the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Mar 2025
Covariability of dynamics and composition in the Asian monsoon tropopause layer from satellite observations and reanalysis products
Abstract. We describe three leading modes of deseasonalized water vapor variability in the tropopause layer (147–68 hPa) above the Asian summer monsoon (ASM) based on Aura Microwave Limb Sounder (MLS) satellite observations and five meteorological and composition-focused reanalyses. The first mode, which describes regional-scale moist or dry anomalies on interannual scales, is separated into a linear trend and detrended interannual variability. Although the reanalysis products all show an increasing trend in tropopause-layer water vapor over 2005–2021, the spatial pattern and sign of the trend disagree between Aura MLS and the reanalyses. The regional water vapor budget indicates that the reanalysis trend originates outside the monsoon region, beyond the domain of our analysis. Interannual variability is otherwise consistent, arising mainly from the pre-monsoon influence of the quasi-biennial oscillation. The second mode features anomalies arcing from the southwestern to northeastern quadrants of the anticyclone coupled with weaker opposing anomalies in the southeast, while the third mode features a horizontal dipole oriented east-to-west. These two modes often vary in quadrature due to the influences of quasi-biweekly waves on deep convective activity, but also appear independently when other modes of convective variability manifest in similar centers of action. Although questions remain regarding the linear trend, mean biases, and the weak and possibly adverse influence of data assimilation, the consistency between Aura MLS and reanalysis-derived modes of variability in UTLS water vapor in this region shows that atmospheric reanalyses are increasingly able to capture the processes controlling water vapor near the tropopause.
- Preprint
(13199 KB) - Metadata XML
-
Supplement
(13237 KB) - BibTeX
- EndNote
Status: open (until 06 May 2025)
-
RC1: 'Comment on egusphere-2025-543', Mengchu Tao, 15 Apr 2025
reply
This paper focuses on analyzing water vapor variability in the Upper Troposphere–Lower Stratosphere (UTLS) region of the Asian Summer Monsoon (ASM) area. The authors use satellite data from Aura MLS and five atmospheric reanalysis datasets, including MERRA-2, M2-SCREAM, CAMS, ERA5, and JRA-3Q, to conduct a spatiotemporal mode analysis. The main results reveal three key modes of variability, including PC1, large-scale regional water vapor wetting or drying anomalies, PC2 & PC3, intraseasonal oscillations linked to quasi-biweekly variability.
A key conclusion in the paper, from my point of view, is the discrepancies between reanalysis products and Aura MLS data regarding spatial distribution and sign (positive/negative) of water vapor trends. While most reanalysis products show an increasing water vapor trend in the southeast of Asian monsoon during warm season, their spatial characteristics differ significantly from those derived from Aura MLS.
Overall, the paper is novel in design, well-written, and thorough in its data analysis, comparisons of methods, and results. I recommend minor revisions to address the following points for further improvement.
General comments
The analyses, particularly those focused on interannual variability and intraseasonal oscillations, are convincing and well-executed. My concerns are mainly about the analysis of PC1 trends. While the discussion clearly identifies the differences in trends between Aura MLS and reanalysis products, I would suggest a deeper exploration of the reasons behind these discrepancies.
Another question from my side, whether the second reason to doubt the reanalysis-based trends are robust: “trends in cold point tropopause (CPT) temperatures are negative in the southeastern quadrant of the anticyclone where the reanalyses show the largest positive trends in water vapor” (in conclusion). If reanalysis WV increase mainly under the tropopause and thus increase the PCWV, it seems be consistent with OLR/cloud trend (convection increase pattern shown in Figure 6). It thus meets “criteria 3: a plausible physical mechanism”. And WV decrease due to local CPT decrease is not a main driver since the mass of WV in the LS is much less than that in the UT. Surely, long-term warming of the tropical cold point tropopause increasing WV outside the monsoon region can be another reason. But I don’t see the contradiction between CPT cooling and PCWV increasing over one region.
To shed more lights on this point, my suggestion could be:
1) the trends (and their spatial characteristics) be further decomposed into contributions above and below the tropopause;
2) further analysis of the "dyn" and "phy" terms individually (specifically their behaviours above and below the tropopause) in Fig. 7. This could potentially reveal whether convection is the primary driving factor in the trends observed. And this could uncover systematic patterns or consistencies within these two terms from reanalysis datasets.
Specific comments
- Following EQ (2) in Line 160, the terms "Sphy" and "Sana" should be briefly explained. Specifically: What key physical processes related to water vapor are captured by "Sphy" (e.g., condensation, deposition, subsidence, etc.)? What is the role of "Sana," particularly in relation to the data assimilation process? I also wonder whether subgrid-scale mixing is included in the "Sres" term?
- Figure 1: The titles of the three subplots for MERRA-2 (panels am-cm) are incorrect; they should refer to "PC1" instead of "PC2."
- Figure 3 (panel b): Please reduce the y-axis range to between -2 and 2 to allow for better visualization. Consider setting the y = 0 reference line to gray for improved clarity.
- I wonder why MLS trend show larger positive trend than merra-2 (Fig. 3)? It seems conflict with Fig.1. Does that mean MLS PC1 trend for the whole region is positive? And the positive trend is highly contributable from 35-45N latitude band according to Fig.1 (b)?
Citation: https://doi.org/10.5194/egusphere-2025-543-RC1 -
RC2: 'Comment on egusphere-2025-543', Anonymous Referee #2, 18 Apr 2025
reply
Review of “Covariability of dynamics and composition in the Asian monsoon tropopause layer from satellite observations and reanalysis products” by Shenglong Zhang et al. [2025]
This work focuses on three leading modes of interannual variability in water vapor in the tropopause layer using the measurements from the MLS satellite and multiple reanalysis products. The first mode is linear trend and interannual variability in regional-scale anomalies, which show some differences in the satellite data and reanalysis. The second and third modes are related to anomalies within the monsoon anticyclone, such as, variabilities within the quadrants and a horizontal east-to-west dipole structure. The results show reanalysis captures the modes of variability in water vapor in the upper troposphere and lower stratosphere and the physical processes controlling them. The results shown here are based on comprehensive, thorough and very detailed analyses. However, I did not see clear motivation and goal of the work. Below are my comments for the authors might take into consideration.
General Comments
L1 (Abstract): Rather than starting with ‘we describe’, I recommend start with some background information including why the Asian summer monsoon is important and what the goal of this study is. This will make the abstract more appealing.
L16 (Introduction): It was mentioned that the goal of this study is to provide further insight into the mechanisms governing variations in water vapor. More specific information could be added here. What are examples of the mechanisms that we need to understand further? What variations in water vapor is discussed here? What is the science goal? A clear motivation and some scientific context of this work will be necessary. This work maybe relevant to the fact that there will be less observations of stratospheric water vapor available from satellites in the near future.
In depth and very detailed analyses and descriptions of the results are presented in this work. I found it rather hard to understand all the detailed descriptions of figures. Many of the sentences are long and the description of results contains some speculation, besides facts. It would be helpful if some of the long sentences are split into multiple short sentences and simplify the descriptions.
It would also be helpful to include some context of the results from this work relative to previous studies throughout the main text. How is the result shown here different from previous work? Are they consistent with or different from previous work? This will help understand the results more scientifically. Is EOF analyses giving us new information that has not been discovered?
I think it would be helpful to provide some outlook. For instance, information about which reanalysis products represent dynamical or thermodynamical processes near the monsoon region well so that we can trust?
Specific Comments
L39 – “The smooth boundaries and distinct shape of the climatological anticyclone” can be explained further with specific descriptions here. What does ‘smooth’ mean? Does ‘distinct shape’ refer to eddy shedding event?
L44 – ‘constrained’ could be replaced with ‘understood’.
L51 – Here ‘associations’ sound vague. I am wondering this could be replaced with some other word.
L57 – Seasonal dilution of ozone in the UTLS above the monsoon anticyclone…
L60 – air to the -> air into the
L65 – A brief mention of why all these species are analyzed together will be useful here.
L66 – Here ‘further insight’ sounds vague. Consider replacing it with more specific terms. For instance, ‘analyzing seasonal behaviors or interactions between various processes that have not been analyzed before’.
L68 – Instead of ‘pay attention’, mentioning what is new in this work compared to the related work (Tegtmeier et al. and Wright et al.) would be recommended.
L77- A couple examples of the reanalysis products included in the supplement material would be helpful.
L81 – What horizontal grids are used in the gridding?
L100 – Does this sentence mean that the metrics are used to distinguish the effects of convection, thermodynamic conditions and transport?
L112 – Is the ‘replay’ technique commonly used or specific to this study?
L121 – I find it hard to understand this sentence.
L140 – The meaning if this sentence is unclear as well. Are specific humidity tendencies produced by parameterized physics and data assimilation?
L159 – Here ‘mechanisms behind’ could be replaced with a sentence with more specific terms. For instance, ‘contributions from physical processes in determining temporal variability of water vapor’.
L171 (paragraph) – This paragraph contains many steps. It is hard to follow each step based on the writing. It would be easier if there is a simple diagram showing the flow.
L176 - What does ‘weighted equally’ mean here?
L181 – What does ‘reordered for consistency’ mean?
L184 – ‘underlying mechanisms’ can be further explained here.
L207 – Instead of stating ‘no prior expectation’, one can try to find if there is a known trend in water vapor.
L218 – Can be more specific in ‘indirect indication’ here.
L239 – I wonder what is causing ‘moistening and drying’ in the anticyclone?
L245 – Also what processes contribute to this dipole structure?
L267 & L274 – PC2 & PC3: Are these results consistent with what Randel et al. (2015) have shown?
Figure 5 – It is striking to see how the left panel (MLS) and the right panel (CAMS) look very different for all the species. It is not easy to understand the explanations of this figure.
L303 – I am not sure how much we can assume that the drift correction for Aura MLS water vapor contributed to the trend. Is it just a speculation or based on some findings?
L307 – ‘mechanisms behind’ sounds vague. This sentence could simply be ‘it is important to understand the reanalysis-based trend’.
L330 – ‘The robust…southeast.’ could be split in two sentences.
L340 – ‘largest increasing trends’ -> Is this trend based on reanalysis?
L363 – Need citations for this sentence. Also please include the sources for the ONI and the QBO time series.
P22, Figure 10 – This is a busy figure. I am wondering if this figure could be split into two, or the longitude range could be reduced to cover only the monsoon area.
L507 – The paragraph starting with ‘Oscillations of the jet…’ is complicated. I would suggest read it again and revise it if possible.
L571 – ‘Concurrent variations…’ -> This sentence seems to be based on speculations.
L580 – ‘which is…local dehydration…vortex’ -> I am not sure what this means. Does this mean condensation due to cold temperature?
L582 – ‘highly coherent’ could be replaced with ‘coherent’. What does ‘dynamical reorganization’ mean?
L611 (Discussion) – I would like to suggest adding a few sentences at the beginning explaining why EOF analyses were used in this work. Comparing with previous work which did not use EOF analyses, and their limits would be useful.
L627 – Can ‘mechanisms behind’ be just ‘role of’?
L682 – ‘mechanisms behind’ could be just ‘physical processes?
L684 – ‘The results serve…’ -> Here, I would like to know what we have learned from this paper not what was done in this paper.
Citation: https://doi.org/10.5194/egusphere-2025-543-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 103 | 27 | 9 | 139 | 15 | 6 | 6 |
- HTML: 103
- PDF: 27
- XML: 9
- Total: 139
- Supplement: 15
- BibTeX: 6
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 57 | 39 |
| China | 2 | 24 | 16 |
| Germany | 3 | 13 | 9 |
| undefined | 4 | 9 | 6 |
| France | 5 | 5 | 3 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 57