the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 23 Feb 2026
Co-occurrence of gravity waves, vertical wind shear and turbulence in the lowermost stratosphere over the North Atlantic
Abstract. This study focuses on the spatial and temporal co-occurrence of gravity waves (GWs) and of vertical shear and turbulence in the extratropical lowermost stratosphere (LMS). For this, one year of ERA5 reanalysis data is used to analyse the occurrence and variability of resolved GW momentum fluxes, vertical shear, and turbulence indices in the region of the North Atlantic. In the tropopause region GWs can lead to strong vertical wind shear and ultimately to the generation of turbulence, which may result in troposphere-stratosphere exchange and mixing and is a risk for commercial aviation. The occurrences of GWs, shear, and turbulence are analysed in terms of vertical, geographical, and seasonal occurrence frequency distribution and put in relation to the large-scale circulation and to processes related to GW appearances. The contribution of resolved GWs to shear is found to be notable, especially in the North Atlantic winter lowermost stratosphere, where vertical momentum flux convergence provides a peak zonal GW forcing up to −2 m s-1 day-1 around 45° N at tropopause altitudes. The prominent vertical propagation in the wintertime mid-latitudes substantially leads to the formation of belt-like structures of GW activity, as evident by momentum fluxes, and further contributes to the pronounced occurrence of shear in the LMS. Ultimately, this study discusses the role of small-scale dynamics in shaping a quasi-permanent layer of elevated shear above the extratropical tropopause and its potential to generate turbulence in this region.
- Preprint
(23319 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (extended)
- RC1: 'Comment on egusphere-2026-413', Anonymous Referee #1, 01 Apr 2026 reply
-
AC1: 'Quick reply and clarification on egusphere-2026-413', Madhuri Umbarkar, 16 Apr 2026
reply
Please see the attachment for our general response and clarification on the manuscript.
-
RC2: 'Comment on egusphere-2026-413', Anonymous Referee #2, 19 Apr 2026
reply
Comments on “Co-occurrence of gravity waves, vertical wind shear and turbulence in the lowermost stratosphere over the North Atlantic” by Umbarkar et al.
The manuscript investigates the role of atmospheric gravity waves in inducing vertical shear layers in the lowermost stratosphere during Boreal winter. Much of the methodology used in the work has been established in past works (but all or a subset of the authors). The authors have studied the spatio-temporal variability of gravity wave activity, shear inversion and strong shear occurrences, instability conditions, and turbulence in the UTLS, using established metrics and indices, to conclude strong co-location between gravity waves and vertical shear, implying that during Boreal winters, gravity waves can play a substantial role in shaping the small-scale shear layers found in the extratropical lower stratosphere.
While reading the manuscript, my main concern was that simply looking at the climatology and drawing conclusions about dynamical connections can be misleading. However, this concern was taken care of in Figure 4.1 where the authors considered case studies for each season to show dynamical connections between enhanced gravity wave activity, distortion of PV and local shear. I still consider this a statistical evaluation and connection between gravity wave perturbations and atmospheric stratification (missing dynamical mechanism and evidence). However, I agree that a full dynamical explanation using idealized simulations or otherwise would be beyond the scope of the study.
The dataset and methodology is overall sound, and while a longer time period of ERA5 would have been ideal for the evaluation (as there are some springtime or summertime deviations), I find the conclusions drawn in the manuscript convincing. The work would add to our understanding of what processes shape the stratification I the UTLS, which also highlights an important way in which gravity waves contribute towards shaping the climate mean state. I have some concerns, which I would classify as minor. I would suggest accepting the manuscript for publication after these comments are satisfactorily addressed.
Major Comments
1) Writing: I admire the length to which the authors have gone to clearly and logically establish the connection between gravity waves and the enhanced shear in the lower stratosphere, but the manuscript is unnecessarily longer than it needs to be and can be significantly trimmed (without losing content). Moreover, the manuscript would be much clearer and enjoyable to read if the authors could reframe some sentences in the results section and move some plots to the SI. For instance: - I would suggest moving Figures 6, 8, 10, and 14 to the SI and referring to them briefly in the text. Figure 6 has been discussed briefly. T1 and T2 have similar structures, so one of the two figures can be moved to the SI. Similarly, only one of the VMFC and GW momentum flux maps is needed in the main body. Figure 14 (histogram) is only needed for a quick reference as the crux of your argument is illustrated in Figure 15 (scatterplot). I found having this many plots very distracting. This was aggravated by the unclear writing in multiple places (see next point)
- Every subsection begins and ends with redundant text, which in my opinion, disturbs the reader’s flow. For example, L395-400 is not needed. Section 4.3.1, the first 4-5 lines are not needed. The final two lines of section 4.3.1 are redundant. Likewise, I would say the first 7-8 lines of section 4.3.2 are not needed. Similarly, L175-180: part of this can be moved to the conclusion section, and the remaining half removed. I found these forced 5-6 line segues, subsection after section quite distracting.
I appreciate the effort that went into writing this, but I would recommend the authors to revise the text. It is easily possible to cut 2-3 pages off the text (apart from moving above said figures to SI) while improving readability.
Minor Comments:
- The introduction can benefit from a revision. For instance, L54-73 read mixed up, with a couple of sentences on dynamics, followed by a couple of sentences on applications, followed by a couple of sentences on data, followed again by a couple of sentences on dynamics. This is easily seen on lines 63-74.
- L89: Are sponge layers and hyperdiffusion applied on all levels?
- L113: Asian continent? Please specify longitudes?
- L115: Rossby or gravity? Which one is it?
- L145: provide citation?
- L174: Where exactly is this shown?
- L180-185: Do the authors have an intuition on why March (springtime) is stronger than February? Such anomalies are also on L188 (May is stronger than Feb). Similarly, Figure 12 and the discussion around L414-419.
L186: I wouldn’t use the word “band” for this, since the structure is localized for some months and quite patchy. Don’t you agree? Possibly, multi-year averages produce a band?
Figure 2: If is possible to superimpose the zonal mean zonal winds in Figure 2 itself? Then the reader can zee the connection in the shear and the mean wind from the get-go
Figure 3: Possible to adjust the pink shading? Difficult to assess the zonal wind map for some months due to the overpowering pink above it.
L231: You mean GW dissipation? Because GW activity is high even down below in regions with high Mfs
L231: If the gradients are high, does it mean limited transport (and not substantial transport)?
Figure 6: Isn’t the climatological shear layer above the region of potential instability? So, is the argument that mixing due to GW-induced turbulence in the region below can create a highly stable region above it (as shown in Figure 2)
L265: Can you clarify what “tropopause-based coordinates” are? Are you simply referring to the dynamic tropopause and the 380 K isentrope? This phrase has been used repeatedly. Sorry if I missed the definition.
L288: Analysis? You mean this is your hypothesis?
L305: Maybe I missed it somehow, but isnt he vertical extent of TI2 broader than TI1, rather than more sharply confined as mentioned?
L314: suggests the substantial → suggests the possibility of substantial.
L316-319: Maybe move this to the conclusion section?
L331: 25 → 30N?
L428: turbulence is a rare event: I am not sure I understand this statement. Are you referring to wave-induced turbulence? Otherwise, the atmosphere is highly turbulent with a high Reynolds number and ample perturbations. In fact, the prevailing turbulence all the way to the Rhines scale maintains the midlatitude jets. Isn’t it?
Figure 15: Interesting scatterplot. Does this mean gravity waves need to carry MF above a threshold to be able to induce inversion and turbulence in the UTLS? Which means even low-amplitude waves can still manage to induce inversions provided their vertical wavelengths are large enough and/or their horizontal wavelengths are small enough?
Citation: https://doi.org/10.5194/egusphere-2026-413-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 155 | 164 | 16 | 335 | 17 | 35 |
- HTML: 155
- PDF: 164
- XML: 16
- Total: 335
- BibTeX: 17
- EndNote: 35
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript addresses a highly relevant topic: the relationship between gravity wave (GW) activity and the formation of tropopause shear layers (TSLs). The research question is well-posed and has the potential to contribute significantly to our understanding of upper-tropospheric and lower-stratospheric (UTLS) dynamics. However, in its current form, the link between the observed correlations and the physical mechanisms remains somewhat speculative. I believe the paper would be much stronger if the authors could move beyond describing "co-occurrence" and provide a more detailed mechanical explanation. I recommend a reject with an encouragement to resubmit to allow the authors time to incorporate high-resolution data and deepen the discussion on dynamics. Following are my comments for the authors to increase the impact and clarity of the work.
Major Comments:
C1.
The manuscript frequently highlights the "importance" of GWs in TSL formation. To make this argument more compelling, it is necessary to explore the specific physical processes involved. For instance, are these TSLs a result of GW breaking and subsequent momentum deposition, or are they the result of the superposition of large-amplitude waves? Additionally, providing information on the characteristics of the waves (e.g., their sources and properties) would greatly enhance the scientific depth of the study. What kind of GWs are responsible for the formation of TSLs?
C2.
While ERA5 offers hourly resolution, the current analysis relies on once-daily data. Since both gravity waves and turbulence are transient and evolve on much shorter timescales, I encourage the authors to utilize the high-frequency (hourly) data. This would allow for a more rigorous assessment of the temporal evolution of these phenomena and provide more robust evidence for the study's conclusions.
C3.
At present, some of the conclusions regarding the "role" of GWs appear to precede the physical evidence. I suggest re-evaluating the causal links to ensure that the data clearly supports each step of the argument. Framing the results more cautiously or providing the missing physical links will help avoid the impression of a circular reasoning.
Specific Points:
l. 29: Since the topic shifts after the phrase “On both hemispheres,” I suggest starting a new paragraph here.
l. 101: Is there a specific reason for defining the dynamical tropopause at 3.5 PVU instead of the more conventional 2.0 PVU (e.g., Hoskins et al., 1985; Kaluza et al., 2021)?
ll. 137-138 It appears there is a grammatical error; the sentence should likely read: “The shear perturbations are calculated…”
e.g., l. 173: It is difficult to characterize a monthly variation observed only in a single year (2017) as an “annual cycle,” as such a variation likely includes both the mean annual cycle and interannual variability. I recommend using a more appropriate term to describe this specific variation.
l. 175: This sentence (“This finding…”) is somewhat abrupt and lacks clarity. Please specify which large-scale dynamics are being referred to and what roles they play. Furthermore, there appears to be a logical leap regarding the definition of the Richardson number. It would be necessary to present the shear (or Ri) associated solely with the large-scale wind to demonstrate the contribution of large-scale processes to TSL formation.
Fig. 3: The magenta shading is difficult to distinguish from the shading used for the 200 hPa zonal wind.
l. 184: “A intensity” is grammatically incorrect. Did the authors mean “intensification of meridional baroclinicity” or “intense meridional baroclinicity”?
l. 188: The phrase “A planetary circulation feature” is too vague. Could the authors briefly elaborate on the discussion in Kaluza et al. (2021) to provide better context?
ll. 222-224: The interpretation of Fig. 5 is not entirely clear at this point of the manuscript. Why do the authors examine GWMF specifically in the vicinity of high-frequency strong shear events rather than showing the overall GWMF? Also, please clarify what is meant by “tropopause-relative,” given that the tropopause height itself varies in the figure.
l. 223: For clarity, please consider changing “shorter GWs” to “GWs with shorter horizontal wavelengths.
ll. 227-230: There is an inconsistency in the list format: point (i) is a full sentence, while (ii) and (iii) are noun phrases. Additionally, the latter half of point (i) lacks a verb (perhaps the authors meant “could also be related to”?). Regarding the interpretation, enhanced absolute GWMF “indicates” or “shows” intense upward propagation rather than just “hinting” at it. Furthermore, I am not convinced that convection directly contributes to the MF above the tropopause in the manner described, even if MF is shown as an absolute value.
ll. 239-240: A potential reason for the strong MF in shear regions is that GWs may be generated by, or have longer vertical wavelengths near, the jet stream where strong shear exists. This indirect relationship, which is a correlation rather than a causal "role" of GWs, should be explicitly noted.
l. 255: The term “smaller hotspots” is unclear. Furthermore, attributing them to “specific dynamical events” without further evidence feels speculative.
ll. 263-265: The last sentence does not make lots of sense. The vertical shear term S2 in the definition of the Richardson number Ri≡N2/S2 already includes contribution from shear associated with GWs. Therefore, the difference between the distribution of S2 and Ri is solely attributable to N2. While GWs may contribute to the distribution of N2, this difference alone does not provide sufficient evidence that GWs play an important role.
ll. 288-289: Could the authors clarify which specific results support the argument made in the sentence starting with “In fact, … GW activity”?
ll. 318-319: The statement "Nevertheless, …” is a fair acknowledgment, and it highlights the very question I had throughout this section. I recommend that the authors avoid overstating the causal link before this point is reached.
ll. 549-558: The final paragraph of the Summary seems slightly disconnected from the main discussion of the paper.