Overall Notes

This paper introduces a python tool aimed at associating in situ aerosol measurements obtained during aircraft field campaigns with remote sensing aerosol retrievals, by addressing the need to compensate for the limited coarse-mode throughput of aircraft inlets and to hydrate in situ samples when comparing with ambient (remote sensing) observations. 

As these calculations must be made if in situ field data are to be used to validate remote sensing retrievals quantitatively, the algorithm presented here represents a useful tool for such applications.  The observations used to test this approach were acquired during the ACTIVATE field campaign, and the scope of the present study is limited to fine-mode sulfate and organic aerosol, and a coarse mode taken as sea salt. They impose assumptions that limit considerably the applicability of the current implementation – constant refractive indices over the spectral range, spherical particles, parameter assumptions required to calculate the hydrated CRI, etc. However, these are stated clearly, which is as much as one can ask in an AMT paper. The remote sensing data were obtained from the HSRL-2 and RSP aircraft instruments, which avoids some of the sampling differences that arise when in situ measurements are compared with spacecraft measurements.  As such, the approach seems most applicable for validating aircraft field measurements.

In summary, the paper develops a useful tool and presents a thorough analysis of its performance. For general application, there are significant limitations in the assumptions made, but given that the analysis is circumscribed to a narrow set of relatively favorable conditions, I think this is acceptable for publication in AMT, perhaps with minor modifications as suggested below.

A Few More Specific Notes

Line 244.  It might be useful to mention how far the aircraft travels in 45 seconds, to provide a sense for the horizontal resolution of the SMPS and other, aggregated measurements.

Line 270.  Might be worth noting that remote sensing is more sensitive to volume than number concentration specifically for particles smaller than the observing wavelength.  For particles larger than the observing wavelength, sensitivity is greater to particle cross-sectional area.

Line 454.  Comparing Fig. 1 with Fig. 5, and taking the y-axis scales into account, I would say “… overall, much less variance.”

Line 469.  As this is synthetic data, doesn’t the statement here just mean that the numerical coding was done correctly?  Not a bad thing to mention, but the statement here makes the observation sound more fundamental.

Line 494.  A word seems to be missing from this sentence. 

Lines 498-500. By way of explanation, wouldn’t the 700 nm channel likely be the most sensitive to coarse-mode particles, for which many of the assumptions might be less applicable?

Section 3.1.3.  Just wondering how representative of the entire column the in situ data sampling might be.  I realize the HSRL-2 data are height-resolved, which can help assess the vertical heterogeneity compared to the in situ sampling.

Figure 10.  There appears to be a lot of scatter in the data, which the text does not seem to acknowledge. This is probably not surprising – in addition to the limitations discussed in the paragraph about this figure, given the likely horizontal and vertical variability in particle concentration combined with differences in sampling.

This manuscript presents a methodology to compare and reconcile aerosol observations from various platforms, so called ISARA algorithm, developed for the ACTIVATE mission data. It provides a useful and rather comprehensive example of attempting to achieve closure between diverse in-situ and remote sensing measurements. The study clearly addresses a complex problem and contributes relevant information for aerosol measurement harmonization. However, it would benefit from clarifying its broader motivation, better articulating its limitations, and improving the language and structure in some sections.

My main concern is that the manuscript abstract, and introduction do not seem to precisely describe the motivation of the study. The introduction frames the problem mainly as a data closure challenge, potentially giving the impression that a general solution (i.e. ISARA) will be offered. For example the sentence “Despite the important findings from these studies, systematic and streamlined closure of aerosol data sets has not been yet achieved.“  However, the ISARA algorithm appears heavily tuned to specific conditions, relying on a priori information about aerosol composition, size, and shape, additionally being very limited in the atmospheric and aerosol conditions where it can be applied. This is perfectly understandable, but a clearer explanation of how broadly applicable the algorithm is would strengthen the outcome. The introduction could therefore more broadly discuss the complexity of in-situ vs remote-sensing aerosol validations, compilations and simulations – as for setting the scheme. I hope that authors consider this, and could slightly streamline the introduction and structure of the paper to better reflect the content.

In summary, this manuscript addresses a complex and relevant topic in atmospheric aerosol measurement by providing a detailed example of integrating multiple observation platforms. The methodology is simple but thoroughly described, and the analysis is well presented, but the applicability and limitations of the ISARA algorithm require clearer discussion. The manuscript would benefit from a refined motivation and improved language in certain sections. I recommend minor revisions before acceptance at AMT. 

Minor Rather Technical Comments:

Abstract:

• L6: Suggest removing “aircraft” as the methodology can apply to other in-situ data platforms as well. Or can it? Please consider this also when writing introduction.

Introduction:

• L33: Remove the word “parameter.”
• L51: Consider adding 1–2 sentences summarizing key findings from past studies for context.
• L52: Clarify “these instrument data.” Which instruments are being referred to?
• L57: Suggest rephrasing to “...using a Nafion membrane dryer in the sampling line.”
• L86: Check for a possible extra “is.”
• L91: Sentence structure is unclear—both content and grammar could be improved. Also, please clarify if the methodology is expected to be broadly applicable or limited to specific platforms and conditions.

Measurements:

• L198–202: Some grammar and sentence construction issues—please revise for clarity. Also, consider using micrometers (µm) instead of nanometers (nm) for >1 µm sizes.
• L202: Nephelometers and absorption measurements are introduced here without prior explanation. Reorganizing or cross-referencing earlier sections may improve flow.
• L213: Was the PSAP measurement also conducted at <40% RH? Please specify.
• L218: Replace “variety of errors” with a more descriptive phrase identifying specific artifacts or correction needs.
• L223: How much data was excluded by the >1 Mm⁻¹ cutoff for scattering and absorption? Justify the threshold.
• L245: Clarify what is meant by “most useful analysis” of profiles extending >1 km.
• L257: Consider rephrasing “1-second data” as “native time resolution data” for clarity.
• L282, L286, L334: “ACTIVATE region” should be defined more precisely—preferably with coordinates or a map. Also specify which areas or conditions were excluded and why.
• L289: Clarify “mid-point particle diameter.” Why is geometric mean diameter not used?
• L285–295: The assumption of sulfate-only aerosol seems oversimplified. Given the limited RRI range and the potential presence of organics or other compounds, the assumptions behind the ISARA-derived optical properties need better justification. Also, specify how wavelength conversions were done and what values of AAE or SAE were used.
• L326: Missing preposition—please review.
• L334: Again, clarify the threshold inconsistency between L223 and here. How much absorption data were actually usable above 1 Mm⁻¹?
• L615: The manuscript states that the region was chosen for “diversity of aerosol and meteorological conditions,” yet the assumptions made to enable closure seem to contradict this. Consider rephrasing or qualifying this claim.

The paper by Dmitrovic et al describes a python toolkit (ISARA) aimed at achieving closure between lidar and polarimeter observations of ambient aerosol properties on one hand, and in-situ measurements of the PSD, scattering and absroption on the other hand, which are limited by drying and by inlet effects. The toolkit is validated through in-situ observations (internal consistency), simulations (synthetic consistency) and remote sensing observations (external consistency).

It is my feeling that the stated goals of the paper are of high importance, but that they are not sufficiently developed beyond the technical stage to be proposed for a publication. I would invite the authors to consider a more scientific and less technical approach in writing this paper, and to allow the time and the effort that this can represent at this stage I suggest to REJECT the manuscript. I will be willing to help with reviewing a revised manuscript in case the the below MAJOR points are addressed in full.

MAJOR POINTS:

1) The paper takes a technical stance to the task, e.g. insisting on unnecessary details such as the use of the python programming language and the MOPSMAP scattering code. I would invite to present an algorithm, and to discuss its assumptions, limitations and uncertainties, rather than a piece of software. Concerning the scattering calculations, given that the authors only assume spherical particles, MOPSMAP is only a library allowing Mie scattering calculations and any other Mie scattering code would achieve the same results, therefore the scientific approach is to indicate that the framework of Mie theory is being used, and not to indicate which scattering library is used. The usage of the MOPSMAP code could most probably be of interest in the case of non-spherical scattering, but non-spherical scattering is not envisaged here.

2) There has been previous literature attempting the same task, and this needs to be accounted for and mentioned. I bring here for example the IRRA approach of Tsekeri et al (AMT, 2017). I am sure that by searching the literature the authors can also find other references on the same topic. The authors should explain how their paper fits within the existing literature, and how similar or different it is from other approaches, indicating pros and cons of each. Note that "previous works" are also cited at line 547, but no references are given.

3) The stated goal is a "rigorous external closure" (see line 4). I find this to be a real contradiction with the state-of-the-art achieved by the paper itself, given that it is a "preliminary consistency" and a "preliminary effort" (see lines 19 and 72). This may question whether more work is needed before a paper is submitted to AMT. The sentence at lines 546-548 seems to confirm this impression.

4) The computations are done for a specific atmospheric scenario only, with a mixture of sulfate and organics in the fine mode and sea salt in the coarse mode (lines 104-109 with a clear mention that "this will limit the scope of this study"). Whereas this is fine in itself, it removes generality away from the ISARA approach and from its stated goal of a generally applicable approach. It is important to state the limited scope from the onset of the paper, rather than add the limitation further down the line. Moreover, given that ACTIVATE operated also in Bermuda (line 136) why is the influence of dust aerosols not also investigated?

5) After clarifying the issues affecting in-situ observations, the authors claim (lines 89-90) that "the ISARA attempts to overcome these limitations by estimating the contribution of coarse-mode particles". It is unclear how this can be achieved, and I would state that this is not possible without a dedicated measurement for the coarse particles (e.g. using an open-path OPC). A CDP is later mentioned at lines 226-234, therefore the problem may perhaps simply be of being explicit and honest frpm the onset about the fact of using a dedicated measurement for the coarse mode.

6) Figure 9: ISARA shows underestimation. Figure 10: low level of closure. This does not look good for the ISARA method.

7) The sentence "a potential reason for detecting low-RH aerosol is the presence of smoke from fires" (lines 588-589) is in contradiction with the assumed marine environment, but is also presented in a way that makes it appear unjustified. Please note that water vapour is a combustion product, hence fires should contribute to increasing absolute humidity and not to decrease it. The whole narrative at lines 590+ appears incoherent: note that figure 13 shows fires and not smoke aerosols as stated here. It is suprising also to read that aerosols "are becoming drier due to colder temperatures" given that lowering the temperature (with constant WVMR) has the effect of increasing the RH. I agree with the sentence on line 593 that "further work will be done" and I feel that it must be done before re-submitting the paper.

8) Overall good agreement (line 602) feels an unjustified statement here. The same applies to the "successful retrieval" (line 622). The evidence is not so clear-cut.

MINOR COMMENTS:

9) "In-situ instruments cannot efficiently sample coarse-mode particles due to limitations in the inlet cutoff diameter" (line 65). This statement is incorrect given that there are several open-path airborne instruments such as the CDP, the CAPS, etc. that overcome these limitations (initially developed for cloud particles and later extended to use in aerosol layers). It is well-known that the FAAM research aircraft has successfully sampled coarse and giant dust particles up to 300 um diameter with such probes (see e.g. Ryder et al, ACP, 2015; Marenco et al, ACP, 2018).

10) The definition of the fine-mode (0.09-1 um dry diameter) and coarse-mode (> 1 um ambient diameter) regimes (lines 83-84) is weird given that there could be an overlapping zone between fine and coarse (particles with dry diameter < 1 um and ambient diameter > 1 um would be in both regimes). I would suggest to consistently refer to either the dry or the ambient diameter for discriminating the two modes.

11) Line 6: the symbol \kapa is undefined. At the end of page 14, the relationship between \kapa and f(RH) must be explained.

12) Joint flights (line 135): it is unclear which two airplanes performed joint flights.

13) "The novel vertically-resolved aerosol particle number concentration" (line 184): I suggest to add the word "estimate" to this statements, given that number concentration from lidar is estimated based on assumptions.

14) Inlet cutoff at 5,000 nm (line 199): indicate how the cutoff diameter was determined.

15) For random variables such as the scale factor (line 365), the \kapa (line 367) and IRI (line 370), the random distribution used must be given.

16) Nephelometers (line 202) are being mentioned before being introduced (line 213 and following). I suggest that the full instrument set must be introduced before discussing their details such as installing a cyclone.

17) equation 1 using RH of 20 and 80% is inconsistent with having observations at 40 and 85% (line 215). Please clarify.

18) Line 248: "the methods standardized by the merging tool". It is unclear what the authors refer to.

19) Symbols: there is an unclear use of symbols N_a and n_0 (seeming to refer to the same variable). Also, in lines 293 and 294 C_calc refers to different variables, hence I suggest using different symbols.

20) Equations 3 and 7: the bounds of the integral should be log(Dmin) and log(D_max), and not dlog(Dmin) and dlog(Dmax).

21) Line 270: remote sensing sensors are normally considered mor sensitive to the surface area and not the volume. Please correct.

22) Stitched data (line 279): this must be documented.

23) Full range of particle sizes (line 279): this is incompatible with the inlet cutoff. Please clarify.

24) Figure 1 caption: I suggest to use percentiles instead of minimum and maximum to reduce the influence of outliers.

25) CRI averaging (line 291): it is unclear over which dataset the averaging is being done.

26) Co-location (line 401). As observations are done on-noard the same platform, they are certainly co-located, therefore this is unclear.

27) Line 404: I suppose that data are removed when the cost function is larger than a threshold, not lower. Please correct.

28) "coarse mode AOD is limited to < 0.1": why?

29) LDR (line 407): clarify if you refer to VLDR or PLDR, because they bare not the same. The threshold at 0.13 (line 410) is not so small, at least for VLDR, and could indicate dust presence.

30) Successful retrieval rate (line 482): please clarify if the retrieval rate means that a solution is found, or that the solution found is close to the observations).

31) Lines 483-486. The authors raise an important limitation and they should perform a more in-depth analysis to disentangle the causes. Note that for measurement noise (mentioned by the authors) this should be possible to address with the current data. Concerning the issu of non-sphericitye, this can be run using MOPSMAP.

32) Acronyms (NRMSD, MRB, NMAD) must be explained when first used.

33) Line 565: specify that you refer to case #10 of table 5, because there are two flights on this date.

34) "are likely organic and sulfate-dominated mixtures": this sentence is unexplained and unsuported by evidence

35) "marine environment" (line 588): This statement seems inconsistent with the Hysplit trajectories showing continental influence.

36) "the data from case 12 are also shown in Table 5" (line 597): why?

37) "1064 m": it is actually 238-4499 m

38) Table 5 caption should read "case studies 10 and 12"