Interannual variations of water vapor in the tropical upper troposphere and the lower and middle stratosphere and their connections to ENSO and QBO

Tian, Edward W.; Su, Hui; Tian, Baijun; Jiang, Jonathan H.

doi:https://doi.org/10.5194/acp-19-9913-2019

Articles | Volume 19, issue 15

https://doi.org/10.5194/acp-19-9913-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/acp-19-9913-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 19, issue 15

Research article

|

07 Aug 2019

Research article |

| 07 Aug 2019

Interannual variations of water vapor in the tropical upper troposphere and the lower and middle stratosphere and their connections to ENSO and QBO

Edward W. Tian, Hui Su, Baijun Tian, and Jonathan H. Jiang

Download

Final revised paper (published on 07 Aug 2019)
Preprint (discussion started on 05 Dec 2018)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

SC1: 'Reference needed', Michaela I. Hegglin, 05 Dec 2018
RC1: 'Review', Anonymous Referee #2, 11 Dec 2018
RC2: '... some major issues', Anonymous Referee #1, 04 Feb 2019
AC1: 'Final Response', Baijun Tian, 17 Feb 2019

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Baijun Tian on behalf of the Authors (19 Feb 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (06 Mar 2019) by Gabriele Stiller

RR by Anonymous Referee #2 (14 Mar 2019)

Suggestions for revision or reasons for rejection

Review of the manuscript "interannual variations of water vapor..." by E. . Tian et al.

This is a review of the revised manuscript. The two most critical issues of the original version of the manuscript were the lack of new results and the weak predictive power of the linear regression presented. I think that these problems have not been solved; thus I cannot recommend publication of this manuscript in its present form. The material presented, however, seems to contain more information than exploited by the authors and I think that another major revision could make the paper acceptable.

1. The weak explanatory and predictive power of the regression analysis:

In the abstract (p1, l20) the authors claim that interannual tropical water vapor anomalies are MAINLY determined by ENSO... is clearly false. This wording suggests that a major fraction of the variability can be explained, but the R**2 values suggest (assuming Gaussian statistics) that not even half of the variation is explained. At the very least a more modest wording is needed throughout the paper in a sense that ENSO and QBO do make contributions (of varying
weight) but that more than half of the variation remains unexplained. The authors refuse to strive for a more complete explanation of the water vapor variabilities and they offer arguments to justify this decision. I agree with the authors insofar as a multilinear regression is indeed not needed if the goal is not to explain the water vapor variability. The fact that only a fraction of the variability is explained needs to be clearly stated in the paper. The question remains what, if not the explanation of the water vapor variability, shall be main the scientific content of the paper. This leads me to the second major point:

2. The lack of new results:

The authors claim that the goal of the paper is to determine the relative weight of QBO and ENSO as a driver for water vapor concentrations. This, as such is not new, and the authors argue that they use a longer time series and a better altitude coverage than preceding work. With this, the added value is only incremental. The authors analyze the time lag between the QBO proxy as a function of altitude but fall short of interpreting this result in a sense of providing a scientific explanation of this relation. This relation, however, may be the key to new insights into the driving mechanisms and may help to deviate from worn paths. My proposed chain of arguments runs as follows:
1. The slope of the time lag with altitude (Fig 3) matches that of the water vapor anomaly (Fig 2) pretty well. This suggests that the QBO does NOT directly affect the water vapor concentration at altitudes higher than 100 hPa but that the QBO signal is imprinted at about 100 hPa and is transported upward with the tape recorder. This does not come as a surprise from a theoretical point of view but I am not sure if any empirical evidence has yet presented for this. It is clear that the tape recorder transports any imprinted signal upwards, but a specific evidence that this holds for the QBO part of the variation may be new.
If not such things, what else would be the benefit that the analysis covers more altitudes than previous studies?
2. Between about 120 hPa and about 40 hPa R**2 seems to decrease with altitude. This is plausible an can be understood by mixing, dilution, a leaky pipe or whatever.
3. Interestingly, R**2 has a maximum at about 14-20 hPa. This cannot easily be explained by transport of an imprinted 100 hPa signal by the tape recorder. This phenomenon begs for an explanation and deserves a thorough discussion of related literature in order to identify possible explanations.
I think the other studies still can remain in the paper. There is nothing wrong with them but they seem to me to be quite a routine application of research tools without leading anywhere. In combination with the suggestion above, they still provide useful additional incremental knowledge.

3. Specific comments:
p2 l15 shouldn't the references be listed in chronological order?
p9 l13/14 but what if the biases included in the uncertainty estimates are time-, temperature- or whatever dependent?
p12 l15-18 is there any evidence that these assumptions are justified? Histograms of similar? If the residual variance is not a random effect but caused by unknown forcing, the assumption of a Gaussian distribution may not be adequate.
p21 l5 I find it audacious to claim that ESO and QBO explain most of the variation, facing the fact that R**2 hardly exceed 0.5. "Make significant contributions" would be more adequate.
Fig 1: Axes captions too small
Fig 4/5: Legend too small

Hide

RR by Anonymous Referee #1 (01 Apr 2019)

ED: Reconsider after major revisions (16 Apr 2019) by Gabriele Stiller

AR by Baijun Tian on behalf of the Authors (14 May 2019) Manuscript

ED: Referee Nomination & Report Request started (01 Jun 2019) by Gabriele Stiller

RR by Anonymous Referee #2 (09 Jul 2019)

ED: Publish subject to minor revisions (review by editor) (09 Jul 2019) by Gabriele Stiller

AR by Baijun Tian on behalf of the Authors (09 Jul 2019) Author's response Manuscript

ED: Publish as is (11 Jul 2019) by Gabriele Stiller

AR by Baijun Tian on behalf of the Authors (11 Jul 2019) Manuscript

Short summary

We study the interannual (2–7-year) water vapor variations in the tropical upper troposphere and the lower and middle stratosphere and their connections to El Nino–Southern Oscillation (ENSO) and quasi-biennial oscillation (QBO) using the Aura Microwave Limb Sounder (MLS) data and time-lag regression analysis and composite analysis. We found that ENSO is more important in the upper troposphere and near the tropopause, while QBO is more important in the lower and middle stratosphere.