1. Introduction
With the rapid development of China’s industrialization and urbanization, the living standards of residents have improved significantly, and the attention to ambient air quality has also risen to a new level. The 2016–2018 “ Bulletin on the State of China’s Ecological Environment” shows that while the pollution situation of fine particulate matter (PM2.5) is still severe, the problem of ozone (O3) pollution in China’s cities has become increasingly prominent, and the number of days with O3 as the primary pollutant exceeding the standard is showing an upward trend year by year, which has become an important factor affecting the number of days with good air quality after PM2.5 [1,2]. Precursors such as volatile organic compounds (VOCs), nitrogen oxides (NOx) and carbon monoxide (CO) undergo a series of photochemical reactions to generate secondary air pollutants such as O3. High near-ground ozone concentrations contribute to an increase in overall non-accidental human mortality, death from cardiovascular system diseases, and mortality from respiratory system disorders, since ozone is strongly linked to many human diseases, particularly cardiovascular diseases [3,4]. On the one hand, the increasing concentration of O3 will trigger urban photochemical smog pollution and endanger the health of residents [5]. On the other hand, it can also interfere with the normal physiological processes of plants and affect the growth of vegetation, resulting in large-scale crop yield reduction [6].
The Yunnan Plateau is located in the southwest of China, close to Myanmar, Laos, and other countries, with the characteristics of a plateau climate and dense vegetation. Compared with other regions of China, the ground can receive more ultraviolet radiation, which is conducive to the production of O3 [7,8,9]. Some studies have shown that: Vehicle exhaust emissions and coal combustion contributed significantly to the generation of O3 in Jinghong and Zhaotong city, respectively [10]. O3 pollution is not only related to local generation, but is also affected by regional transmission [11]. There is significant seasonal variation in the distribution of lower tropospheric ozone around the Qinghai-Tibet Plateau [12]. Finally, the annual O3 in urban sites of Zhengzhou city were affected by air temperature and relative humidity [13]. Based on the O3 concentration data and meteorological data (temperature, precipitation, relative humidity, sunshine hours and wind speed) of 46 monitoring stations in 16 cities (prefectures) in Yunnan from 2015 to 2020, the paper analyzed the spatial and temporal variation in O3 concentration, sources and influencing factors. It is expected to provide reference for air quality forecast, early warning and prevention of O3 pollution in the Yunnan Plateau and the whole southwest region of China.
2. Materials and Methods
2.1. Data Sources
The O3 monitoring data used in this paper are divided into hourly and daily data, which are obtained from 46 state-controlled automatic air quality monitoring stations (hereafter referred to as state-controlled stations) in 16 cities or prefectures (Kunming, Qujing, Yuxi, Baoshan, Zhaotong, Xishuangbanna, Dehong, Diqing, Lijiang, Puer, Lincang, Chuxiong, Honghe, Wenshan, Dali, Nujiang) in Yunnan Plateau from 2015 to 2020. The state-controlled stations in each city are laid out according to the Technical Specification for Ambient Air Quality Monitoring Locations (Trial) (HJ 664-2013), which covers all major functional areas of the city and has good representativeness, comparability and integrity, and can reflect the air quality conditions of each city more accurately. The concentration of O3-8h is derived from the daily maximum 8 h sliding average of O3 at each national control station. The meteorological data for the same period are obtained from the national basic meteorological stations, and the elements include average temperature, annual precipitation, sunshine hours, relative humidity, average wind speed, etc. The geographical distribution of the stations is shown in Figure 1.
2.2. Research Methods
The urban environmental assessment standard strictly adopts level 2 standard for O3 in the “Ambient Air Quality Standard” (GB3095-2012), that is, the arithmetic average concentration of O3-8h exceeds the standard of 200 μg/m3, and the maximum O3-8h/day moving average concentration standard is 160 μg/m3. The annual evaluation of urban ambient air quality is based on the “Technical Specifications for Ambient Air Quality Evaluation (Trial)” (HJ663-2013), and the annual average and daily average percentiles of the evaluation items are implemented to meet the requirements at the same time. The main methods are as follows:
Spatial difference between O3 concentrations is analyzed by using Kriging interpolation. Spearman’s analysis of the correlation between meteorological factors and O3 concen trations in SPSS software was used to analyze the mean correlation.
The spatial dispersion coefficient coefficient of divergence (COD) was used to assess the degree of difference in O3 concentrations between cities. The closer the COD value is to 0, the smaller the difference between the two sites, and the closer it is to 1, the larger the difference COD. The formula is:
(1)
where: and represent city (prefecture) and city (prefecture), respectively; representscity (state) and city (state) spatial dispersion coefficient; is O3 concentration value; and is the total number of data to compare.The HYSPLIT model is an integrated model system jointly developed by the National Oceanic and Atmospheric Association (NOAA) and the Australian Bureau of Meteorology (BOM). It can be used to calculate and analyze processes such as airflow movement, deposition, atmospheric pollutant transport and diffusion trajectories [14,15]. It has been widely used in the study of the transport pathway and source analysis of air pollutants [16,17].
TrajStat follow-up software [18] is used to perform backward trajectory analysis and potential source analysis of O3. Given that the wind field at 500 m height can accurately reflect the mean flow field characteristics of the boundary layer [17], the simulation height is chosen as 500 m. The backward trajectory is calculated for four time periods (00:00, 06:00, 12:00, 18:00, UTC) per day with a simulated duration of 72 h, using a typical city as the simulated receiver point. The 72 h backward trajectories and potential sources are calculated for day-by-day arrival at the receiver point from 2015 to 2020 to facilitate the analysis of O3 sources.
3. Results and Analyses
3.1. Status and Spatial Distribution of O3 Pollution
The number of days that O3-8h in 16 cities in Yunnan Plateau exceeded the level 2 of Ambient Air Quality Standard (GB3095-2012) in 2015–2020 was 371 days, which was 1.1% of the total number of monitoring days (i.e., the total number of monitoring days in 16 cities, 33,655 days). Among them, Xishuangbanna exceeded the level 2 standard for 62 days, followed by Kunming for 58 days, while Dali, Chuxiong and Diqing only exceeded the level 2 standard for 6 days, 3 days and 1 day, respectively. Only the cities of Lijiang and Nujiang did not exceed the standard.
It can be seen from Figure 2 that annual mean O3-8h increased slightly from 2015 to 2019, and dropped significantly in 2020. This is due to the fact that COVID-19 began to spread rapidly in the whole province and cities in late 2019. Most areas of China have implemented traffic control, factory shutdown, banning of gatherings, school closures and company closures [19] since 2020, and the duration and scale of impact are previously incomparable. The development of epidemic prevention work has reduced the 2020 annual emissions of air pollutants.
Figure 3 gives the spatial distribution of O3-8h, NO2 and CO in Yunnan Province averaged over 2015–2020. It can be seen that ρ(O3-8h) shows distribution characteristics that are high in the northeast, i.e., Zhaotong, Kunming and Qujing, high in Baoshan and Dehong in the west, and low in the northwest, mid-west and south. The highest value is found in Zhaotong with 93.06 μg/m3, while the lowest value is found in Nujiang with only 69.58 μg/m3. This may have a strong relationship with the differences in climatic environment and economic development level in different cities (prefectures) in Yunnan Province. Due to the combined influence of the southwest monsoon, strong solar radiation in the plateau region and topography, the northeast has a higher temperature, sufficient sunshine, lower water vapor, lower wind speed and relatively sparse vegetation compared to the northwest, resulting in faster photochemical reaction rate and naturally higher ρ(O3-8h). In addition, Kunming and Qujing in the east-central region have a larger population base, motor vehicle ownership, and GDP, resulting in more anthropogenic O3 precursors and higher ρ(O3-8h). The higher O3 concentrations in Baoshan and Dehong in the west may be related to the horizontal transport of pollutants carried by the external atmosphere.
O3 distribution was quantitatively analyzed in the Yunnan Plateau; COD was used to evaluate the difference in O3 concentrations between each two cities, and to take the annual average concentration of O3 in different cities as x data sets, quantitatively assessing and comparing differences between cities. As can be seen from Figure 4, showing Diqing Prefecture and the other 15 cities (prefectures), the COD values of all cities are relatively large, which indicates that the O3 pollution in Diqing Prefecture is quite different from the other 15 cities, and is less affected by other cities. There may be two reasons for this: On the one hand, as a result of the dominance of the cultural tourism, highland-specific modern agriculture, green energy, and green mining industries in Diqing, the state’s ambient air quality status is excellent for 275 days and good for 90 days, and has an excellent rate of 100% in 2019. In Diqing Prefecture, the daily variation in O3-1h has a mild “unimodal” distribution from 2015 to 2020, with the lowest mean value of 47.84 μg/m3 at roughly 9:00. and the greatest mean value of 71.17 μg/m3 at roughly 17:00. When the light intensity increases at 9:00, the O3-1h climbs quickly and peaks between 15:00 and 16:00. The light response contributes greatly to the production of ozone.
On the other hand, Figure 5 shows that westerly air currents, with Myanmar as Yunnan’s upwind side and Diqing Prefecture in northwest Yunnan as the upwind side of northern Yunnan, govern the atmospheric circulation in Yunnan and the surrounding middle 500 hPa (5500 m above sea level). At the same time, compared with the upwind side of central Yunnan, south of central and southern Myanmar, and northern Laos (circular area in the middle figure), the annual average number of fire points in north Myanmar where the windward airflow passes over northwest Yunnan is 0–50/100 km2, which belongs to the low value area of fire points. Additionally, the high mountains that line the Nujiang River valley on both sides [20], the mountain range’s north–south orientation, the topographic relief of more than 1000 m (the rectangle on the right), and the upwind side of the impact of pollutant transport on northwest Yunnan all function as natural barriers. The above may be the reason for the large difference between the O3 pollution status of Diqing Prefecture in northwest Yunnan and the remaining 15 cities.
3.2. O3 Concentration Time Variation Characteristics
3.2.1. Seasonal Variations in O3-8h
Figure 6 gives the spatial distribution of ρ(O3-8h) averaged over various seasons in the Yunnan Plateau from 2015 to 2020. It can be seen that ρ(O3-8h) has obvious seasonal variation. The ρ(O3-8h) was significantly higher in spring, followed by winter and summer, and lowest in autumn. On the seasonal scale, unlike most regions in northeastern [21], northern [22] and northwestern [23] China where O3 concentrations are highest in summer, the peak O3 concentrations in Yunnan cities (prefectures) are concentrated in March–May (spring). This may be due to the combined effect of precursors and light radiation intensity. Yunnan is located in a low-latitude plateau region, with thinner air and strong solar radiation, and the near-surface temperature rises rapidly after entering into February, with 41% of the days above 30 °C occurring in spring. The rapid increase in photochemical reaction rate makes O3 generation faster, and Yunnan belongs to the southwest monsoon region, with a distinct wet and dry climate, low precipitation and low humidity in the dry season. The rainy season starts in June with more cloudy and rainy weather in Yunnan. In winter, when the sun shines directly on the Tropic of Capricorn, the intensity of light radiation on the Yunnan Plateau is weak and O3 precursors are mainly accumulated; in March–May (spring), the intensity of radiation on the plateau is high, which makes the atmospheric photochemical reactions significantly enhanced and accelerates O3 production. Subsequently, June to August (summer) is the period of decreasing O3 concentration, during which Yunnan welcomes the rainy season and rain has a wet scavenging effect on atmospheric pollutants, and during which there is cloud cover and solar radiation is significantly weakened, which can effectively reduce atmospheric O3 concentration. September to February (autumn and winter) is the period of calm O3 change, when solar radiation weakens, temperature decreases, atmospheric photochemical reaction is slower, and O3 concentration tends to level off. It is noteworthy that the number of fire spots in Southeast Asia begins to slowly increase in November, which is why western Yunnan experiences high ozone concentrations in the winter. Additionally, Southeast Asian countries conduct a lot of biomass burning during March–April (according to the SNPP/VIIRS satellite fire point data provided by the National Oceanic and Atmospheric Administration (NOAA), about 70% of the fire points are concentrated in March–April), and the emitted pollutants may be transported long-distance through the atmospheric circulation to the downstream over Yunnan, increasing the local O3 concentration.
3.2.2. Daily Variations in O3-1h
From 2015 to 2020, the daily variation in O3-1h in Yunnan Province shows a “single-peaked” distribution. In general, the levels in the daytime are higher than those in the nighttime, and the O3-1h is at the lowest value at 8:00, and the O3-1h rises rapidly with the increase in light intensity at 9:00, and reaches the peak at 15:00–16:00, and then decreases gradually, with a rapid decrease at 17:00–23:00, and the decrease in O3 O3-1h tends to level off after 23:00 and remains at a low level until the next morning (Figure 7). The O3-1h decreases rapidly from 17:00 to 23:00 in the evening, and after 23:00 in the evening, the decline tends to level off and it remains at a low level until the early morning of the second day (Figure 7). This is because after sunset, a stable boundary layer begins to form at the base of the residual layer as a result of the inversion of surface radiation, and as turbulent kinetic energy dissipates, the residual layer’s height decreases even further [23]. The PBLH in Yunnan was examined using the fifth generation ECMWF Re-Analyses (ERA5) data that were published by ECMWF in 2016. It was discovered that the multi-year average of PBLH in the province from 5:00 to 9:00 (BJT) is 136 m (annual average 498 m), and the lower PBLH results in less vertical exchange of pollutants in the early morning due to weak turbulence activity and the concentration of O3.Additionally, 8:00–9:00 is the morning peak of travel, in which motor vehicles emit a large amount of NOx, and the atmospheric NO2 concentration rises to consume O3, while the solar radiation is weak or there is even a lack of solar radiation at this time, resulting in a slow photochemical reaction. Additionally, existing O3 is consumed and new O3 is not generated, resulting in the lowest O3 concentration at this time. After 9:00, the solar radiation strengthens and O3 concentration gradually rises. After a period of photolysis and chemical reaction of O3 precursors, the O3 concentration reaches its peak at 15:00–16:00, then the sunlight weakens and O3-1h decreases rapidly, mainly in the evening when the photochemical reaction weakens significantly, and superimposed on the combined effect of NO titration, the O3 consumption increase accelerated reduction of its concentration level. This is basically consistent with the daily variation pattern in other regions.
3.3. Correlation Analysis of O3 Concentration and Meteorological Factors
Studies on the relationship between O3 and meteorological factors vary slightly from region to region, but both domestic and international studies have roughly the same conclusions on the effects of solar radiation, air temperature, relative humidity, and precipitation on O3 concentrations [24]. The analysis of O3 pollution characteristics in Xi’an showed that solar radiation is a key factor in determining O3 production [25]. O3 concentration is negatively correlated with relative humidity and visibility, and positively correlated with air temperature, wind speed and sunshine hours [26,27]. Meteorological factors cause changes in near-surface O3 concentrations by influencing the occurrence of photochemical reactions; therefore, studying the effects of meteorological factors on O3 pollution will help to reveal the patterns of meteorological factors on O3 pollution and their intrinsic changes and connections. In order to provide more accurate results of meteorological influences in different regions, while analyzing the correlation between O3 concentrations and meteorological factors in the province, Zhaotong in northeast Yunnan, Kunming in east-central Yunnan and Pu’er in southern Yunnan were selected as representative cities to analyze the differences in correlation between O3 concentrations and meteorological factors in each region according to the distribution characteristics of O3 pollution in Yunnan Plateau. In the following, all non-parametric Spearman correlation analysis was used to calculate the correlation coefficients between O3 concentrations and conventional meteorological factors, and significance tests were performed.
3.3.1. Air Temperature
Temperature is directly influencing the concentration of O3 by affecting its photochemical reaction production efficiency [28]. As can be seen from Table 1, the representative cities of Zhaotong and Kunming show a significant positive correlation between annual O3 concentration and temperature at the 0.01 level, and Pu’er shows a significant negative correlation between annual O3 concentration and temperature at the 0.01 level. The differences among cities are more obvious, with the correlation being better in the northern cities than in the southern cities. The correlation coefficients of O3 concentration and temperature in Zhaotong and Kunming were significantly lower in summer and autumn than in spring and winter, and the correlation coefficients of O3 concentration and temperature in Pu’er were lower in spring and winter than in summer and autumn, and they showed a significant negative correlation in autumn. Among them, Zhaotong in summer, Kunming in autumn and Pu’er in spring did not pass the significance test at a confidence level (two-sided) of 0.01.
Representative cities were selected to analyze the O3 concentrations and exceedances (primary standard 100 μg/m3, the same below) at different temperatures. From Figure 8, it can be found that the frequency of O3 exceedance in Zhaotong and Kunming is relatively low when the temperature is below 10 °C, but it increases significantly with the rise in temperature. The maximum value of O3 exceedance frequency occurs at 25–30 °C in Kunming and 15–20 °C in Zhaotong. In Pu’er, the exceedance frequency at 15–30 °C was significantly higher than that at 0–15 °C, but the correlation between the two was not significant, which might show a negative correlation in autumn, and was influenced by rainfall and wind speed. The minimum temperatures for daily O3 concentrations to reach the primary standard pollution were 0.6 °C, 4.3 °C and 8.5 °C for Zhaotong, Kunming and Pu’er, respectively, and 14.0 °C, 13.6 °C and 17.6 °C for secondary pollution. In general, the temperature required for O3 exceedances to occur in the northern region is much lower. This may be due to the richer vegetation in the northern region and more VOCs emissions from industrial and traffic sources than in the southern region.
3.3.2. Relative Humidity
The correlation coefficients of relative humidity and O3 concentration in the three representative cities were −0.551, −0.621 and −0.693 from north to south (Zhaotong, Kunming and Pu’er), and the correlation degree of O3 concentration and relative humidity in each city varied in different seasons. The correlation between O3 concentration and relative humidity varies in different seasons, most prominently in winter, with the absolute values of the correlation coefficients ranked as Pu’er > Kunming > Zhaotong, and the correlation increases from north to south, the same as the annual ranking. This is due to the fact that the relative humidity causes a decrease in O3 concentration mainly by weakening solar radiation and forming wet deposition, and the suppression effect is more significant in winter. In addition, the seasonal characteristics of the correlation between O3 concentration and relative humidity (absolute value) differed among cities. Zhaotong and Pu’er showed: winter > autumn > spring > summer, while Kunming showed: winter > autumn > summer > spring, and the correlation coefficient of Pu’er was more different, especially between winter and autumn and summer. This indicates that the effect of relative humidity on O3 concentrations is more pronounced in the southern region than in the central and northern regions in terms of seasonal differences.
Figure 9 shows that when the relative humidity is less than 50%, the average concentration of O3 is the highest and the frequency of exceedance is the greatest; with the further increase in relative humidity, the concentration of O3 and the frequency of exceedance both decrease significantly; when the relative humidity is greater than 90%, there is no exceedance of O3 in Kunming. This indicates that the higher relative humidity is not conducive to O3 production. This is mainly due to the influence of atmospheric water vapor on O3 concentration changes in three ways: first, solar radiation, as one of the important conditions for the occurrence of photochemical reactions, will be attenuated by the extinction mechanism under the action of water vapor [29]; second, high humidity is conducive to the wet deposition of O3, thus achieving the effect of O3 removal [30]; third, under higher humidity conditions, the photochemical reactions that consume O3 reaction process dominates under higher humidity conditions (Equations (2)–(6)) [31].
O3 + NO→ NO2 + O2(2)
O3 + OLE(olefins) → products(3)
O3 + OH→ HO2 + O2(4)
O3 + HO2→ OH + O2(5)
HO2 + NO→ NO2 + OH(6)
RO2 + NO→ φ NO2 + HO2(7)
NO2 + hυ→ NO + O3(8)
When the relative humidity is less than 70%, the frequency of O3 exceedance in Pu’er is significantly higher than that in Zhaotong and Kunming, which indicates that within a certain relative humidity interval, O3 pollution also shows significant geographical differences, especially in the lower relative humidity, and that the southern region is more prone to high concentration of O3 than the northern regions (Table 2).
3.3.3. Wind Direction and Wind Speed
Table 3 shows that, for each city (state), O3 concentration and wind speed are roughly positively correlated; seasonally, except for summer and autumn in Kunming, the correlation is mainly positive, and is most significant in winter in Kunming. Figure 10 shows the wind direction versus O3 The wind rose diagram of the effect of concentration, obviously on the daily O3 in Zhaotong, Kunming and Pu’er southwesterly winds have the greatest influence on the concentration. Analysis of meteorological data found that from 2015 to 2020, the southwesterly wind prevailed in the Yunnan Plateau in the prevailing wind direction for the concentration is southwesterly. The overall seasonal distribution of wind speed is: spring > winter > autumn > summer, and the wind speed is mainly distributed in 0–3 m/s.
As shown in Figure 10 and Figure 11, in the wind speed range of 0–3 m/s, the O3 exceedance rate and ρ(O3-8h) in Zhaotong and Kunming generally showed a decreasing trend with increasing wind speed, while the O3 exceedance rate and ρ(O3-8h) in Pu’er generally show an increasing trend with increasing wind speed, which may be related to the high vegetation coverage in Pu’er near Southeast Asia and the heterogeneous wind direction in different boundary layer heights. In Zhaotong and Kunming, the O3 exceedance rate and ρ(O3-8h) fluctuate at wind speeds higher than 3 m/s, and the values rebound, which may be due to the transport effect of high O3 concentrations upstream of the wind direction. Usually, an increase in wind speed implies an increase in air mass transport power, which is favorable to the transport of O3 and its precursors, and this causes an increase in O3 concentration when accumulation is predominant in local areas.
3.4. Analysis of Potential Sources of O3 Pollution
In order to further understand the influence of the trajectory of pollutant transport air mass on the change of O3 concentration in the Yunnan Plateau, the backward trajectory and PSCF were used to analyze the O3 concentration in the Yunnan Plateau. The annual average ρ(O3-8h) was the only one in Baoshan city with a higher concentration exceeding 90 μg/m3 in 2018 for analysis and discussion.
Figure 12 and Table 4 show the results of backward trajectory clustering and WPSCF analysis in the four seasons of Baoshan city in spring, summer, autumn and winter in 2018.
As shown in the figure, the westerly and southwesterly airflows accounted for the largest proportion of the airflow in 2018, and Baoshan city was mainly affected by the westerly or southwesterly winds. In spring, the northeast airflow (track 3) accounted for less (1.65%) of the fastest moving speed and the longest transport distance; the west airflow accounted for the most (64.01%) and the transported O3 concentration was the highest (94.81 μg/m3), indicating that most air masses originate from or pass through Southeast Asia (Myanmar, Laos, and Thailand), and only a few air mass trajectories originate in China (Sichuan and Guizhou provinces). In summer, the contribution of air masses from the South Asian continent decreases, and most trajectories originate from the Bay of Bengal and Myanmar. In autumn, the trajectories starting from the Bay of Bengal are less than in summer, and there is an air mass transmission in each direction, and the most air currents are still the western and southwest airflows. In winter, almost all trajectories come from South and Southeast Asia, where the air masses passing through northern Myanmar (Track 1) accounted for the largest proportion and they transported the highest concentration of O3 in general; the air masses arriving at the study site were mostly from South and Southeast Asia.
To avoid double counting the contribution of the same source, potential sources were calculated for only one year (2018). India (mainly north-central India), Myanmar, Thailand and central China were identified as high potential sources of NOx, probably due to biomass burning in the region [32,33]. O3 is a secondary pollutant formed by photochemical reactions between NOx and volatile organic compounds. In addition to photochemical reactions, inputs from the stratosphere and troposphere are another important source of ozone, especially at high altitudes [34], which leads to the frequent occurrence of high stratospheric O3 intrusion on the Tibetan Plateau [34,35,36]. Calculations of potential sources of O3 suggest that the Tibetan Plateau region may concentrate the impact on O3 concentrations in southwestern China (mainly influenced by regional stratospheric intrusion). In addition to the influence from the Qinghai-Tibet Plateau, the influence from local and neighboring regions is also evident in spring. In this study, local air pollutant emissions from Baoshan City make a limited contribution to high air pollutant concentrations, and cross-border transport of air pollutants from South and Southeast Asia perhaps affect O3 concentrations at the southwestern border of China.
The cluster analysis results show that the long-distance migration of air pollutants has a profound impact on the local air quality in Baoshan, especially the air mass from Myanmar.
4. Conclusions
O3 has become the main pollutant affecting urban air quality in Yunnan Plateau. The spatial characteristics of O3 concentrations show high values in the northeast and west, and low values in the northwest, central and west, and south of the Yunnan Plateau. The O3 pollution in Diqing is significantly different from other cities, and perhaps it is less influenced by O3 transport from other regions.
The correlations between O3 concentration and temperature as well as relative humidity on the Yunnan Plateau was good, showing significant positive and negative correlations. The correlation between temperature and O3 concentration was most significant in winter, and the correlation between relative humidity and O3 concentration was better in autumn and winter. The lowest O3 concentration was found at the average wind speed of 3 m/s, which was mainly influenced by the southwest wind.
O3 pollution in the Yunnan Plateau is mainly influenced by local emissions and external transport, with industrial and traffic sources such as Kunming contributing the most to O3; southwestern and western air currents are the most important transport pathways for pollutants from outside, and most of the high concentrations of O3 observed in southwestern Yunnan are perhaps associated with transboundary transport of air pollution from South and Southeast Asia.
Conceptualization, J.S., P.Z. and X.H.; investigation, Z.W. and C.Z.; resources, J.S.; data curation, C.Z., J.W. and X.Y.; writing—original draft preparation, Z.W.; writing—review and editing, P.Z. and X.H.; supervision, H.X. and P.N. All authors have read and agreed to the published version of the manuscript.
Not applicable.
Not applicable.
The data used in this paper can be provided by Jianwu Shi (
This work was supported by the National Natural Science Foundation of China, the National Key R&D Program of China, and the Science and Technology Special Project of Demonstration Zone for National Sustainable Development in Yunnan.
The authors declare that there are no competing financial interests that could inappropriately influence the contents of this manuscript.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Enlarge this image.Figure 1. Spatial distribution of 46 O3 stations and 16 meteorological observation stations in Yunnan Province.
Enlarge this image.Figure 2. Variation trend of annual mean O3-8h in Yunnan Province from 2015 to 2020.
Enlarge this image.Figure 3. The average spatial distribution of O3-8h, NO2 and CO in Yunnan Plateau from 2015 to 2020.
Enlarge this image.Figure 4. The COD values of O3-8h in Yunnan Plateau from 2015–2020.
Enlarge this image.Figure 5. For the majority of Southeast Asia (90° to 107° E, 15° to 29° N), Yunnan Province’s topographic relief and average annual biomass combustion fire points are distributed spatially from 2015 to 2020.
Enlarge this image.Figure 6. Average distribution of O3-8h concentration in four seasons in Yunnan from 2015 to 2020.
Enlarge this image.Figure 7. Daily trend of O3-1h in Yunnan from 2015 to 2020.
Enlarge this image.Figure 8. Zhaotong, Kunming and Pu’er at different temperatures. Changes in O3 concentration and exceeding standard frequency.
Enlarge this image.Figure 9. Changes from relative humidity and O3 concentration, exceeding standard frequency between Zhaotong, Kunming and Pu’er.
Enlarge this image.Figure 10. Changes from O3 Concentration and wind rose illustration, exceeding standard frequency between Zhaotong, Kunming and Pu’er.
Enlarge this image.Figure 11. Different changes from wind speed and O3 concentration between Zhaotong, Kunming and Pu’er.
Enlarge this image.Figure 12. O3 backward trajectory clustering and WPSCF analysis of four seasons in Baoshan city in 2018.
Correlation Analysis of O3 Concentration and Temperature in Different Seasons.
| Season | Zhaotong | Kunming | Pu’er | |||
|---|---|---|---|---|---|---|
| r | P | r | P | r | P | |
| Spring | 0.268 * | 0.0000 | 0.559 * | 0.0000 | 0.055 | 0.1996 |
| Summer | 0.144 * | 0.0007 | −0.098 | 0.0219 | 0.498 * | 0.0000 |
| Autumn | −0.052 | 0.2335 | 0.380 * | 0.0000 | −0.382 * | 0.0000 |
| Winter | 0.384 * | 0.0000 | 0.438 * | 0.0000 | −0.160 * | 0.0005 |
Note: “*” indicates a confidence level (two-sided) of 0.01, the correlation is significant.
Correlation Analysis of O3 Concentrations and Humidity in Different Seasons.
| Season | Zhaotong | Kunming | Pu’er | |||
|---|---|---|---|---|---|---|
| r | P | r | P | r | P | |
| Spring | −0.470 * | 0.0000 | −0.349 * | 0.0000 | −0.489 * | 0.0000 |
| Summer | −0.383 * | 0.0000 | −0.481 * | 0.0000 | −0.280 * | 0.0000 |
| Autumn | −0.615 * | 0.0000 | −0.477 * | 0.0000 | −0.630 * | 0.0000 |
| Winter | −0.657 * | 0.0000 | −0.670 * | 0.0000 | −0.674 * | 0.0000 |
| Annual | −0.551 * | 0.0000 | −0.621 * | 0.0000 | −0.693 * | 0.0000 |
Note: “*” indicates that the confidence level (two-sided) is 0.01, the correlation is significant.
Correlation Analysis of O3 Concentration and Wind Speed in Different Seasons.
| Season | Zhaotong | Kunming | Pu’er | |||
|---|---|---|---|---|---|---|
| r | P | r | P | r | P | |
| Spring | 0.062 | 0.1439 | −0.007 | 0.8644 | 0.292 * | 0.0000 |
| Summer | 0.135 * | 0.0000 | −0.285 * | 0.0000 | −0.050 | 0.2441 |
| Autumn | 0.162 * | 0.0000 | −0.122 * | 0.0047 | 0.055 | 0.2083 |
| Winter | 0.180 * | 0.0000 | 0.376 * | 0.0000 | 0.241 * | 0.0000 |
| Annual | −0.033 | 0.1336 | 0.211 * | 0.0000 | 0.223 * | 0.0000 |
Note: “*” indicates that the confidence level (two-sided) is 0.01, the correlation is significant.
Clustering results of backward trajectories and corresponding average concentrations of O3 in four seasons of Baoshan city in 2018.
| Quarter | Track | Frequency | O3/(μg/m3) |
|---|---|---|---|
| Spring | 1 | 6.59 | 75.93 |
| 2 | 64.01 | 94.81 | |
| 3 | 1.65 | 92.92 | |
| 4 | 27.75 | 91.31 | |
| Summer | 1 | 75.55 | 49.01 |
| 2 | 8.52 | 63.25 | |
| 3 | 9.34 | 53.11 | |
| 4 | 6.59 | 72.64 | |
| Autumn | 1 | 22.22 | 51.28 |
| 2 | 63.33 | 45.42 | |
| 3 | 9.72 | 41.90 | |
| 4 | 4.72 | 58.60 | |
| Winter | 1 | 60.47 | 66.11 |
| 2 | 2.33 | 60.07 | |
| 3 | 22.67 | 49.31 | |
| 4 | 14.53 | 52.88 |
References
1. Gong, S.; Zhang, L.; Liu, C.; Lu, S.; Pan, W.; Zhang, Y. Multi-scale analysis of the impacts of meteorology and emissions on PM2.5 and O3 trends at various regions in China from 2013 to 2020 2. Key weather elements and emissions. Sci. Total Environ.; 2022; 824, 153847. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2022.153847] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35189213]
2. Wang, Q.; Zhu, H.; Xu, H.; Lu, K.; Ban, J.; Ma, R.; Li, T. The spatiotemporal trends of PM2.5-and O3-related disease burden coincident with the reduction in air pollution in China between 2005 and 2017. Resour. Conserv. Recycl.; 2022; 176, 105918. [DOI: https://dx.doi.org/10.1016/j.resconrec.2021.105918]
3. Liu, Y.; Qiu, P.; Li, C.; Li, X.; Ma, W.; Yin, S.; Yu, Q.; Li, J.; Liu, X. Evolution and variations of atmospheric VOCs and O3 photochemistry during a summer O3 event in a county-level city, Southern China. Atmos. Environ.; 2022; 272, 118942. [DOI: https://dx.doi.org/10.1016/j.atmosenv.2022.118942]
4. Sather, M.E.; Cavender, K. Update of long-term trends analysis of ambient 8-hour ozone and precursor monitoring data in the South Central US; encouraging news. J. Environ. Monit.; 2012; 14, pp. 666-676. [DOI: https://dx.doi.org/10.1039/c2em10862c]
5. Marenco, A. Variations of CO and O3 in the troposphere: Evidence of O3 photochemistry. Atmos. Environ.; 1986; 20, pp. 911-918. [DOI: https://dx.doi.org/10.1016/0004-6981(86)90275-1]
6. Zhu, J.; Cheng, H.; Peng, J.; Zeng, P.; Wang, Z.; Lyu, X.; Guo, H. O3 photochemistry on O3 episode days and non-O3 episode days in Wuhan, Central China. Atmos. Environ.; 2020; 223, 117236. [DOI: https://dx.doi.org/10.1016/j.atmosenv.2019.117236]
7. Shen, Z.; Cao, J.; Zhang, L.; Zhao, Z.; Dong, J.; Wang, L.; Wang, Q.; Li, G.; Liu, S.; Zhang, Q. Characteristics of surface O3 over Qinghai Lake area in Northeast Tibetan Plateau, China. Sci. Total Environ.; 2014; 500, pp. 295-301. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2014.08.104]
8. Crutzen, P. Photochemical reactions initiated by and influencing ozone in unpolluted tropospheric air. Tellus; 1974; 26, pp. 47-57. [DOI: https://dx.doi.org/10.3402/tellusa.v26i1-2.9736]
9. Li, M.; Mao, J.; Chen, S.; Bian, J.; Bai, Z.; Wang, X.; Chen, W.; Yu, P. Significant contribution of lightning NOx to summertime surface O3 on the Tibetan Plateau. Sci. Total Environ.; 2022; 829, 154639. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2022.154639]
10. Wang, C. Study on the Causes of Ground-level Ozone in Typical Cities of Yunnan Province. Master Thesis; Kunming University of Science and Technology: Kunming, China, 2018.
11. Kim, B.U.; Kim, H.C.; Kim, S. Effects of vertical turbulent diffusivity on regional PM2.5 and O3 source contributions. Atmos. Environ.; 2021; 245, 118026. [DOI: https://dx.doi.org/10.1016/j.atmosenv.2020.118026]
12. Yang, Y.F.; Bian, J.C.; Wang, H.Y.; Wang, W.G.; Zhang, W.C.; Fan, W.X.; Yu, K. Seasonal vailation of atmospheric ozone in the lower troposphere surrounding the Tibetan Plateau based on OMI satellite date. J. Yunnan Univ. -Nat. Sci. Ed.; 2013; 35, pp. 183-192.
13. Wang, M.; Zheng, Y.; Liu, Y.; Li, Q.; Ding, Y. Characteristics of ozone and its relationship with meteorological factors in Beijing-Tianjin-Hebei Region. China Environ. Sci.; 2019; 39, pp. 2689-2698.
14. Draxler, R.R.; Hess, G.D. An overview of the hysplit-4 modeling system for trajectories. Aust. Meteorol. Mag.; 1998; 47, pp. 295-308.
15. Stein, A.; Draxler, R.R.; Rolph, G.D.; Stunder, B.J.; Cohen, M.; Ngan, F. NOAA’s HYSPLIT atmospheric transport and dispersion modeling system. Bull. Am. Meteorol. Soc.; 2015; 96, pp. 2059-2077. [DOI: https://dx.doi.org/10.1175/BAMS-D-14-00110.1]
16. Bera, B.; Bhattacharjee, S.; Sengupta, N.; Saha, S. Variation and dispersal of PM10 and PM2.5 during COVID-19 lockdown over Kolkata metropolitan city, India investigated through HYSPLIT model. Geosci. Front.; 2022; 13, 101291. [DOI: https://dx.doi.org/10.1016/j.gsf.2021.101291]
17. Escudero, M.; Stein, A.; Draxler, R.; Querol, X.; Alastuey, A.; Castillo, S.; Avila, A. Source apportionment for African dust outbreaks over the Western Mediterranean using the HYSPLIT model. Atmos. Res.; 2011; 99, pp. 518-527. [DOI: https://dx.doi.org/10.1016/j.atmosres.2010.12.002]
18. Shan, W.; Yin, Y.; Lu, H.; Liang, S. A meteorological analysis of ozone episodes using HYSPLIT model and surface data. Atmos. Res.; 2009; 4, pp. 767-776. [DOI: https://dx.doi.org/10.1016/j.atmosres.2009.03.007]
19. Fan, X.; Lu, J.; Qiu, M.; Xiao, X. Changes in travel behaviors and intentions during the COVID-19 pandemic and recovery period: A case study of China. J. Outdoor Recreat. Tour.; 2022; 100522. [DOI: https://dx.doi.org/10.1016/j.jort.2022.100522]
20. Huang, X.; Shao, T.; Zhao, J.; Cao, J.; Song, Y. Influence factors and spillover effect of PM2.5 concentration on Fen-wei Plain. China Environ. Sci.; 2019; 39, pp. 3539-3548.
21. Li, X.; Li, S.; Liu, P.; Kong, Y.; Song, H. Temporal and Spatial Variation of Ozone Concentration in Chinese Cities in 2016. J. Environ. Sci.; 2018; 38, pp. 1263-1274.
22. Gao, L.; Wang, T.; Ren, X.; Ma, D.; Zhuang, B.; Li, S.; Xie, M.; Li, M.; Yang, X. Subseasonal characteristics and meteorological causes of surface O3 in different East Asian summer monsoon periods over the North China Plain during 2014–2019. Atmos. Environ.; 2021; 264, 118704. [DOI: https://dx.doi.org/10.1016/j.atmosenv.2021.118704]
23. Wang, L.; Li, M.; Wang, Q.; Li, Y.; Xin, J.; Tang, X.; Du, W.; Song, T.; Li, T.; Sun, Y. Air stagnation in China: Spatiotemporal variability and differing impact on PM2.5 and O3 during 2013–2018. Sci. Total Environ.; 2022; 819, 152778. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2021.152778] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34990676]
24. Paraschiv, S.; Barbuta-Misu, N.; Paraschiv, S.L. Influence of NO2, NO and meteorological conditions on the tropospheric O3 concentration at an industrial station. Energy Rep.; 2020; 6, pp. 231-236. [DOI: https://dx.doi.org/10.1016/j.egyr.2020.11.263]
25. Hecht, A.; Galo, R.; Fellows, S.; Baldez, P.; Koonath, P. Radiolytic ozone yield G (O3) from 210Po alpha-particle radiation in air. Radiat. Phys. Chem.; 2021; 183, 109387. [DOI: https://dx.doi.org/10.1016/j.radphyschem.2021.109387]
26. Plocoste, T.; Calif, R.; Jacoby-Koaly, S. Temporal multiscaling characteristics of particulate matter PM10 and ground-level ozone O3 concentrations in Caribbean region. Atmos. Environ.; 2017; 169, pp. 22-35. [DOI: https://dx.doi.org/10.1016/j.atmosenv.2017.08.068]
27. Li, C.; Liu, Y.; Cheng, B.; Zhang, Y.; Liu, X.; Qu, Y.; An, J.; Kong, L.; Zhang, Y.; Zhang, C. A comprehensive investigation on volatile organic compounds (VOCs) in 2018 in Beijing, China: Characteristics, sources and behaviours in response to O3 formation. Sci. Total Environ.; 2022; 806, 150247. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2021.150247]
28. Wu, J.; Wang, Y.; Liang, J.; Yao, F. Exploring common factors influencing PM2.5 and O3 concentrations in the Pearl River Delta: Tradeoffs and synergies. Environ. Pollut.; 2021; 285, 117138. [DOI: https://dx.doi.org/10.1016/j.envpol.2021.117138]
29. Liu, J.; Ding, Y.; Huang, Y.; Huang, L. Correlation analysis between solar ultraviolet radiation intensity and meteorological elements. Plateau Meteorol.; 2003; 22, pp. 45-50.
30. Kavassalis, S.C.; Murphy, J.G. Understanding ozone-meteorology correlations: A role for dry deposition. Geophys. Res. Lett.; 2017; 44, pp. 2922-2931. [DOI: https://dx.doi.org/10.1002/2016GL071791]
31. Gallimore, P.; Achakulwisut, P.; Pope, F.; Davies, J.; Spring, D.; Kalberer, M. Importance of relative humidity in the oxidative ageing of organic aerosols: Case study of the ozonolysis of maleic acid aerosol. Atmos. Chem. Phys.; 2011; 11, pp. 12181-12195. [DOI: https://dx.doi.org/10.5194/acp-11-12181-2011]
32. Qin, Y.; Xie, S. Historical estimation of carbonaceous aerosol emissions from biomass open burning in China for the period 1990–2005. Environ. Pollut.; 2011; 159, pp. 3316-3323. [DOI: https://dx.doi.org/10.1016/j.envpol.2011.08.042] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21911273]
33. Qin, X.; Yan, H.; Zhan, Z.; Li, Z. Characterising vegetative biomass burning in China using MODIS data. Int. J. Wildland Fire; 2013; 23, pp. 69-77. [DOI: https://dx.doi.org/10.1071/WF12163]
34. Luo, J.; Liang, W.; Xu, P.; Xue, H.; Zhang, M.; Shang, L.; Tian, H. Seasonal features and a case study of tropopause folds over the Tibetan Plateau. Adv. Meteorol.; 2019; 2019, 4375123. [DOI: https://dx.doi.org/10.1155/2019/4375123]
35. Chen, X.; Ma, Y.; Kelder, H.; Su, Z.; Yang, K. On the behaviour of the tropopause folding events over the Tibetan Plateau. Atmos. Chem. Phys.; 2011; 11, pp. 5113-5122. [DOI: https://dx.doi.org/10.5194/acp-11-5113-2011]
36. Yin, D.; Zhao, S.; Qu, J. Spatial and seasonal variations of gaseous and particulate matter pollutants in 31 provincial capital cities, China. Air. Qual. Atmos. Health; 2017; 10, pp. 359-370. [DOI: https://dx.doi.org/10.1007/s11869-016-0432-1]
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.