Dear Dr Youwen Sun,

I have reviewed the revised version of your manuscript. Over all, I find that you have considered the comments of the reviewers properly. I have some (minor) comments myself, that I list in the following. The language of the paper will need some copy-editing that will be done in any case. I have not checked the language carefully for this reason. Please consider my comments and resubmit a revised version of your manuscript (with marked changes), together with a list of changes and replies to my comments.

Kind regards,
Gabriele Stiller

Comments to the revised version (Page and line numbers refer to the version of the manuscript included in the replies to the reviewers):
Section 2.1.3. “ … HCN averaging kernels only have one maxima at 10 km. Both CO and HCN retrievals .” => “… HCN averaging kernels have only one maximum at 10 km. Both CO and HCN retrievals …”

Remove “We can see inTable 2, …” => “ The typical degrees of freedom (DOFS) obtained at Hefei (32°N) over the total atmosphere for CO and HCN are about 2.8 ± 0.3 (1σ) and 1.3 ± 0.2 (1σ), respectively (Table 2).”

Figs. 3,4, Table 4: The name of the station is Jungfraujoch, and it is separated as Jungfrau-joch

page 20: line 3: … and their relative contributions …

end of line 5: vice versa

page 22, starting in line 3:
“ … follows. First, the GEOS-Chem CO VMR profilesof alltracersin the grid box containing the Hefei (32°N)site were converted to partialcolumn profiles and linearly interpolatedand regridded onto the FTIR vertical retrieval grid. This was necessary in order toaccount for the differences in the verticallevelsof the model and the FTIR (Barret et al., 2003).Then, The GEOS-Chem CO partial column profiles are smoothed by the normalized FTIR CO total column averaging kernel followingRodgers and Connor (2003).”
In my opinion, the reversed order of steps would make more sense: First smooth the vertically well resolved model profiles with the retrieval averaging kernels; this provides you with the profile the instrument/retrieval scheme would have seen it if the model profile was the truth. Then calculate the partial columns in the same way as calculated from the observations.
I strongly suggest you do a test to quantify the difference between the two approaches.

Fig. 8/9/10: Indicate in the Fig. caption that these data belong to Hefei alone.

Fig. 11: what are the red marks in the maps?

Fig. 12: What is the unit of the color scale?

Page 24, second-but-last para: I do not quite understand the following sentence:
“Driven by the Asian monsoon anticyclone (AMA), the dominant circulation feature in the Indian–Asian upper troposphere–lower stratosphere region during the Asian monsoon, these emissions can be transported to eastern China which contributed to the tropospheric HCN peak in May. “
Do you indicate here that the May peak observed in Hefei is linked to the AMA? I don't believe so; the Asian monsoon just starts in May, and the most pronounced transport patterns occur in July/August.

page 26, line -8: abnormally => abnormal

++++End of editor comments++++