This study presents new measurements of VOC fluxes near the coast. Interestingly, the coastal fluxes are comparable to previous measurements in open ocean of similar latitude. Flux magnitudes were dependent on the wind-sector, with generally larger fluxes for acetone and acetaldehyde from the terrestrial sector. The authors compare their observed fluxes to those in global models and discuss some of the similarities and differences.

Overall, I think this paper is well written and organized. The research of this paper is novel and helps fill a void in our current scientific understanding of air-sea flux of VOCs. I recommend it for publication after some minor comments are addressed.

Line 121-122: Can a quantitative number be placed on this? For example, how much does the humidity impact the background and overall, what is the correction?

Line 160: Is there a statistical difference in the day versus night fluxes?

Line 179: Again, what effect does humidity have? Is it a few % or tens of percent?

Line 242: Table 2?

Line 246-248: It would be helpful to provide the concentration data here from these other studies in addition to the percent differences reported. Precents depend on magnitude so smaller concentrations could be largely different in percent terms.  It would just help the reader not have to dig up those papers to find values.

Line 250-265: This section is a bit dense with all the comparison and numbers. It would benefit from a (supplementary) figure that shows the different means from the various studies for each compound.

Line 250: Were all the atmospheric concentrations observed just over one day or even a few days? If so, this should be highlighted more than just the brief note made in the first sentence. This could explain quite a bit of differences between the various studies.

Section 3.5: The first paragraph is strange. I was expecting a discussion about flux and then it is about a lifetime calculation. Which atmospheric lifetimes are calculated? The ones in the model? A bit vague.

Section 4: What research needs to be done to solve some these discrepancies? What would the next research project look like? “Future work should…”

General comment: Models are only as good as the data going into them. This study provides more constraints for the models. Thank you for your contributions.

Figure 2: Please avoid using red & green together for color-blind accessibility.

Figure 3:

• I understand it’s hard to keep all the subpanels on the same vertical axis range, but it is also then hard for the reader to compare the different fluxes. Panels B & C can at least be on the same range (possibly -50 to 20)
• I was confused having the blue dashed lines mean different things between panel A and panels B-D. Maybe a shaded gray area would be better for the LOD?
• The gridlines between the subpanels are of different increments.

Figure 4: The gridlines are strange in this figure as well. (Maybe it is just how the graphics were saved in my PDF) but double check for publication.

Figure 5:

• Please avoid using red & green together for color-blind accessibility. (Not as critical here since the markers are different, but a general comment).
• Gridlines are inconsistent

Figure 6:

• Gridlines are inconsistent

Figure 7:

• The omission of color in panels B and C seems like a mistake.
• Panels B and C are heard to read because the marker sizes of the triangles and circles are small.
• What are the dashed black lines, the uncertainty bands? Please provide clarity in the figure caption.
• Poly fit is to the observed data, I assume?

Table 1:

• Please add bottom borders under “open-water” and “Plymouth Sound” to make it easier to know which columns go under each sector
• I appreciate the raw values especially for future comparisons.

Table 2:

• Provide the sampling date(s) for the atmospheric mixing ratios.
• Please add bottom borders under “open-water” and “Plymouth Sound” to make it easier to know which columns go under each sector

Review of acp-2021-108 Air-sea exchange of acetone, acetaldehyde, DMS and isoprene at a UK coastal sit by Phillips et al.

This manuscript reports the air-sea fluxes of several short-lived trace gases computed using the bulk method and measured with the eddy covariance technique. The measurements are performed from a fixed point and values from two wind sectors, representing open waters and estuary waters, are reported. The authors compare the fluxes derived using the two methods with each other and with those previous estimates (including from models). They also compute the gas transfer velocities and compare to open ocean values. The results are robust and the manuscript well written. I recommend accepting the manuscript after the following minor revisions are adequately addressed.

Specific comments:

Line 134 – I think this table should be A1 and not A3.

Lines 156-157 – How does this method compare to computing the covariance directly over a one-hour time period?

Paragraph starting at line 189 – It seems that it would be simpler to use the buoy winds for the TL calculations. Was this correction done so that the fluxes from the two wind sectors were computed using the same corrected winds?

Line 242 – table not specified

Line 265 – typo: observation should be observations

Paragraphs starting on line 334 and 345 – could there be any influence from vertical gradients in the directly measured fluxes? Did the authors consider that?

Line 345 – please add the units of the flux in parentheses

Lines 378-379 – The fluxes computed in Booge et al. (2016, supplement) should be used instead of Palmer and Shaw.

Appendix A – Twice the tables are not specified.

Figure 7 – The caption and the legends are bit difficult to reconcile with each other. The black lines described for b are also in c, correct? Or which black lines are meant exactly?

Table 2 – It is stated in the text that the EC didn’t work for isoprene (lines 226-229), yet it is reported in the table. Why?

Table A1 – H2O mass flow controller should say H2O mass flow; what is meant by source valve %?

Table A3 – caption typo: should say for or in calculations; perhaps would be better to state in the caption what the parameters are instead of only showing H and V, etc.