| 12 Aug 2026
Meso to sub-mesoscale origins of strong surface winds and influence of surface waves in an extra-tropical cyclone
Abstract. Extra-tropical cyclones can bring extreme surface winds. Several mesoscale jets can be associated with these winds originating from the upper, mid, and lower troposphere. Sub-mesoscale processes have been shown to play a major role in the downward transport of jets, but few modeling studies represent them explicitly. This study sheds light in the relative contribution and storm-relative location of jets present in the explosive cyclone Alex which made landfall in southern Brittany in October 2020 causing extensive wind damage in Belle-Ile and further inland. A series of stand-alone atmospheric simulations and coupled wave-atmosphere simulations using the Meso-NH and WAVEWATCH-III® models were run. Their horizontal grid spacing was incrementally halved by two going from 1.6 km (numerical weather prediction mode) down to 100 m (large eddy simulation mode). The simulated storm is consistent across resolutions at the mesoscale with little differences in extreme wind values between the kilometric and hectometric resolutions. Waves coupling is shown to decrease surface wind extremes by 0.5 to 1.5 m s-1. Impacts on the surface drag coefficient and winds are not uniform spatially with winds decreasing ahead of the cold front and within the high wind core while decreasing elsewhere. Online trajectory calculations allow for a Lagrangian air mass tracking in Meso-NH. In the kilometric simulation, four distinct jets are shown to be responsible for the strongest winds. The evolution of state parameters along these trajectories helps match the jets to the classical conceptual model for extra-tropical cyclones. Evidence hints to the presence of a Sting Jet associated with Alex’s extreme winds, along with the more common Cold Conveyor Belt, and Dry Intrusion. The fourth jet is labeled as Potential Sting Jet on account of its similarities with the Sting Jet prior to subsidence. The Cold Jet is shown to be responsible not only for the majority of the strong wind trajectories in the lower 500 m, but also for the strongest wind speeds. Spatial distributions of these jets is further investigated in the hectometric simulation. These reveal that the Sting Jet descends towards the surface at the rear of the high wind core in the frontal fracture region. This location differs from that reported for previously studied storms where the sting jets are typically found at the forefront of the high wind core. Two distinct Dry Jets are shown to coexist, one descending to the surface around the Sting Jet and another one remaining above the Cold Jet. The Potential Sting Jet is located above the cold conveyor belt and shown to be brought down towards the lower troposphere by coherent structures invigorated by moist processes and spanning several kilometers in width and height at the outer edge of the cloud head. Unlike in previous high-resolution numerical studies, boundary layer rolls are not found to play a key role at the studied instant. Different sting jet location and processes associated with potential sting jet subsidence compared to literature, encourage more systematic studies of extra-topical cyclones through common diagnostics adapted for both kilometric and hectometric resolutions.
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.
Summary
The present study investigates meso-scale airstreams within an extratropical cyclone at kilometric and hectometric model resolutions. Storm Alex serves as a case study to evaluate the performance of such simulations, study the role of ocean wave coupling, investigate the resolution dependence of peak surface winds, analyse the origins of different airstreams that result in strong low level winds, and inspect sub-mesoscale structures involved in mixing these to the surface. The authors find good agreement between the model simulations and observations across highest and lowest resolution. Wind speeds converge at 400m resolution and wave coupling introduces varying responses in different regions within the cyclone. A trajectory analysis reveals the spatiotemporal evolution of four characteristic airstreams which is largely consistent across the considered resolutions. Mesoscale structures consisting of coherent up- and downdrafts are found to bring sting jet and so-called potential sting jet air parcels closer to surface. Thereby, this study constitutes a first contribution towards understanding sub-mesoscale mechanisms behind high surface winds in extratropical cyclones. Finally, this paper highlights the importance of high-quality observations to evaluate such case studies.
Review
This paper represents an important contribution to the field of high-resolution modeling of midlatitude dynamics. The adopted resolution is unprecedented for the regional scale of interest. The simulations reveal highly interesting aspects of (sub-)mesoscale dynamics that are relevant for understanding cyclone-related impacts and may spark many ideas for follow-up research. The scientific quality of the work is high – as indicated by the use of established observation datasets, a state-of-the-art numerical model, and online trajectories to study the short-lived airstreams – and I only have a few important topical comments and questions below. The figures are neat and the structure of the result sections feels natural. The introduction, in turn, deserves some more attention with respect to streamlining towards and specifically defining the research objectives. Moreover, the paper contains numerous linguistic shortcomings. Overall, I think that addressing my comments will require a major effort.
Major comments
• In general, the introduction contains a lot of background and the current state of research on cyclone airstreams (and the modeling of it). A few key research gaps are identified. From the conclusion section, the specific aims of the study are clear (as I would summarize them above), but in my view these should be more clearly defined in the introduction. How exactly may be a question of style, but after reading the introduction I would have liked to know better what exactly to expect from this work. I therefore suggest streamlining the introduction section more towards the objectives and remaining concise. I have a few more comments related to this.
◦ L69: “For SJs in particular, knowledge about boundary-layer processes is missing and is considered a key limitation (Clark and Gray, 2018)” Here for instance, I suggest highlighting that you aim to improve this knowledge with your unique and state-of-the-art set of simulations.
◦ The choice of storm Alex still comes a bit out of the blue. It is common practice in the field to perform case studies, but any selection criteria that go beyond societal impact deserve to be mentioned. For instance, do you prefer a recent case for any reason? Are there unique observations of this storm (satellite overpass)? Did it fulfill SJ precursor criteria?
◦ L47 Here I suggest the remaining sentences of the paragraph to be rephrased more concisely and aiming at the open research gaps. For instance, “Instead, …” to e.g. “Thus, identifying SJs in current global models is limited to diagnosing their precursor mesoscale instabilities within the cloud head (Gray et al, 2024).”
◦ L52 Conciseness: “to be explicitly represented by current NWP models (Sheridan, 2018). For instance, near-surface winds and gusts are not reliable in reference data sets such as the ERA5 reanalysis (Minola et al., 2020)” could be shortened to “and not represented by current NWP models or reanalyses such as ERA5.”
◦ In the current framing, L55 and L61 somewhat contradict each other (or did I miss a subtle difference?). Some meso-scale processes are mentioned, but then they are “largely unknown”. This part can be smoother.
• The analysis of the trajectories in A200 is highly interesting. I have a few scientific comments and follow-up questions to better understand the mixing processes in the boundary layer, which I understand as being one of your key research objectives.
◦ Wouldn’t it be worth (and only a low to moderate effort) to further separate the CJ trajectories into those that remain in the lowest few hundreds of meters, those that are above it, and those in the bands which reach close to the surface eastward of km 80 in Fig. 8? Do the CJ trajectories in the MABL westward of km 80 start at a similar height or get mixed down? This may help understanding differences in the role of CJ vs. SJ airstreams for surface winds.
◦ To go even further along these lines, I am wondering whether the signal that you see in the trajectories, vertical winds, and rain evaporation in Fig. 9 eastward of km 17 could be the imprint of a Kelvin-Helmholz wave. These have been found in LES simulations of real weather cases, e.g. in Umek et al. 2020. Maybe the authors find a simple way (e.g. looking at wind shear, stability, Froude number) to address this aspect (in a way that fits within the scope of the work), which could affect the interpretation of the results.
◦ In general, would it help the computational cost of trajectory calculations if the sampling is somehow coarse-grained (e.g. tracing only every third gridpoint)? Could you otherwise outline in the discussion how you could perform longer trajectory calculations in future analyses (other than a larger computer)?
◦ As you write in L446, the “study sheds >some< insights in sources of high winds”. In case you decide not to follow-up on the evolution of the CJ air parcels above, I suggest to specify the airstream in this sentence or otherwise enhance precision here.
• Section 3.1: How do the AROME simulations used to force Meso-NH compare to the Meso-NH simulations? For instance, are the 30-60min delays in winds better captured in AROME, or does Meso-NH inherit this bias from the boundary conditions?
• The readability of this manuscript is repeatedly hindered by some grammatical errors, imprecise phrasing, and non-standard sentence structures. I suggest a thorough English-language edit to improve fluency. I provide a non-exhaustive list of examples from the introduction and conclusions below.
◦ L29: “In the Northern Hemisphere, they...”
◦ L39: “Some intense maritime cyclones developing a bent-back front further involve an additional airflow that may result in the strongest winds: the Sting Jet (SJ) first documented by Browning (2004).” Suggestion: “For some maritime cyclones that develop a bent-back front, the strongest winds are related an additional air stream referred to as the Sting Jet (Browning, 2004).”
◦ L93 Scientific writing style: “a cyclone is referred to as”, “a cyclone propagates”
◦ L448 "Jets are harder to distinguish over shorter periods over which high resolution trajectories are feasible" Please rephrase this sentence for more clarity.
◦ L450 “is necessary” → “remains necessary”
◦ L454 “What is more, …” e.g. “Moreover, …”
◦ L462 “To assess the veracity of these model efforts measurements are key” Suggest e.g. “Measurements are key to evaluate the performance of these model simulations.”
◦ L468 “The NAWDIC-DICHOTOMI … will allow..." The campaign already happened, but arguably the measurements “will allow ...”.
Minor comments
• Airstream identification criteria:
◦ Did you perform any sensitivity analyses with respect to the adopted height thresholds? Perturbing these parameters could for instance yield a min/max range on the relative contributions of the different airstreams. This could also be of interest when comparing the values stated in line L365 between the different simulations. I am pressing on this because these numbers reappear in the conclusion section.
◦ Reconciling L278 "trial and error" + "the most coherent grouping based on visual inspection of main jet properties and origins and conclusion" and L294 "The altitude evolution of the SJ and its humid character match the definition put forward by Clark et al. (2005) and Clark and Gray (2018) as a jet descending from the cloud head.", is the latter actually a conclusion (signal emerging on its own) or simply the criterion (signal emerging by design)?
◦ Is there a particular motivation behind the 80% RH threshold?
• I suggest adding the model resolution as subpanel titles in Fig. 7.
• L246 and Fig. 3: Within the red box, on first sight the values look white and thus in perfect agreement. If you mean the slight orange within this black contour, maybe declare this more precisely or stress that this relatively small discrepancy is very important for peak winds. Other than that, the comparison of wave age in Fig. 3 is nicely illustrated.
• L262: Since the wave-coupling section and the jet section are not closely related to each other, I suggest adding a short summary sentence like in L222 to round off the section. This transition to me was the least natural.
• L324 "DJ (square) and SJ (dots) and associated time stamps reveal that both join each other and become undistinguishable shortly after 21:30 UTC" refers to A1600. In the cross section of A200, the jets still still seem to be mostly coherent (not mixed) at 22.30 UTC. Are they thus actually indistinguishable? Or does the same median location arise from the two separate DJs discussed in L377?
• L373: Clarification question: "Three regions of high winds (shading in Fig. 8(b)) descending from the mid troposphere are discernible at the rear of P1–P2." Here, do you mean the three vertically tilted maxima between 0-60 km in Fig. 8b?
• L389 Is there any literature that has described such bands (in observations) before or are you the first ones?
• Figure 9d: I am surprised to find low cloud water content in the central ascent region (approx 1.5km height, km 18) since I would have naively expected moist adiabatic ascent there. Is it not, or is it just a matter of contour levels that hide cloud formation in this region?
• Please add in-text specifications of the grey, white, and yellow contours in Fig. 9f.
References
Umek L, Gohm A, Haid M, Ward HC, Rotach MW. Large-eddy simulation of foehn–cold pool interactions in the Inn Valley during PIANO IOP 2. QJR Meteorol Soc. 2021; 147: 944–982. https://doi.org/10.1002/qj.3954