This manuscript reviews the AerChemMIP effort and procedures, some of the science resulting from the work, and concludes with recommendations for the next round of such a MIP. It is valuable and appropriate to reflect on this community chemistry-climate endeavour within ACP and the authors are certainly qualified to do so. It seems like an opinion article may have a different standard of peer review than a research article. For example, not every statement needs to be referenced or supported, as it may represent an opinion, not a fact. I didn’t find any specific guidelines on the ACP website in this regard, so I have relied on my own interpretation. Thus, I focus primarily on the scope and readability of the manuscript in my review. I have a few overall suggestions that I believe would improve the manuscript in this regard:

1. The title does not seem to really capture the content of this article. It rather seems a review of the approaches used in AerChemMIP (which does include some sprinkled opinions) and then a set of recommendations for the future (which is wholly opinion). I did not truly see much on “the impact” of the research, beyond a quantification of number of papers and citations and the general use in IPCC. Perhaps the authors might consider modifying the title to better reflect the content and goals of the manuscript, with more of an emphasis on process and outputs of AerChemMIP?
2. The manuscript as currently written includes quite a bit of jargon and an expectation that the reader has a good grasp on the chemistry-climate modeling protocols and procedures. This doesn’t seem appropriate for an article targeting the entire Atmospheric Chemistry and Physics community. I would recommend that the authors insert a section after the AerChemMIP objectives to describe the methods and experiments of AerChemMIP (and then contrast/compare to other MIPs). This would include a consolidation of existing text (lines 105-137, 146-162, 174-184,260-264, 289-302, 337-348, 359-367, 395-397, 736-741) and the table on page 8. It would be particularly helpful if the authors could include some sort of schematic to describe the different types of simulations (transient, time slice, histSST vs hist, piClim, etc.) and what exactly is meant by the nomenclature used to describe the simulations, as well as the different Tiers of experiments. This would enable the reader to better interpret which simulations were more easily performed, and why some were performed by a limited number of modeling groups.
   1. Examples of jargon that require a bit more explanation in the text: “entry card” (line 30), “PI-to-PD” not defined (line 49), “everything but” (line 127, 132)
   2. Figure 1 x-axis is not easily parsed without more detailed explanation of each experiment or fuller discussion of how to read the experiment names (e.g. with schematic suggested above)
3. The current Section 3 would benefit from some sub-sectioning to improve readability.
4. I recommend that the authors elevate/expand the comments made on lines 600-609. There is a thread through the manuscript about the importance of communication (between MIPs, between modeling teams, etc.). One aspect of communication that I think should be emphasized more prominently is the importance of clearly communicating the key differential capabilities of all of the models included in the MIPs to the outside world (e.g. summary tables prominently featured on the AerChemMIP website). Some of these differences and varying complexities may be discussed in specific AerChemMIP papers to explain identified differences, but the information is not clearly/easily visible to the outside world who may want to use AerChemMIP output (e.g. which models included tropospheric halogen chemistry? included prognostic methane? included aerosol-cloud interactions? included nitrate aerosols? How do models treat aerosol size? etc), especially the more innovative aspects (i.e. advances over previous CMIP rounds). It is critically important that the user community be informed about the major elements that are or are not included in a given simulation, so that they don’t infer false equivalence between simulations.

As an aside: such a summary table for AerChemMIP would be a tremendous use to the community if the authors were to put this together in this manuscript!

Minor comments/corrections:

1. Lines 191-194: text is a bit confusing as NOx is also an aerosol precursor, but it’s unclear if changing in nitrogen-containing aerosol also impacted the cloud radiative effects.
2. Line 227: specify if soil NOx also included and if all of biomass burning was treated as “natural”
3. Line 234: “radiative efficiencies” not “forcing efficiencies”
4. Line 351: were NH3 and AVOCs mitigation not included in this scenario?

This manuscript presents a detailed overview of AerChemMIP results and it assesses the added value and advancements in our understanding for the role of aerosols and chemistry in the earth’s climate system with regard to its stated objectives, as well as the contribution to the CMIP6 project. Furthermore, it discusses the challenges, the gaps and the future perspective in the design of the numerical experiments and the related research. Overall, it is well structured and presented although there is space for improvements (e.g. organize section 3 into sub-sections and add  a Table with the description of the chemistry and aerosol set up of the models used for AerChemMIP simulations). I think this manuscript is useful and thoughtful contribution on the topic of global scale climate and air quality research. I suggest acceptance of the manuscript for publication but I have a few minor comments to be considered before publication.

Comments

Line 49: Define acronym for PI and PD and first place it shows in the text.

Lines 50-51: The use of the word "similar" could be somehow misleading. The combined effect of ERF from CH4 and O3 is roughly half of that of CO2. Maybe the word “comparable” fits better.

Line 79: “All model bar one …”. Please refer in parenthesis which model is excluded. 

Lines 126-127: Please modify the sentence to show more clearly the difference between RFMIP and AerChemMIP approach in transient experiments.

Line 126: Maybe it would be helpful for a non-expert reader to describe shortly at this point what is the “everything but” approach applied in AerChemMIP.

Line 138, Section 3: After the overview of the simulations given up to line 184, it would be helpful for the reader to follow if the section is organised into sub-sections in terms of the kind of simulations (piClim-X, hist, ssp).

Table 1: The Table 1 caption is missing. Please add.

Line 165: I would rather suggest “cumulatively” instead of “together”

Line 169-170: I would rather suggest “The work added value in a) quantification and improved understanding of model biases (Morgenstern et al., 2020), b) surface ozone biases (Liu et al., 2022), c) historical temperature biases (Zhang et al., 2021a) and d) the development of emulators (ref?). Please add also a reference for the development of emulators.

Line 220: “large low bias” is rather confusing wording. Please clarify or modify accordingly.

Lines 221-222: Please check is the sentence needs rephrasing.

Line 260: Concerning the calculation of ERF from the transient historical prescribed SST (histSST) experiments you may consider the consistency in the comparison of aerosol ERF and its components (ARI, ACI, Albedo) among piClim (aer vs control) and histSST (histSST vs hiSSTpiAer) experiments pointed out by Kalisoras et al., 2024 (already included in the reference list).

Line 337: You may consider referring in the CMIP6 future projection studies some recent ones assessing air quality health impacts under different SSP experiments, e.g. Turnock et al., 2023 (https://doi.org/10.1029/2023GH000812) and Akritidis et al., 2024 (https://iopscience.iop.org/article/10.1088/1748-9326/ad2162).

Lines 356-357: Please clarify where this number (30-50% reduction) comes from? Is it based on Figure 1 of Wilcox et al., 2023? Why SSP3-7.0-lowNTCF is compared with SSP2-4.5 and not with SSP3-7.0? I guess maybe a reader would rather expect a straightforward comparison in the SSP3_7.0 scenario with additional air quality measures of non-methane NTCFs.

 Line 387: Is Figure 4 based on the analysis in the work of Allen et al. (2021)? If this is the case, I would suggest to mention this.

Lines 395-397: As I pointed earlier it would be helpful for the reader to split section 3 into sub-sections in terms of the kind of simulations (piClim-X, hist, ssp). I suggest moving this sentence in the sub-section discussing hist experiments.

Lines 398-408: Following my previous comment I suggest moving this paragraph into hist or piclim sub-section.

Line 459: Please define the acronym ECLIPSE (Evaluating the Climate and Air Quality Impacts of Short-Lived Pollutants) and provide a reference.

Line 465 (Section 4.4): I would suggest to add a Table (even as supplementary material) with the description of the chemistry and aerosol set up for the 11 models used for AerChemMIP simulations. This would provide a synoptic overview of model differences and characteristics with regard to their chemistry and aerosol set up.

Lines 511-513: I would suggest to split the sentence and state clearly that the use of ozonepause may be advantageous to perform whole atmosphere evaluation and assessment in the future.