The author report measurements of HCHO on a ship cruise. Though this is a measurement report, the discussion is partly not very satisfying and it is not very clear, if there is any deeper meaning in the analysis. It remains unclear, if there is anything to learn from the HCHO yield that is calculated, specifically if the inorganic fraction of the OH reactivity is included. It may make it easier to exclude inorganic species that do not produce HCHO from the beginning. Specifically, the dependence on NOx drastically change, because NO and NO2 are OH reactants, but what the authors want to point out is that the fate of RO2 and therefore the product distribution depends on the availability of NO. In addition, the authors want to connect the HCHO production with the chemistry and not just with the presence of OH reactants that do not produce HCHO (e.g. HCHO would be very low in an environment with high NO2 though there might be a lot of efficient HCHO from the chemistry of the organic compounds). The argument that subtracting the inorganic part reduces the data set is weak because the inorganic fraction reducing the chemical meaning of the derived yield. In the discussion the authors mention also the results, if the inorganic fraction is subtracted, but there is no meaningful interpretation.

P3 L2: “reactions” instead of “reaction”

P2/3: It is confusing that Equation 1 is defined for the sum of all reactants, but that later the authors distinguish between different type of OH reactants. This should be consistent specifically regarding the “yield”.

P3 Equation 4: Could transportation play a role?

P4 Equation 5: It would be easier for the reader, if Eq. was written using the same terms as in the previous Equations.

P7 L11: Are the assumptions about the photo-stationary state plausible, if maximum values were around noontime, when photooxidation is at its maximum and the chemical lifetime is 2.5h? Wouldn’t the maximum expected to shift into the afternoon?

P8 L18: The reason for using of a constant value for the BHL instead of 30 min values is not plausible. Why does an interpolation of values lead to a lower data coverage compared to using a constant value?

P9 L13: A compact linear relationship would only be expected, if chemical conditions do not change and transportation does not play a role. This should be made clear.

Figure 4: Error bars would help to see, how much of the scatter is explained by the statistical errors of data points.

P14 L12-221 and Figure 5: I do not see the meaning of plotting the yield against reactivity. Why should a yield linearly scale with the concentration of a family species? I would rather expect a constant number and a discussion, if the numbers are in any way expected from chemical oxidation mechanisms of these family species. In addition, the slopes in plots in Fig. 5 are essentially determined by one data point, because all other points are scattered in the low range. Please comment.

P16: In the discussion of the impact of NOx on the HCHO yield, the impact of NO2 as an OH reactant is missing, if the inorganic part is included. It is hard to see, if there is any meaning in the way the dependence is discussed. This part of the discussion needs significant improvement.

The manuscript titled “Measurement report: Observation-based formaldehyde production rates and their relation to OH reactivity around the Arabian Peninsula” reported the shipborne observations of HCHO, OH, and OH reactivity from the Air Quality and Climate Change around in the Arabian Basin (AQABA) campaign in summer 2017. The authors calculated the effective yield of HCHO with total HCHO loss rates, OH and OH reactivity, by assuming a photochemical steady-state of HCHO and analyzed factors (e.g., NOx, VOC speciation, OH reactivity) that potentially impact the effective yield of HCHO. The measurements in the unique environment and the simultaneous measurements of HCHO, OH and OH reactivity will add value to the community but the analysis is not clear and needs major modifications. I’ll recommend publication of this manuscript after the authors address my comments below.

Page 1 Line 28: Biogenic is the natural emission from living organisms such as plants and does not include biomass burning. Most HCHO from biogenic sources is secondary produced.

Page 3 eqn 3 and eqn 4:  If you did not use eqn3 to calculate p(HCHO), please state it clearly especially when you provide the detailed information about each term. when you used L(HCHO) in eqn 4 to calculate P(HCHO), would the steady-state assumption hold at zero OH ? It is not clear that the intercept can represent additional HCHO sources to VOC oxidation by OH and major question 1 can be addressed. The intercept looks like to be the dry deposition term actually. Please also provide more information about the uncertainties in the dry deposition term.

Page 3 line 18: Do you mean photochemical steady-state instead of photo-stationary state? I think photo-stationary state can mean recombination of the photolysis products to form HCHO. Please replace all “photo-stationary state” terms by “photochemical steady-state” throughout the manuscript. The lifetime of HCHO is 2.5 hrs or more in the study. What is the HCHO lifetime at sunset and sunrise? Are they short enough to assume to be in steady-state?

Page 4 eqn 5: Please define P0(HCHO)

Page 8  line 17: No stickler et al. (2007) in the reference section

Page 9 line 15-18: Is direct emission of HCHO very small compared to unsaturated hydrocarbons in the fresh pollution plume? Please specify the region and time when the fresh pollution plume occurred. The HCHO and ROH in AG leg seem to be high.

Page 8 line 18-20: Specify which region and time had rainout and which region and time had primary emission.

Page9 line20- page 10 line 1-2: The intercept may not represent additional HCHO sources because the steady-state assumption does not work at very long lifetime. The intercept can be equal to the dry deposition term.

Page 10 line 9: How do primary emissions and transport affect the alpha derived here?

Page 11 line 12-17 Can primary emission of HCHO also play a role in the highest calculated pHCHO on 30/07/2017 from figure 2?

Page 12 line 20: which are these three regions mentioned? Please be specific.

Page 12: As mentioned before, the intercept likely cannot be interpreted as Padd(HCHO) because of the long lifetime at OH=0. Do the intercepts correlate with dry deposition?

Page 13 line 13-23. “individual trace gases (R(OH)x) from alkanes (R(OH)alkanes), alkenes (R(OH)alkenes), OVOCs without HCHO” is confusing. Does x represent individual group (e.g., alkanes) instead of individual trace gas? I would expect less VOC species and higher measured R(OH)x to total R(OH) ratio in cleaner or lower ROH environments. No information about the VOCs measurement methods is provided in Section 2. A table of the measured VOC species is needed to understand R(OH)x/R(OH).

Figure 5: Do you assume the unmeasured VOC groups are proportional to the measured R(OH)alkene, R(OH)alkanes, R(OH)OVOCs,  and R(OH)(aromatics? What about using the ratio R(OH)x /ROH in the x axis ? This sounds the way to go if you want to evaluate the impact of the different VOC groups on HCHO yield alpha.

Page 15 line 19-page 16 line 1-2: I think the statement “the role of NOx in HCHO production is ambiguous” does not sound correct based on Wolfe et al. (2016) and Valin et al., (2016). Do you mean HCHO yield instead?  The following explanation is also not clear. It is more evident that NO influencing the amount of OH radical can impact the production of HCHO. Explain how NO influencing the amount of OH radical can impact the yield of HCHO.

Page 17 line 10-11: again “photochemical steady-state” instead of “photo-stationary state”.

Page 17 Line 11-12: Could you provide potential reasons why the yield of HCHO depends largely on the presence of absolute VOCs levels? This does not have a clear physical meaning to me.