This study focuses on collecting the recent advances in the chemistry of DMS with OH from laboratory and chamber studies and building from this an updated chemical model for the OH-initiated DMS oxidation. The updated mechanism is built starting from what available in the MCM adding data from NASA JPL and/or IUPAC when available, published experimental data, SAR and theoretical calculations. For some reactions where no data is available and/or where disagreement between published data is present, are tuned to match the observations. The updated mechanism is compared with experimental results and model results for four different chamber experiments performed at different conditions and is then used to model a selected number of sulfur compound in the marine boundary layer. Here a comparison is performed with the different chemical mechanisms from the chamber studies tested.

Overall the paper is well written and clear and offers a valuable addition to the current knowledge on OH-initiated chemistry of DMS by trying and summarizing together the recent discoveries.I do recommend publication after the following point is considered.

I do though have one main comment that I think should be addressed previous publication. The chemical mechanisms highlighted from the Jernigan et al. (2022) study is, in my opinion, incomplete. The authors mention that there are only five reactions added in the study by Jernigan et al. (2022) which are then used to simulate their experiments in the chamber. This is not correct as a detailed chemical mechanism was developed in the study by Jernigan et al. (2022) by use of theoretical calculations and SAR. Indeed, nine of the reactions from the extended mechanism from the Jernigan et al. (2022) work are then included in the updated mechanism generated within this study. As it is then clear that the authors are aware of this detailed mechanism, why is it not tested for the experiments in Jernigan et al. (2022) study but only five (very tuned) reactions are used? I would be interested to see how the detailed mechanisms would compare with other chamber studies and with the updated new mechanisms developed. At the very least, it should be mentioned that the Jernigan et al. (2022) study has a very extensive mechanism mostly based on theory and SARs, modifying the MCM by dozens of reactions and a reason on why the authors have chosen though to only use the five tuned reactions should be given.

Is there a reason why SAR seems to be preferred in comparison to explicit theoretical calculations? For example, for reaction 134 a SAR for CH₃S (a non-oxygenate) is a dubious source to represent HOOCH2_S (an oxygenate) compared to a high-level theoretical calculation (Jernigan et al., 2022) as the SAR would not represent H-bonding or electron-withdrawing effects.

Specific comments:

Page 12 Line 252. I wouldn’t say that major uncertainties are there for the isomerization rate of CH₃SCH₂O₂. By now there is a direct temperature dependent kinetic study (Assaf et al., 2023) in very good agreement with theory (Jernigan et al., 2022). As pointed out in the text, much less is known about the self reaction of CH₃SCH₂O₂ or the heterogenous losses of HPMTF. I would also argue that photolysis of HPMTF is also not known.

Page 13 Line 279. In the study by Jernigan et al. (2022) also a theoretical value for the rate coefficient of HPMTF with OH for each different channels is available.

Page 13 Last paragraph. I guess also here it could be worth mentioning that also in the study by Jernigan et al. (2022) the H-shift for HOOCH₂SCH₂O₂ was explicitly theoretically calculated. The value is quite faster than what found by Veres et al. (2020), ~ 3 s^-1 at 298K.

Figure 8. Although I like this figure and find it useful, I think the addition of a table with the percentage of each species listed for each different mechanisms would give even more information in particular for a more detailed comparison.

Page 19 Line 419. Wouldn’t it make more sense to use theoretical calculations before estimates?

Page 20 Line 430. Here it is stated that theoretical calculations were used if SAR was not available. This seems in disagreement with the statement from page 19 line 419.

Page 20 Line 437. That is not correct, theoretical calculations for that very H-shift are also available in the study by Jernigan et al. (2022).  

References

Assaf, E., Finewax, Z., Marshall, P., Veres, P. R., Neuman, J. A., and Burkholder, J. B.: Measurement of the Intramolecular Hydrogen-Shift Rate Coefficient for the CH3SCH2OO Radical between 314 and 433 K, J. Phys. Chem. A, https://doi.org/10.1021/acs.jpca.2c09095, 2023.

Jernigan, C. M., Fite, C. H., Vereecken, L., Berkelhammer, M. B., Rollins, A. W., Rickly, P. S., Novelli, A., Taraborrelli, D., Holmes, C. D., and Bertram, T. H.: Efficient Production of Carbonyl Sulfide in the Low-NOx Oxidation of Dimethyl Sulfide, Geophys Res Lett, 49, e2021GL096838, https://doi.org/10.1029/2021GL096838, 2022.

Veres, P. R., Neuman, J. A., Bertram, T. H., Assaf, E., Wolfe, G. M., Williamson, C. J., Weinzierl, B., Tilmes, S., Thompson, C. R., Thames, A. B., Schroder, J. C., Saiz-Lopez, A., Rollins, A. W., Roberts, J. M., Price, D., Peischl, J., Nault, B. A., Møller, K. H., Miller, D. O., Meinardi, S., Li, Q., Lamarque, J. F., Kupc, A., Kjaergaard, H. G., Kinnison, D., Jimenez, J. L., Jernigan, C. M., Hornbrook, R. S., Hills, A., Dollner, M., Day, D. A., Cuevas, C. A., Campuzano-Jost, P., Burkholder, J., Bui, T. P., Brune, W. H., Brown, S. S., Brock, C. A., Bourgeois, I., Blake, D. R., Apel, E. C., and Ryerson, T. B.: Global airborne sampling reveals a previously unobserved dimethyl sulfide oxidation mechanism in the marine atmosphere, Proc. Natl. Acad. Sci. U. S. A., 117, 4505, 2020.

Jacob and co-workers present a new improved chemical mechanism for DMS oxidation. For the most part, this study is well-designed and well-realized, taking account of much of the existing literature measurements and models in a thorough intercomparison exercise. Given the importance of DMS oxidation to the atmospheric sulphur budget, I expect that this study will be of interest to the atmospheric chemistry community. Therefore, I recommend that it is appropriate for publication in Atmospheric Chemistry and Physics after the authors have considered the following points:

Line 17: I am not sure what the authors mean, when they say that DMS is the largest natural source of sulphur. There is a certain unappealing circularity to the logic of this statement as it stands.

Line 29: The halogen oxides do not, I presume, “undergo” hydrogen abstraction, but rather participate in/ initiate hydrogen abstraction.

Line 73: This is admittedly minor, but since the authors have done such a careful job otherwise, I would suggest that they ought to hyphenate “gas-phase” in this sentence.

Figure 1: This is an interesting overview figure that helped me to quickly understand the general comparison between the various treatments of DMS oxidation that are available, so for that, I thank the authors. Despite the complexity of this figure, there are some pieces of information that I would have liked to have seen in addition: 1. Which of the molecules have been measured in lab/ field experiments (perhaps these molecules could be coloured accordingly?). 2. Which of these reactions have been measured/ estimated/ calculated (perhaps the perimeters of these arrows could be coloured accordingly?). I think doing this would give the reader a sense of the state of the knowledge in this subject.

Figure 2: I didn’t really enjoy the combination of chemical abbreviations and chemical line notation. It asks a lot of the reader to keep the structures of all these abbreviations in mind, and I would advise against their use.

Figure 3: It would be useful, where possible to include the chemical structures of the chemicals on each of these panels, so that the reader can quickly grasp which molecule is under consideration.

Figure 4: I found that the upper two panels were quite congested compared with the lower panel. Why don’t the authors retain the same form factor for each of the panels?

Lines 189–217: The terms “useful metric”, “useful summary” and “useful way” seem insubstantial to me. Useful in what way exactly?

Line 207: It seems quite obvious that mechanisms that are informed by laboratory and field observations will likely be more reliable. However, it also seems plausible that the inclusion of erroneous measurements could introduce a undesirable bias into the model. With this in mind, it seems like the best idea to include a wide variety of experimental observations to minimize this effect. This appears to be what the authors have done with this study, but I don’t think that you made this point clearly in the manuscript. I would suggest that you comment on this idea.

Lines 220–230: The reliability of the experimental data is a key consideration. Would it be possible to provide experimental error bars on some of the time-series in Figure 3 for example? Some of these are going to be highly uncertain, I imagine (e.g. TPA and HPMTF).

Line 239: many O2 additions are reversible to some extent. Do we know for sure that this O2 addition is irreversible?

Line 244 (and many other examples): It is incorrect to describe a bimolecular rate constant as being “slower” or “faster” than another. They are perhaps best described as being “smaller” or “larger”. I would suggest that the authors carefully look through this manuscript for these and related words in order to assess in each case whether they are referring to these constants or (correctly) referring to the rate of chemical change with time.

Figure 5: I don’t understand why the units of concentration vary in the panel related to the experiments of Shen et al. I suggest that it is easier to make a direct comparison between studies when the units are consistent throughout.

Line 258 (and other similar examples): where possible, it would be best to include uncertainties in rate constants, such that the reader can assess whether these are well-known quantities (or otherwise).

Figure 7: Again, this is a comment that relates to chemical abbreviations and chemical line notation. The authors should make more of an effort with the latter. Where are the radical centres? Where are the unsaturated bonds?

Figure 8: Again, chemical line notation is sub-optimal. OCHSOH is a good example, wouldn’t it be better to refer to it as O=CHSOH?

General comments on mechanism:

I am surprised that photolysis reactions are not considered to be a key uncertainty in this mechanism. When inspecting your Figure 1, I was left unconvinced by your treatment of HOOCH₂SCH₂OOH and HPMTF. Firstly, why would you expect that HOOCH₂SCH₂OOH breaks across a C–S bond to yield the S-centred radical SCH2OOH? Isn’t it more likely that it breaks across one of the peroxidic bonds to yield an alkoxy that goes towards HPMTF? Secondly, it is possible that its carbonyl moiety will be more photolabile than the hydroperoxide functionality. This would serve to increase the amount of SCH2OOH that you are forming in your model. Thirdly, what about those molecules that possess carbonyl and hydroperoxide groups, but possess no photolysis reactions in your model (e.g. TPA and O=CHSOH)?

It would help the reader to understand the mechanism better if more of an attempt were made to balance the chemical equations that are presented. There are lots of examples of this, but one such example is reaction 128.

Reactions 132–134: Do reactions 132 and 133 ever contribute, and do reactions like this need to be included in your mechanism?