Please see the attached commentary.

Summary

The authors perform four sets of model simulations over China: a base case, a 50% NOx emission reduction, a 50% AVOC emission reduction, and a combined 50% NOx and 50% AVOC emission reduction.  The science presented is largely sound.  However, the paper currently reads as a lengthy report rather than a scientific manuscript.  There is extensive repetition of conclusions in different sections of the paper, and a lot of the more impactful conclusions get lost in the details.  I recommend that the authors restructure the paper around their main conclusions and science questions, rather than organizing by metric as is done currently.  I also suggest that many of the minor details presented in the main text be moved to the supplement so that only text in support of the main conclusions of the paper is presented in the main text.

Major comments

As it is currently written, a lot of the main conclusions and the primary storyline get lost in the details presented.  In addition, a lot of conclusions get repeated in different sections.  For example, a lot of the same conclusions are drawn when assessing ozone production regime (Section 3.1), odd oxygen production and destruction (Section 3.2.2), and ozone concentrations (Section 3.3.2).  Uniting these sections will reduce the overall manuscript length, help highlight whether these different lines of analysis lead to consistent conclusions, and help to bring the overall conclusions of the study to the forefront.

I wonder if the authors could comment on the relevance of broad NOx and AVOC reductions in China, in contrast to reductions that vary by sector or by region.  For example, are transportation sector reductions more/less likely than stationary emissions, and what might this imply for the chemistry discussed?  In addition, a lot of the analysis presented highlighted changes in radical cycling related to changes in OVOC emissions.  Do we expect AVOC emission reductions to be consistent across classes of VOCs, or could the effectiveness of emission reductions of OVOCs vs hydrocarbons differ? And, similarly, do we expect consistent emission reductions for VOCs with higher and lower HCHO yields?  How might this impact the conclusions presented here?

Minor comments

Line 110: typo (nitration vs titration)

Line 236: Zhang et al, 2009 get these numbers from Tonnesen and Dennis, 2000, so Tonnesen and Dennis, 2000 should be cited here as the original citation.