3.3.4. The Synoptic Causes of PM2.5 Distribution Variations

From previous analysis, we could conclude that the spatial disparity of PM2.5 concentration was gradually decreasing in SCB and presented prominent regional characteristics of declining in the western and southern basin, maintaining in other regions and even increasing in the northern and northeastern basin. In this section, possible synoptic causes of this phenomenon are analyzed. Because the PM2.5 concentrations in summer were relatively low (Figure 6) and the occurrence frequencies of identified summertime synoptic patterns varied slightly, except in Type 1, as shown in Table 2, the synoptic causes in summer were not analyzed.

In winter, the two synoptic patterns with the two largest CVs, Type 2 and Type 3, occurred in declining frequencies. As shown in Table 2, the occurrence days of Type 2 and Type 3 decreased from 17 days and 42 days to 9 days and 20 days, respectively, during 2016–2019. On the contrary, the synoptic patterns with the smallest CV (Type 6) occurred more frequently, from 5 days to 26 days. Along with the metric of CV, the average PM2.5 concentration in the northern basin (GY, BZ, NC and DZ) was largest in Type 6, reaching 79 μg/m3, and smallest in Type 3, only 58 μg/m3. Therefore, the growth in Type 6 and the reduction in Type 3 and Type 2 could be the reasons for the decrease in spatial disparity in SCB and the increase in RAs in the northern basin in winter.

In spring, the synoptic patterns with higher CVs, Type 5 and Type 1, were increasing and the synoptic patterns with lower CVs, Type 3 and Type 2, were decreasing. Type 1 occurred in 40 days in 2016 and 26 days in 2019. The occurrence days slightly declined from 18 days to 15 days, while Type 2 and Type 3 grew from 17 days and 9 days to 28 days and 16 days, respectively. Specifically, the RAs in the northern basin (NC, DZ and BZ) for Type 2 and Type 3 were relatively larger than those for Type 1 and Type 5. Hence, the reduction in Type 1 and Type 5 and the growth in Type 2 and Type 3 could reduce the spatial disparity of PM2.5 in SCB, and explain the increasing RAs in the northern basin in spring.

In autumn, the occurrence frequencies of all synoptic patterns in 2019 remained almost the same as those in 2016. This was consistent with the fact that the distribution in RAs varied slightly, as shown in Figure 4. In detail, the difference in RAs between the western and northern basin began narrowing from 2017. Correspondingly, Type 2, in which the RAs in the northwestern basin (CD, DY and MY) were higher than other cities, occurred in more days, and Type 3 and Type 4, in which RAs in the northwestern basin were lower, occurred in fewer days.

#### **4. Conclusions**

In this study, the spatial disparity of PM2.5 concentrations in SCB and its variation characteristics were explored, and the possible synoptic causes of these variations were analyzed. It was found that the spatial disparity of PM2.5 concentrations in SCB narrowed from 2016 to 2019. This tendency towards conformity of PM2.5 distribution was the result of a decreasing trend in cities with high concentrations, maintaining trend in cities with moderate concentrations and increasing trend in cities with low concentrations. Spatially, the main feature was that PM2.5 pollution was improved in the western and southern basin and deteriorated in the northern basin, especially in GY and BZ.

The regional characteristics of PM2.5 distribution variations could be partly interpreted by the occurrence frequencies of typical synoptic patterns, including weak low-pressure type (Type 1), weak high-pressure system and uniform pressure fields (Type 2), northern high-pressure type (Type 3), eastern weak high-pressure type (Type 4), southwest weak lowpressure type (Type 5), and weak low-pressure with northern high-pressure type (Type 6). Type 1, Type 2 and Type 6 were related to more polluted weather and Type 2, Type 3 and Type 5 were linked to cleaner days. The synoptic patterns influenced the PM2.5 distribution by modulating the diffusion conditions through PBLH and wind fields. The reduction in Type 2 and Type 3 (Type 1 and Type 5) and the growth of Type 6 (Type 2) led to a decrease in spatial disparity in winter (spring). Moreover, diffusion conditions (PBLH and wind) were the most important meteorological conditions affecting PM2.5 concentration and spatial distribution in SCB.

It was worth noting that the emission control measures were key factors that led to an improvement in air quality, although the impacts of synoptic patterns were manifested in this study. Specifically, the fact that PM2.5 concentration declined at a faster rate in more polluted cities might be the result of easier and more effective emission reduction in these areas. However, the regional maintaining, even rebounding, of PM2.5 concentration in the northern and northeastern basin could not be easily explained by emission variation alone, because a continuous reduction in emissions was expected, considering the implemented policies. Hence, the results of this study provide rational interpretation to this extraordinary trend on one hand. On the other hand, the fluctuation in PM2.5 concentration caused by synoptic circulation implied that the emissions in these areas might be close to the atmospheric capacity already. Implementing more precise and effective emission control measures is urgent to continuously improve the air quality.

**Author Contributions:** Conceptualization, G.S.; methodology, X.X.; investigation, X.X. and X.W.; writing—original draft preparation, X.X.; writing—review and editing, G.S. and F.Y.; supervision, F.Y. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Key R&D Program of China, grant number 2018YFC0214002 and 2018YFC0214001; the Key S&T Program of Sichuan Province, grant number 2018SZDZX0023; the National Natural Science Foundation of China, grant number 22076129; the Fundamental Research Funds for the Central Universities, grant number 41875162 and 22076129; the Young Talent Team Science and Technology Innovation Project of Sichuan Province, grant number 2020JDTD0005.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.
