Firstly, I would like to apologise for taking my time in making my decision. I was out of the office since December 11.

I think you have done a good job in responding to the reviews. Performing the extra sensitivity run with 4x traffic NOx is especially welcome. I think you have integrated this well into the manuscript, with the exception of how this is discussed in the abstract and conclusions.

In lines 25-26, you state the the similarity of the 2NOx and the 4traf_NOx runs means that traffic NOx emissions are likely to be underestimated. But in fact I don't see anything in your results that really supports this. All you can say based on your results is that both of these sensitivity runs are more consistent with the observations than your base run.

The missing piece of information here is the existing body of work pointing to the likelihood of the traffic NOx emissions being underestimated by up to a factor of 4. So in fact it is your 4traf_NOx scenario being more consistent with the previous work which should be leading to this conclusion.

Obviously with hindsight it would have been better to simply use 4traf_NOx consistently, rather than doubling total NOx emissions at all. But rather than asking you to rewrite your manuscript to remove all references to the 2NOx runs (which would also involve you having to re-do the combined NOx and VOC increases), I think it will suffice here for you to come up with a more careful formulation in your abstract (as well as in your conclusions, page 18, lines 27-29). The point you should be conveying more clearly to the reader is that your 4traf_NOx scenario itself is more consistent with the previous work than the 2NOx scenario. In your results section (page 13, lines 1-4) you already do a pretty good job of this.