 3 -8h-90per concentration in NEC decreased markedly until, in winter, the O 3 -8h-90per concentration reached one of its lowest values (approximately 60 µg/m 3 ). By contrast, the distribution of PM 2.5 concentration increased considerably in winter, followed by that in spring and autumn; the lowest PM 2.5 concentration was observed in summer. In spring, the highest PM 2.5 concentrations (approximately 60 µg/m 3 ) were observed in central Liaoning and central Jilin. In summer, the PM 2.5 concentration in NEC decreased. Then, in autumn and winter, the PM 2.5 concentration in central Liaoning increased considerably, reaching approximately 80-90 µg/m 3 . The highest PM 2.5 concentrations in winter were affected mainly by emissions, whereas meteorological conditions such as high temperatures and strong solar radiation were conducive to the formation of O 3 .
The monthly mean O 3 -8h-90per concentration was approximately 70 µg/m 3 , with low concentrations observed in November, December, and January. In May and June, the spatial distribution of the monthly mean O 3 -8h-90per concentration increased substantially, and the highest O 3 -8h-90per concentrations were observed in central and western Liaoning (170-180 µg/m 3 ), followed by central and western Jilin (140-150 µg/m 3 ) and other regions (100-110 µg/m 3 ). By contrast, the PM 2.5 concentrations (approximately 100 µg/m 3 ) were higher in December, January, and February. The highest PM 2.5 concentration was observed in January in central Liaoning and the central part of the Harbin-Changchun urban agglomeration; the highest PM 2.5 concentration was approximately 100 µg/m 3 . In February, the PM 2.5 concentration was lower than that of January, but remained high (approximately 80-90 µg/m 3 ).
From 2013 to 2015, the annual mean O 3 -8h-90per concentration was low (approximately 80-90 µg/m 3 ). In 2016, the annual mean distribution of O 3 -8h-90per in Liaoning began to considerably increase, both in concentration level and in regional distribution. The cities with high PM 2.5 concentration were characterized by industrial activity and high population densities. Since 2012, the interannual variation of tropospheric NO 2 has been decreasing, which is consistent with PM 2.5 , indicating the influence of anthropogenic emission sources on the level of PM 2.5 ; on the contrary, the temperature at 2 m showed a significant increasing trend, which had an important impact on the increase of O 3 -8h-90per concentration The objective of this study was to elucidate the multi-year variation and current levels of O 3 and PM 2.5 pollution in NEC and provide crucial scientific support for regional air pollution prevention and control measures in China.  University's Tracking Air Pollution team is gratefully acknowledged for making the fine particulate matter concentration data and 10 km resolution near-surface ozone concentration data publicly accessible. The authors would like to thank the anonymous reviewers and the editor for their constructive suggestions and comments.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.