I think that the authors have addressed the two sets of referee comments that were in open discussion in a sufficient detail.

I did, however, receive a third set of referee comments which has a few more issues not covered by the two earlier reviews. I have inserted them below this message. Please have a look at these comments and make the necessary revisions to the text (note that the pages and lines below refer to the ACPD version of this paper as the referee had no access to the revised paper version). Please also remove any comments written in Chinese that were present in your previous author response letter.

I will personally check out your re-revised manuscript (the comments are minor, so I see no reason to consult the referee in this respect).

Comments:

1. Page 2, line 5: The authors cited previous study (Guo et al., 2014): “… and contribute to the particle mass enhancement even in polluted regions.” Is this conclusion/statement correct? In my opinion, although many NPF events were regularly followed by an increasing of particle mass, this only indicates a high abundance of condensable vapors in the atmosphere. Actually, the newly formed particles can only grow up to 60~100 nm. The contribution by particles smaller than 100 nm to aerosol mass is not significant. Maybe the large number of grown particles can, through the coagulation process, grow into a larger size range, where they could contribute more efficiently to the particle mass concentration or particle extinction. But I think it is not proper, or at least it may result in certain misunderstanding, to directly connect the NPF (or continuous growth of nucleated particles) with haze formation. Please consider this issue.

2. Page 2, lines 6-7: What is “possible particle source”? Do you mean where it come from? Please explain.

3. Page 2, lines 20-21: Any explanation? Is it true for the NPF event in polluted area?

4. Page 2, lines 25-27: Please cite the recent two overview papers: Kulmala et al., (2016) and Wang et al., (2017).

5. Page 4, line 15: 3080 is the type of the classifier, not SMPS. Should be 3936, I guess. Please check.

6. Page 5, line 14: How you consider the size dependence of GR. What is the uncertainty?

7. For the section 2.3 Nanomap, I would suggest that several issues should be explained more clearly:
1. Based on the Kristensson et al., 2014, it seems that the maximum distance is 500 km in this method. I assume this value is determined by the air mass trajectory calculated for Finland case. So could we directly use this value for other studies? Have you compared the air mass trajectory patterns between Finland and China? Also, as one reviewer mentioned, 500 km in China contains a lot of megacities with heavily emissions, so the first assumption in Nanomap should be reconsidered.
2. It seems that this method is also closely related to the calculation of air mass trajectory, so why you use 72 h, have you tried others (48- and 96-h), any differences? In Section 3.2, you want to retrieve where does the NPF take place, but I think it also depends on that how you calculate the air mass trajectory.
8. Section 3.3: How you calculate the occurrence probability? Please classify.

Kulmala, M., Petäjä, T., Kerminen, V.-M., Kujansuu, J., Ruuskanen, T., Ding, A., Nie, W., Hu, M., Wang, Z., Wu, Z., Wang, L., and Worsnop, D. R.: On secondary new particle formation in China, Front. Environ. Sci. Eng., 10, 1-10, 10.1007/s11783-016-0850-1, 2016.

Wang, Z., Wu, Z., Yue, D., Shang, D., Guo, S., Sun, J., Ding, A., Wang, L., Jiang, J., Guo, H., Gao, J., Cheung, H. C., Morawska, L., Keywood, M., and Hu, M.: New particle formation in China: Current knowledge and further directions, Sci. Total Environ., 577, 258-266, http://dx.doi.org/10.1016/j.scitotenv.2016.10.177, 2017.

The re-revised paper is acceptable for publication in its present form.

Editor