General comment

This papers aims to investigate the effect of vertical distribution of ship plumes on modelling of ground level concentrations at small spatial scale, i.e. a few kilometres from the source. The topic is of interest even because the impact of ship emissions at berth in port cities is actually concentrated at small distance from the port. Considering that specific parameterizations to be included in models are provided, I believe that it is a topic suitable for the Journal. It must be said that in a previous paper of the same authors, a very similar input dataset was used and the same modelled ship, I believe that there are elements of novelty in this analysis, however authors are invited to better clarify this aspect. In addition, some aspects of discussion and interpretation is not very clear, see my specific comments, and a revision step will furnish benefits to this paper.

Specific comments

Authors should explain better what are the elements of novelty compared to the previous paper because a similar modelling approach and the same input ranges are actually used.

Lines 28-32. Here it would be useful to put a priority in nitrogen oxides that could be significantly impacted by shipping leading to overcomes of the legislation threshold in specific areas of port cities. SOx is certainly a ship-related pollutant but less relevant for maintaining local air quality standards. However, it could lead to formation of secondary aerosol (sulphate) that could have non negligible impacts at local and regional scales.

In the introduction, chain citations of a lot of papers together such as in lines 50-54 is not a good practice because it does not give to the reader any real clue of why these papers have been cited.

In the introduction is given a brief overview of the impact of ships to local air quality. I would like to suggest to consider the recent review Contini and Merico (Atmosphere 2021, 12, 92) in which an global overview of these impacts is provided.

Lines 89-93. I believe that the over prediction effect is not only due to plume rise, rather, the uniform distribution of emissions on such a large grid could influence results.

Lines 264-265. This is not clear. Probably you mean that you modelled gases as passive tracers because you analyse small spatial and temporal scales so that transformations are limited?

I suggest to include a discussion relative to the applicability to real cases. All the paper is based on modelling of a single ship and the influence of considering (or not considering) the plume rise is discussed for this specific cases. However, in real cases, there would be a mixture of ships in which several parameters relevant to plume rise (like exhaust temperature and momentum flux) are not known. The total emission itself is generally rather uncertain as you also mentioned. So my question is if this approach to take into account vertical distribution would have a relevant practical implication in real cases or the uncertainties are large enough that this is a second order effect?

Title of section 5.5. Better “Comparison of the effects of different input variables”.

Line 444. There is a non-necessary parenthesis.

The paper is a prosecution  of a previous work (Badeke et al. 2021) aimed to  investigate the vertical distribution of ship emissions. In this paper the authors study the influence of the vertical distribution of emissions on the ground-level pollutant concentrations at a few kilometres distance from the mooring point. The topic is of interest for an accurate assessment of the impact of ship emissions at berth in port cities. The paper deals with the development of an atmospheric dispersion model chain, for this reason it fits with the arguments treated in GMD journal.

I suggest the publication of the paper after the authors will reply to the following comments.

Methodology 116-120. The authors would better explain which obstacles have considered using MITRASv2.0. I guess they considered only the shape of the ship. What happens if ships with different shape are considered? Could the eventual presence of  buildings near to the dock influence the presented results?

Line 145 “No chemical reactions occur in the simulations.”  This assumption can be considered valid also in case of very low wind speed?

Line 185 “No clear correlation was found for µ against the atmospheric stability, but a negative dependency has been found for stability against σ”. The absence of correlation of µ against the atmospheric stability is a logical consequence of the range of vertical height considered. If it contains all the plume also in case of high convective conditions  then the result is logical. If the authors agree with this interpretation they could include it in the text. Otherwise give their interpretation.

Line 195 “Especially in cases of strong winds and stable atmospheric conditions, the simple Gaussian distribution delivers good results.” But cases of strong winds are of  less interest for the impact on air quality. This reduces significantly the value of the simple Gaussian distribution.  It would be evidenced.

Line 315. It is not clear how initial vertical concentration profiles were converted into vertical emission profiles in EPISODE-CityChem. Could the author explain this point?

It would be useful to introduce the definition of upper plume boundary height.

I did not find in the paper the exact definition of  initial concentration profiles. It is necessary to introduce this definition  at the first time the authors discuss about “initial concentration profile”

The profile for the single cell emission model is reported in the caption of Fig. 7 but is not present in the diagram.

A final observation. The authors evaluated, among others, the impact of surface roughness on pollutant ground-level concentration. Since the interest is focused on the local scale (few kilometres from the source) do they consider important or necessary a precise description of the buildings and the streets instead of the use of a simple parameter like the surface roughness?