the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Sep 2025
Global transport of stratospheric aerosol produced by Ruang eruption from EarthCARE ATLID, limb-viewing satellites and ground-based lidar observations
Abstract. The Atmospheric LIDar (ATLID) instrument of the ESA’s Earth Cloud, Aerosol and Radiation Explorer (EarthCARE) satellite mission launched in May 2024 provides high-resolution vertical profiling of aerosols and clouds at 355 nm. Fully operational since July 2024, ATLID has been witness to a significant perturbation of stratospheric aerosol budget following the eruptions of Ruang volcano (Indonesia) in late April 2024. Using ATLID together with limb-viewing satellite instruments (OMPS-LP and SAGE III), we quantify the stratospheric aerosol perturbation generated by the Ruang eruption and characterize the global transport of volcanic aerosols. To evaluate the ATLID performance in the stratosphere, its data are compared with collocated ground-based lidar observations at various locations in both hemispheres and overpass-coordinated balloon flights carrying AZOR backscatter sonde. The intercomparison with suborbital observations suggests excellent performance of ATLID in the stratosphere and proves its capacity to accurately resolve fine structures in the vertical distribution of stratospheric aerosols. Using various satellite observations, we show that Ruang’s eruptive sequence in April 2024 produced eruptive columns reaching 25 km altitude, and resulted in a doubling of the tropical stratospheric aerosol abundance for several months. The eruption timing in austral Fall and its high-altitude reach fostered efficient poleward transport into the southern extratropics during austral Winter 2024. By the time of the austral Fall 2025, the sulphate aerosols from Ruang have spread across the entire Southern hemisphere and were most probably entrained by the 2025 Antarctic polar vortex, potentially enhancing the polar stratospheric cloud occurrence.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1951 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1951 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Review for egusphere-2025-4377', Anonymous Referee #1, 12 Oct 2025
- AC1: 'Reply on RC1', Sergey Khaykin, 16 Dec 2025
-
RC2: 'Review of egusphere-2025-4377', Anonymous Referee #2, 13 Nov 2025
This is nice paper that should be published after a few modifications. The one difficult claim, which is not well supported by the discussion or the figures, is that there was transport from the tropics into the Northern Hemisphere along the 750 K isentrope. Here are the detailed comments.
313-315 This claim by the authors is not substantiated by either Figs 3 e) or f). The northward extension of scattering ratios near 1 (purple) are not at all obvious and certainly not along the 750 K contour.
412-414 This seems like a stretch given there was no clear evidence of transport from the tropics along the 750 K isentrope. Could this be a filament from the polar vortex as it dissipates in the spring? Where did the air come from? Were there any back trajectory calculations?
Other comments/corrections.
Figure 1a) b) Where is the triangle marking Ruang?
276 … Antarctic …
280-282 This statement is perhaps reasonable at SAOD > 15e-3, but less than this the difference is a factor of two or more, very obvious in the NH mid latitudes. Might be worth mentioning the regions of particularly larger discrepancies.
281-283 This claim is questionable. Yes there is a faint signal in OMPS-LP, but this is not really supported by either SAGE III or ATLID.
Figure 2 Why does the SAGE III data stop in July. Just not available yet?
319 Probably the reason is because it wasn’t there.
338 Isn’t it a black dotted line?
341 The structure of the zonal mean is not vastly different, it just doesn’t capture the fine scale structure, nor would a zonal mean be so expected.
351 Isn’t it the opposite with the zonal mean exhibiting somewhat lower scattering ratios than the local measurements?
455 Is there a reference for 0.4 Tg of sulfur?
Citation: https://doi.org/10.5194/egusphere-2025-4377-RC2 - AC2: 'Reply on RC2', Sergey Khaykin, 16 Dec 2025
Interactive discussion
Status: closed
-
RC1: 'Review for egusphere-2025-4377', Anonymous Referee #1, 12 Oct 2025
The paper by Khaykin et al. describes the effects of the Ruang volcano eruption on the stratospheric aerosol conditions. It used several data sources and methodologies and describes the event and the dispersion of the associated aerosol plume with time. It furthermore related its impact to previous major events. It is of high scientific interest and therefore in general suited for ACP.
The transport of the volcanic aerosols is in general well described. However, I am struggling with some of the methodologies applied and also some of the conclusions drawn. Partly it appears to be a “business as usual publication” after a new strong volcanic eruption without accounting for novel knowledge and data sources. This needs to be significantly improved before the manuscript can be accepted. Therefore, I recommend major revisions for this work.
I will describe in the following some of my general criticism, followed by more specific comments:
General issues:
Derivation of stratospheric AOD from Lidar observations
The paper applied methodologies which obviously have been applied since a long time to investigate stratospheric aerosol distortions by volcanic aerosol. The scattering ratio of the ground-based and space borne lidars is used together with an assumed lidar ratio of 50 sr to estimate the stratospheric AOD. However, is this methodology still up to date?
First of all, the assumption of a lidar ratio of 50 sr at 355 nm is not discussed by means of any reference. Furthermore, it is in my opinion not justified per se as it assumes the existence of volcanic sulphate aerosol only and neglects any influence of wildfire smoke which was recently observed to be entered frequently in the stratosphere (see e.g. Khaykin et al., 2020; Ansmann 2024, Peterson 2025). Thus, justification for this lidar ratio and an uncertainty estimate for the derived products need to be done and are key for publication.
Second and maybe more important, many of the lidars, especially the novel EarthCARE lidar ATLID which is used intensively, can provide extinction measurements directly. Why don’t you use these products to derive the stratospheric AOD? While the ground based lidar may struggle with low-signal-to-noise ratio, EarthCARE at least is for sure capable to retrieve extinction profiles in the stratosphere. It is not clear to me why the authors use the detour via the scattering ratio and the assumed lidar ratio instead of using the directly measured extinction coefficient profiles.
This is especially important also in view of the other data sources they use, as many provide extinctions measurements (e.g. NOAA-21 OMPS-LP, ISS SAGE III, GloSSAC..….) and not scattering ratio.
Third, the computation of the scattering ratio from EarthCARE Level1b data is not understandable for me. As level 1b signal still suffer from molecular and particular attenuation (therefore attenuated backscatter coefficient), the simple integral of the sum of two signals multiplied with a lidar ratio seem to be not appropriate in my opinion. Please justify and describe more intensively.
Thus, my main recommendations with respect to this topic are: Use direct extinction measurements, provide error estimates (in case this is not possible), discuss and consider always the influence of stratospheric wildfire smoke.
Some other methodological weaknesses are:
The estimated mass of the emitted SO2 is from personal communication only without giving any reference. In my opinion, this is not sufficient as the value is key for Fig. 6 and the respective discussion. Thus, either you describe properly how the mass is estimated or give proper reference. Otherwise, you need to cancel the whole discussion about the impact of the recent eruptions compared to previous ones.
The ground-based instruments and data retrieval descriptions are very heterogenous. Please unify and justify the different methodologies you use for each lidar system. Currently it seems that each ground-based lidar has its own retrieval algorithm with its own assumptions. This might be ok, but needs to be discussed. As for EarthCARE (described above) the same question comes up for many of the derived products from the ground based lidars: How do you account for attenuation of the backscatter when calculating the SAOD from scattering ratio.
Other general issues:
The discussion often (not always) assumes Ruang aerosol to be the only reason for increased aerosol in the stratosphere neglecting other potential sources like wildfires, other volcano eruptions etc… this needs to be discussed better.
Related to this: In Fig. 6, you show major events for stratospheric AOD distortion, but I wonder if it is complete or is missing some of the severe Northern hemispheric fire events having a stratospheric impact (e.g. see in Ohneiser et al., 2023 for the Canadian fires in 2017 or many others as listed in Peterson et al 2025.). You state it by yourself in line 453: “Since 2017, the significant SAOD perturbations, caused by either volcanic eruptions or wildfire outbreaks in both hemispheres, occurred at least once per year, maintaining the global stratospheric aerosol loading well above the background levels.” But I can see only one fire event in the NH in Fig. 6.
Please also broaden your view in the discussion and do not only cite yourself for explanation of certain events. There are many other scientists working on these topics and this would give the paper a broader justification.
Figs 4 and 5: If you would provide extinction profiles from the lidars, you could even compare these to the other data sources you use (e.g. NOAA-21 OMPS-LP, ISS SAGE III, GloSSAC..….) which would value the paper even more.
Many abbreviations are not explained, e.g., JPL, LA, JRA, PNE, AZOR … It is good practise to write the whole name at the first instance.
Specific comments:
- 100: calibrated backscatter profiles is not specific enough, please be more precise, the official nomenclature is “fully processed, calibrated, and geolocated attenuated backscatter signals“ according to ttps://www.esa.int/Applications/Observing_the_Earth/FutureEO/EarthCARE/EarthCARE_data_products
- 108: Again, it is not the Mie backscatter, but the attenuated Mie backscatter (coefficient)
- 109: As written above, the use of a constant lidar ratio of 50 sr need to be justified and uncertainty estimates give unless it is obsolete because you use extinction profile observations.
- 110: The formula used is not clear to me. How do you account for example for the attenuation of the molecular extinction? In Level 1B data there is not a particle backscatter coefficient which but the attenuated one. Can you justify your formula? But maybe I am also wrong.
- 126: If you assume a lidar ratio of 50 sr for Maido as well, at least you cannot validate your assumptions - as you make the same ones - and it is not clear if the retrieved SAOD is valid. But as far as I see you do not use the AOD from the ground-based observations generally?
- 142: “The backscatter coefficient is derived from the backscatter ratio using the 3-hourly atmospheric density output from MERRA-2 interpolated at the site.“ it is not clear for me how this works and which assumptions are made. Please describe more explicitly.
- 162: Why using lidar ratios of 33 to 58? Is this the natural variability? What does this mean for your SAOD retrievals?
- 167: How can you justify the backscatter Angstroem exponent of 1.8 ?
- 192: please describe more explicitly what the application of a “5 point boxcar filter“ means. I.e., what final effective vertical and horizontal and temporal resolution you have.
- 202: Please state which wavelength you use for the comparison.
- 239 ff. Where does this information come from? Any reference?
- 244: What is the extinction ratio? It’s not explained in the methodology. Why not using extinction coefficient profiles?
- 250: “DU“ never explained as many other abbreviations – I assume DU is not a SI unit and thus need to be explained.
- 251: The estimate of the mass of the SO2 plume is given without justification or reference, this is inappropriate given its importance.
- 264: Fig. 1 Date for 1d is wrongly stated in the caption.
- 276: You state that ATLID might have captured PSC, but then the assumed Lidar ratio is not valid anymore, right?
- 283ff: The SAOD perturbation could be also partly affected by stratospheric wildfire smoke - please discuss this. Such events not only occurred in 2025 as exemplarily stated in line 291 ff. Furthermore, can you exclude any other major source of stratospheric aerosol? I guess yes, but you need to discuss this.
- 306: Again, when using ATLID and linking the observations to PSC the used Lidar ratio might not be appropriate. At least, error estimates should be made.
- 310: “strong PSC signal”: Can you exclude other sources for this strong signal? Please discuss.
- 319: “However, in the northern extratropics ATLID shows reduced sensitivity to the volcanic aerosol layer, for which the reason remains to be investigated” – this statement is just an assumption. I would rephrase it to e.g.:”… ATLID shows lower aerosol load compared to OMPS…”.
Concluding on a reduced sensitivity is in my opinion not valid from the figures you show. - 355: “The upper part of ATLID profiles, above ~25 km tends to be noisier and prone to noticeable deviations with respect to the ground-based data within 1-2 km-thick layers”: this is too strong statement. Maybe the ground-based reference is also not seeing some of the aerosol peaks? I recommend to formulate it more neutral. At least I cannot find evidence for the statement in the figure.
- 385: “remnants of aerosols produced by the Hunga eruption in January 2022,” à or other aerosols like smoke….
- 386: Why can you undoubtedly link this to Ruang aerosol?
- 394-395: Why can the lower aerosol layer be linked to Ruang and the upper one to Hunga volcano? Please explain in more detail.
- 398: What do you mean with regional feature? Can you describe in more detail?
- Sec 3.4 The balloon launched are temporally and geographically partly very close to OHP observations. Maybe you can use this to discuss? E.g., Figure 5 h and i.
- 412: “upper enhancement around 26 km may be the result of the poleward transport of the volcanic aerosols from the tropics“ what other volcanic aerosols? Please discuss.
- 460: Based on this statement (factor 1.8 increase), can you exclude any other event contributing (other volcanos, wildfires)? Please discuss.
- 464: The introduction of ACE comes a bit out of nothing. Here some more explanation is needed about this experiment and what was done.
- 475: Ansmann et al 2022, Ohneiser et al 2022, and Solomon et al. 2023, discussed the influence of smoke on the ozone hole. You should mention this as well in view of your future investigations. As marked in the general statements, the disentangling of smoke and volcanic aerosol is still not clear and need to be discussed as well
- 480: “In Spring…“: one has the feeling the discussion is still wrt northern hemisphere. If so, the reference to the figures is not appropriate. If this is not the case, the sentence should be rephrased.
References:
- Ansmann, A., Veselovskii, I., Ohneiser, K., & Chudnovsky, A. (2024). Comment on “stratospheric aerosol composition observed by the atmospheric chemistry experiment following the 2019 Raikoke eruption” by Boone et al. Journal of Geophysical Research: Atmospheres, 129, e2022JD038080. https://doi.org/10.1029/2022JD038080
- Ansmann, A., Ohneiser, K., Chudnovsky, A., Knopf, D. A., Eloranta, E. W., Villanueva, D., Seifert, P., Radenz, M., Barja, B., Zamorano, F., Jimenez, C., Engelmann, R., Baars, H., Griesche, H., Hofer, J., Althausen, D., Wandinger, U., Ozone depletion in the Arctic and Antarctic stratosphere induced by wildfire smoke, Atmos. Chem. Phys., 22, 11701-11726, doi:10.5194/acp-22-11701-2022
- Khaykin, S., Legras, B., Bucci, S., Sellitto, P., Isaksen, L., Tence, F., Bekki, S., Bourassa, A., Rieger, L., Zawada, D., Jumelet, J., and Godin-Beekmann, S.: The 2019/20 Australian wildfires generated a persistent smoke-charged vortex rising up to 35 km altitude, Commun. Earth Environ., 1, 22, https://doi.org/10.1038/s43247-020-00022-5, 2020.
- Ohneiser, K., Ansmann, A., Kaifler, B., Chudnovsky, A., Barja, B., Knopf, D. A., Kaifler, N., Baars, H., Seifert, P., Villanueva, D., Jimenez, C., Radenz, M., Engelmann, R., Veselovskii, I., Zamorano, F. Australian wildfire smoke in the stratosphere: The decay phase in 2020/2021 and impact on ozone depletion, Atmos. Chem. Phys., 22, 7417–7442, doi:10.5194/acp-22-7417-2022
- Ohneiser, K., Ansmann, A., Witthuhn, J., Deneke, H., Chudnovsky, A., Walter, G., Senf, F. Self-lofting of wildfire smoke in the troposphere and stratosphere: Simulations and space lidar observations, Atmos. Chem. Phys., 23, 2901-2925, doi:10.5194/acp-23-2901-2023
- Peterson, D.A., Berman, M.T., Fromm, M.D. et al. Worldwide inventory reveals the frequency and variability of pyrocumulonimbus and stratospheric smoke plumes during 2013–2023. npj Clim Atmos Sci 8, 325 (2025). https://doi.org/10.1038/s41612-025-01188-5
- Solomon, S., Stone, K., Yu, P. et al. Chlorine activation and enhanced ozone depletion induced by wildfire aerosol. Nature 615, 259–264 (2023). https://doi.org/10.1038/s41586-022-05683-0
Citation: https://doi.org/10.5194/egusphere-2025-4377-RC1 - AC1: 'Reply on RC1', Sergey Khaykin, 16 Dec 2025
-
RC2: 'Review of egusphere-2025-4377', Anonymous Referee #2, 13 Nov 2025
This is nice paper that should be published after a few modifications. The one difficult claim, which is not well supported by the discussion or the figures, is that there was transport from the tropics into the Northern Hemisphere along the 750 K isentrope. Here are the detailed comments.
313-315 This claim by the authors is not substantiated by either Figs 3 e) or f). The northward extension of scattering ratios near 1 (purple) are not at all obvious and certainly not along the 750 K contour.
412-414 This seems like a stretch given there was no clear evidence of transport from the tropics along the 750 K isentrope. Could this be a filament from the polar vortex as it dissipates in the spring? Where did the air come from? Were there any back trajectory calculations?
Other comments/corrections.
Figure 1a) b) Where is the triangle marking Ruang?
276 … Antarctic …
280-282 This statement is perhaps reasonable at SAOD > 15e-3, but less than this the difference is a factor of two or more, very obvious in the NH mid latitudes. Might be worth mentioning the regions of particularly larger discrepancies.
281-283 This claim is questionable. Yes there is a faint signal in OMPS-LP, but this is not really supported by either SAGE III or ATLID.
Figure 2 Why does the SAGE III data stop in July. Just not available yet?
319 Probably the reason is because it wasn’t there.
338 Isn’t it a black dotted line?
341 The structure of the zonal mean is not vastly different, it just doesn’t capture the fine scale structure, nor would a zonal mean be so expected.
351 Isn’t it the opposite with the zonal mean exhibiting somewhat lower scattering ratios than the local measurements?
455 Is there a reference for 0.4 Tg of sulfur?
Citation: https://doi.org/10.5194/egusphere-2025-4377-RC2 - AC2: 'Reply on RC2', Sergey Khaykin, 16 Dec 2025
Peer review completion
Post-review adjustments
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 2,073 | 145 | 39 | 2,257 | 43 | 32 |
- HTML: 2,073
- PDF: 145
- XML: 39
- Total: 2,257
- BibTeX: 43
- EndNote: 32
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Sergey Khaykin
Michael Sicard
Thierry Leblanc
Tetsu Sakai
Nickolay Balugin
Gwenael Berthet
Stéphane Chevrier
Fernando Chouza
Artem Feofilov
Dominique Gantois
Sophie Godin-Beekmann
Arezki Haddouche
Yoshitaka Jin
Isamu Morino
Nicolas Kadygrov
Thomas Lecas
Ben Liley
Richard Querel
Ghasssan Taha
Vladimir Yushkov
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1951 KB) - Metadata XML
The paper by Khaykin et al. describes the effects of the Ruang volcano eruption on the stratospheric aerosol conditions. It used several data sources and methodologies and describes the event and the dispersion of the associated aerosol plume with time. It furthermore related its impact to previous major events. It is of high scientific interest and therefore in general suited for ACP.
The transport of the volcanic aerosols is in general well described. However, I am struggling with some of the methodologies applied and also some of the conclusions drawn. Partly it appears to be a “business as usual publication” after a new strong volcanic eruption without accounting for novel knowledge and data sources. This needs to be significantly improved before the manuscript can be accepted. Therefore, I recommend major revisions for this work.
I will describe in the following some of my general criticism, followed by more specific comments:
General issues:
Derivation of stratospheric AOD from Lidar observations
The paper applied methodologies which obviously have been applied since a long time to investigate stratospheric aerosol distortions by volcanic aerosol. The scattering ratio of the ground-based and space borne lidars is used together with an assumed lidar ratio of 50 sr to estimate the stratospheric AOD. However, is this methodology still up to date?
First of all, the assumption of a lidar ratio of 50 sr at 355 nm is not discussed by means of any reference. Furthermore, it is in my opinion not justified per se as it assumes the existence of volcanic sulphate aerosol only and neglects any influence of wildfire smoke which was recently observed to be entered frequently in the stratosphere (see e.g. Khaykin et al., 2020; Ansmann 2024, Peterson 2025). Thus, justification for this lidar ratio and an uncertainty estimate for the derived products need to be done and are key for publication.
Second and maybe more important, many of the lidars, especially the novel EarthCARE lidar ATLID which is used intensively, can provide extinction measurements directly. Why don’t you use these products to derive the stratospheric AOD? While the ground based lidar may struggle with low-signal-to-noise ratio, EarthCARE at least is for sure capable to retrieve extinction profiles in the stratosphere. It is not clear to me why the authors use the detour via the scattering ratio and the assumed lidar ratio instead of using the directly measured extinction coefficient profiles.
This is especially important also in view of the other data sources they use, as many provide extinctions measurements (e.g. NOAA-21 OMPS-LP, ISS SAGE III, GloSSAC..….) and not scattering ratio.
Third, the computation of the scattering ratio from EarthCARE Level1b data is not understandable for me. As level 1b signal still suffer from molecular and particular attenuation (therefore attenuated backscatter coefficient), the simple integral of the sum of two signals multiplied with a lidar ratio seem to be not appropriate in my opinion. Please justify and describe more intensively.
Thus, my main recommendations with respect to this topic are: Use direct extinction measurements, provide error estimates (in case this is not possible), discuss and consider always the influence of stratospheric wildfire smoke.
Some other methodological weaknesses are:
The estimated mass of the emitted SO2 is from personal communication only without giving any reference. In my opinion, this is not sufficient as the value is key for Fig. 6 and the respective discussion. Thus, either you describe properly how the mass is estimated or give proper reference. Otherwise, you need to cancel the whole discussion about the impact of the recent eruptions compared to previous ones.
The ground-based instruments and data retrieval descriptions are very heterogenous. Please unify and justify the different methodologies you use for each lidar system. Currently it seems that each ground-based lidar has its own retrieval algorithm with its own assumptions. This might be ok, but needs to be discussed. As for EarthCARE (described above) the same question comes up for many of the derived products from the ground based lidars: How do you account for attenuation of the backscatter when calculating the SAOD from scattering ratio.
Other general issues:
The discussion often (not always) assumes Ruang aerosol to be the only reason for increased aerosol in the stratosphere neglecting other potential sources like wildfires, other volcano eruptions etc… this needs to be discussed better.
Related to this: In Fig. 6, you show major events for stratospheric AOD distortion, but I wonder if it is complete or is missing some of the severe Northern hemispheric fire events having a stratospheric impact (e.g. see in Ohneiser et al., 2023 for the Canadian fires in 2017 or many others as listed in Peterson et al 2025.). You state it by yourself in line 453: “Since 2017, the significant SAOD perturbations, caused by either volcanic eruptions or wildfire outbreaks in both hemispheres, occurred at least once per year, maintaining the global stratospheric aerosol loading well above the background levels.” But I can see only one fire event in the NH in Fig. 6.
Please also broaden your view in the discussion and do not only cite yourself for explanation of certain events. There are many other scientists working on these topics and this would give the paper a broader justification.
Figs 4 and 5: If you would provide extinction profiles from the lidars, you could even compare these to the other data sources you use (e.g. NOAA-21 OMPS-LP, ISS SAGE III, GloSSAC..….) which would value the paper even more.
Many abbreviations are not explained, e.g., JPL, LA, JRA, PNE, AZOR … It is good practise to write the whole name at the first instance.
Specific comments:
Concluding on a reduced sensitivity is in my opinion not valid from the figures you show.
References: