|The authors have done considerable work to incorporate the comments of the referees, leading to an improved version of the manuscript.|
However, there are still major and minor points that need to be addressed before the paper can be published (see list below).
Additional points mentioned below in comparison to the first referee comment result from the changes of the revised manuscript in comparison to the ACPD version.
Comments to the authors:
- You write that lightning NOx is included in the simulations, which answers one major point raised by the referees. However, I do not agree with the statement (p.7 l.153-155) that: "Therefore NOx anomalies obtained from the ... do not have an impact of lightning or soil emissions as they are same in all the simulations.". This statement might be true for the soil emissions, if they are simply fed into the model. But for lightning NOx, which I assume is parametrized as in Fadnavis et al 2015, this is not true for a fully-coupled model. A changed circulation (initially due to different anthropogenic NOx emissions) has an effect on convection (raised already in my first comment: General Comments first point) and hence on lightning NOx production. An example of this can be probably seen in Fig.2 d) at the Equator and south of it at ~500-200hPa (blue blob), which is likely connected to less (convective precipitation) in that region (Fig. 9 d). This affects the discussion of the results at various parts in the manuscript.
- With respect to this. Please mention the parametrization scheme of lightning NOx you are using as it is a key aspect of your simulation. Also the content of Table 1 could be stated in the text as it contains mostly repetitive statements. Instead - in my view - a table showing the VOC, CO, NOx emission totals during JJAS (maybe split by sector) over India and China for the CTRL run would be appropriate.
- Sometimes cross TP transport is indicated with arrows in the Figures, while the differences are really small and might not be important/significant (especially for the Ind38 simulation).
- I was asking for significance of the results in the first comment. Unfortunately, I don't understand the contour lines in Figs. 3 and S6, which are supposed to relate to the 95% confidence levels (using black for the tropopause, as well, is not optimal). I was expecting some transparent shading out of not significant results, or are these contours to various levels. Also have you checked the significance with respect to the other figures? This relates also to the question about cross TP transport.
- On p. 7 l156-159 you mention 4 additional "lightning-off" simulations. However, you only discuss the CTRL "lightning-off" simulation in the whole manuscript. Either only introduce this simulation or make a statement about the other simulations in the manuscript.
- p. 15 l. 356-357: Please include a reference to the Figures that support your statement.
- p. 1 l.14: add "emissions" after "(NOx)"
- p 5. l106-108: You state that MLS data was interpolated to potential temperature levels but you are using MLS data on pressure levels (Fig. 1)?
- Sorry about the mistake regarding the MLS kernels, I was thinking about Fig 1b) , i.e. the model data. However, I thought that the MLS AKs are defined for the MLS levels, which do not include 90hPa (https://mls.jpl.nasa.gov/data/ak/MLS-Aura_L2AK-O3-LAT0N_v04-2x_0000d000.txt).
- p. 9 l. 210: "demonstrated" rather than "quantified"?
- p. 10 l 231-235: You attribute the enhanced values of NOx in Fig. 2f to the southern flank of the Himalaya, however, the major sources over China are located east of the Himalaya (Fig. S1)...
- In two Figures the descripition within the graphic and the description in the caption do not match (Figs. 7 and 9).
- Please add the Tibetan Plateau in (one of) the map plots to make the discussion easier to follow. Also, including the outline of the AC in lat/lon vs. height plots, e.g. via winds or just a box, would be helpful to follow some of the statements (e.g. in Sect. 3.3).