the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 May 2023
ENSO Modulation of the QBO Periods in GISS E2.2 Models
Abstract. Observational studies have shown that the El Niño–Southern Oscillation (ENSO) exerts an influence on the Quasi-Biennial Oscillation (QBO). The downward propagation of the QBO tends to speed up and slow down during El Niño and La Niña, respectively. Recent results from general circulation models have indicated that the ENSO modulation of the QBO requires a relatively high horizontal resolution, and that it does not show up in the climate models with parameterized but temporally constant gravity wave sources. Here, we demonstrate that the NASA GISS E2.2 models can capture the observed ENSO modulation of the QBO period with a horizontal resolution of 2° latitude by 2.5° longitude and its gravity wave sources parameterized interactively. This is because El Niño events lead to more vigorous gravity wave sources generating more absolute momentum fluxes over the equatorial belt, as well as less filtering of these waves into the tropical lower stratosphere through a weakening of the Walker circulation. Various components of the ENSO system such as the SSTs, the convective activities, and the Walker circulation are intimately involved in the generation and propagation of parameterized gravity waves, through which ENSO modulates the QBO period in GISS E2.2 models.
-
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
(3211 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3211 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-774', Anonymous Referee #1, 04 Jun 2023
Comments on “ENSO Modulation of the QBO Periods in GISS E2.2 Models” by Zhou et al.
Summary
In this study, the authors focus the possible impact of the ENSO on the QBO cycle using the observations and the model simulations from GISS E2.2. However, the modulation of the QBO cycle by the El Nino and La Nina is not consistent among the model configurations. The authors concluded that the model physics are important to simulate the impact of the ENSO on the QBO period. In general, the authors well explored the possible impact of ENSO on the QBO period. However, I also found some relevant concerns, which should be well addressed. Therefore, I suggest a major revision at the present time.Major comments
1. The relationship between the ENSO and QBO should be well reviewed in the introduction. The possible impact of ENSO on the QBO amplitude and phase should mentioned. Further, ENSO and QBO phase coincidence should also be mentioned. Before 1980s, El Nino tends to appear during EQBO, and La Nina tends to appear during WQBO. After 1980s, El Nino tends to appear during WQBO, and La Nina tends to appear during EQBO (DOI: 10.1175/JCLI-D-19-0087.1; 10.1175/JCLI-D-20-0024.1).2. Previous studies also found that the relationship between ENSO and QBO is not universal among models (DOI: 10.1175/JCLI-D-20-0024.1). This paper only emphasizes the possible impact of ENSO on the QBO, is there a possibility that the QBO can impact the ENSO occurrence and amplitude.
3. Further, this paper is too long and too dispersal and include too many contents. This paper is not aimed to evaluate the model configurations. However, the comparison between AMIP and CMIP simulations and that between SP and AP physics accounts for a large portion of the paper. I suggest to remove the experiments that fail to reproduce the impact of ENSO on the QBO period. Including those results that do not simulate a significant difference between the QBO periods during El Nino and La Nina, the paper is not convinced at all.
4. The paper should provide a section named “Data and method” or something like. Without a data description and experiment introduction, this paper reads weird and readers fail to find the experimental setup.
Other comments
1. L30, L43-45: There are too many papers concerning the possible impact of ENSO and QBO on the climate (DOI: 10.1175/JCLI-D-19-0663.1; 10.1029/2020GL089149; 10.1175/JCLI-D-20-0960.1). I suggest to include more recent publication in the citations.2. L59-60: There are also some studies that focus on the possible impact of the QBO on ENSO. The authors should present some review.
3. L145-146: The ENSO amplitude in observations and CMIP models are not identical (DOI: 10.3878/AOSL20140055). If you use the same criterion, will the results be not convincing.
4. L153-154: Other studies also performed the EOF analysis for the QBO wind profiles (See Figure 3 in Rao and Ren 2018CD, doi: 10.1007/s00382-017-3998-x).
5. L168: Here is programming language. I suggest to use the science language: φ =atan(PC2/PC1)
6. L233: You should provide a detailed introduction for the calculation steps in a method section.
7. L254: modulates of => modelate (remove “of”)
8. L259: Section 3.1, and 3.2 : Should move to a method section.
9. L404-406: I also download historical runs from the GISS-E2 models for CMIP6. But I did not see the spontaneous QBO. Is the model in this paper same configured as for CMIP6?
10. Section 4.1: Should move to the method section.
11. L439-441: Can you explain why the relationship is not stable?
12. L475: help => helps
13. L495-497: Please also see Domeisen et al. 2019 (RG) and references therein.
14. L531-532: Most CMIP models simulate a smaller ENSO amplitude as compared with the observations (DOI: 10.3878/AOSL20140055). This model is different and simulate a stronger ENSO.
15. L588-593: This paragraph describes the ERA5 reanalysis and should be moved to the method section.
16. L591: during in => during
17. L605-606: This sentence repeats many times. You can provide a section describing the data and methods. There is no need to repeat the methods time by time.
18. L732-735: The ENSO amplitude has so larger a bias. To what extent can we trust the results?
19. L743, Figure 14: This figure is redundant and fail to connect with the topic of this study.
20. L760: Do you mean that none of the results are robust in this study? I also did not see that the QBOi also explore the ENSO modulation of the QBO cycle.
Citation: https://doi.org/10.5194/egusphere-2023-774-RC1 - AC1: 'Reply on RC1', Tiehan Zhou, 19 Sep 2023
-
RC2: 'Comment on egusphere-2023-774', Anonymous Referee #2, 06 Jun 2023
General comments
The work by Zhou et al. deals with the ENSO modulation of the QBO phase speed in the GISS model. This seems strongly related to the parameterized gravity wave forcing, which is not described in much detail (line 93 and following). Basic information should be provided (e.g., how convection changes the parameterized wave spectrum) and possible model-dependence of the results stressed.
The text is well-written, but there are some repetitions which could be avoided (see specific comments).
For example, the data processing used for calculating ONI and the QBO could be given only once in the methods.
The observational analysis part could probably be shortened.There are several long sentences which are not very readable. It is in some cases not easy to follow the reasoning or motivation or some of the analysis, please improve the connections between paragraphs where needed.
It could be useful to provide more information on the ENSO characteristics in coupled experiments (e.g., references or number/intensity of events), briefly mentioned around line 530 and shown in the spectra plots.
Opposite results for some simulations (line 738 and following) are interesting but should be discussed further: is this suggesting a very important role for internal variability, at least in simulations?Differences in some of the plots are small, it would be good to add some significance estimate.
Specific comments
18 'gravity waves parameterized interactively' -> 'interactive GW parameterization'?
41 may refer at QBO zonal asymmetry
61 all -> how many. When first introducing 'T' for truncation, please clarify what it means
80 using which model?
114 may add a reference like Naujokat, 1986
131 please add a reference, e.g. Salby, 2012
148 are you referring to the ONI for ERSST or the simulations?
171 I do not see why 'now', as to me this is unrelated to the previous paragraph
206 I'd say that N1 and N2 do not result from calculations
207 you could introduce as done for A the meaning of the overbar for both quantities
344 What else could be done, since you stated that you are not considering the amplitude already?
424 To avoid confusion with capitalised psi for phase speed, please ensure the latter is uppercase elsewhere.
436 Not sure to understand this sentence
450 do you need to scale by the respective variance?
462 Why the focus on Coupled-NINT-Ap? Can you motivate and remind the reader about this configuration?
543 La Niña does not have a well defined peak, suggest rephrasing
554 in any season? in both hemispheres?
588 please explain why you discuss this now. ERA5 should be introduced in the Methods section.
1311 figure could be improved by using logarithmic ordinate and/or putting spectra in a single plot
Technical corrections
133 NOAA undefined
143 CDC=CPC ?
209 repeated v and t
221 be consistent in the use of lowercase psi
267 Why 'according'? Unclear
288 NCCS undefined
338 'baseline'
434 'Andrews'
441 maybe 'configuration'?
632 'nether'?
688 'Earth'
1146 more shades in Fig 5 right would be better?
Or maybe using the same levels to ease comparison
1275 the varying levels across plots should be fixedAdditional references
Naujokat, 1986 https://journals.ametsoc.org/view/journals/atsc/43/17/1520-0469_1986_043_1873_auotoq_2_0_co_2.xml
Salby, 2012 https://doi.org/10.1017/CBO9781139005265Citation: https://doi.org/10.5194/egusphere-2023-774-RC2 - AC2: 'Reply on RC2', Tiehan Zhou, 19 Sep 2023
-
RC3: 'Comment on egusphere-2023-774', Anonymous Referee #3, 15 Jun 2023
Review of " ENSO Modulation of the QBO Periods in GISS E2.2 Models” by Zhou et al.
The authors have done a lot of detailed work to compile these results. The study covers both model and observational parts. However, much of the text describes the figures extensively without giving the reader a clear road map that how these results differ importantly from previous studies. The repetition of text can be seen at some places. My suggestion is for the authors to perhaps shorten the description of results to reflect only their essential messages. I have the following major concerns before any recommendation on the manuscript:
- Sometimes, the description of results is too extensive and needs to be shortened. The description related to the mechanics in the results sections should be moved in the discussion section. The repeated text requires curbing. Sometimes, the figures are described randomly within the results section.
- In the observation part, the authors are only confirming the finding of Taguchi (2010) with the same methodology. I could not observe any advantage of their observation analysis in compare to Taguchi (2010).
- This study is using 5-month moving averaged deseasonalized data. Is such huge smoothed data (a nearly half-year window) suitable to study the gravity wave generation? I suggest using the monthly data instead of the 5-month moving average. Using the monthly will also be an advantage of your study against Taguchi (2010).
- A separate section is required for the data and method; currently, there is mixing of data information and result description.
- All the figures are plotted very causally and not suitable for publication. There is a lot of scope for improvement in almost all the figures. Please see the specific comments for details.
Specified comments & suggestions:
L47-48 “the tropospheric subtropical jet (Garfinkel and Hartmann, 2011a, 2011b)..”, can be updated with more recently citation ( e.g. DOI: 10.1029/2022JD036691).
L48 “the boreal summer monsoon (Giorgetta et al., 1999)”, can be updated with more recently citation, Yoden et. al 2023 which shows the QBO modulation on global monsoon system (https://doi.org/10.54302/mausam.v74i2.5948).
L103-108: I suggest to add some sentences related to the motivation of this study.
L121 “We further smooth the deseasonalized zonal winds using a 5-month moving average (for more details, refer to Taguchi, 2010)”. It will be better to use the monthly deseasonalized zonal winds instead of the 5-month moving average.
L 132, I suggest to the authors that the data can be extended for seven more years, i.e., 1953 to 2022.
L139-142, Why the different base periods? Only one base period can be used, i.e., a de-seasonalized anomaly for the whole time period of the data set.
L143 “CDC” to “CPC”
L144 As suggested in the comments line 121, if authors consider the use of monthly data, then monthly ONIs can be used to define periods of El Niño and La Niña whenever it exceeds the threshold values ± 0.5 K (+ El Niño, − La Niña). Sometimes the SST lies in ENSO phase for 2 to 3 months, and the generation of gravity waves for such a short period will be washed out in the 5-month moving average and cannot be ignored.
L149 "identified 21 El Nino and 15 La Nina events between 1953 and 2015”. Definitely, using monthly data, the El Nino and La Nina events will increase by two to threefolds. The same can be applied on the model part too.
L154-156 Not a justified reason. If we go beyond the 2015 period, the QBO disruptions (2016 and 2019/20) will not have a significant impact on the total variance of the leading two EOFs. In our own analysis for the period 1979–2022, the two leading EOFs account for 94.73% of the total variance (58.07% by EOF1 and 36.66% by EOF2). If authors are worried about these QBO disruptions, then the time period of disturbance can be excluded if lies in El Nino and La Nina sampling.
L157 Instead of two data sets (FUB and ERA5), the authors may also think of using only the ERA5 data for all observational analysis.
L168: This is programming language. Proper mathematical expression should use here.
L207: If possible, unit “radians/month” to Km/month. Same in sequent text.
L241: This is a clear mixing of analysis description and data information. The data information should be in a separate section.
L336 -345: As mentioned in the above comment (L139–142), why the different base periods to calculate the SST anomalies?
L356: For the reader's convenience, it will be nice to include the Fig.1 EOFs vectors in Fig.4 also (same line format as in the Fig.1).
L530 -532: “Comparing Figs. 8a and 8b with Figs. 3a and 3b”. I suggest to add one more row at the bottom of Fig. 8 for the difference between the model and observed amplitude of QBO during El Nino and La Nina.
L 588-590: The description of the ERA5 reanalysis should be moved into the new suggested data and method section.
L593: Is the composite difference in Fig.11c passes the statistically significant test (>95% confidence)?
L604-628: This paragraph can be rewritten more precisely by focusing on the comparison between ERA5 and models.
L687-693: This paragraph seems unfit here and can be shifted to an appropriate place. The discussion part can start with Paragraph 2.
Comments of Figs.
All the figures are plotted causally and not suitable to be considered for publication.
Fig.1 It is too elongated along the x-axis, for the best view, the aspect ratio X:Y should be ~=1:1.3
Fig.2 This figure is also too elongated along the x-axis, the aspect ratio X:Y should be =4:1. I suggest to interchange the panels (a) and (b) for the systematic representation, i.e., top (a) should be for El Nino and (b) La Nina and (c) same (El Nino- La Nina). It is an inconvenience to compare all the panels in the current color scale as it is different for different panels. The color bar should have the same scaling on both the positive and negative sides for all panels (here for this figure -45 to 45 Wm-2). The fine and coarse contour intervals can be used for visualization of smaller and larger signals (e.g. please see Fig. 21 of Hitchman et. al 2020, https://doi.org/10.2151/jmsj.2021-012). The same can apply for other color figures too.
Fig.3 Same comments as for Fig.2
Fig.4 Same comments as for Fig.1
Fig.5 Same comments on color scale as for Fig.2. The representation of panel numbers should have the same order in the caption of all figures. In Fig. 3, it is before starting the description [ see L 1111 (a) La Nina and (b) El Nino] but here it is after starting the description [see L1152 La Nina (a) and El Nino (b)]. The same corrections must be applied for other figures too.
Fig.6 Same comments on color scale as for Fig.2.
Fig. 7. The aspect ratio X:Y should be similar to Fig.5 and 6. Same comments on color scale as for Fig.2.
Fig.8 too compressed along x-axis. the aspect ratio X:Y should be =4:1.
Fig.9 and 11: Same comments as for Fig.2.
Fig. 12, and Fig. 13: Same comment on the color scale as for Fig.2. Fig13. the aspect ratio X:Y should be =4:1.
Citation: https://doi.org/10.5194/egusphere-2023-774-RC3 - AC3: 'Reply on RC3', Tiehan Zhou, 19 Sep 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-774', Anonymous Referee #1, 04 Jun 2023
Comments on “ENSO Modulation of the QBO Periods in GISS E2.2 Models” by Zhou et al.
Summary
In this study, the authors focus the possible impact of the ENSO on the QBO cycle using the observations and the model simulations from GISS E2.2. However, the modulation of the QBO cycle by the El Nino and La Nina is not consistent among the model configurations. The authors concluded that the model physics are important to simulate the impact of the ENSO on the QBO period. In general, the authors well explored the possible impact of ENSO on the QBO period. However, I also found some relevant concerns, which should be well addressed. Therefore, I suggest a major revision at the present time.Major comments
1. The relationship between the ENSO and QBO should be well reviewed in the introduction. The possible impact of ENSO on the QBO amplitude and phase should mentioned. Further, ENSO and QBO phase coincidence should also be mentioned. Before 1980s, El Nino tends to appear during EQBO, and La Nina tends to appear during WQBO. After 1980s, El Nino tends to appear during WQBO, and La Nina tends to appear during EQBO (DOI: 10.1175/JCLI-D-19-0087.1; 10.1175/JCLI-D-20-0024.1).2. Previous studies also found that the relationship between ENSO and QBO is not universal among models (DOI: 10.1175/JCLI-D-20-0024.1). This paper only emphasizes the possible impact of ENSO on the QBO, is there a possibility that the QBO can impact the ENSO occurrence and amplitude.
3. Further, this paper is too long and too dispersal and include too many contents. This paper is not aimed to evaluate the model configurations. However, the comparison between AMIP and CMIP simulations and that between SP and AP physics accounts for a large portion of the paper. I suggest to remove the experiments that fail to reproduce the impact of ENSO on the QBO period. Including those results that do not simulate a significant difference between the QBO periods during El Nino and La Nina, the paper is not convinced at all.
4. The paper should provide a section named “Data and method” or something like. Without a data description and experiment introduction, this paper reads weird and readers fail to find the experimental setup.
Other comments
1. L30, L43-45: There are too many papers concerning the possible impact of ENSO and QBO on the climate (DOI: 10.1175/JCLI-D-19-0663.1; 10.1029/2020GL089149; 10.1175/JCLI-D-20-0960.1). I suggest to include more recent publication in the citations.2. L59-60: There are also some studies that focus on the possible impact of the QBO on ENSO. The authors should present some review.
3. L145-146: The ENSO amplitude in observations and CMIP models are not identical (DOI: 10.3878/AOSL20140055). If you use the same criterion, will the results be not convincing.
4. L153-154: Other studies also performed the EOF analysis for the QBO wind profiles (See Figure 3 in Rao and Ren 2018CD, doi: 10.1007/s00382-017-3998-x).
5. L168: Here is programming language. I suggest to use the science language: φ =atan(PC2/PC1)
6. L233: You should provide a detailed introduction for the calculation steps in a method section.
7. L254: modulates of => modelate (remove “of”)
8. L259: Section 3.1, and 3.2 : Should move to a method section.
9. L404-406: I also download historical runs from the GISS-E2 models for CMIP6. But I did not see the spontaneous QBO. Is the model in this paper same configured as for CMIP6?
10. Section 4.1: Should move to the method section.
11. L439-441: Can you explain why the relationship is not stable?
12. L475: help => helps
13. L495-497: Please also see Domeisen et al. 2019 (RG) and references therein.
14. L531-532: Most CMIP models simulate a smaller ENSO amplitude as compared with the observations (DOI: 10.3878/AOSL20140055). This model is different and simulate a stronger ENSO.
15. L588-593: This paragraph describes the ERA5 reanalysis and should be moved to the method section.
16. L591: during in => during
17. L605-606: This sentence repeats many times. You can provide a section describing the data and methods. There is no need to repeat the methods time by time.
18. L732-735: The ENSO amplitude has so larger a bias. To what extent can we trust the results?
19. L743, Figure 14: This figure is redundant and fail to connect with the topic of this study.
20. L760: Do you mean that none of the results are robust in this study? I also did not see that the QBOi also explore the ENSO modulation of the QBO cycle.
Citation: https://doi.org/10.5194/egusphere-2023-774-RC1 - AC1: 'Reply on RC1', Tiehan Zhou, 19 Sep 2023
-
RC2: 'Comment on egusphere-2023-774', Anonymous Referee #2, 06 Jun 2023
General comments
The work by Zhou et al. deals with the ENSO modulation of the QBO phase speed in the GISS model. This seems strongly related to the parameterized gravity wave forcing, which is not described in much detail (line 93 and following). Basic information should be provided (e.g., how convection changes the parameterized wave spectrum) and possible model-dependence of the results stressed.
The text is well-written, but there are some repetitions which could be avoided (see specific comments).
For example, the data processing used for calculating ONI and the QBO could be given only once in the methods.
The observational analysis part could probably be shortened.There are several long sentences which are not very readable. It is in some cases not easy to follow the reasoning or motivation or some of the analysis, please improve the connections between paragraphs where needed.
It could be useful to provide more information on the ENSO characteristics in coupled experiments (e.g., references or number/intensity of events), briefly mentioned around line 530 and shown in the spectra plots.
Opposite results for some simulations (line 738 and following) are interesting but should be discussed further: is this suggesting a very important role for internal variability, at least in simulations?Differences in some of the plots are small, it would be good to add some significance estimate.
Specific comments
18 'gravity waves parameterized interactively' -> 'interactive GW parameterization'?
41 may refer at QBO zonal asymmetry
61 all -> how many. When first introducing 'T' for truncation, please clarify what it means
80 using which model?
114 may add a reference like Naujokat, 1986
131 please add a reference, e.g. Salby, 2012
148 are you referring to the ONI for ERSST or the simulations?
171 I do not see why 'now', as to me this is unrelated to the previous paragraph
206 I'd say that N1 and N2 do not result from calculations
207 you could introduce as done for A the meaning of the overbar for both quantities
344 What else could be done, since you stated that you are not considering the amplitude already?
424 To avoid confusion with capitalised psi for phase speed, please ensure the latter is uppercase elsewhere.
436 Not sure to understand this sentence
450 do you need to scale by the respective variance?
462 Why the focus on Coupled-NINT-Ap? Can you motivate and remind the reader about this configuration?
543 La Niña does not have a well defined peak, suggest rephrasing
554 in any season? in both hemispheres?
588 please explain why you discuss this now. ERA5 should be introduced in the Methods section.
1311 figure could be improved by using logarithmic ordinate and/or putting spectra in a single plot
Technical corrections
133 NOAA undefined
143 CDC=CPC ?
209 repeated v and t
221 be consistent in the use of lowercase psi
267 Why 'according'? Unclear
288 NCCS undefined
338 'baseline'
434 'Andrews'
441 maybe 'configuration'?
632 'nether'?
688 'Earth'
1146 more shades in Fig 5 right would be better?
Or maybe using the same levels to ease comparison
1275 the varying levels across plots should be fixedAdditional references
Naujokat, 1986 https://journals.ametsoc.org/view/journals/atsc/43/17/1520-0469_1986_043_1873_auotoq_2_0_co_2.xml
Salby, 2012 https://doi.org/10.1017/CBO9781139005265Citation: https://doi.org/10.5194/egusphere-2023-774-RC2 - AC2: 'Reply on RC2', Tiehan Zhou, 19 Sep 2023
-
RC3: 'Comment on egusphere-2023-774', Anonymous Referee #3, 15 Jun 2023
Review of " ENSO Modulation of the QBO Periods in GISS E2.2 Models” by Zhou et al.
The authors have done a lot of detailed work to compile these results. The study covers both model and observational parts. However, much of the text describes the figures extensively without giving the reader a clear road map that how these results differ importantly from previous studies. The repetition of text can be seen at some places. My suggestion is for the authors to perhaps shorten the description of results to reflect only their essential messages. I have the following major concerns before any recommendation on the manuscript:
- Sometimes, the description of results is too extensive and needs to be shortened. The description related to the mechanics in the results sections should be moved in the discussion section. The repeated text requires curbing. Sometimes, the figures are described randomly within the results section.
- In the observation part, the authors are only confirming the finding of Taguchi (2010) with the same methodology. I could not observe any advantage of their observation analysis in compare to Taguchi (2010).
- This study is using 5-month moving averaged deseasonalized data. Is such huge smoothed data (a nearly half-year window) suitable to study the gravity wave generation? I suggest using the monthly data instead of the 5-month moving average. Using the monthly will also be an advantage of your study against Taguchi (2010).
- A separate section is required for the data and method; currently, there is mixing of data information and result description.
- All the figures are plotted very causally and not suitable for publication. There is a lot of scope for improvement in almost all the figures. Please see the specific comments for details.
Specified comments & suggestions:
L47-48 “the tropospheric subtropical jet (Garfinkel and Hartmann, 2011a, 2011b)..”, can be updated with more recently citation ( e.g. DOI: 10.1029/2022JD036691).
L48 “the boreal summer monsoon (Giorgetta et al., 1999)”, can be updated with more recently citation, Yoden et. al 2023 which shows the QBO modulation on global monsoon system (https://doi.org/10.54302/mausam.v74i2.5948).
L103-108: I suggest to add some sentences related to the motivation of this study.
L121 “We further smooth the deseasonalized zonal winds using a 5-month moving average (for more details, refer to Taguchi, 2010)”. It will be better to use the monthly deseasonalized zonal winds instead of the 5-month moving average.
L 132, I suggest to the authors that the data can be extended for seven more years, i.e., 1953 to 2022.
L139-142, Why the different base periods? Only one base period can be used, i.e., a de-seasonalized anomaly for the whole time period of the data set.
L143 “CDC” to “CPC”
L144 As suggested in the comments line 121, if authors consider the use of monthly data, then monthly ONIs can be used to define periods of El Niño and La Niña whenever it exceeds the threshold values ± 0.5 K (+ El Niño, − La Niña). Sometimes the SST lies in ENSO phase for 2 to 3 months, and the generation of gravity waves for such a short period will be washed out in the 5-month moving average and cannot be ignored.
L149 "identified 21 El Nino and 15 La Nina events between 1953 and 2015”. Definitely, using monthly data, the El Nino and La Nina events will increase by two to threefolds. The same can be applied on the model part too.
L154-156 Not a justified reason. If we go beyond the 2015 period, the QBO disruptions (2016 and 2019/20) will not have a significant impact on the total variance of the leading two EOFs. In our own analysis for the period 1979–2022, the two leading EOFs account for 94.73% of the total variance (58.07% by EOF1 and 36.66% by EOF2). If authors are worried about these QBO disruptions, then the time period of disturbance can be excluded if lies in El Nino and La Nina sampling.
L157 Instead of two data sets (FUB and ERA5), the authors may also think of using only the ERA5 data for all observational analysis.
L168: This is programming language. Proper mathematical expression should use here.
L207: If possible, unit “radians/month” to Km/month. Same in sequent text.
L241: This is a clear mixing of analysis description and data information. The data information should be in a separate section.
L336 -345: As mentioned in the above comment (L139–142), why the different base periods to calculate the SST anomalies?
L356: For the reader's convenience, it will be nice to include the Fig.1 EOFs vectors in Fig.4 also (same line format as in the Fig.1).
L530 -532: “Comparing Figs. 8a and 8b with Figs. 3a and 3b”. I suggest to add one more row at the bottom of Fig. 8 for the difference between the model and observed amplitude of QBO during El Nino and La Nina.
L 588-590: The description of the ERA5 reanalysis should be moved into the new suggested data and method section.
L593: Is the composite difference in Fig.11c passes the statistically significant test (>95% confidence)?
L604-628: This paragraph can be rewritten more precisely by focusing on the comparison between ERA5 and models.
L687-693: This paragraph seems unfit here and can be shifted to an appropriate place. The discussion part can start with Paragraph 2.
Comments of Figs.
All the figures are plotted causally and not suitable to be considered for publication.
Fig.1 It is too elongated along the x-axis, for the best view, the aspect ratio X:Y should be ~=1:1.3
Fig.2 This figure is also too elongated along the x-axis, the aspect ratio X:Y should be =4:1. I suggest to interchange the panels (a) and (b) for the systematic representation, i.e., top (a) should be for El Nino and (b) La Nina and (c) same (El Nino- La Nina). It is an inconvenience to compare all the panels in the current color scale as it is different for different panels. The color bar should have the same scaling on both the positive and negative sides for all panels (here for this figure -45 to 45 Wm-2). The fine and coarse contour intervals can be used for visualization of smaller and larger signals (e.g. please see Fig. 21 of Hitchman et. al 2020, https://doi.org/10.2151/jmsj.2021-012). The same can apply for other color figures too.
Fig.3 Same comments as for Fig.2
Fig.4 Same comments as for Fig.1
Fig.5 Same comments on color scale as for Fig.2. The representation of panel numbers should have the same order in the caption of all figures. In Fig. 3, it is before starting the description [ see L 1111 (a) La Nina and (b) El Nino] but here it is after starting the description [see L1152 La Nina (a) and El Nino (b)]. The same corrections must be applied for other figures too.
Fig.6 Same comments on color scale as for Fig.2.
Fig. 7. The aspect ratio X:Y should be similar to Fig.5 and 6. Same comments on color scale as for Fig.2.
Fig.8 too compressed along x-axis. the aspect ratio X:Y should be =4:1.
Fig.9 and 11: Same comments as for Fig.2.
Fig. 12, and Fig. 13: Same comment on the color scale as for Fig.2. Fig13. the aspect ratio X:Y should be =4:1.
Citation: https://doi.org/10.5194/egusphere-2023-774-RC3 - AC3: 'Reply on RC3', Tiehan Zhou, 19 Sep 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 383 | 132 | 26 | 541 | 12 | 14 |
- HTML: 383
- PDF: 132
- XML: 26
- Total: 541
- BibTeX: 12
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
1 citations as recorded by crossref.
Kevin J. DallaSanta
Clara Orbe
David H. Rind
Jeffrey A. Jonas
Larissa Nazarenko
Gavin A. Schmidt
Gary Russell
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(3211 KB) - Metadata XML