Early evolution of the ozone mini-hole generated by the Australian bushfires 2019–2020 observed from satellite and ground-based instruments

Belhadji, Redha; Sellitto, Pasquale; Eremenko, Maxim; Bucci, Silvia; Nguyet, Tran M.; Schwell, Martin; Legras, Bernard

doi:10.5194/amt-18-7465-2025

Articles | Volume 18, issue 23

https://doi.org/10.5194/amt-18-7465-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/amt-18-7465-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 18, issue 23

Research article

|

08 Dec 2025

Research article |

| 08 Dec 2025

Early evolution of the ozone mini-hole generated by the Australian bushfires 2019–2020 observed from satellite and ground-based instruments

Redha Belhadji, Pasquale Sellitto, Maxim Eremenko, Silvia Bucci, Tran M. Nguyet, Martin Schwell, and Bernard Legras

Download

Final revised paper (published on 08 Dec 2025)
Preprint (discussion started on 03 Apr 2025)

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-1453', Anonymous Referee #1, 29 Apr 2025

This is an excellent contribution to atmospheric science and should therefore be published in ACP, and not in AMT. In this manuscript, the authors discuss the lofting of an Australian wildfire smoke layer with a likewise low (tropospheric) ozone content into the stratosphere up to heights above 25-30 km height. The discussion is based on carefully analyzed different satellite observations.
My comments (minor revision) may help to improve the manuscript.
List of comments:
Abstract and throughout the article: I have a general problem with the following wording: ozone depletion! For me, ozone depletion is always linked to chemical (ozone decreasing) processes, and not linked to a vertical ozone transport. I would avoid the word ‘depletion’ in this manuscript dealing with an ozone transport phenomenon.
Page 2, line 42: Solomon et al. 2022 is not a good reference concerning wildfire smoke, it is more related to ozone depletion by smoke. Better references are Ohneiser et al (JGR, 2022) and all the Australian-smoke-related papers cited in this Ohneiser paper, and the recently published paper of Sakai et al. (JGR, 2025). Sakai et al. analyzed long-term lidar observations at Lauder, New Zealand. The time series covers the Australian bushfires.
Page 2, line 56: Solomon et al. (2022) describes the impact of wildfire smoke on mid latitude stratospheric ozone, but there was even a more severe impact of the smoke on polar ozone as Ansmann et al. (ACP, 2022) highlighted. Should probably be mentioned.
Page 3, line 64: vertical, not vortical.
Page 5, line142: The smoke bubble was nicely observed over the tip of South America by Ohneiser et al. (ACP, 2022).
Page 5, line 144: Many papers modelling the lofting of the smoke disk came to the conclusion that the BC content was just 2-3%.
Page 6, line 166: The most intensive pyroCB phase was on 31 Dec 2019 and 1 Jan 2020, and not around 4-5 Jan. 2020 according to the report of Peterson et al. (npj, 2021) and summarized in the Ohneiser paper.
Page 12, Figure 7: The vortex was transported to the east with the main westerly winds at heights below 20-25 km until 20-25 January, and after lofting to heights above about 25 km the vortex was transported to the west with the dominating easterly winds. Why is then the ‘ozone hole’ observed at heights around 22 km (in the west wind zone) over New Zealand on 17 February in Figure 7? The vortex was above 30 km height at that time…., in the easterly wind zone, and the ozone hole should be located around 30 km height.
Page 14, Figure 8: The smoke bubble crossed New Zealand on 17 February, but not on 15 February, according to Figure 1. What is the reason for the pronounced local minimum in the ozone profile at about 26 km height on 15 February? The small ozone minimum in the ozone distribution around 30 km height observed on 17 February makes sense, and is linked to the wildfire smoke disk, but why is there a mini ozone hole on 15 February? Should be discussed!

Citation: https://doi.org/10.5194/egusphere-2025-1453-RC1
- AC2: 'Reply on RC1', Redha Belhadji, 09 Sep 2025
  
  Thank you for your very constructive feedback and for the time you dedicated to reviewing my work.
  
  Please find attached a file containing all my responses to the comments provided.
  
  Citation: https://doi.org/10.5194/egusphere-2025-1453-AC2
RC2:
'Comment on egusphere-2025-1453', Anonymous Referee #2, 26 May 2025

This manuscript explained the ozone mini-hole generation and shows the observational evidence.
The story of the manuscript is well described. However, most of the figures are difficult to check the key point.
In addition, the satellite dataset explanation is not sufficient. More detailed explanation is needed.
In detail...
1) Page 3 L67: The sentence is not essential. Please delete it.
2) Page 3 L80: The LISA algorithm for IASI ozone was used in this study. However, the IASI ozone algorithm also has the FORLI total ozone algorithm. What is the difference between the two methods? In addition, why is the LISA algorithm only used in this study?
2) Section 2.2: Could you write the detailed specification of ECMWF-IFS? In the result section, the ECMWF-IFS specification (especial to the vertical resolution) is essential to understand the result.
3) Figure 2: This figure is too hard to see. Especially the right column of the figures are very difficult to check the detailed structure. Please correct it.
4) Figure 3: The author marked the 'star' in the figure to identify the center of the mini-hole. However, due to the 'star' mark, the intensity of the mini-hole is difficult to check. Please delete or change the color.
5) Page 8 L204-206: As mentioned before, the ECMWF-IFS vertical resolution is coarse. Also, the LISA ozone retrieval algorithm has its own vertical resolution. Could you clarify the vertical resolution of the LISA ozone result? In addition, how to calculate the exact '3km above and 3 km below' the center of the vortex? Could you explain the interpolation(?) method?
6) Page 8 L212: latitude and longitude ranges are too broad. Please explain the reasons why this coarse horizontal resolution is used in this study.
7) Page 10 L247: Overall, I agreed that the aerosol plume affects the noise of ozone retrieval by IASI. However, this explanation needs a reference. Please include the related references that the IASI ozone retrieval is affected by the aerosol plume existence.
8) Figure 5: Is this the temporal ozone anomaly? Please rephrase the exact physical variable (not simple O3 col).
9) Section 4.2: The explanation related to Tables 1 and 2 is not supported to the paragraph in L271-L286. Please rephrase it. In addition, I recommended that the difference be used to the 'absolute difference', not the 'relative difference'.
10) Tables 1 and 2: What is the 0-113 km range?
11) Figure 8: Could you check the partial degree of freedom of vertical distribution?

Citation: https://doi.org/10.5194/egusphere-2025-1453-RC2
- AC1: 'Reply on RC2', Redha Belhadji, 09 Sep 2025
  
  Thank you for your very constructive feedback and for the time you dedicated to reviewing my work.
  
  Please find attached a file containing all the responses to the comments provided.
  
  Citation: https://doi.org/10.5194/egusphere-2025-1453-AC1

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

AR by Redha Belhadji on behalf of the Authors (09 Sep 2025) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (13 Sep 2025) by Sandip Dhomse

RR by Anonymous Referee #3 (25 Sep 2025)

Suggestions for revision or reasons for rejection

The manuscript is well written and has clearly improved following the referees’ comments. The evolution of the Australian wildfire smoke vortex is well described in its early stages and supported by satellite and ground-based observations, as well as modelling evidence.
The points below refer to minor improvements that may help before submission. These are all minor revisions that will improve clarity and precision but do not alter the scientific content.
These are:
1) Abstract (lines 24–26): It may help to add that 30–40% ozone depletion is observed immediately after the event in the partial stratospheric column. This would emphasize the large values and help readers understand the contrast with the situation at the end of January and in the total column.
2) Page 2, lines 50 and 53: Replace who with which.
3) Page 2, line 52: Replace in his way with in its way.
4) Page 2, line 42: As RC1 suggested, Solomon et al. (2022) is a strong reference describing chemical perturbations on chlorine and ozone. I suggest using it as a second reference. For the unprecedented magnitude of the event, however, the Sakai et al. (2025, JGR) paper may be more suitable: https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JD041329?af=R
5) Page 2, line 58: Replace where also performed with were also performed.
6) Figures (Page 6, Fig. 1; Page 11, Fig. 6): The text refers to panels a and b, but these labels are not shown in the figures. Adding them would improve clarity.
1) Page 6, line 159: use within instead of with. For example: “vertical evolution of the ozone mini-hole formed within the smoke vortex.” Is the meaning intended to be: vertical evolution of the ozone mini-hole formed within the smoke vortex?
7) Page 8, line 174: Replace 22th January with 22nd January.

Hide

ED: Publish subject to minor revisions (review by editor) (25 Sep 2025) by Sandip Dhomse

AR by Redha Belhadji on behalf of the Authors (30 Sep 2025) Author's response Author's tracked changes Manuscript

ED: Publish as is (01 Oct 2025) by Sandip Dhomse

AR by Redha Belhadji on behalf of the Authors (09 Oct 2025)

Short summary

The 2019–2020 Australian wildfires triggered massive Pyro-cumulonimbus clouds, injecting smoke aerosols into the stratosphere and forming a self-sustaining vortex that reached 35 km altitude. This vortex created a transient ozone mini-hole. Using satellite and ground-based observations, we tracked a 30–40 % initial ozone depletion, which decayed to ~7 % within a month. These findings highlight the impact of extreme wildfires on stratospheric dynamics and ozone composition.