This manuscript can be published after the authors make the 2 following changes to their revised version.
In response to reviewer's 3 comment:

The manuscript says that an evaluation of the WRF meteorology will be
the subject of a separate paper. I think that there should be at least some evaluation of
WRF in the current paper. This could be simply be some summary statistics of model
Performance.

The authors answered:
“WRF performance for the simulation of hourly surface temperature (T), relative humid-
ity, wind speed and direction was evaluated through comparison with observations from
NOAA’s National Climatic Data Center (NCDC) Integrated Surface Data (ISD with lite-
format) which provides hourly (or with 3-hour interval) meteorological observations over
a long historical period across the globe. The mean bias of T, wind-speed and direction
over the simulation domain is -0.4 K, 0.4 m s-1 and -3 degree respectively, within the
benchmark range suggested by Emery et al. (2001) for retrospective regional-scale
model applications which is ≤± 0.5 K, ≤± 0.5 m s-1 and ≤± 10 degree respectively.”

This answer is very factual and too short. What does this evaluation imply for the results presented in the paper? Does such comparison insure that the meteorological fields used by the model have no bias? Please give 2 or 3 sentences that give credence in your comparison.

In response to the following comment of reviewer 3:
p. 25459, lines 16-17: lightning NOx emissions are said to be from Price
et al., 1997. This paper indicates the total global emission is 12.2 Tg/yr. This amount
is well above the most well-accepted values of 2 - 8 Tg/yr (Schumann and Huntrieser
et al, 2007, ACP). Please provide some indication of what the impact of this likely too
large emission value is on NOx, O3, and nitrate.
[Response]: “We have included a brief discussion about this bias in the revised
manuscript (P25 L20-P26 L3), as below: “However, the model estimates still suffer from
uncertainties in emissions (in regards to temporal variation and speciation), coarse
spatial resolution and subsequent impacts on representation of non-linear atmospheric
chemistry. The lightening NOx emissions used in this study (Price et al, 1997) are
likely overestimated by 0.5 to 5 times times compared to more recent study (Schumann and Huntrieser et al., 2007) and may contribute to some extent to the overestimation of
NOx, O3 and nitrate concentrations.”

Please change the sentence:
“The lightening NOx emissions used in this study (Price et al, 1997) are
likely overestimated by 0.5 to 5 times times compared to more recent study (Schumann and Huntrieser et al., 2007) and may contribute to some extent to the overestimation of
NOx, O3 and nitrate concentrations.”
To
“The lightening NOx emissions used in this study (Price et al, 1997) are
likely overestimated compared to more recent study (Schumann and Huntrieser et al., 2007) and may contribute to some extent to the overestimation of
NOx, O3 and nitrate concentrations.”
It is far fetched to give this range of 0.5 to 5 times, I do not understand where it comes from.

After these 2 modifications are included, this paper can be published in ACP.

Best regards,

Yves Balkanski