the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
The role of the Indian Ocean Dipole in modulating the Austral Spring ENSO teleconnection into the Southern Hemisphere
Abstract. The combined influence of El Niño-Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) on the extratropical circulation in the Southern Hemisphere (SH) during austral spring is examined. Reanalyses and the large ensemble of CFSv2 model outputs, were used to compute composites and linear regressions for relevant variables. The results show that positive IOD can reinforce the El Niño-induced circulation by merging the Indian Ocean wave train with the PSA-like pattern over the Pacific Ocean. In addition, the results obtained with the CFSv2 model output shows that strong positive IODs can contribute to enhancing the circulation signal of the El Niño anomalies and the Indian Ocean wave train. On the other hand, negative IODs in combination with La Niña do not have that combined circulation response. While there is a moderate intensification of the circulation anomalies associated with La Niña, accompanied by some changes in the location of their main action centers, results vary considerably between linear regression, the observed and model composites. Regarding the influence of the IOD activity (independent of ENSO), reanalysis-based results show that the IOD positive phase has a significant impact over the entire SH, while the negative phase is associated with weaker anomalies and a large inter-event variability.
- Preprint
(17783 KB) - Metadata XML
-
Supplement
(3185 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-1812', Anonymous Referee #1, 16 Jul 2024
Review for “The role of the Indian Ocean Dipole in modulating the Austral Spring ENSO teleconnection into the Southern Hemisphere.” by Adrian et al.
General Comments:
The study analyzes how the IOD contributes to the ENSO-generated extratropical wave train to the Southern Hemisphere. They use observations and an ensemble of CFSv2 analysis to calculate the environmental anomalies generated by ENSO and IOD, in isolation and combined to each other, using composite and linear regressions. They find that positive IOD intensifies the wave train anomalies associated with El Nino in the Southern Hemisphere. On the other hand, no consistent anomalies in the wave train were found during negative IOD and their modulation of La Nina is not as clear, as it is dominated by noise.
The analyses in the manuscript are well thought and executed. The topic is especially difficult to address due to the high correlation between ENSO and IOD, leading to few publications in the area. I would recommend the manuscript to the journal after revisions. Please find my comments below.
Specific Comments:
- Main points to be addressed:
- Be careful when you mention the intensity of the Walker circulation throughout the text. I would mention the intensity of the anomalies in the fields you are analyzing and then argue that it is due to changes in the Walker circulation. If you want to attribute it directly to changes in the Walker circulation, then it would be better to use an analysis designed specifically for that (e.g. Vecchi et al. https://psl.noaa.gov/data/20thC_Rean/timeseries/monthly/Walker/, Kosovelj et al. https://www.mdpi.com/2073-4433/14/2/397, Sohn et al. https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2009JD013713).
- I liked the analysis of temperature and precipitation. Have you checked if these responses are similar in the CFSv2 model?
- Methodology:
- Line 75: Nino 3.4 is mentioned throughout the text, but the link goes to ONI. As you are applying a running mean to DMI, ONI would be more similar to the method you are applying. Either way, please clarify the index used.
- Line 80: it was not clear to me if the DMI is calculated before or after extracting the linear trend. Was the linear trend filtered out from the IOW and IOE, or from DMI? There may be a difference between the methods due to the spatial pattern of the warming in the basin.
- Please add somewhere how many cases of IOD+, IOD-, ENSO+, ENSO-, and their combination exists in the observational record.
- Results:
- Line 130: add some references to papers showing the difference in strength between positive and negative IOD.
- Figure 1: Shouldn’t the point around DMI ~ -1 and ENSO ~ 2 be classified as El Nino & IOD- too? I may be wrong, but there is another point classified as only El Nino that looks to have IOD- from this figure. I suggest double-checking the classification plot.
- Figure 2: Please clarify which color refers to divergence and which to convergence in the figure caption. Same for Fig. 5.
- Line 140, last phrase: “The Walker circulation is weaker compared to the full regression and is restricted to the IO basin.” – do you mean the anomalies are weaker?
- Line 150: they are not negligible for the IO.
- Line 160: I cannot see the weakening of the divergence in figure 2. Could you add more contours and/or add a supplementary figure with the divergence plotted in colors?
- Line 175: I assume you mean the magnitude of the anomalies in the Walker circulation? Aren’t the signs opposed?
- Line 175: also, clarify you are talking about the anomalies being more intense, not the Walker circulation itself.
- Line 186: what do you mean by “not very tidy”?
- Line 239: maybe rephrase this… “the opposite sign” implies that it would be opposite of what we expect. However, if I got this right, “the opposite sign” would be expected in this case.
Technical Corrections:
- Line 80: Extra “a” in Saji and Yamagata reference
- Line 120 (last paragraph): were calculated? I think it would be worth rephrasing this.
- Be careful with the commas. There are a few commas separating subjects from the verbs throughout the text.
- Figure 1: the two tones of orange for El Nino and IOD+ and El Nino and IOD- are too similar.
- Line 150: equatorwards
- Line 159: rephrase this first sentence to be clear what responses you are comparing. Something like “The IO wave train is less intense and less significant for Nino34|DMI (Fig 3c) than in the Nino3.4 full regression (Fig. 3a), as evidenced by …”
- Line 182: remove the first “IO”
- Line 213: intensity
- Fig. 9: the titles of the panels are wrong.
- Link for CFSv2 is broken
Citation: https://doi.org/10.5194/egusphere-2024-1812-RC1 - Main points to be addressed:
-
RC2: 'Comment on egusphere-2024-1812', Anonymous Referee #2, 21 Aug 2024
Review of “The role of the Indian Ocean Dipole in modulating the Austral Spring ENSO teleconnection into the Southern Hemisphere” by Andrian, Osman and Vera
General comments:
This manuscript investigates the interaction between the extratropical wave train forced by the Indian Ocean Dipole (IOD) and the wave train induced by the El Niño-Southern Oscillation (ENSO) during austral spring. The authors use a combination of linear regression and composite analysis of observational data and a model large ensemble to show that the positive phase of the IOD consistently reinforces the Southern Hemisphere circulation response associated with El Niño. In contrast, the circulation response induced by the negative IOD is weaker and exhibits large inter-event variability, showing a less consistent connection with La Niña. The authors have conducted a thorough analysis and I recommend the manuscript for publication after some revisions. Overall, I would like to see the authors improve the flow of the manuscript by reinforcing the main messages throughout, review and strengthen the interpretation of changes in the Walker circulation, and better link the temperature and rainfall analysis in Section 3.3 to the circulation analysis in Section 3.1.Specific comments:
Lines 20-21: There have been some important studies examining ENSO and IOD teleconnections into the Southern Hemisphere. For instance, Cai et al. (2011, 2012), as reference in this manuscript, and McIntosh and Hendon (2018). Several studies have also attempted to disentangle the impacts of the IOD and ENSO on rainfall or temperature, particularly in Australia (e.g. Liguori et al. (2022) and references therein).
Lines 51-52: It would be helpful to state the correlation between ENSO and the IOD earlier in the manuscript and include references.
Lines 61-62: In this paragraph, please add more detail describing how you will build on earlier studies and increase understanding of IOD/ENSO teleconnections.
Data and Methodology section: The organisation of this section could be improved with some sub-headings e.g. “Observational data”…
Lines 67-68: Add some references that show ENSO and IOD have the greatest influence on SH climate during SON.
Line 83-84: Is a threshold of half a standard deviation commonly used in other studies to identify IOD events? Do your results change substantially if only stronger events (1 standard deviation) are selected?
Line 87: I like that you have used a large ensemble to help overcome the issue of small size in the observations. I’m wondering why you chose to use an ensemble of initialised predictions, rather than a single model initial condition large ensemble (SMILE), such as from CESM2.
Line 92: Can you very briefly explain the reason behind the abrupt shift in the climatology in this model?
Lines 97-98: To be consistent with the model, the observational ENSO events should also be defined using the same standard deviation threshold.
Line 128-130: It would be interesting to know when these two instances of positive IOD/La Niña and negative IOD/El Niño occurred.
Lines 136-138: A large part of the discussion of the circulation anomalies involves inferring changes in the Walker circulation from divergence and vertical velocity anomalies at 200 hPa. I am unsure about using 200 hPa divergence as an indicator of Walker Circulation strength. Commonly used approaches are outlined in Kosovelj and Zaplotnik (2023). Can you more clearly describe the link between these anomalies and the Walker circulation, and provide relevant references?
These findings for Niño 3.4 also seem to suggest a weaker Walker circulation, not a stronger one. The typical Walker circulation has upward motion over the eastern Pacific and downward motion over the Maritime Continent. The interpretation of the DMI also appears to the opposite.
Line 149: Showing surface-level anomalies, instead of at 750 hPa, would be more relevant to link with surface impacts in Section 3.3.
Lines 180-181: What is meant by “less significant”? The centres of the extratropical anomalies are significant. The tropical response at 200 hPa is more widespread, as expected.
Lines 181-183: The significant parts of the positive IOD wave train are not really similar to the El Niño wave train though.
In general: Throughout the text, perhaps at the end of a paragraph or section, it would help to include some sentences summarising the results and describing the implications, e.g. if the Walker circulation is enhanced, what does this mean for the teleconnections? When the composite does/does not resemble the regression, what does this say about the linearity of the response? While some of this discussion comes in the final section, incorporating it throughout the manuscript would improve the narrative.
Lines 197-198: The large variability in the circulation response across negative IOD events is interesting. To strengthen this argument and demonstrate that negative IOD events have a more varied response than positive IOD events, you should also include a version of Figure S1 (and possibly Figure S2) for positive IOD events.
Given this point is mentioned in the abstract (lines 10-11), I think Figure S1 should be included in the main manuscript.
Lines 199-200: The idea that the negative stationary wavenumber Ks has a role in explaining the large variability across negative IOD events is discussed, but what about the strength of negative events compared to positive events?
Lines 217-220: Can/will these hypotheses be tested with the large ensemble, or is the purpose of the large ensemble simply to increase the sample size? Do you have plans to test these hypotheses?
Line 240: Do you have an idea why the negative IOD composite in the model is so different to the observed composites? Could this be related to the weak SST anomalies in Figure 9d?
Lines 246-247: The signal to noise figure is very informative!
Wouldn’t we expect the negative IOD to have a lower signal to noise ratio than the positive IOD based on the discussion of Figure S1 (lines 197 onwards)? Including a version of Figure S1 for the positive IOD would be helpful here.
Lines 262-263: The Indian Ocean wave train is still clear for strong negative IOD events in Figure 12g.
Lines 266-270: I would argue that there’s more of a difference between the moderate La Niña combinations with negative IOD than for the strong La Niña combinations, particularly over the Indian Ocean.
Line 295: Figure 15d --> Figure 15c? There is only one dashed line indicating significance. Please clarify how this temperature composite is consistent with the regression results.
Lines 299-314: Are the rainfall results in Australia consistent with Cai et al. (2011, 2012)? Can the Africa results be compared with other studies?
Section 3.3: It would be great if you could link the changes in rainfall and temperature to the circulation changes in Section 3.1. E.g. if the wave train shifts, how does this affect regional climate?
Section 3.3: Do the rainfall and temperature regressions and composites in the model look similar to the observations? Perhaps the model results could be included in the supplementary material.
Figure 2: Are positive contours for divergence green or pink? These contours are hard to see.
Figure 3: The vectors are difficult to see. Consider showing only significant contours of geopotential height (i.e., remove black contours) and increase the colour bar scale so that the shading is less saturated.
Figure 5: How many samples were included in each composite?
Figure 9: Add labels to all panels. State the number of samples in each composite.
Figures 15 and 16: The colour bars have changed from Figure 14, i.e., Figure 15 uses brown and green for temperature, and Figure 16 uses blue and red for rainfall.
Technical corrections:
Ensure all acronyms are defined throughout the manuscript (e.g., CFSv2 and PSA in the Abstract, and SST, z200, and CFSv2 in the main text).
Try to refer to specific panels in the figures as much as possible rather than using descriptions like “middle row” or “upper right panel”.
Line 122: “where” calculated --> “were” calculated.
Line 187: Replace “not very tidy” with another phrase.
Line 300: “territory” --> “country”
References:
Kosovelj and Zaplotnik (2023): https://doi.org/10.3390/atmos14020397
Liguori et al. (2022): https://doi.org/10.1029/2021GL094295
McIntosh and Hendon (2018): https://doi.org/10.1007/s00382-017-3771-1Citation: https://doi.org/10.5194/egusphere-2024-1812-RC2
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
212 | 64 | 21 | 297 | 28 | 11 | 12 |
- HTML: 212
- PDF: 64
- XML: 21
- Total: 297
- Supplement: 28
- BibTeX: 11
- EndNote: 12
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1