*3.2. Seasonal Variations of Pollutants*

Figure 4 shows the seasonal variations of the six pollutants in each city. For almost all pollutants (except O3), the highest concentrations were observed in winter and the lowest in summer. It is speculated that the continuous adverse weather conditions in winter include smaller wind speed and rainfall, lower temperature and atmospheric boundary height, which are not conducive to the diffusion of pollutants. And compared with other seasons, the consumption of coal and biomass fuel for heating in winter is higher [41,42]. On the contrary, in summer, the wind speed and planetary boundary layer is higher, the rainfall is abundant, the rain removal effect is obvious, and the pollutant concentration is lower [43,44].

Since NO2 is the main man-made pollutant emitted from vehicles and transportation facilities and fuel combustion, these activities are more frequent in the two megacities of Chengdu and Chongqing than in other places [45]. In 2018, the total number of motor vehicles in Chengdu and Chongqing accounted for 28.2% and 36.5% of the entire basin area respectively. Therefore, the highest NO2 concentration was observed in these two

cities. The average concentration of NO2 in winter was between 111.87 μg/m3 (Chengdu) and 50.93 μg/m3 (Aba Prefecture). Similar characteristics were observed for CO, namely the highest and lowest concentrations were observed in winter and summer, respectively. The opposite trend of ozone occurred. The highest concentration happened in spring and summer, and the lowest concentration occurred in winter. This is related to the formation mechanism of ozone. Many studies showed that under sufficient light, volatile organic compounds (VOCs) and nitrogen oxides (NOx) underwent a photochemical reaction to generate O3 and at the same time produced secondary pollutants in the atmosphere [46–48]. High temperature, strong ultraviolet and high photochemical reaction rate were common phenomena in Sichuan Basin during spring and summer. The weaker solar radiation in winter inhibited the photochemical reaction, which was not conducive to the production of O3. Therefore, the O3 concentration in the Sichuan Basin had the highest trend in spring and summer.

**Figure 4.** Seasonal variation in concentrations of air pollutants in Sichuan Basin. The vertical error bars represent the standard deviation values. Spring (March to March), Summer (June to August), Autumn (September to November), Winter (December to February of the following year).

Regions with high SO2 concentrations were mainly located in the plateau areas of the western Sichuan Basin, such as Panzhihua, Liangshan, Ganzi and other cities. Far more than the cities such as Chengdu and Mianyang in the basin, the winter SO2 concentration of Panzhihua was 38.7 μg/m3, about 3 times of the average SO2 concentration in whole Sichuan Basin (12.8 μg/m3). On the one hand, the SO2 of cities in the basin such as Chengdu was mainly derived from industrial emissions. In these areas, the government took strict desulfurization measures, which greatly reduced the concentration of SO2. On the other hand, coal combustion for household heating due to low temperature in high altitude regions, led to more SO2 emissions, and the implementation of desulfurization measures in these areas were not yet fully completed.

Different with northern China, due to the warm temperature in Sichuan Basin (about 10 ◦C on average), there was no widespread coal or wood burning for household heating in winter; therefore, atmospheric processes and meteorological conditions played an important role in the seasonal changes of particulate matter effect [49]. Almost all regions had the highest concentration of particulate matter in winter, about 1.8–2.5 times that of the other three seasons. The concentration was similar in spring and autumn, and the lowest concentration occurred in summer.

#### *3.3. Analysis of City's Pollutant Ratio*

The average PM2.5/PM10 ratio of all cities in Sichuan Basin was 0.61, and the monthly ratio was between 0.43 and 0.69, appeared in April and January respectively. The average ratio in Chengdu and Chongqing was 0.60 and 0.63, respectively. In 2017, a study reported the average ratio of 0.58 for 31 provincial capital cities, and Zhang (2015) reported an average ratio of 0.56 for 190 cities in China [7,50]. However, in Beijing (0.80), Shanghai (0.70) and Guangzhou (0.72), the ratio was much higher than that observed in this study [51]. These findings indicated that, compared to developed cities in China, the air quality in Sichuan Basin was more affected by coarse particles. Figure 5 showed the monthly average ratios of different cities in Sichuan Basin during the study period. The lowest average ratio was found to be 0.43 (in Guangyuan), while the highest average ratio (0.69) was observed in Zigong and Luzhou.

**Figure 5.** The monthly average PM2.5/PM10 ratio of each city.

In winter, all cities had the highest PM2.5/PM10 ratio, while in spring and summer, the ratio decreased rapidly. This was due to the high emissions of coarse particulate matter from sand and soil during the spring when it is very dry, windy, and dusty in Sichuan Basin [52]. Dust emitted from desert areas in Xinjiang (such as Taklimakan) may be transported towards the Qinghai-Tibet Plateau in the northwest of Sichuan Basin, thereby affecting the atmosphere and ecosystems of the basin area.

SO2 can be used to normalize PM2.5 to exclude the effects of coal combustion and meteorological conditions. It can be seen from Figure 6 that during the study period, among the cities in Sichuan Basin, the city with the highest PM2.5/SO2 was Bazhong (6.39), followed by Deyang (5.19) and Nanchong (5.18), which reflected the contribution of nonindustrial source to PM2.5. The average ratio in Sichuan Basin is 3.45, which was close to

the national average (2.92) in the previous study [53]. It is worth noting that in Panzhihua, Aba, Liangshan and other areas, the value of PM2.5/SO2 has always remained at a low level throughout the year, which may be because the industry in Sichuan Basin is mainly concentrated in the western region, and industrial sources contribute more to fine particles.

**Figure 6.** The monthly average PM2.5/SO2 ratio of each city.

PM2.5/SO2 also exhibited a U-shaped mode in most cities, which reflected that nonindustrial sources such as power and residential contributed more to PM2.5 in winter. Multi-resolution Emission Inventory for China (MEIC) was often used to estimate emissions from various sectors in China [54]. In order to determine the relationship between this dynamic change and the emission trends of key sectors involved in the air pollution process, we collected PM2.5 emission information from key sectors in the 2017 MEIC in-ventory of Sichuan Basin (Figure A1). Among them, non-industrial source emissions showed a similar U-shaped trend, which was consistent with the previous conclusions.

CO is an indicator of the primary combustion source. The secondary formation of fine particles in the basin can be studied by calculating the ratio of PM2.5 to CO [55]. From 2015 to 2020, the value of PM2.5/CO was higher in the southern areas such as Luzhou and Zigong, and the lowest in the western plateau areas such as Panzhihua (Figure 7). This indicated that the secondary sources in the southern cities of the basin had a higher contribution to the generation of fine particles.

Previous research reported that the sulfur dioxide emissions were much lower than the emissions of nitrogen oxides for motor vehicles in China, and the ratio of [SO2]/[NO2] in motor vehicles was usually between 0.0084 and 0.042. Both NOx and SO2 were emitted from stationary sources, but the emissions of SO2 was relatively more. The ratio of [SO2]/[NO2] in fixed sources was usually between 1.25 and 5 [56]. Therefore, the SO2/NO2 ratio was often used as an indicator of air pollution caused by stationary sources and mobile sources [57]. Figure 8 showed the monthly average ratio of SO2/NO2 in each city. In study area, Liangshan and Panzhihua had the highest SO2/NO2 ratios, indicating that the air pollution in these western plateau cities mainly came from local industrial sources and coal combustion. Bazhong was the lowest (0.21), followed by Chengdu (0.24) and Chongqing (0.26). These results confirmed that there was a strong correlation between air pollution and automobile exhaust emissions in Chengdu and Chongqing.

**Figure 7.** The monthly average PM2.5/CO ratio of each city.

**Figure 8.** The monthly average SO2/NO2 ratio of each city.

From 2015 to 2020, the ratio of PM2.5/SO2 in Sichuan Basin had shown a continuous upward trend, and the ratio of SO2/NOx had shown a continuous downward trend (Figure A2). It showed that the contribution of industrial sources to fine particulate matter con-tinued to decline. This was related to the pollutant emission reduction measures that the government had introduced. PM2.5/CO also showed a downward trend, reflecting the de-cline in the contribution of secondary sources to fine particulate matter, which was related to the decrease in the concentration of SO2 and NOx in the regional atmosphere. In 2015, the executive meeting of the State Council of China decided to implement ultra-low emis-sion and energy-saving retrofits for coal-fired power plants before 2020. Sichuan

Province had successively formulated the implementation rules for the Air Pollution Prevention and Control Action Plan from 2014 to 2017, and proposed a series of measures to improve the atmosphere environmental quality, including the elimination of coal-fired boilers be-low 10 tons per hour and the prohibition of new coal-fired boilers below 20 tons per hour. At the same time, we also noticed that during the study period, the ratio of PM2.5/PM10 showed an overall upward trend. This reflected the effectiveness of current dust removal measures to a certain extent, because the existing dust removal measures had far greater removal effects on coarse particles than fine particles.
