the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 01 Jun 2026
Surface impacts of Sudden Stratospheric Warmings (SSWs): Comparison of 2018 and 2019 SSWs in SNAPSI experiments
Abstract. Stratospheric Sudden Warmings (SSWs) significantly affect surface climate in boreal winter. However, their impacts vary considerably from one event to another: more than one-third of SSWs are not followed by expected tropospheric anomalies. To isolate the forced responses to SSWs, this study analyzes model experiments from the Stratospheric Nudging And Predictable Surface Impacts (SNAPSI) project, where the stratospheric zonal-mean state is nudged using either observations or climatology. The differences between the two experiments are examined within a multi-model ensemble framework. Nudging experiments conducted for the 2018 SSW, which featured significant tropospheric responses, and the 2019 SSW, which showed none, reveal that both SSWs consistently drive a negative Northern Annular Mode over time. The forced tropospheric response is primarily driven by an increase in Arctic surface pressure resulting from poleward mass fluxes in the stratosphere and upper troposphere. The poleward mass fluxes are initially induced by the zonal wind nudging in the middle stratosphere and subsequently by the eddy heat and momentum fluxes in the stratosphere and upper troposphere. This result suggests that while SSWs intrinsically drive tropospheric anomalies, the internal variability in the troposphere strengthens or suppresses the forced anomalies from the stratosphere, which may determine the existence of expected tropospheric anomalies following SSWs.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Weather and Climate Dynamics.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(1598 KB) - Metadata XML
-
Supplement
(632 KB) - BibTeX
- EndNote
Status: open (until 13 Jul 2026)
- RC1: 'Comment on egusphere-2026-2798', Anonymous Referee #1, 06 Jul 2026 reply
-
RC2: 'Comment on egusphere-2026-2798', Anonymous Referee #2, 08 Jul 2026
reply
Review of “Surface impacts of Sudden Stratospheric Warmings (SSWs): Comparison of 2018 and 2019 SSWs in SNAPSI experiments” by Dong-Chan Hong et al.
Overall Assessment
In this study, the authors analyse multi-model forecast experiments, initialised around about the onset of two different major midwinter SSWs that occurred in February 2018 and January 2019 in the Northern Hemisphere, conforming to the SNAPSI protocol (detailed in Hitchcock et al., 2022). They compare tropospheric circulation response following each event by examining the difference (SSW signal) between simulations nudged to stratospheric observed conditions and those nudged to climatological conditions and further compare results against observations (ERA5). The detailed assessment is greatly facilitated through decomposition of the origin of tropospheric anomalies (temperature versus surface pressure induced), elucidation of the polar-cap averaged surface pressure budget and diagnosis of the meridional circulation vertically through the stratosphere-troposphere.
A key message is that the polar-cap averaged evolution of the 2018 and 2019 SSWs was not so different, and the expected tropospheric circulation response and attendant changes in surface weather might have been expected to be similar, probabilistically speaking. Although not a focus of this study, the authors point to the likely role of modulating tropospheric factors in shaping the real-world response (downward propagating 2018 event versus non-downward propagating 2019 event) and find supporting evidence that the muted 2019 tropospheric response from ERA5 lies within the ensemble spread of the experiments (range due to internal variability).
The study is well executed, and the findings constitute a significant advance in our understanding of the different processes and mechanisms associated with stratosphere-troposphere coupling, in explaining the expected tropospheric/surface response. I have a few comments that should be taken into consideration by the authors, but suggest the manuscript is largely already in an acceptable state for publication in Weather and Climate Dynamics. I have just one possible concern over potential confounding influence on the results due to initialisation date differing between the two events, so suggest a minor revision round is completed first before publication.
General Comments
Reference to Tropopause: Should the definition of tropopause used by the authors be stated? Presumably, a WMO standard lapse-rate definition was used, but many other tropopause definitions do exist. I suggest the authors clarify this from the outset.
References to Hong and Son (2025): Although I appreciate the frequent reference to this study, given it builds upon existing authors’ work, I would have expected to see more reference to other literature such as Baldwin et al. (2024), which explore the likely mechanisms for surface amplification of the stratosphere-troposphere coupled response.
For instance, the authors make no reference to the likely role of compression on the polar tropospheric air column, due to polar vortex weakening, aided via lowering of the tropopause height. In relating their results to theory, the combined roles of PV anomaly and stretching should be discussed (Fig. 8b in Baldwin et al., 2024).
Forecast Initialisation Dates (2018 versus 2019 event): I couldn’t help but notice and wonder, if the results shown would in anyway be sensitive to the initialisation dates for each event falling on opposite sides of the SSW onset (4 days before 2018 event and 6 days after 2019 event).
I appreciate the authors make use of the forecast experiments as a community dataset, thus did not perform the experiments themselves and had no control over the initialisation dates. But I would have expected at the very least some discussion over the likely importance (or not) of this difference when comparing between the two events.Could this also explain (if only in part), why the 2019 NUDGED – CONTROL signal is somewhat weaker than for 2018? Because the initialisation date for 2019 is after the event, the control U10 at 60°N wind ramps up from ~0 to 35 m s-1, so a signal from the SSW state at the beginning of the forecast will also be present in the CONTROL simulations, at least potentially impacting the troposphere during the period of the forecast examined by the authors?
Specific Comments
L34: “This leads to surface extremes…” --> Perhaps not always, certainly on a regional level, so maybe: “This leads to enhanced risk of surface extremes…”
L74-75: The authors should add/cite here that zonally symmetric nudging has been found in free-running models to reproduce closely the modelled tropospheric response to SSW events (Hitchcock and Simpson, 2014). And importantly, zonally symmetric nudging protocols in the stratosphere have been verified in earlier studies to not induce any significant artificial properties in tropospheric circulation (Hitchcock and Haynes, 2014).
Fig. 1 & 2: I wonder if a vertical line as opposed to a small gray triangle on x-axis might help the timing of SSW onset to stand out better, certainly for Fig. 1. The transparency/linewidth can be customised to ensure this does not obscure the filled contour field?
L219-220: As mentioned under ‘General Comments’, the mechanisms for surface pressure increase over the Arctic, in response to downward coupling of SSW-induced anomalies, was the subject of much attention in Baldwin et al. (2024). This study and ideally other relevant literature should be cited and discussed.
L363-371: I found myself a little confused when reading this, the pressure ranges the features of note correspond to. I suggest the authors could improve by specifying pressure ranges in this paragraph where possible or alternatively ‘upper/mid/lower stratosphere or troposphere’ to help the reader out.Technical Corrections
L94: “SSW” --> “SSWs”
L139: “…one-moth-averaged…” --> “…one-month-averaged…”
L318: “(Figs. S4)” --> Singular as its just one Figure?: “(Fig. S4)”Additional References
Baldwin, M. P., Birner, T., & Ayarzagüena, B. (2024). Tropospheric amplification of stratosphere–troposphere coupling. Quarterly Journal of the Royal Meteorological Society, 150(765), 5188-5205, https://doi.org/10.1002/qj.4864.
Hitchcock, P., & Haynes, P. H. (2014). Zonally symmetric adjustment in the presence of artificial relaxation. Journal of the Atmospheric Sciences, 71(11), 4349-4368, https://doi.org/10.1175/JAS-D-14-0013.1.
Hitchcock, P., & Simpson, I. R. (2014). The downward influence of stratospheric sudden warmings. Journal of the Atmospheric Sciences, 71(10), 3856-3876, https://doi.org/10.1175/JAS-D-14-0012.1.Citation: https://doi.org/10.5194/egusphere-2026-2798-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 55 | 16 | 6 | 77 | 7 | 4 | 4 |
- HTML: 55
- PDF: 16
- XML: 6
- Total: 77
- Supplement: 7
- BibTeX: 4
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of “Surface impacts of Sudden Stratospheric Warmings (SSWs): Comparison of 2018 and 2019 SSWs in SNAPSI experiments” by Hong et al.
General comments
This work systematically analyses and compares the dynamical pathways through which two SSW cases influence the surface, using the SNAPSI experiments. I find the manuscript interesting, and the analysis itself is potentially useful. However, I think the current structure frames the paper rather generally as a study of SSW surface impacts, and the central scientific question is not yet made clear enough. I therefore recommend that the authors reorganize the structure and highlight the key question and new contribution more clearly.
Major comments
1. The framing of the title, abstract, and introduction is too general.
Currently, the first parts of the manuscript give the impression that this work mainly aims to understand the impacts of the two SSWs using the SNAPSI experiments. However, in my reading, the paper is not simply asking whether SSWs affect the surface, as some previous studies have already addressed this question. Rather, the more specific and interesting question is whether two contrasting SSW cases have different or similar forced downward responses and dynamical pathways when the stratospheric state is controlled. This is an important result, because it provides clearer evidence that the observed difference between the two events may mainly reflect tropospheric internal variability rather than fundamentally different stratospheric downward pathways. I therefore think this key question should be stated much more clearly in the Introduction, also the title of this paper should be modifed to better reflect it.
I also suggest moving Fig. 1, or a simplified version of it, to the Introduction. Fig. 1 mainly establishes the observed contrast between the 2018 and 2019 events. This is important motivation for the study, but it is less a new result. Introducing the two cases with this figure would help the reader understand early on why these two events are being compared.
Related to this, there is some repetition between the Introduction, Section 2.1, and Section 3.1. The Introduction discusses the general issue of SSW surface impacts and previous nudging studies; Section 2.1 again introduces SNAPSI and the two cases; and Section 3.1 again establishes the observed contrast between the two events. I suggest streamlining these sections: use the Introduction to introduce the two cases and the key scientific question, keep the Methods focused on the experimental design and diagnostics, and start the Results with the SNAPSI-based comparison.
2. The nudging logic and tendency interpretation should be clarified earlier.
This is especially important for this manuscript because the later analysis relies on detailed KE diagnostics and tendency decompositions. At present, some of the relevant explanations appear later in the interpretation of Figs. 8–11, but they are somewhat buried there. I suggest at least adding a brief explanation in the Methodology of what tendency terms are available from the model output, which terms are diagnosed as residuals, and how direct nudging tendencies, parameterized tendencies, and resolved eddy-flux responses should be interpreted. This would provide the necessary basis for understanding the later diagnostics and for distinguishing the directly imposed nudging response from the subsequent dynamical adjustment of the model atmosphere.
Specific comments
L123: Please clarify what “these terms” refers to.
L133–142: As noted in my major comment 1, the contrasting impacts of the 2018 and 2019 SSWs have already been introduced several times. I suggest integrating this part, together with Fig. 1, into the Introduction.
Figs. 1 and 2: The grey triangles are not very visible. I suggest changing them to another colour or making them more prominent.
L187: Please specify that “the absence of SSW-related signal in the observations” refers to the 2019 SSW.
L205–209: The transition to the next section is too general and does not fully reflect the core question. The analysis is not simply about how SSWs drive tropospheric anomalies, but about diagnosing the dynamical pathway of the forced response and comparing whether this pathway differs between the 2018 and 2019 cases. I suggest revising this sentence accordingly to better reflect the storyline.
L212–217: The statement here seems too strong, as the conclusion is drawn from Fig. 5, where the geopotential height response has been standardized. I suggest either softening statements such as “only in the stratosphere” and “instead of a time-lagged direct downward propagation” or also checking the results showing the unstandardized decomposition.
L273–276 / Fig. 9: Please clarify how the direct nudging tendency is separated from the parameterized zonal wind tendency. It would be useful to state whether these tendencies are directly available from the model output or diagnosed indirectly (as in my major comment 2).