Review of Effect of rainfall-induced diabatic heating over southern China on the formation of wintertime haze on the North China Plain (acp-2021-402):

An and the co-authors found when extreme rain fell over southern China, the probability of a haze event over the NCP during our research period of 1985−2015 was 59.09%. Further analysis revealed that the secondary circulation, associated with rainfall over southern China, in conjunction with Rossby wave trains along the waveguides may apt to the haze pollution over the NCP. This is important for understanding of haze pollutions over China. Although this manuscript fit into the scope of ACP, there are still several concerns must be addressed before publication.

Major comments:

1. The major concern is the small samples and half (13) and half (9) category. I suggest the authors to carefully illustrate the necessity of research and the robust of your revealed relationship. 50.9% is not a high probability.
2. Possibly, to discuss the differences between SR-NH and SR−noNH is a helpful way. That is, to answer why there is no haze pollution in North China form the perspective of physical mechanisms.
3. It is difficult to say the anomalous large-scale atmospheric circulations over North China were simulated by the extreme rainfall. Possibly, it is a dipole pattern of atmospheric anomalies, which simultaneously influenced the haze in North China and rainfall in South China. The diabatic heating might be an attendant phenomenon. Thus, more robust evidences are required to make your arguments stand.
4. Related to the above comment, I suggest the authors the re-plot the schematic diagram or provided more evidence to enhance the story line.
5. Figure 9, 11 and related texts: I did not find anticyclonic responses over North China.
6. Why show visibility in Figure 2, but PM₅ concentrations in Figure 15? Can you kindly show me both of the anomalies visibility and PM_2.5 concentrations in the reply letter (relative to climatology), associated with SR-NH, SR−noNH and the left samples?

Specific comments:

1. Abstract: Some abbreviations, that occurred only once (or less than 3 times), are not necessary in abstract. Too much abbreviations are not easy to read. Possibly, the authors could also check the main texts throughout.
2. Line 26: The ‘diabatic heating’ was defined as Q much later (Line 107).
3. Line 30: The range of NCP should be added in the text, and the others are similar.
4. Line 46-47: “Overall, the role of meteorology in the generation of haze is crucial but uncertain, and may be closely related to the regulation of the large-scale circulation”. This sentence must be rephrased, because the meanings are confused.
5. Line 51: The abbreviation of SSTAs may be not necessary.
6. Line 83: PM2.5 here has a format error.
7. Line 83-85: Because the climate scientist must carefully choose the data series, can you show the readers about the quality assessment of Yang’s PM₅ datasets for the period 1980–2019? In my opinion, excellent agreement with ground measurements during 2013–2019 did not illustrate good performance for the period 1980–2019.
8. Line 162: Add the full name of 'EU' if it is the first time it appears.
9. Line 251: “the appearance of haze over the NCP” may only be some of haze event, which related to the rainfall over southern China.
10. Figure 1: the color of the circular arrow was confusing.
11. Figure 6: ω>0 (Pa s^–1, shading) means descending motion, why is it an ascending motion here? Is it multiplied by −1?
12. Figure 7: The positive velocity potential may represent divergence.
13. Figure 10: “As Fig. 5,” may be “Fig. 6”?
14. How did you composite the maps associated with 13 SR-NH events. I did not find it in any of the Figure captions.

General comments:

The manuscript by An et al. investigates the effects of rainfall-related diabatic heating over southern China on the wintertime haze events over northern China plain (NCP). The authors suggest that the NCP haze event is modulated by the Rossby wave train emanating from the North Atlantic and the secondary circulation induced by the heavy rainfall over southern China. Specifically, the authors argued that the diabatic heating associated with the heavy rainfall over southern China leads to descending motions over NCP, which reinforce the anticyclonic anomaly produced by the Rossby wave train and thus favor the formation of haze events.

Overall, the flow of the paper and the figures used to support the arguments are cohesive. However, I am not fully convinced about the role of diabatic heating over southern China in the haze events over NCP. I recommend that the paper be considered for publication after addressing the major comments below.

Major comments:

1. The authors found that the NCP haze is modulated by the Rossby wave train emanating from the North Atlantic and the secondary circulation induced by the heavy rainfall over southern China. The diabatic heating associated with the heavy rainfall over southern China leads to descending motions over NCP, which reinforce the anticyclone resulting from the Rossby wave train. I am not fully convinced by this argument because (1) in observations, it is difficult to disentangle the effect of diabatic heating over southern China on the anticyclone from the Rossby wave train; (2) the LBM simulation doesn’t reproduce the observed anticyclonic anomaly over NCP (compare Fig. 9a with Fig. 4b; Fig. 11).
2. L131: Could the authors elaborate on the definition of extreme winter rainfall events? Is the definition based on monthly rainfall or daily rainfall averaged over southern China? Why there are 22 rainfall events during 1985-2015 and why the durations of each individual event are different (Tables 1 and 2)?
3. L137: When heavy rainfall fell over southern China, the probability for haze to occur over NCP is ~59% (13 out of 22 extreme rainfall events). Although the authors have compared the atmospheric circulations between SR-NH (13 events) and SR-noNH (9 events) events, I feel that the upper-tropospheric Rossby wave trains look very similar (c.f., Fig. 4a and Fig. 14a). Instead, the significance of the Rossby wave train reduces in the SR-noNH events, which might suggest more variabilities in the wave train. Could the authors explicitly show the differences between Fig. 4 and 14?
4. While the authors have compared the atmospheric circulations during SR-NH events and SR-noNH events, how many NH events occur without SR? Is the atmospheric circulation during NH-noSR events also controlled by Rossby wave train similar to the one shown in Fig, 4b? This might help illustrate the importance of SR rainfall in observations.

Technical corrections:

1. L51-56: I wouldn’t call the Eurasian snow cover, ENSO, and Arctic sea ice changes as atmospheric conditions
2. L108: add a space between QG and w
3. L222: “The NSC simulated by the LBM bears a striking resemblance to the observed spatial pattern of the NSC (Figs. 5 and 10)”. Should “Figs. 5” be “Fig. 6”? Also, for the caption of Fig. 10, should be “as Fig. 6” instead of “as Fig. 5”.
4. L222: Given the substantial differences between Figs. 6 and 10, I personally wouldn’t say the NSC simulated by LBM bears a striking resemblance to observation.
5. L205: Fig. 9a should be Fig. 8a
6. L208: Fig. 9b should be Fig. 8b
7. L216: Fig. 10a, may be Fig. 9a
8. L218: Fig. 8b, may be Fig. 9b
9. Fig. 12: the first and second rows show the diabatic heating and dynamic forcing terms, of which the sign is opposite to the omega. For ease of comparison with the omega shown in the bottom row, I suggest reversing the sign of diabatic heating and dry forcing terms.
10. Is the omega shown in Fig. 6 also multiplied by -20 as that in Fig. 10? If not, the magnitude of omega is significantly different in the LBM model and observation.
11. L47-49: “Large-scale circulation, and the related external forces derived via exciting the teleconnection pattern, regulate meteorological conditions, reduce dispersion, and facilitate the accumulation of haze pollutants (Zhang et al., 2020).” Please consider rephrasing this sentence.