Detailed comments can be found in PDF file.

Review of 

"Retrieving UV-VIS Spectral Single-scattering Albedo of Absorbing Aerosols above Clouds from Synergy of ORACLES Airborne and A-train Sensors" by Hiren Jethva et al.

This paper introduces a synergy algorithm to improve the retrieval of SSA of aerosols over clouds, which is dearly needed to improve estimates of aerosol direct effects and aerosol forcings. The method is based on a 'color ratio' approach which relies on the spectral absorption of aerosols in the UV and visible, which is true mostly for smoke. Using two spectral measurements, two pieces of information can be retrieved, like AOD and COD when assuming the spectral properties of the aerosols. In the current paper, a method is proposed to instead constrain the AOD above the cloud and retrieve the COD and SSA of the aerosols, which can be done if independent above-cloud AOD is available, in this case using airborne measurements from three ORACLES campaigns over the tropical Atlantic. This area is interesting for the smoke over persisting stratocumulus, providing an obvious focus for the application provided analyses.

The presented approach is interesting since SSA retrievals, especially over clouds, are scarce. The application of the ORACLES aircraft campaign data is an interesting approach to test the satellite retrievals with a proper constraint on the aerosol properties, which is almost always missing. However, there are some major issues related to the presentation of the work.

First, the introduction is quite extensive, but it reads like an advertisement for the color ratio approach, while many pioneering work has been neglected. 

Some examples: 

l.55-56: retrieval of elevated aerosol from satellite: add Hsu, et al (2003)

l 77. Satellite observations have unambiguously shown the presence of absorbing aerosols above clouds over several regions of the world on a monthly to seasonal scale:  add Peers, et al (2019).

l. 533. The magnitudes of bias in the apparent COD depend on the strength of aerosol absorption and backscattering as well as on the actual value of COD (Jethva et al., 2018). The proper reference is: Haywood, et al. (2004)

Figure 1 and l.133-147. The physical basis has been explained in the original papers on the color ratio (as amply referenced in the introduction) and are rather illegibly repeated in Figure 1 (the six panels are difficult to read). 

l. 122. "this kind of situation produces a strong ‘color ratio’ effect, which can be seen in the TOA measurements made by satellite sensors such as OMI (Torres et al., 2012) and MODIS (Jethva et al., 2013).": Actually shown in TOA reflectances from OMI and MODIS combined in De Graaf, et al. (2019). 

Some more examples referencing own work 

l 530. "The lofted layers of absorbing aerosols over the clouds attenuate light reflected by the cloud top through scattering and absorption. This effect reduces cloud-reflected upwelling UV (Torres et al., 2012) and VIS-NIR radiation (Jethva et al., 2013; Meyer et al., 2015), "

Under the pretence of "benchmarking" ones own RTM: 

l. 366 The model results have been benchmarked against the results published in the literature, as well as used in the operational ACAOD product of OMI (Jethva et al., 2018), research-level ACAOD retrievals from MODIS (Jethva et al., 2013, 2016), DSCOVR-EPIC (Ahn et al., 2021), and S5p-TROPOMI (Torres et al., 2020). 

Repetition of references:

l. 85 These techniques have shown the potential to retrieve ACAOD using measurements from different A-train sensors, including (...), Aura/OMI (Torres et al., 2012), Aqua/MODIS (Jethva et al., 2013; ....), 

l 102. This work builds upon the ‘color ratio’ (CR) technique previously applied to OMI (Torres et al., 2012; Jethva et al., 2018) and MODIS (Jethva et al., 2013, 2016) sensors. 

l. 122. In other words, this kind of situation produces a strong ‘color ratio’ effect, which can be seen in the TOA measurements made by satellite sensors such as OMI (Torres et al., 2012) and MODIS (Jethva et al., 2013).

l. 364. We employ the VLIDORT code (Spurr, 2006) (...) - This should be merged with the info on l. 130. 

. 718 - l.724: For instance, Jethva et al. (2018), through theoretical simulations, quantified (...) - Remove from the summary or merge with results in 5.4. 

Second, the manuscript lacks a comparison with existing work. Section 6 is interesting, if it would be compared to what we already know. This could easily be a simulation of the change in retrieved COD as a function of the aerosol absorption (in whatever form, SSA or AEE), which should nog be too difficult with the RTM and LUTS already available. Alternatively, and more interesting would be to compare the actual measurements of the COD retrievals with and without ORACLES constraint aerosol above the cloud to other estimates of the change in COD, like e.g. in Haywood, et al (2004) or Peers, et al (2019). Now we learn nothing new. The main conclusion of the section is this, which should be put in perspective:

l. 556 - 560. "These results are significant and further signify the importance of aerosol absorption above clouds in the UV to VIS-NIR spectral region in at least two ways. First, aerosol absorption above clouds, if not accounted for in the remote sensing inversion, can potentially introduce a negative bias in the retrieved cloud optical depth retrieval, whose magnitude depends on the strength of aerosol 560 absorption (AAOD) and cloud brightness (COD) underneath the aerosol layer. " It would be nice to see how significant actually the results are, by comparing it with the expectations and results from earlier studies.

Also, the sensitivity section is interesting and the amount of work is appreciated, it provides the necessary feel of the accuracy of the proposed method, and AMT is the right place for this kind of information. However, the provided tables of increased or decreased numbers of quantities are tediously repeated from the table into the text, and not compared to anything. It would quite interesting to have a few sensitivity studies from the RTM be compared to the retrievals. Then we could actually see if what is expected is also being found from the satellite measurements when aerosol properties are properly and sufficiently constrained from the aircraft campaigns, or that still more information is needed.

Last, the manuscript lacks a proper motivation of the proposed synergy method. Obviously, the ORACLES data can only be used to test a proper constraint of the aerosol during the flights. CALIOP is mentioned as a replacement of the ACAOD measurements, noticing that (l776.) "CALOP [sic], OMI, and MODIS sensors fly in formation and make measurements within a few minutes of time difference.", in the present tense. The Calipso mission has ended and OMI is hardly producing useful measurements any more. Is the purpose to derive a SSA-over-cloud climatology from the past 10-15 years? Or can this also be applied to existing and future missions?

I feel there is an increased interest in aerosol-cloud-radiation interaction and upcoming mission focus exactly on this topic, like 3MI/Sentinel-5 and EarthCare. The results presented here would be interesting for those missions as well and deserve a discussion, especially in a section named FUTURE IMPLICATIONS.

Some textual issues are listed below and should be addressed, but more importantly some terms and phrases must be revisited:

(1)

l 639. "A strong relative spectral dependence of 20% in the imaginary part of the refractive index between 354 nm and 388 nm, resulting in AAE of 2.45-2.60 "

This should be rephrased in proper physical terms. It is unclear what is meant, but from the context one can expect that the spectral variation of the refractive index is meant, which cannot be "strong". It also does not result in AAE of 2.45, it is expressed in AAE.

Similar arguments hold for:

l. 753. "reduced spectral dependence of absorption". Please, state the reduction in AEE (if this is meant).

l. 757. "stronger absorption strength" Please, rephrase.

(2)

Replace all instances of 'spacetime' with spatiotemporal (or something better). 

Minor issues:

l. 64-67: A better (or additional) reference of the critical cloud fraction is Chand, et al (2009). 

l. 181-184. This information seems out of place for the current paper. 

Section 5.1 should be merged with section 3.1.1. and repetition of information removed.

l. 493: remove "noted"

l. 507. RMSD is not defined.

The AERONET station Mongu is repeatedly mentioned and explained in the text. Remove all repetition. 

l. 607. Start a new paragraph here.

l. 630 lesser errors -> smaller errors.

l. 669 define PSD.

l. 730-735: Merge with the methods section (or simply remove this repetition).

l. 774. one reason -> One reason of the bias 

l. 776 CALOP -> CALIOP

References

de Graaf, M., Tilstra, L. G., and Stammes, P.: Aerosol direct radiative effect over clouds from a synergy of Ozone Monitoring Instrument (OMI) and Moderate Resolution Imaging Spectroradiometer (MODIS) reflectances, Atmos. Meas. Tech., 12, 5119–5135, https://doi.org/10.5194/amt-12-5119-2019, 2019.

Haywood, J. M., Osborne, S. R. and Abel, S. J. (2004) The effect of overlying absorbing aerosol layers on remote sensing retrievals of cloud effective radius and cloud optical depth. Quarterly Journal of the Royal Meteorological Society, 130 (598). pp. 779-800. ISSN 1477-870X

Hsu, N. C., Herman, J. R., and Tsay, S.-C. (2003), Radiative impacts from biomass burning in the presence of clouds during boreal spring in southeast Asia, Geophys. Res. Lett., 30, 1224, doi:10.1029/2002GL016485, 5.

Waquet, F., Riedi, J., Labonnote, L. C., Goloub, P., Cairns, B., Deuzé, J., and Tanré, D.: Aerosol Remote Sensing over Clouds Using A-Train Observations, J. Atmos. Sci., 66, 2468-2480, https://doi.org/10.1175/2009JAS3026.1, 2009.

Peers, F., Francis, P., Fox, C., Abel, S. J., Szpek, K., Cotterell, M. I., Davies, N. W., Langridge, J. M., Meyer, K. G., Platnick, S. E., and Haywood, J. M.: Observation of absorbing aerosols above clouds over the south-east Atlantic Ocean from the geostationary satellite SEVIRI – Part 1: Method description and sensitivity, Atmos. Chem. Phys., 19, 9595–9611, https://doi.org/10.5194/acp-19-9595-2019, 2019.