The manuscript presents a modeling investigation of SO₂ conversion to sulfate in tropospheric volcanic plumes. As a case study, the authors simulate the explosive eruption of Mount Etna on Apr. 12, 2012, using the CHIMERE model and analyze the sensitivity of SO₂ oxidation to water and transition metal contents, plume diffusion and plume altitude.

The results suggest that the dominant oxidation pathway is gas-phase oxidation by OH (about 70 %), followed by transition-metal catalyzed oxidation in aqueous phase (about 25%). The contribution of aqueous phase pathway increases with H₂O/SO₂ ratio and shows sensitivity to plume height as it modulates the liquid water content. The authors argue that H₂O₂ is rapidly diminished in volcanic plumes and cannot contribute to SO₂ oxidation substantially.

This is a timely study with potential relevance for the atmospheric chemistry and volcanology communities. The manuscript is well structured and written. Nevertheless, there are some critical points:

• It seems that there is a significant flaw in the model as it allows negative mass for OH radicals. If the model allows this (which is fundamentally wrong), then it is no surprise that OH becomes the dominant pathway as it never diminishes. The authors are all well-known experts and should definitely check this in the experiments and results. 
• Methods lack some important information about the modelling setup, which complicates the reading and understanding of the results. Are you using online-coupled WRF-CHIMERE? Besides, there is no information about AOD calculation. I assume the plotted sulfate mass is calculated in the gas phase. But how does it come to aerosol phase? Are the aerosol microphysical processes in place? Or there is a simple assumption based on some particle size? Another principal question would be the role of AOD for model validation. When there are no observations to compare with or not radiative transfer calculations, why do we need AOD in the output? Why AOD at 200 nm? 
• Figures need substantial improvement. For e.g. captions of figures 3 and 4 are wrong. Figure 2 is impossible to follow. Lines and colors need to be more visibal and distinguishable. 
• Design of experiments is fine, but the paper would be more beneficial if authors would present the chemistry parts only and leave out the sensitivity to the vertical transport scheme. The transport has to be validated first (based on available observations) 
• One critical aspect is the sensitivity to pH. The aqueous pathways strongly depend on pH. The authors should vary the pH in a range like 1-6 and then calculate the rate of sulfate formation for different pathways/experiments. I would like to see the plots for pH and SO₂.

Overall, I find the topic very important but am disappointed by the low quality of the presentation. Especially the negative OH mass shown in different figures challenges the scientific soundness of the study and the key outcome of the paper with respect to oxidation pathways. Therefore, I reject the manuscript. I hope this gives authors enough time to rigorously go through the methods, assumptions and experiments, critically validate their results and come up with a much better presentation quality. 

Other points:

P2L24: “volcanic particle size distribution is evolving to a coarser distribution as time goes by” not always. Volcanic PSD can move to finer modes if ash is involved. Upon ash removal from the atmosphere, the volcanic PSD moves from coarse to fine but increases later due to new particle formation.   

P2L35: It is not clear if the model takes into account aerosol-radiation interactions and its impacts on photolysis. This can substantially affect the OH generation (R1 and R2).

P9L9: this title is odd. Use something like “numerical experiments”.

P9L25-30: Because of the low quality of Fig 2, it is not easy to follow the arguments here. Perhaps one option would be to superimpose IASI data on both plots (7 and 8 PM and use larger dots so one can better compare them.

P11L14: “understand”

Figure3: The caption seems to be wrong. It should be sensitivity tests for chemistry, or? Lines are impossible to distinguish. Use thicker lines with colors that are better distinguishable.

Figure 3d: What is the use of “sulphate volume”?

Figure 3e: Negative mass?! This is wrong and challenges the whole study. 

Review of Lachatre et al., “Modelling SO₂ conversion into sulphates in the mid-troposphere with a 3D chemistry-transport model: the case of Mount Etna’s eruption on April 12, 2012.”

In this paper, Lachatre et al. attempt to test a model of SO₂ chemical conversion in a volcanic plume, focusing on a particular eruption of Mount Etna in Italy. The authors focus on one day, and follow the plume with the CHIMERE chemical transport model. Results show that sulfate is produced mainly through oxidation of SO₂ by OH in the gas phase (70%) and also by aqueous-phase oxidation of O₂, catalyzed by Mn²⁺ and Fe³⁺ ions (25%). The authors test the model with different plume heights, water availability, and transport algorithms. Given the low abundance of H₂O₂ in volcanic plumes, better characterization of these SO₂ oxidation pathways is of interest. The work also has importance for better understanding the role of volcanic particles as cloud condensation nuclei and for better constraining radiative forcing of preindustrial climates.

While not breathtakingly novel, the paper merits publication with minor revisions.

Minor issues.

1. The reader would like to learn a bit more about why explosive volcanic eruptions emit large quantities of water. (Page 2, Line 30).
2. More information on how plume heights were determined would be helpful. (Page 5, Lines 25-31).
3. Figure 1. The figure is not interesting and should be move to the Supplement.
4. The text states that “useful information” can be extracted from a comparison of model output and observations from IASI. (Page 9, Lines 26-27). Could a more quantitative comparison be made? True, the middle panel of Figure 2 looks something like the model results, but a more substantive comparison would be appreciated.
5. Table 2 and all tables. Please identify the acronyms in a footnote. Also state in the caption of the Tables which simulation is the most realistic.
6. Figure 3 and most of the figures. There is an abundance of figures in the paper that do not seem to contribute much to the overall message. I recommend that the authors choose one panel from each Figure like Figure 3 and keep these panels in the main text. (Likley these will be the panels showing the timeseries of volcanic sulfate mass.) Put the remaining panels, if deemed necessary, in the Supplement. Also, the thin lines could be fattened to improve the visual message, and less white should be retained in the panels. Having so many figures with so much white space dilutes the main messages of the paper. The figures in the appendix can also go in the Supplement.
7. Captions for Figure 3 and similar figures. Please write out, without symbols or acronyms, what is being presented. Many readers will just browse through the paper and look most closely at the captions.
8. The Conclusions section mentions the need to better constrain emissions of Fe³⁺ and Mn²⁺ from volcanoes. (Page 20, lines 10-13). Do such emissions vary greatly among different volcanoes? What steps can be taken to constrain these emissions?