General comments

The work by Torres et al. aims to demonstrate the feasibility of fully automated sun-photometer measurements aboard ships that meet AERONET standards, ensuring that shipborne data is consistent with existing land-based AERONET observations. This advancement supports the development of a shipborne AERONET-compatible network, addressing current observational gaps in aerosol measurements over remote maritime regions, enabling reliable assessments of aerosol optical depth measurements and other aerosol-related properties. The study details the adaptation of the CIMEL CE318-T Sun photometer for shipborne autonomous operation, and analyses data collected over a three-year period in the southwestern Indian Ocean aboard R.V. Marion Dufresne. The aerosol optical depth is validated through intercomparisons with co-located instruments and the nearby Saint-Denis AERONET site, and the first shipborne quality-assured AERONET aerosol retrievals are presented.

The manuscript is quite dense, with a large amount of background information provided in the early sections. While this provides useful context, it would benefit from streamlining and improved organization to enhance clarity and readability. 

Additionally, I suggest including a systematic cost-benefit analysis (space and power requirements, maintenance demands, personnel needs, etc.) and explicitly address the operational feasibility of broader deployment, useful to assess scalability.

Overall, I consider the study to be scientifically significant and well-aligned with the scope and objectives of Atmospheric Measurement Techniques (AMT), therefore I recommend publication after minor revisions.

Specific comments

The introduction is generally well-written, effectively establishing the scientific context and motivation for the study. However, it presents a lot of foundational information before transitioning to the study’s focus. A more direct introduction to the specific objectives of the research would make the introduction more engaging and accessible to a broader audience. For example, the detailed discussion of system configurations and preliminary tests aboard various research vessels, while informative, could be more effectively integrated later in the manuscript. Consider summarizing this content in the introduction and relocating the technical details to either subsequent sections or a new dedicated section. Also, the final part of the introduction could benefit from a more explicit presentation of the research objectives.

In section 2.2, I suggest improving the structure by breaking it down into clearer subsections corresponding to key stages (e.g., calibration, cloud screening, and quality control) to enhance readability. Additionally, consider reducing redundancy, as some of the information presented in section 2, overlaps with content in the Introduction.

In Section 3.1, the AOD averages presented appear to correspond to the complete dataset of valid measurements. In Section 3.2, it is stated that from April to June 2023, the R.V. Marion Dufresne was operating along the Brazilian coast. Were these data included in the averages shown in Tables 1 and 2? If so, the discussion regarding comparisons of average AOD conditions over the Indian Ocean with other studies should be revised, or the averages in the tables recalculated to exclude these data. Additionally, were there any other periods during the campaign when the vessel operated in regions outside the Indian Ocean? Please clarify this.

Section 3.2 could benefit from further use of subsections. For instance, it could be subdivided into parts that separately present the instrument consistency analysis (intercomparison between sun photometers #1273 and #1243) and the comparison with the Saint-Denis AERONET site.

The systematic cost-benefit analysis mentioned above could be included in section 5. Additionally, detailing the unique challenges associated with shipborne sun-photometer measurements in maritime environments would enhance this section.

Technical corrections

- Lines 164 to 168: there’s some problem with the text here, please check and correct.

- The Ångström exponent is mentioned several times, but the wavelength range used for its calculation is not clearly stated. I infer that it is 440–870 nm, as in the MAN dataset, but this should be explicitly specified in the manuscript. Consider including this information also in Tables 1 and 2, in the same way the AOD wavelengths are indicated.

This work presents the results of what can be considered a great step forward in the use of photometry for the study and monitoring of atmospheric aerosol, extending to the marine domain what until now has been a technique possible only in fixed locations.

The work is very well written and clear, also from the point of view of English.

My only concerns are the following:

- the introduction seems a bit too long containing perhaps too many details on the description of aerosols, their role in the climate, on measurement techniques. Furthermore, in the second part the history of the development of this system suitable for operating on ships is reconstructed. This information is then repeated in section 2 (where it is certainly more suitable)

- section 2.2 reports in too much detail what is the standard analysis procedure of AERONET. I think it is enough to refer to the article that is in fact cited several times (Giles et al. 2019)

- the results are illustrated in detail in section 3. It is therefore not clear to me the usefulness of repeating them in section 5

Specific comments:

L27 Which is the meaning of specifying the percentage of international waters?

L209 Not only the Sun, but also the Moon, so I would say “locked onto the target”

L222-223 What about sea spray?

L230-231 It’s not clear to me what this means “SUN, MOON, (Sun and Moon direct measurements)”. Maybe the second comma is not necessary?

L234 “identical to those applied at regular fixed ground-based sites”. This concept has been repeated many times.

L231-234 120+220 days doesn’t make a full year

L264 Why 5% percentile is not reported as you did with 95%?

L401-407 Could you give more details or put a reference on the evaluation of indetermination of Angstrom Exponent at very low AOD?

L425-432 Is not clear to me which is the meaning of give all this numerical details (e.g. different percentiles) on the AOD and AE values during the BB event

L482-484 Could you provide a reference for this?

L517 How is calculated this bias? I don’t find it

L538 “aerosol retrieval” sounds too generic in my opinion. Maybe you can “optical-physical”?

L596 “exceeding the 95% percentile value of 1.46” of the total period?

L595-598 In these sentences there are repetitions of AOD and SZA values.

Typos:

L67 analyzes

L94 platformS

 L164 andber, imprimproved greatlypared

L167 for all alls

L170-171 “This solution was shown inefficient for our proposed automated independent final solution”. Repetition

L186 I would suggest “identified” instead of “discovered”

L190 I would write “the first instrument fully compatible with AERONET” or “the first fully AERONET compatible instrument”

L192 “marking a significant milestone in achieving 100% AERONET compatibility” is a repetition of what stated at line 190

The study performed by B. Torres et al. presents three years of solar and lunar AOD measurements aboard a research vessel in the area of the Indian Ocean, using a Cimel CE318T automatic sunphotometer, and following the standard procedure kept by the Aerosol Robotic Network (AERONET). The analysis also includes for the first time the measurements of sky radiance, performed in the almucantar and also using hybrid escenarios. The results show good performance, comparable to standard measurements taken at ground sites from AERONET. It is therefore considered kind of foundational paper for the future establishment of a network of instruments deployed at vessels. This is an important obljective given the huge gap of data found in vast oceans.  The text also points at future further developments in order to improve current limitations.

The results are of scientific interest, well within the scope of the journal. The English usage is very good to my understanding, and it has been written and composed with care. However, I would recommend some major changes (on the structure mainly) before its acceptance. 

General comments:

The introduction is interesting and informative. However, I think many paragraphs should be moved to section 2. In fact, some of the information is redundant in section 2. Please, keep the introduction shorter, and integrate the removed paragraphs in section 2. I do not recomment elimination of information, but integration in the next section. 

I also think that section 5 (discussion) should be integrated in section 4 (results). In fact, the initial paragraphs in section 5 are redundant again. By moving section 5 to 4 you can eliminate them.

It would be ilustrative to include an image of the new system and the platform in section 2.

Similar ship version developments of Prede POM instruments were tested in japanese R.V. Shirase. It would be interesting to cite as an example in the introduction. See for example Kobayashi et al. (DOI: 10.1117/12.2195691).

Specific comments and corrections:

- Line 10: Angström exponent is writte differently in the text. Please correct them accordingly (for example, it appears incorrectly at line 10, 100, table 1 and 2)

- Line 54: I would say that "preindustrial" conditions is not the best term to use, as the earth is already affected by anthropogenic emissions, even in remote areas. Maybe using remote oceanic conditions, or natural background conditions would suit better?

- Line 101: Does AERONET use least-sqyares method over 440-870 nm wavelength range? Can you confirm? I thought the Angstrom exponent was calculated by using rations of channels 440 and 870 nm. 

- Line 155: attempts

- Lines 164-167: there a series of typos and words sticked together that look caused by editor software problems: andber, imprimproved, greatlypared, shoshowing, squaredferences, all alls...

- Line 182-183: revise the sentence please.

- Line 194: Why not using a different name for the version of Cimel CE318T? 

- Line 398: characteristic for the whole indian ocean or only SW?

- Figure 3: why not merging together the two plots? Is there a problem in readibility?

- Line 484: Please add a reference for last sentence.  

- Line 494-499: this paragraph is redundant. 

- Line 503: state kilometers as km, as done in other appearances.

- Line 529: It is a pity not to include a period in which the instrument was installed at ground in the port to estimate the differences with Saint-Denis, so effect of the vessel could be removed. 

- Figure 6: The black cross symbol is difficult to find...

- Figure 7: The two M.D.Alm colors used for left figure are somewhat difficult to distinguish, mainly in printed material.

- Line 666-668: please review the sentence.

- Line 696-697: Repeated sentence

- Line 716: SSA already defined in 625 

- Line 760: extra dot

- Footnote 9: It would be interesting to add the corresponding histograms for comparison. 

- Line 835-836: repeated