This paper presents important new data and modelling related to improving our understanding of ozone depletion in near-source tropospheric volcanic plumes. It covers an interesting range of different aspects of the plume chemistry and the direct comparison between plume aircraft measurements and modelling outputs make it innovative. There is however still some work to do to really clarify the study and to realise its full potential.

* In general:

The quality of presentation of the graphs could be improved. For example, text sizes are often small and proper scientific notation rather than 1E-4.

I also wonder whether the authors have considered other SO2 oxidation routes other than OH? Eg metal catalysis or aqueous reactions. This would be useful to do especially if more comparison was included between the modelled SO2 lifetime here and that measured and modelled at Kilauea and Icelandic plumes like Holuhraun (Kroll et al 2015, Schmidt et al, Ilyinskaya et al, Whitty et al etc).

The concentration of ash or fine silicate in the plume is an important parameter that requires some discussion.

More effort to put the extent of these processes in a global context would really enhance the impact of the paper.

* Specific comments:

L7-10: Add some more nuance that not suggesting a direct chemical relationship between ozone and SO2.

L41: Might nuance by saying S species rather than SO2 as H2S can be significant and in some circumstances dominant.

L60-65: There several different measurements of volcanic HCl into the stratosphere and some of BrO also via satellite. See Table 4 and section 4 of Mather, 2015. UTLS/stratospheric injection is an important issue re future impacts of large-scale eruptions.

L127: ‘Another difficulty lies in attributing observed ozone losses to halogen chemistry when volcanic bromine emissions and/or bromine radical levels are not well-known.’ I am not quite sure what the authors mean here. Is the implication that Br emissions will not be ubiquitous?

L201 Useful to add here some overview of how ozone varies with altitude

L293-307 How sensitive are the results to this HBr/Br radical partitioning? Similarly with the NO/SO2 ratio and Hg/SO2 emissions? Some sort of sensitivity analysis seems rather vital. These are actually input parameters that you could use the model to ‘invert’ from the measurements in order to put bounds on them. This I stared with Table 2 (and Table 6) but could do with integrating with the other measurements and extending to systematic tests with some of the more poorly constrained inputs.

Table 2 is hard to follow. Please consider how better to present this information.

Figures 4 and 5 are hard to read. What are the possible explanations for some of the structures in the data (e.g., systematic curving)?

Figure 6: more details are needed in the legend and/or caption explaining what the different traces reference and which axis they relate to.

Figure 7: refer to table 2 in figure caption

L367: This aerosol/SO2 ratio and the flux measurement is interesting and could be compared to measurements on Etna and other plumes using techniques like sun photomers.

L384: but in contrast to the HOx results at Hekla in Rose et al., 2006. This seems especially relevant given the hal00 run results although the differences in plume altitudes could be important.

L393: ‘Although the instantaneous lifetime of SO2 (with respect to oxidation by OH) is substantially increased in the halogen-free model plume, we note that the addition of halogen emissions to the model further suppresses OH, increasing the SO2 lifetime and having a reductive effect on secondary aerosol production both in terms of mass and surface area (Figure 7).’ I am really not clear about the authors’ meaning here. Wouldn’t higher OH in the hal00 model be expected to decrease SO2 lifetime? I think that more detail is needed about the issues with comparing tropospheric and stratospheric chemistry here too.

L413 Comment on why the aerosol surface area is lower on this day?

L425-8: There are other studies of this variability of BrO with met conditions etc that should be cited. E.g., Dinger et al. study at Masaya (https://acp.copernicus.org/preprints/acp-2020-942/).

Figure 10: I found this figure quite hard to interpret and maybe it could be made to work better with Figure 1. I am not sure what more it brings beyond Figure 9, which is much clearer.

Figure 11: Mark on the field of measured BrO/SO2 ratios over the years at Etna.

L445 As are the many downwind measurements made by Bobrowski et al over the years.

Section 4.2.2 I am not clear how the fit to reality of the run in Figure S2 compared to other runs is being assessed.

L514 remind reader the reason for the different plume density on 30^th

Figure 16: Can some attempt be made to get into the time frame rather than distance frame?

L593 So does the volcano need to be a source of NOx to account for the observed HNO3 levels in the plume according to the model?

L704 And more lab experiments focused on parameterising Hg atmospheric chemistry reactions.

* Typos

The article needs a thorough proof read for language and typos

L8, 471 subscript on SO2

L58 Spurious ‘.’

L148 ‘withing’

L168 reproduce

L270 come close

L275 Mount Etna’s

L561 through

L603 halogen repeated

References (and references therein)

Ilyinskaya, E., et al. (2017). Understanding the environmental impacts of large fissure eruptions: aerosol and gas emissions from the 2014–2015 Holuhraun eruption (Iceland). Earth Planet. Sci. Lett. 472, 309–322. doi: 10.1016/j.epsl.2017.05.025

Kroll, J. H., Cross, E. S., Hunter, J. F., Pai, S., et al. (2015). Atmospheric evolution of sulfur emissions from Kılauea: real-time measurements of oxidation, dilution, and neutralization within a volcanic plume. Environ. Sci. Technol. 49, 4129–4137. doi: 10.1021/es506119x

Mather, Volcanoes and the environment: lessons for understanding Earth’s past and future from studies of present-day volcanic emissions, Journal of Volcanology and Geothermal Research, 304, 160-179, 2015. (doi:10.1016/j.jvolgeores.2015.08.016)

Schmidt, A., et al. (2015). Satellite detection, long-range transport, and air quality impacts of volcanic sulfur dioxide from the 2014–2015 flood lava eruption at Bárðarbunga (Iceland). J. Geophys. Res. Atmos. 120, 9739–9757. doi: 10.1002/2015JD023638

Whitty et al, Spatial and Temporal Variations in SO2 and PM2.5 Levels around KÄ«lauea Volcano, Hawai’i during 2007-2018, Frontiers in Earth Science, 8, 36, 2020. (doi:10.3389/feart.2020.00036)

General comments:

The manuscript by Surl et al., presents a new dataset of airborne measurements made at Mount Etna volcano during summer 2012 and a new version of the WRF-Chem model, named “WRF-Chem Volcano” (WCV), that have been modified to incorporate volcanic emissions and multi-phase halogen chemistry. WCV shows good skills in reproducing the observed volcanic plume. This paper shows the importance of including halogen chemistry to model the chemistry-climate impacts of volcanic events. The paper is well-written, however, it would be very helpful if the authors include a paragraph summarizing the uncertainties in their model developments (e.g. uptake coefficient for heterogeneous reactions, surface area density, mercury chemistry). In addition, some of the model results need to be quantified in section 4.2. This study is scientifically relevant and I recommend the publication to Atmospheric Chemistry and Physics, after addressing general comments and, specific and technical comments listed below.

Specific comments:

1) The Abstract is long and the aim and the main results are not clear. Please shorten the abstract to include the overall purpose, the design and the relevant findings of this study.

2) Please enlarge the font size of the axis labels, keys and legends as appropriate (e.g. Figure 4 and Figure 11).

3) Consider mention the recent study of Hirtl et al., 2020 that presents the update of the WRF-Chem volcanic emission pre-processor towards more complex source terms and evaluates the results for the eruption of the Grimsvötn volcano in Iceland in May 2011 in the literature review (section 1.2.2).

Hirtl, M., Scherllin-Pirscher, B., Stuefer, M., Arnold, D., Baro, R., Maurer, C., and Mulder, M. D.: Extension of the WRF-Chem volcanic emission preprocessor to integrate complex source terms and evaluation for different emission scenarios of the Grimsvötn 2011 eruption, Nat. Hazards Earth Syst. Sci., 20, 3099–3115, https://doi.org/10.5194/nhess-20-3099-2020, 2020.

4) It will be useful to add another panel to Figure 3 with a terrain height map for the domain 4 (d04).

5) Line 258: 24 hours of spin-up seems a short period to initiate the chemistry in a regional model. Did you make any tests to make sure 24 hours was enough?

6) Line 323: Could you explain why encounter 10 and 12 displayed in Figure 4 and 5 give a positive correlation?

7) Figure 6, 7: Are the numbers “30, 31, 01” the days of the analysis? Please clarify this in the caption.

8) Lines 557-358: “The model SO2 chemistry includes gas-phase oxidation by OH, generating secondary sulfate aerosol.” Was this chemistry already in the chemical mechanism or did you update it? If you updated these chemical reactions please mention where can we find their description.

9) Line 363: Please quantify this statement “Notably, the average mixing ratio of SO2 in the plume of the 30th July is significantly less than on the other two days, although the trend is similar. “

10) Line 381-382: Please quantify the statement “There is a substantial depletion of OH within the plume, and a moderate depletion of HO2 (Figure 8).”

11) Line 382: Could you give more details about the study of Jourdain et al., 2016: “This result is consistent with model findings for the Ambrym plume (Jourdain et al., 2016)“

12) Line 394: Please quantify this statement “increasing the SO2 lifetime”

13) Line 460: Please quantify “whereas HCl remains abundant.”

14) Have you check changes in the O3 Total Vertical Column between the main and novolc model runs? If so, are these changes significant that could be mention in section 4.4?

15) Line 553: Please quantify this statement with values from Table 6 and describe more in detail “In the absence of halogens (only), the plume is slightly ozone productive“.

16) Line 571: Please quantify this statement “the plume is nearly totally depleted in NOx (Figure 17a).”

17) Line 574:  Please quantify this statement “ the plume is elevated in HNO3 compared to the background (17b)."

Technical corrections:

Line 8: Change SO2 to SO₂

Line 216: Change “verion” to “version”