the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Sep 2025
Evaluation of factors affecting TOC and its trend at three Antarctic stations in the years 2007–2023
Abstract. This study assesses trends in the total ozone column (TOC) and the atmospheric factors influencing ozone variability at three Antarctic stations (Marambio, Troll/Trollhaugen, and Concordia) from 2007 to 2023. Ground-based TOC measurements were used, supplemented by satellite observations from the Ozone Monitoring Instrument on NASA's Aura satellite. TOC trends were derived using a multiple linear regression model provided by the Long-term Ozone Trends and Uncertainties in the Stratosphere (LOTUS) project. The selected LOTUS model was able to explain 94–97 % of the TOC variability at all three stations. The regression analysis showed that ozone variability at these stations is mainly driven by the lower stratospheric temperature, eddy heat flux, and the Quasi-Biennial Oscillation. A statistically significant increasing trend was found at the Marambio station (3.43 DU/decade), while statistically insignificant trends were detected at the other two stations. Using MERRA-2 reanalyses, the LOTUS model was applied to each grid point in the 40–90° S region, which effectively illustrates the spatial distribution of the impacts of individual predictors. It was found that warmer conditions in the Antarctic stratosphere in September 2019 caused TOC to be up to 100 DU higher than normal, especially over East Antarctica. The results improve understanding of regional TOC trends and how the Antarctic ozone layer responds to changes in ozone-depleting substances.
- Preprint
(2391 KB) - Metadata XML
-
Supplement
(947 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-3963', Anonymous Referee #1, 31 Oct 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-3963/egusphere-2025-3963-RC1-supplement.pdfCitation: https://doi.org/
10.5194/egusphere-2025-3963-RC1 -
RC2: 'Comment on egusphere-2025-3963', Anonymous Referee #2, 06 Nov 2025
General Comments:
This manuscript describes a long-term time series analysis of ground-based records from three stations in Antarctica. The authors use the multiple regression model developed by the LOTUS group, with additional parameters optimized for variability at polar latitudes. The authors additionally analyzed MERRA-2 reanalysis output to assess the spatial distribution of the various parameters. The study is very relevant and the manuscript clearly written and well-referenced. The figures, tables and supplemental material are clearly presented. I recommend publication after the following issues are addressed.
Specific Comments:
L68: The sentence starting “While the B199 instrument offers very high accuracy…” is somewhat confusing. I believe the 0.15% from Scarnato et al. refers to the precision rather than the accuracy and is for the double Brewer instrument in general. Also, the wording should clarify that the direct sun measurements are the most precise. Something like: “To assure the highest precision, only direct sun measurements were utilized.”
L99: Including a different data set, particularly at the endpoints, can have a notable impact on the trend even if the offsets are small. It will be difficult to compare with other studies because of the specific period of the fit, starting later in the recovery time period, in 2007. To address this, I think it would be instructive, either in the paper or in the supplemental material, to show a figure similar to Figure 5 but use the OMI overpass and MERRA-2 overpass data at each station (time series shown in Figure 2) to get an idea of the possible range of values from different data sources. There are notable differences in the time series as shown in Figure 2, including a small drift at Troll and Concordia.
L235: Here and throughout the manuscript when discussing the QBO fits, can the authors explain the relevance of the individual QBO terms. I understand why multiple EOF principal component time series are used, but I do not believe these terms can be physically interpreted individually. For example in Figure 4, rather than show the individual terms, I believe this result is more easily understood if the terms are re-added to represent the net QBO variability. I would also be interested to know if the net QBO fit was statistically significant at each station. I believe the authors can apply a joint F-test to determine this. Also in Figure 10, a panel can be added that shows the reconstruction of the net QBO for the two years. I realize the figures and discussion are set up to address the individual terms, I would just like to see the examples of the full QBO signal expressed in DU in Figures 4 and 10, and if possible an estimate of the significance of the full QBO signal.
Line 245: This is a nice figure, can the trend term be added as well? It is a little out of place because it covers all three stations, it might fit better after the regression fits for each station are presented, but this change is not mandatory. The same results using the OMI and MERRA-2 overpass time series would be very interesting as mentioned before, this plot could be part of the supplemental information but referred to in the text. Such a plot would also make it easier for the reader to compare the results in Figure 9 to the station results.
Line 392: I would caution the authors not to assign too much causality to some of the fits. For example, when comparing 2019 and 2020, the QBOa and QBOd terms switch signs, as do the ENSO and IOD terms, but this is due to the proxy signals changing sign over this period, not due to the conditions of a warmer or colder polar stratosphere (see Figure S1). The temperature and EHF are related to the polar dynamics, but the QBO/ENSO/IOD terms vary according to the time scale of those forcings and just happen to change sign from 2019 to 2020. It is possible that the QBO phase impacts the wave activity and thus the vortex, but to show this the authors would need to assess the QBO phase over a series of cold and warm polar conditions. The text reads as though a warm vortex produces a QBO signal that is one sign, and a cold vortex a QBO signal of the other sign. This may not be the intention, but it should be carefully worded to avoid inaccurate (or at least unproved) associations.
Conclusions: It would be interesting to see comparisons of the derived trends with trends from other studies. This might be difficult because of the variable time periods between studies. But again, a comparison with the satellite and reanalysis overpass time series would be useful in place of outside comparisons.
Technical Corrections:
Title: suggest spelling out TOC in the title
L19: any time the trend is given, the uncertainty estimate should be included.
L43: suggest wording change: “a strong wave-1 disturbance developed”
L50 Jonson -> Johnson
L116: SBUV can be removed here, it is not included in the description and SBUV is not relevant to MERRA-2 after October 2004.
L156: please clarify in the text, the equatorial zonal mean wind at seven pressure levels between 70-10 hPa were used
L189: the lowest mean deviation
L212: suggest removing “and the lowest in September-October and January -April” I think it is sufficient to say the largest variability is in November-December.
L230: Can the authors say more here about whether the ENSO results agree with the study by Lin and Qian (2019).
L235: suggest adding tick marks for each year to make the plot easier to read.
L263: Is this because Marambio is sometimes in the collar region?
L284: The last sentence in this paragraph is largely repetitive. I suggest removing it or revising the last sentence of the previous paragraph to include this information.
L311-312: The Marambio trend in November is positive, and the trend for September is increasing but also not statistically significant. Suggest: “Interestingly, the trends for October and November and decreasing but not statistically significant at all stations except Marambio in November, while the trends for September at all stations are increasing, but also not statistically significant.”
L320: suggest adding tick marks for each year to make the plot easier to read. Also in Figures S1 and S4.
L328-329: suggest simplifying the text here, possibly “Time series analysis at each grid point shows the spatial distribution of the fits to each parameter, which are expressed using standardized coefficients of determination.”
L338: suggest Lin and Qian (2019) shows that …
L346: the equatorial zonal wind at seven pressure levels between 10 and 70 hPa.
Line 359: Jonson -> Johnson
Line 359 and 361: can remove “Studies by” and “A study of” to simplify the text.
L376: suggest wording clarification: “slowdown the heterogenous reactions that activate Cl on the surface of PSCs, thus slowing ozone depletion and suppressing the formation… “
Line 380: Was the further decomposition of the EHF done in the Shen et al., 2020 study or in this study? If in the Sten et al study I suggest “Based on further decomposition of the EHF, Shen et al. (2020) found… “ to make it clear this was not part of the current work.
L413: include trend uncertainty value in text
Citation: https://doi.org/10.5194/egusphere-2025-3963-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,772 | 25 | 10 | 1,807 | 20 | 9 | 14 |
- HTML: 1,772
- PDF: 25
- XML: 10
- Total: 1,807
- Supplement: 20
- BibTeX: 9
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1