This manuscript studies a unique mechanism that the typhoons approaching to Taiwan from the east and the presence of the central mountain range result in a weak counter flow system behind the mountain range (lee vortex) and enhanced air pollution there, particularly at around the mountain slope rather than the coastal regions with emissions, during the summertime. The mechanism is clearly presented with both observations and model simulations (WRF-Chem). As the ACP journal scope is focused on studies with general implications for atmospheric science rather than investigations that are primarily of local or technical interest, my primary concern is about how the interesting results here are to be generalized to other locations of the world. Such a statement is better included. Second, there are several parts where additional justification and explanation are necessary. For example, description on observations is weak, for both TEPA monitoring and campaign observations (for example, in-situ measurement techniques, sampling flow rates, and ion analysis). Also, the roles of biogenic VOCs emitted from the forests near the mountain regions should be discussed. Overall, I would recommend publication but only after major revisions on these issues.

Specific comments:

1, line 113-114: Specify measurement techniques for PM2.5 and O3, whose data are heavily used.

2. Lines 135-136: Were high-volume air samplers used for sampling and subsequent chemical analysis? What was the flow rate? Measurement methods for sulfate, nitrate, ammonium, and EC/OC should be described. Maybe the sampling periods are 11 h (not 12 h)?
3. What is this number (46734) for?
4. Line 217: For the buildup of OC, what are the roles of biogenic VOCs emitted from the forests near the mountain regions? I thought model simulations could tell the importance.
5. Lines 220-222: Chemical aging was not important for the buildup of sulfate?
6. Line 229: Which species are emitted from agriculture?
7. Line 267. Bifurcated air flows meeting in the western Taiwan may form a kind of convergence. Why does the convergence result in subsidence rather than updraft?
8. Line 285: Do the authors mean change over time?
9. Line 306-307: Titration of O3 by freshly emitted NO might be another reason for the low ozone near the sources?
10. Line 335-336: Are the correlation coefficients as high as 0.72 and 0.81 from hourly values?
11. Lines 398-400: Here mentioning a branch flowing over the ocean to the south? Please clarify.
12. Table 2: July 15, 2018 is Sunday. Any emission change on this day?
13. Figure 4. Add explanation of the colored lines in the caption or legend.
14. Figure 5a. Land-sea breeze is well resolved with the model with a resolution of 3 km. This should be better highlighted, with some comparison to coarser resolution model cases. Are there any reasons why the simulated wind speeds are often higher? And any potential influence of the stronger wind on the model results of O3 and PM2.5?
15. Figure 16. Mention unit of the PM2.5 mass concentrations in the caption.
16. Figures 8 and 9: Which altitude are the shown streamlines for?

Major Comments:

This manuscript investigated air quality deterioration episode associated with typhoon over the complex topographic environment in central Taiwan. The schematic summary in Figure 12 is interesting and clearly presented the diurnal process during the episode. This manuscript is well organized and is helpful to understand the interactions between ambient flow, sea-land breeze, boundary layer development with topographic in Taiwan and other regions.  I recommend it to be accepted after minor revision. The following comments are refered as follows,

1. Mechanism: It seems the easterly flows interacted with the central mountain range (CMR) is an important process and easily to deterate air quality over western Taiwan. Although typhoon circulations provided a stronger easterly flow, I am just wondering if a typhoon is not a necessary condition. In other words, perhaps other weather conditions will also form such an air quality deterioration, if so, you should also mention it in your article?
2. The authors presented the sources of meteorological initial and boundary conditions for WRF-chem model and simulated well for the pollutants’ variations in Fig. 5. Which anthropogenic emission inventory in Taiwan do the authors? The information on emission inventories used in WRF-Chem simulation is not introduced.
3. The authors presented the aerosol composition analysis over western Taiwan and showed the differences in coastal, urban, and mountain sites. I am curious about what kind of instruments (models, ..) you used?
4. L135-137 The sampling period of each sample was 12 h; daytime samples were collected from 08:00 to 19:00 LST, while nighttime sampling was conducted from 20:00 LST to 07:00 LST. Please check the sampling period since nighttime sampling is only 11 h.
5. Line 181: Figure 2a-c, the monthly mean results for PM₁₀, PM_2.5 and daytime ozone. What’s the analysis period? It seems to me has a typo error for the starting in 1994, Please double check it.
6. Line 294-295, 18:00 LT……..,; Line 487 “ daytime ozone (08:00-17:00 LT);  Please consistently present the abbreviation of local standard time (LST) or LT,
7. Line 318, Is it “Maoli county”,  or “ Miaoli county” ? In Figure 2b, it is “Miaoli.” Please double check it.
8. Summary: It is wordy, loss of focus, and needs to be more concise.
9. References: Line 83:  Lin Y. L. (1993) was not shown in the reference list. ; Line 48 and Line 601: it should be Huang et al. (2021). Line 635:” Atmos. Environ., 41, 3684–3701, https://doi.org/10.1016/j.atmosenv.2006.12.050, 2007b” , it should be 2007.