This opinion paper communicates views on an important subject: the use of current and future satellite data to constrain global troposphere OH.  Given the central role of OH in removing pollution (including CH4), this is an important opinion paper that aims to advise the upcoming 2027-2037 Earth Science Decadal Survey (ESDS) for the National Aeronautics and Space Administration (NASA).

The paper contains valuable views. However, the paper is rather wordy (saying sensible things), and it would help to provide some more structural elements like tables listing available and future missions. The figures presented in the paper seem to merely advertise personal work, rather than helping the reader to grasp the research agenda concerning OH-proxies.

Scattered around the paper, different observing strategies are mentioned (Polar orbiting, Geostationary, UV/vis, NIR, IR, lighting observations, etc.). Here it would really help to provide a table of existing strategies to monitor OH proxies, right at the beginning of the paper.

In that sense, the order is rather odd. I would expect first an overview of the role of OH in tropospheric chemistry, and historic developments e.g (Lelieveld et al., 2004, 2006). Currently, some of this information is provided later in the paper (in the “thought experiment”). This order might be OK for an opinion paper, but I was distracted by the lack of proper introduction of the subject.

What is also a missing element in the paper is a thorough discussion about the quality of the products. Observing isoprene, H2O2, and formaldehyde is exiting, but using these quantitively is a different game. The same holds for tropospheric ozone. They author mention the 2006 Ziemke approach, and indeed since then not much happened to reliably determine tropospheric ozone. The reason is that it is simply a difficult problem. Although mentioned at places, the authors should be more specific what approaches are feasible to improve on the quality of the OH proxies. For tropospheric ozone, for instance, multi-wavelength satellite observations (IR, vis/Uv) could bring the scientific community further (there are existing studies in this field). As written now, the paper seems to argue that “all” mentioned proxies are of the same quality. Maybe adding a column in table one about the current accuracy would help to guide future needs for scientific research.

I understand the US-central approach, since this paper aims to inform the 2027-2037 Earth Science Decadal Survey (yet, to be published in a European journal?). However, some developments in Europe could help the US developments, and have played a vital role. As an example of the US-central approach of the paper, I would like to point to: As another example, discussion surrounding continuity of the NO2 VCD (Section 5.1.1) started with OMI (launched in 2004), though such observations actually began in 1996 with ESA’s Global Ozone Monitoring Experiment.

Other European elements that are missing are the ESA efforts in the AQ4ECV program (e.g. for NO2, CH2O). Focus should be on synergy, e.g. with the developments of satellites in geostationary orbits (TEMPO, GEMS, S4).

In the final recommendations I miss the further development of techniques (ML, data-assimilation, …) that help the interpretation of the data. Models are needed because they act as an integrating operator that moves around the longer-lived gases towards the “next” satellite observation. Moreover, models provide first guess profiles for retrievals, and may fill in unobserved parts of the atmosphere. Observing system development should be paired with model developments and the developments of techniques to integrate satellite data in models. In that sense, again, no mention is made of the European Copernicus program, which in my opinion is a missed opportunity to guide the US developments.

Minor comments (see also annotated PDF).

Figure1: Although interesting, this paper should not be used to present new results. At most, it could be used as an uncertain exploration of newly developed techniques. Now there is no uncertainty quantification, the legend fails to mention “tropical ocean”.

Line 30: suborbital observing strategy: unclear (in an abstract).

Line 36: “will be an integral part of a comprehensive observing strategy”

Maybe “should be an integral part of a comprehensive observing strategy.” Is better?

Line 70: contributes significantly to the overall uncertainty in the budget, interannual variability, and trends of CH4 (Saunois et al., 2020).

I agree here with uncertainties in the budget, but the interannual variability and trend in methane are well constrained by observations. Of course, the impact of OH on trend and variability in methane is large (what you intend to write), but this phrasing suggests that trend in methane itself is uncertain.

Line 164: “In contrast, decreases in CO can lead to decreased OH and longer CH4 lifetimes.”

I guess this should be “increased OH” and “shorter CH4 lifetimes”.

Lelieveld, J., Brenninkmeijer, C., Joeckel, P., Isaksen, I., Krol, M., Mak, J., Dlugokencky, E., Montzka, S., Novelli, P., & Peters, W. (2006). New Directions: Watching over tropospheric hydroxyl (OH). Atmospheric Environment, 40(29), 5741–5743. https://doi.org/10.1016/j.atmosenv.2006.04.008

Lelieveld, J., Dentener, F. J., Peters, W., & Krol, M. C. (2004). On the role of hydroxyl radicals in the self-cleansing capacity of the troposphere. Atmospheric Chemistry and Physics, 4, 2337–2344.

Summary

The following manuscript provides a review of the current state of retrieving OH from space, and provides recommendations for going forward. OH is the most critical chemical species in the atmosphere, as it dictates the oxidative capacity, i.e., the atmospheric lifetime before chemical removal of the majority of air pollutants and reactive greenhouse gases. I find no issues with the scientific content or accuracy, and agree with the recommendations, so I recommend publication following addressing the following general comment.

General Comment

In Line 72, the manuscript states that it is not possible to retrieve OH directly via remote sensing, but then later in Line 228 states that such technologies could be developed. As a review paper, I would appreciate a greater summary of the physical reasons that OH spectrometry is unable to be retrieved from passive methods, and how active methods like LIDAR may be developed. Also, is it really hopeless that there are no OH absorption lines that might be used in light of the recent work with IASI for VOC retrievals as discussed in section 2.2?