This manuscript studied the relationship between surface PM2.5 and AOD, and also the influence of meteorology and topography on the relationship in high spatial resolution. The topic is critical the improved knowledge can contribute to the research field, enhancing air pollution research using satellite-retrieval technology. I suggest to accept this work for publication after revisions.

Title: AOD should be replaced by aerosol optical depth.

Line 159: the word “vs.” should be revised.

Table 1: Why the result is missing in Northern Tianshan in winter? Is it because cloudy weather over there during the season?

Figure 1: the texts in the legends are small. They should be enlarged.

Line 623: The figure should be the spatial distribution of annual mean ground-level PM2.5 and satellite AOD in China in 2019. It is better to add “annual mean” to make the meaning clearer.

Figure 2: the texts of the legend are too small.

Line 174: What is the reason about the U-shape trend? Why is the PM2.5 concentration lower during Jun-Aug?

Line 188: the sum of the stations (559+519+50) is not equal to 1494. Please check the number of the stations.

Line 171: It should be explicitly mentioned Figure S3 is shown in supplementary information.

Figure S9 in supplementary information: the word of “elevatior” should be “elevation”

The caption of Figure S9: the second sentence should be revised: Note that the asterisk after r values represent statistically insignificance (p>0.05).

Satellite-based retrieval of PM2.5 shows great advantage over traditional ground-level observations, which largely relies on the well-established statistical relationship between AOD and PM2.5. This manuscript by He et al. investigated the relationship between PM2.5 and AOD in China using the high-resolution AOD products from MAIAC, and the major influencing factors, including relative humidity, PBLH, and terrain, have been discussed. Overall, this work is well organized, and the analysis methods are scientifically sound. I am almost in a position to accept it for publication at ACP after the authors adequately address the following concerns, even though most of them are minor.

Specific comments:

L11: “to monitoring surface PM2.5 concentration”can be revised to “surface-based PM2.5 concentration observations”

L13: “ground” -> “ground-level ”

L29: the article is missing before “Findings”

L38-43: This is a pretty long sentence and can be shortened or rephrased.

L47: “heavily” -> “heavy”

L52: “tuning” needs to be corrected.

L62-63: “further exploring the relationship between near-surface PM2.5 and MAIAC AOD for China during a long-term period is necessary.” can be rephrased as “further exploring the relationship seems critically imperative between near-surface PM2.5 and MAIAC AOD over China based on much finer AOD products.”

L65: “suggested” -> “recognized”; also, “different definitions” can be revised to “the definition difference”

L70: “The vertical structure” of aerosols? Please clarify it.

L80: The reference of Guo et al. 2017 is wrongly cited here, since the effect of aerosol diurnal variability on precipitation did not involve at all. Instead, the author may refer to Kim et al. 2010 (doi: 10.1007/s00382-010-0750-1); Guo et al., 2016 (doi:10.1002/2015JD023257) and Lee et al. 2016 (doi: 10.1002/2015JD024362), and Zheng et al. 2020 (doi: 10.1088/1748-9326/ab99fc), among others.

L303-309: Except for the different number of available observations for AOD and PM2.5, the main causes for the mismatch between AOD and PM2.5 lie at the fundamental discrepancy in physical concept. AOD is a unitless variable that denotes the total extinction induced by aerosol in the whole atmospheric column, whereas PM2.5 represents the aerosol concentration measured at the ground level.

L378-379: The diurnal variability of PBLH needs to be specified from a climatological perspective, and the authors are suggested to refer to Liu and Liang 2010 (doi:10.1175/2010JCLI3552.1); Zhang et al. 2018 (doi: 10.1175/JCLI-D-17-0231.1)

L395: “by PBLH correction” -> “corrected for PBLH”

L399: “of ” is redundant.

L421: “Corresponding to” -> “Generally consistent with”

L455: “As far as the meteorology” -> “As far as the confounding impact of meteorology”

L460: “despite of” -> “despite”

L465: “The vertical structure” of aerosols? Please clarify it.

L465-466: “Even though the individual effect…., the synthetic impact of these factors has been recognized to be core…” seems problematic in terms of logic. The joint effect of these factors could be revealed using the multi-regression analysis in the future. This point is suggested to be taken into account when revising this sentence.   

The variables appeared in Eqs. 1-2 should be clarified, including the units.

The quality of Figure 6 is a little low for its blurred x-axis and y-axis titles, and I strongly recommend the authors redraw it.  

The authors investigated the spatiotemporal relationship between ground-level PM2.5 measurements and satellite derived AOD data over China for the year of 2019. Compared with previous research with similar topic, this study used high-spatial-resolution AOD data with combination of AERONET data which provide high-temporal-resolution aerosol data, which is important for understanding the relationship of PM2.5-AOD and useful for ground-level PM2.5 estimation, especially when deriving PM2.5 from satellite remote sensing data is becomes more popular. The findings of this study are worth of publication in the journal after minor revision as followings:

1. My major concern is the way that you matched PM2.5 and AOD in space, that is， what is the radius of buffer zone around the site for AOD average? please clarify it in detail.
2. Line 39-44: A long sentence. Please rewrite it.
3. Line 108: It seems no Fig. A1 in supporting document. May be Fig. S1?
4. Line 125-128: Please clarify which level of AERONET data were used in this study.
5. 1 and 2: Please explicitly describe the two equations. What is the meaning for each variable?
6. Line 344: “pollutant acumination”? May be “accumulation”.
7. Table 1: Why there is no value for NTS in winter?
8. Figure 2: the text of “hour” legend is a little small.
9. In figure 5, suggest that the count is replaced by frequency, which can be easily compared among different regions, due to their different samples.
10. Figure 6: The text for latitude and longitude is a little small. Please replot it.
11. Figure S9: it should be “elevation”?