Overall Comments

The paper is well structured and provides a valuable data set with observations of bubble penetration depths. The introduction and background are appropriate and the literature review is up to date and well cited. Methods are clear (although I do request some minor changes below) and Figures are overall appropriate and support the presented narrative.

My only major issue with the paper lies in section 4.4 when the authors address gas transfer in terms of bubble penetration depths and whitecap coverage. In short, some of the main conclusions here are not fully supported by the observations and there is a perhaps too heavy of a reliance on parameterizations which are wind and wave based and addressed by the paper earlier in the previous sections.  I recommend restructuring this section and presenting some of these results with a very clear caveat.

Specific Comments (Line by Line and Section 4.4)

Line 235-240: Why is the bubble radius increasing with depth?

Line 370: suggestion. “different size of the bubble plume” change to “different bubble plume penetration depth or extent … “ or something along the lines. This reads as the bubble size distribution is changing as a function of fetch.

Figure 4. The wind speed variable in the legend is the average wind speed over the storm?

Line 475: Please clarify. Are the authors suggesting that turbulence at the 20s period is the main driver for bubble transport?

Line 465 an onward: Authors talk about bubble-height. Is this the Bubble plume penetration depth?  I would suggest discussing bubble penetration depths or bubble depths.

Line 495: typo “dept”.

Line 495: It is not clear to me how the lifetime of the bubble depth was determined. Also, not sure how to interpret the lifetime of the bubble depth. In my opinion the temporal evolution of depth is one thing, lifetime of the bubble(s) should be addressed in terms of bubble size distributions.

Regarding Figure 9a, the histogram corresponding to bubble penetration depth is for S1, S2 or both?

Line 530: Eqs. 7 and 8 I think it is relevant to state the units needed for the constant of proportionality and the physical meaning if any.

Line 560: define f_win the text.

Line 550-570: Please define fully developed in wave age (or inverse wave age). Also, how do the conditions at the site compare to a fully developed wave field?

Also, I don’t think I fully follow the f_w argument. Are the authors exchanging the R_H term in Brumer et al. 2017 from mean significant wave height to significant wave height? What is plotted in Figure 11 as f_w = f_w(R_H)? Please clarify

Regarding Brumer et al. 2017 are the authors evaluating the steepness parameter in terms of significant wave height or mean significant wave height? Same for the forcing (inverse wave age)?

The overall value here (in my opinion) is attempting to characterize bubble plume penetration depths as a function of forcing and wave statistics more than directly addressing the relationship between observed penetration depths and [observed] fractional whitecap coverage.

Section 4.4

In my opinion it would be valuable to add a table with the gas transfer parameterizations being used (I realize they are in the appendix, but perhaps make them more available). This would also help better interpret Figure 14. This is basically a plot of observed bubble penetration depths and wind, wave forcing already presented in Fig 11. Although a very interesting point the authors attempt to make this might be pushing the dataset. The gas transfer velocity by W14 has a quadratic dependence on wind speed (as presented by the authors) Eq. 13 shows the square root of it, presenting z_ba as a function of wind speed. (this is identified in ~Line 665)

Line 665:670: “The results reveal a rapid increase in transfer velocity with increasing bubble depth penetration depth”. I don’t think the authors can make this claim as the data to support it is missing.

Line 680: consider restructuring this paragraph.

This paper describes measurements and scaling of the depth of whitecap bubble plumes from an offshore platform.  The analysis links the depth to wind speed and a selection of wind/wave parameterizations of gas transfer velocity, which for weakly soluble gases are linked to bubble mediated processes.

The paper is well-written and the background material is thorough.  The average depth of bubble plumes scales roughly as wind speed suggesting it is driven by the speed of the breaking waves or the wave-induced orbital velocity.  The authors suggest that including wave parameters in addition to raw wind speed my improve correlations with plume depth.

In my opinion this is a good paper and it can be published essentially in its present form.   I do have a few thoughts for the authors to consider.  The main issue is section 4.4 linking zb to co2 transfer velocity parameterizations.  I think the link is slightly strained because k has temperature dependencies via solubility and Schmidt number that may map well to zb.  Also, the use of k parameterizations that do not distinguish bubble and nonbubble modes does not make sense to me.  I don’t think there is any doubt that this separation is real, so I think a focus on bubble mode scaling is preferred.  Just my opinion. 

Here are a few editorial questions

Line 471  ‘bubble height’  - do you mean depth?

Line 498  What do you mean by lifetime in %?  Is that the probability of a particular lifetime?  Also, Fig. 9b is referred to as ‘lifetime’. do you mean probability distribution of lifetime?