Dear authors/co-authors,

Some weeks ago you submitted your revised version of the ms as well as your response to the comments provided by two reviewers. Overall, I see that the comments provided by the reviewers have been properly addressed and would like to thank both reviewers for their constructive comments.
There has been some delay in my decision being out of the office for some weeks but can now inform you that I am accepting your ms for publication in ACP. There are only some further suggestions from my side on some other potential modifications/typos as you can find them below and that you should consider for the final version to be published in ACP.

In your reply to the comments that have been provided by reviewer #1 on the uncertainties involved in some of the components involved in the feedback mechanism you address the issue on the dependence of the Iodide concentrations on SST as reported by Chance et al., 2014, and MacDonald et al., 2014. I also noticed that you included in the figure caption now the reference to the Ganzeveld 2009 study in which there was also an evaluation of the inferred and observed latitudinal gradients in oceanic Iodide concentrations and where the inferred Iodide was not based on SST as a proxy but rather using oceanic nitrate concentrations. These alternative approaches to infer Iodide are discussed in more detail by Chance et al (2014) but with respect to this discussion on uncertainties, it would be worthwhile to also shortly mention here that there are these alternative approaches compared to the one you applied in your study and which result in quite comparable Iodide concentrations.

In the model set-up, you indicate that used one and the same SST climatology for your simulations. Since this SST is used to infer your Iodide concentrations, which in turn determines the ISG fluxes, differences between the pre-industrial and present-day SST would also affect this feedback mechanism where you have now only considered the role of changes in O3. I think you should also discuss the potential implications of this experiment set-up. In addition, because of the link between O3 deposition (concentration x deposition velocity) and the Iodide source, I think it is worthwhile to spend one sentence on how oceanic O3 deposition is calculated in the CAM CESM modelling system (presumably using a prescribed constant oceanic uptake rate of ~0.05 cm s-1)

Line 18: pp2, Gómez Martín et al., 2013; Grossman et al., 2013;

I suggest to change the following sentence to:
Page 4, lines 27-28: Note that, for a given SST, the [I-aq] parameterisation according to MacDonald et al. (2014) results in lower iodide concentrations than considering the abovementioned [I-aq]-SST2 correlation according to Chance et al. (2014).

Page 7: “Integrating the tropospheric column, the rate of iodine catalyzed ozone destruction has increased by 90 Tg y-1 since the pre-industrialisation era, yielding a total present day tropospheric ozone removal by iodine of 280 Tg y-1.”. Would it be worthwhile to put these numbers further in perspective, e.g., by comparing this with a global annual ocean O3 deposition flux on the order of 300 Tg y-1.

In the figure caption of Figure 1 there was a typo: (see also Ganzeveld et al., 2009).

Dear author, co-author, I checked your response and last minor edits made in response to my last list of editors comments. Those are all clear/well explained and consequently accept your ms for publication in ACP. Let's see how this interesting analysis on these atmosphere-ocean feedback mechanism is further received and might result in follow-up analyses with models and observations,

Laurens Ganzeveld

(the only small typo that I found is: (estimated range of 200-350 Tg O3 y-1;..); removing "the"