| 13 Jul 2026
Downslope windstorms in southeast Greenland – Part II: Multi-scale dynamics and spatial evolution of piteraq events
Abstract. Severe downslope windstorms in southeast Greenland, known locally as piteraq, dominate extreme weather conditions across the region, including the main settlement Tasiilaq. However, their complex multi-scale dynamics remain poorly resolved in numerical weather prediction and global climate models, leading to inaccurate early warnings and biases in regional heat flux and ocean circulation projections. In this study, we utilize polar adapted, high-resolution reanalysis to trace the spatial, kinematic, and thermodynamic evolution of piteraq events from the synoptic scale down to localized turbulent wave breaking. Validation against regional observational networks shows that the reanalysis captures the broad spatial and temporal variability of the wind speed of these events, but it overestimates the peak intensities with a positive mean bias of 6.5 m s−1 during the winter cases.
We demonstrate that piteraq events are dynamically triggered by persistent upstream westerlies followed by the crossing of a Canadian cold vortex over Greenland. While approaching Greenland the Canadian cold vortex replaces a statistically significant, anomalous near-surface warming over the ice sheet to a rapid surface cooling. This transition establishes a strong cross-barrier surface pressure gradient (on median 7.3 hPa 100 km−1) and a reduction in the Scorer parameter profile that supports wave trapping and resonance. As the stably stratified flow is forced to descend the topography, it accelerates into intense low-level jets that are distinctly localized directly beneath regions of frequent wave breaking and overturning flow aloft. Our analysis further reveals a thermodynamically driven seasonal dichotomy. Winter events are characterized by dense, cold drainage flows that erode low-level coastal stratification and drive strong maritime cold-air outbreaks over the Irminger Sea. In contrast, summer events are characterized by a lack of windward blocking, allowing the upstream westerlies to force moister air masses over the crest. The resulting cloud formation and latent heat release, combined with subsequent adiabatic warming during the leeward descent, manifests as a relatively warm, dry windstorm capable of enhancing local ice melt.
Ultimately, our study defines a distinct, synoptic-scale blueprint for piteraq formation. Identifying these resolvable upstream precursors provides a pathway to train machine learning algorithms for improving the reliability of regional early warning systems, and offers a physical approach to dynamically parameterize sub-grid turbulent heat fluxes for more robust Atlantic Meridional Overturning Circulation (AMOC) projections.
Review of
Downslope windstorms in southeast Greenland - Part II: Multi-scale dynamics and spatial evolution of piteraq events
by Silva and others
General
This is part 2 of a study of the forcing (thermo-)dynamics of piteraq winds om southeast Greenland. The analysis is based on CARRA reanalysis, and a brief evaluation with AWS data is included. the CARRA performance in terms of correlation appears adequate for such a quantitative study. I enjoyed reading the paper, although with almost 30 pages, the paper is very long which makes it a demanding read. But the text is generally clear and the figures generally of good quality. The results are interesting and worthy of publication. As the paper clearly demonstrates, the considerable spatial footprint of associated pressure and temperature fields highlight that piteraq events are not localized, topographically isolated features, but rather the result of a synoptic-scale weather event impacting the entire southern half of Greenland. the study also clearly demonstrates the connection between piteraqs and tip jets in the south. All in all, I recommend publication once the relatively minor comments below have been adequately addressed. I trust this can be checked by the editor and do not need to see the revised version again.
Meso comments
l. 10: Without further explanation, "Canadian cold vortex" comes across as jargon. Please provide some explanation here.
Abstract and Introduction: too many acronymas are introduced, negatively impacting readability: DWE, LLJ, MCAO, AMOC, CARRA...only introduce and define those that are frequently used afterwards.
Same for Section 2.2.
l. 64: "...low spatial resolution fails to resolve sub-grid scale dynamic processes..."; sub grid scale processes are not resolved by definition (independent of resolution).
l. 107: "likely"? Please explain/expand, for CARRA evaluation this is of course important.
l. 117: Are the used wind components (at 10 m and at pressure levels) averaged or instantaneous? If they are averages, absolute wind speed cannot be calculated from the u, v components as in l. 156, this would lead to a negative bias. Same question is valid for the observations.
Fig. 1: The positive bias in CARRA during piteraq events is fascinating. Could it be partly eplxained by the interpolation to the station location, i.e. what does the scatter plot look like is nearest-neighbor is used, without height correction (the use of which is questionable anyways for a variable such as wind speed that increases so quickly from the surface upwards)?
l. 184: Please briefly explain how "CARRA precision" is defined.
l. 192 - 194: This explanation is partly a repetition from previous l. 171 and further.
l. 261: What uncertainty is introduced by using a fixed lapse rate?
l. 267: At the surface the wind speed is zero, please use "Near the surface" or "at the lowest model level (XX m)".
Figure 3 is unclear and must be improved. Figures 3a-d are very similar, please try to make them more distinct by reducing the number of arrows and using another temperature colour scale. Same is true of colour scale in Figures 3 e-h. In the latter four graphs, consider removing MSLP from the elevated parts of the ice sheet where it becomes rather meaningless. Over sea, labels are missing.
Fig. A4 is much clearer when it comes to explaining what causes the piteraq by the advection of a cold air mass. Consider moving to the main text or swapping with Fig. 3?
l. 304: Please explain briefly what you mean by "thermodynamic erosion".
l. 307: I cannot see the "mesoscale surface low", only a larger system?
l. 317: Please expand on what you call "classical katabatic winds". Do you mean that they are forced by locally sourced cold air rather than advection of a cold air mass? The forcing mechanism of the downlope winds that ensue remains the same.
l. 400: " It should be noted that while Alpine pressure gradients are traditionally evaluated across the entire mountain range, our piteraq gradient is calculated specifically over the leeward descent." Please elaborate on the reason(s) for this difference in analysis methods.
Minor and textual comments
l. 95: pressure gradients -> horizontal pressure gradients
l. 107: are -> is
l. 147: As CARRA is a model product, I prefer not to use "validated" but rather "evaluated". See also e.g. Section 3.2 heading and throughout.
l. 387: Remove "to" (?)
l. 479: "occasionally" does not conform to a frequency of 66%.
l. 520: "descent air"; unclear, please reformulate.
l. 521: "warmer temperatures"; please consider using the gramatically correct "lower"/higher temperatures" rather than "colder/warmer temperatures" throughout.