This paper attempts to understand the relationship between AOD and PM2.5. However, after reading through, I feel that the paper is more of a GEOS-Chem validation and uncertainty analysis work, rather than offering physical explanation of the AOD-PM2.5 relationship.  Specifically, could the authors clarify, perhaps with additional analysis, how different factors, such as PBL height, RH, organic matter fraction, etc, contribute to the uncertainty in AOD-PM2.5 relationship? How does the role of each factor vary with region (e.g., Korea vs. China)? The only clear conclusion is that AOD and PM2.5 have reversed seasonality because of seasonally varying PBL height, but this is already well known.

The manuscript investigated the physical relationships between AOD and PM2.5 over East Asia by using the model simulation and comprehensive observation. The results indicate that the aerosols over this region are largely contributed by the sulfate-nitrate-ammonium and organic aerosols within the PBL. Meanwhile, the dust in the free troposphere also has an important contribution to column AOD. The seasonality of AOD and PM2.5 has been specifically discussed. In general, this paper is well-written with a good logical connection.  Thus, I recommend the manuscript for publication in Atmospheric Chemistry and Physics, after addressing the following comments.

Specific Comments:

1. The current introduction section may be insufficient to demonstrate the significance of this paper. The authors need to clearly explain the limitation of previous studies and the advantage of this study.
2. The analyses of this study are closely associated with the model simulation of GEOS-Chem, while the title only mentioned the observations. There are some disconnections between the title and the main text.
3. Line 203, Page 8. The PBL varies significantly during the different periods. It is risky to define the 0-2 km as the PBL. The authors should give more justifications for this definition.
4. The seasonality of AOD and PM2.5 and its association with PBLH have been discussed previously (e.g., Guo et al., 2017; Su et al., 2018). I suggest the authors acknowledge these works.

References:

Su, T., Li, Z. and Kahn, R., 2018. Relationships between the planetary boundary layer height and surface pollutants derived from lidar observations over China: regional pattern and influencing factors. Atmospheric Chemistry and Physics, 18(21), pp.15921-15935.

Guo, J., Xia, F., Zhang, Y., Liu, H., Li, J., Lou, M., He, J., Yan, Y., Wang, F., Min, M. and Zhai, P., 2017. Impact of diurnal variability and meteorological factors on the PM2. 5-AOD relationship: Implications for PM2. 5 remote sensing. Environmental Pollution, 221, pp.94-104.

The relationships between AOD and PM in East Asia are discussed by using ground-based and aircraft observations, but the whole study focus on direct validation and comparison, lacking in-depth analysis and literature support. In addition, there may be some problems in the use of satellite data. I suggest that the authors add more analysis to enrich the study.

Abstract: Line 35: Himawari-8/AHI provides AOD products at 500 nm, not 550 nm.

Introduction

It is too short and the authors are suggested to summarize previous studies on investigating the relationships between PM2.5 and AOD, especially those focusing on Asia.

In addition, studies on PM estimation from satellite AOD products need to be summarized, especially those using geostationary satellites.

Finally, the author should highlight the innovation and difference between the current study and previous related studies, and discuss the importance of understanding the relationships between PM and AOD in these studies.

Lines 85-86: Ångström Exponents at 500 nm? AE refers to a wavelength range. Reference is needed here.

Lines 107-110: Himawari-8/AHI: Which version do you use? Reference is needed. Again, Himawari-8/AHI provides AOD products at 500 nm. I am not sure about GOCI (should be 550 nm). Are they the same? If not, does the wavelength difference be taken into account in the data fusion?

Lines 288-290: What are the potential reasons? Is it the aerosol algorithm or the difference caused by sample matching at different wavelengths?

Lines 295-296: What are the potential reasons?

Lines 297-312: Is there any relevant published literature to support the author's explanations of reasons for these differences between GEOS-Chem and satellites observations?

I also suggest adding some scatter plots to validate and compare the satellite-based and modeled AODs, PM2.5s, and other parameters if possible, so that readers can see their differences more clearly.

Figure 6: I suggest adding some satellite PM2.5 estimated results to see the difference with model simulations since there are many available PM products, especially in China.

Last, the authors should consider the impact of other factors, especially BLH, meteorological conditions, and topography, on surface PM, and to see how much impact can they have on the differences between satellite and model results.