|Review of "Development of a balloon-borne instrument for CO2 vertical profile observations in the troposphere" by Ouchi et al.|
This work describes an atmospheric measurement technique, specifically the measurement of CO2 mole fraction using a balloon-borne sensor, enabling CO2 measurements through the troposphere. There are not a lot of ways to accomplish this measurement using a sensor light-weight enough to be able to be flown without specific FAA approval or waiver. These measurements would be valuable to the CO2 measurement community. I believe it should be published in AMT because for me it satisfies the requirements there: 1) it is a new technique 2) it is described adequately 3) the paper includes estimates of uncertainties.
One reviewer previously objected to the paper because the instrument may not give measurements of sufficient accuracy as to be useful for certain studies (carbon cycle). I do not think it is more suitable as a technical note, and believe it should be published in AMT so it is in the open literature, as have been other papers using low-accuracy / low-precision CO2 measurement systems (e.g. Shusterman et al., 2018; Martin et al., 2017). I would argue that a user would need to decide whether this instrument would have the accuracy to meet their science requirements.
I do agree with the reviewer, however, that the authors must be honest about statements that are made in the manuscript regarding the accuracy and applicability of the measurement to different science questions, and that the uncertainties must be expressed honestly and not misrepresented (which I believe has been done here). I would also recommend the authors insert the uncertainty metrics below 7 km in the abstract itself, so that the reader quickly knows the final result, and that the manuscript includes the cost of the parts for the disposable sonde.
Overall, I believe the authors addressed the original reviewer comments sufficiently to warrant publication. I believe they have represented the accuracy of the system properly, and would only want the cost estimate to be included, because that is another very important piece for these measurements, as they are being offered as an alternative to aircraft flights that have been shown to have higher accuracy.
The English is still a bit awkward in several places. Through the text (e.g. L449, Table 2) sometimes "concentration" is used instead of "mole fraction", this should be checked throughout.
Re L34-35, I agree with the authors' response that they can claim the measurements are "useful". There are certainly applications where they would be useful, the accuracy does not need to be better than that of global models to be considered useful (one might be interested in much smaller-scale variability of CO2 than such a model can achieve, such as within boundary layer variability).
L47 Winderlich is misspelled I believe
L 84-85: The NOAA aircraft program (Sweeney et al) is neither short-term nor limited to near large airports, so this sentence should be modified. (12 flasks are sampled from 0-8 km every two weeks at sites across the US).
L 118: I agree with a previous reviewer that more discussion of cost and recoverability should be made here. If a user must spend $4000 per flight, that is pretty significant. This cost estimate should be in the manuscript (not just in the reviewer response), even if it is a crude estimate of the cost of parts only. I am not sure it can compete with hiring a charter pilot to do the profile and getting permission from the FAA to fly a small airplane is trivial (for a pilot). (L121).