Review of manuscript acp-2021-939 “Smoke in the river: an AEROCLO-sA case study”

https://doi.org/10.5194/acp-2021-939

General comments:

The authors conducted a multifaceted study of an atmospheric long-distance mass (e.g., gas, aerosol) transport feature referred to as “River of Smoke” over southern African sub-continent. River of smoke refers to coherent bands of smoke (from biomass burnings) spanning hundreds of kilometers in width and extending for a few thousands of kilometers. Authors use a set of space-borne, aircraft-based, and ground-based observations in conjunction with global and mesoscale simulations for the period of the Aerosols, Radiation and Clouds in southern Africa (AEROCLO-sA) campaign in September 2017. The study presents a thorough and comprehensive investigation of combined function of tropical temperate troughs (TTTs) and cut-off lows (CoLs) in the formation of the river of smoke. Authors discuss the role of TTTs and CoLs in this context as a novel concept.

The study looks sound, and the paper reads well. I have a few questions and suggestions for possible improvements before publication.

Specific comments:

1. This paper investigates a mass transport feature referred to as the river of smoke, thus the title “Smoke in the river” is somewhat misleading. Also, it might be useful to convey the key features of the article, the function of TTTs and CoLs in the formation of the river of smoke, in the title. This might be useful for the searchability of the article, as this is a novel and interesting concept. Further, and to a lesser extent, the use of the acronym AEROCLO-sA in the title can be confusing for the unfamiliar reader. I suggest a possible revision of the title.
2. This study makes use of high-resolution (5 km) model simulations with Meso-NH and discusses model output in terms of investigating the atmospheric composition and locating the BBA in accordance with observational data. Given the high-resolution of the model and sharp gradients in BBA on the boundaries of the river of smoke, model transport scheme and local mass conservation can have an impact on the results. It has been shown that the inappropriate choice of mass conservation schemes (e.g., global schemes for local case studies) can result in the erroneous creation/transport of mass locally and the use of local mass conservation schemes (e.g. ILMC) can improve the model performance locally and in the presence of sharp mass/concentration gradients (Sørensen et al., 2013; de Grandpré et al., 2016; Fathi et al., 2021). Please comment on and/or provide reference(s) for the performance of the high-res model setup employed in this study in terms of mass conservation and transport schemes.
3. Considering the impacts of transport and distribution of BBA related aerosols and gases (river of smoke) for example “potential important implications for the radiative and the marine productivity of the region” as mentioned in the introduction section of the article, is there value in estimating the aerosol/gas mass-flux using downwind aircraft measurements (Peischl et al.,2010; Gordon et al., 2015; Fathi et al., 2021). Are Dropsonde and Lidar measurement data useful for mass-flux calculations and has this been attempted in the context of AEROCLO-sA? Please comment.
4. Line 174: At what horizontal spacing and frequency were the dropsondes released? Any relevant limitations?
5. Lines 181-183: Regarding the two flights (F06 and F09), is the choice of flight path important (e.g., counter clockwise, clockwise)? If it is important, please explain why.
6. Line 536: Regarding “… thick clouds embedded in the river of smoke”, are there any possible interactions with BBA in terms of cloud formation (e.g., nucleation)?
7. Lastly, a general comment on article structure. Results and discussions are presented simultaneously in three different sections (3, 4 and 5). It might be useful, in terms of readability, if these were grouped together under a results and discussion section. However, this is just a minor suggestion; the article is clear and easy to follow as is.

Technical corrections:

1. Line 24: revise “temperate tropical trough” to “tropical temperate trough (TTT)” to be consistent with the rest of the text.
2. Line 65: cite the final published version of Gaetani et al., 2021
3. Line 108: TTL or TTT?
4. Line 117: acronym ECMWF is not written out in full anywhere in the manuscript
5. Line 226: Figure 2, subplot details (e.g., contour labels) are very hard to read (small).
6. Line 234: Do you mean Potential Vorticity (PV)?
7. Line 258: revise “… area of interest in under …” to “… area of interest is under …”
8. Line 271: Do you mean Figure3d?
9. Line 274: ~20°W or ~20°E?
10. Line 340: Do you mean “west of the continent”?
11. Lines 388,396: Figure 9, wrong figure panel labels are used. References to Figure 9 are generally confusing. Figure 9 shows a height-distance cross-section while the text makes time and geographical references (e.g., morning, afternoon, northern part of the flight). It would be useful if flight legs (e.g., north, south, west, east) were labeled on the graph along the distance axis and a few reference timestamps were also provided for the same.
12. Line 424: Do you mean Figure S5?
13. Line 481: Despite being in the title, there is no mention of TTT in section 5?
14. Line 492: Revise to “The time-height cross-section of PV …”
15. Line 525: Please provide figure/text reference for “… the airborne lidar measurements in the area of Windpoort.”
16. Line 529: Please provide figure/text reference for “…, in accordance with the lidar observations”
17. Line 571: Figure 15 was never mentioned before this line, it is not common to introduce new figures in the conclusion section.
18. Figure 4: End of first line in the caption, do you mean “… at 1200 UTC”?
19. Figure 5: Second last line in the caption, do you mean “… geopotential at 300 hPa is a tidal effect…”? The periodic trend is observable in the 300 hPa curve (blue) more prominently!
20. Figure 6: Panels don't seem to have labels (a,b,c,d,e)
21. Figure 8: Details are hard to read.
22. Figure 9: Panel labels don't match the figure caption and the article text.
23. Figure 11: Why are wind data missing?
24. Figure 13: No panel labels provided?
25. Figure 14: No scale is provided for the contours, maybe label the contours!
26. Figure S5: Repeated panel labels (a,b) in the right column!
27. Line 744: The hyperlink doesn’t seem to be valid.

References

de Grandpré, J., Tanguay, M., Qaddouri, A., Zerroukat, M., and McLinden, C. A.: Semi-Lagrangian Advection of Stratospheric Ozone on a Yin–Yang Grid System, Mon. Weather Rev., 144, 1035–1050, https://doi.org/10.1175/MWR-D-15-0142.1, 2016.

Fathi, S., Gordon, M., Makar, P. A., Akingunola, A., Darlington, A., Liggio, J., Hayden, K., and Li, S.-M.: Evaluating the impact of storage-and-release on aircraft-based mass-balance methodology using a regional air-quality model, Atmos. Chem. Phys., 21, 15461–15491, https://doi.org/10.5194/acp-21-15461-2021 , 2021.

Gordon, M., Li, S.-M., Staebler, R., Darlington, A., Hayden, K., O'Brien, J., and Wolde, M.: Determining air pollutant emission rates based on mass balance using airborne measurement data over the Alberta oil sands operations, Atmos. Meas. Tech., 8, 3745–3765, https://doi.org/10.5194/amt-8-3745-2015, 2015.

Peischl, J., Ryerson, T. B., Holloway, J. S., Parrish, D. D., Trainer, M., Frost, G. J., Aikin, K. C., Brown, S. S., Dubé, W. P., Stark, H., and Fehsenfeld, F. C.: A top-down analysis of emissions from selected Texas power plants during TexAQS 2000 and 2006, J. Geophys. Res.-Atmos., 115, D16303, https://doi.org/10.1029/2009JD013527, 2010.

Sørensen, B., Kaas, E., and Korsholm, U. S.: A mass-conserving and multi-tracer efficient transport scheme in the online integrated Enviro-HIRLAM model, Geosci. Model Dev., 6, 1029–1042, https://doi.org/10.5194/gmd-6-1029-2013, 2013.

General comment:

The manuscript entitled “Smoke in the river: an AEROCLO-sA case study” written by Cyrille Flament presented the formation of a river of smoke over south Africa found in AEROCLO-sA campaign. Based on full dataset of reanalyses data, numerical modeling, ground-based, airborne, and space-borne measurements, this study suggested the interaction between temperate tropical trough (TTT) and cut-off low (CoL) to promote the transport of biomass burning aerosols. This kind of study is essential to interpret “smoke in the river” and I would like to consider the possible publication. However, I have fundamental questions on the numerical modeling used in this study. I would like to request to address my concerns listed below.

Major comments:

Description of Meso-NH: The current description of Meso-NH includes ambiguous statement. Please clarify following specific points.

• Why high resolution simulation is needed in this study?
• For the analysis from 2 September, there is only one day spin-up time. Is it appropriate to adequately remove the initial condition? (Even in Figure 14, 1 September is plotted but is it appropriate?) Related to this point, there is no description for the initial condition data. What kind of initial condition was assumed?
• The proxy for BBA are used by organic matter taken from CAMS whereas black carbon is analyzed from Meso-NH. I guess that black carbon is an important proxy to represent BBA, but the purpose of using black carbon is passive tracer?
• Only GFED emission is used to calculate Meso-NH chemistry? Even though the biomass burning is dominant emission sources to this analysis, available emissions of anthropogenic source and biogenic source are needed to be considered to represent the chemical field over modeling domain.
• As stated, GFED emission’s grid resolution is 0.25 degree and this is approximately five times greater than the grid resolution of Meso-NH. How to interpolate into 5 km horizontal spacing? Without the fine-scale representation of emission itself, what is the advantage to conduct fine-scale Meso-NH simulation?
• In addition, because the treatment of vertical allocation is an important aspect to describe biomass burning emission sources, the information of the vertical grid allocation is required to understand the modeling treatment.

After the clarification of these questions on model configuration, I am wondering the performance of Meso-NH itself. There is no direct comparison for modeled black carbon. In Figure 9, model were indirectly compared to measured attenuated backscatter coefficient. Despite the discussion in line 396-404, I simply impressed that model posed much mixing of black carbon from the surface to 6 km. Observed high coefficients were only found in 3-6 km, and I am suspicious the modeling skill to capture the measurement. Without the appropriate modeling performance, results drawn from Meso-NH model could be also suspicious. I would like to request to include more discussion to reinforce the modeling performance by Meso-NNH to represent the behavior of black carbon.

Specific comments:

1. Line 24: This is not consistent to line 81. Please confirm.
2. Line 94: The campaign period was August-September, and the analyzed event was early September. In my understanding, this analyzed period could be regarded as late winter to early spring and not to fit winter. Is this contradict to the statement in Line 91-92?
3. Line 108: There is no definition for “TTL”. Please clarify.
4. Line 150: If authors use four-digit as HHHH, “000 UTC” should be “0000 UTC”.
5. Line 169: Please confirm the wavelength of AERONET dataset. If 500 nm, this is slightly different to calculated AOT by Meso-NH. How can we understand the difference of wavelength in the comparison between calculation and observation?
6. Line 389: Does “morning” mean small distance because this flight started from 0736 UTC? Sometimes it is ambiguous to use time using this Figure 9, so it would be better to add another x-axis represented by time.
7. Line 571-572: I understand that Figure 15 summarizes and illustrated the finding in this study. However, this should be fully discussed before the conclusion section. Please move these discussions related to Figure 15 into Section 5.
8. Figure 2: The characters for mean sea pressure level is small. Please enlarge, or use color-scale to distinguish them.
9. Figures 3 and 4: The expression of data stated in the caption should be unified through manuscript. “1 September” is used in Figure 2, but “01/09” is used in Figure 3. These are hard to read.
10. Figure 6: There is no indexes to represent (a) to (f) within this figure. In the caption, MODIS is repeated. Please rephrase.
11. Figure 8: Please enhance the black color. This seems as gray color and hard to read.
12. Figures 10 and 12: Because four panels are not unified as altitude levels, it is hard to follow the meaning of this figure. It is much straightforward to align as 1-2, 2-3, 4-5, 6-7 km (or vice versa). Please reconsider the order of panel and also rearrange the discussion main text.
13. Figure 13: There is no indexes to represent (a) to (d) within this figure. Please move the panel of vertical velocity because only this panel is slightly differently positioned.
14. Figure 14: What is the contours level for liquid water (blue) and potential vorticity (brown)?