the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 May 2023
Cold wintertime air masses over Europe: Where do they come from and how do they form?
Abstract. Despite the general warming trend, wintertime cold air outbreaks in Europe have remained nearly as extreme and as common as decades ago. In this study, we identify six principal 850 hPa cold anomaly types over Europe in 1979–2020 using self-organizing maps (SOMs). Based on extensive analysis of atmospheric large-scale circulation patterns combined with nearly two million kinematic backward trajectories, we show the origins and contributions of various physical processes to the formation of cold wintertime 850-hPa air masses. The location of the cold anomaly region is closely tied to the location of blocking; if the block is located farther in the east, also the cold anomaly is displaced eastwards. Considering air-mass evolution along the trajectories, the air parcels are typically initially (5–10 d before) colder than at their arrival in Europe, but also initially warmer air parcels sometimes lead to cold anomalies over Europe. Most commonly the effect of adiabatic warming on the temperature anomalies is overcompensated by advection from regions that are climatologically colder than the target region, supported by diabatic cooling along the pathway. However, there are regional differences: cold anomalies over western Europe and southeastern Europe are dominantly caused by advection, and over eastern Europe by both advective and diabatic processes. The decadal-scale warming in the site of air mass origin has been partly compensated by enhanced diabatic (radiative) cooling along the pathway to Europe. There have also been decadal changes in large-scale circulation patterns and air mass origin. Our results suggest that understanding future changes in cold extremes will require in-depth analyses on both large-scale circulation and the physical (adiabatic and diabatic) processes.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(5117 KB)
-
Supplement
(3813 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(5117 KB) - Metadata XML
-
Supplement
(3813 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-889', Anonymous Referee #1, 21 Jun 2023
Summary of the work: The analysis considers cold air outbreaks from 1979-2020 in Europe using SOMs, to investigate the thermodynamic pathways for the extreme cold events. Despite the general warming, these extreme cold events have not weakened, so the authors partition the 1979-2020 period into two halves to investigate the changes to the thermodynamic contributions to cold events. The analysis uses kinematic back trajectories to investigate the formation of the cold air and makes some assumptions about the sources of advective, adiabatic, and diabatic changes in airmass temperature. The analysis shows that the general warming is compensated by diabatic cooling at decadal scales.
Review discussion: Overall, this analysis is solid, and the methods are sound. The analysis of the thermodynamic changes to parcel/airmass trajectories follows the methods of Röthlisberger and Papritz (2023), which is very appropriate for the goals of the research. However, the motivation for using separate SOM analyses for the interpretation of European cold events is not as well motivated/explained as perhaps it could be. Currently, it is fair but with more clairty and detail, it could be excellent. The SOM analysis for temperature/cold, makes sense, but then attributing the cold clusters to the individual nodes on a separate MSLP SOM that don’t necessarily align with the notes in the temperature SOM is a bit confusing. Some of the confusion comes from the fact that these additional SOMs are only shown as supplemental figs. I see that you get more information about the flow configurations contributing to each type of cold type with the additional SOM analysis, but when the common MSLP or GPH modes represent less than 20% of the cold nodes/cases (e.g., type 1 in Fig. 3 & 4), is this the best way to present the result? Perhaps an analysis more like the chicklet plot in Supplemental Fig. 7 would be better suited for the discussion in the main text. Such a change would allow for broader explanation & description of the variety of flow configurations in each cold type than the current Fig. 3/4 discussion. If you take such an approach, the MSLP and GPH SOM figures would need to be moved from supplemental figures to full figures. With all that being said,
L194-195: Typo – The years for your latter and earlier periods need to be swapped.
L209-212: When you consider the events in which the SOM composite cold anomalies are less than the 10th percentile, how many of the days (or what fraction of days) in each Type/SOM group reach the cold threshold vs have the right pattern but don’t meet the threshold?
L345: The assessment here about wind is based on the composite fields, are the weak gradients a result of the composite smoothing or are the gradients actually weak in most/all the cases?
L553: “The more efficient radiative cooling is most probably related to the generally higher radiative cooling rate of warmer and moister air.” This statement seems important in this discussion and could be examined in more detail, “most probably” is a bit hand-wavy. In general, this paragraph has several locations where suggestive wording is used but could be
Fig. 8: The text on this figure is hard to reach, please make the text bigger or bolder. What do the dots on the trajectories represent. The caption says these are 10 d trajectories, but there are not 10 dots, there are 6 dots per trajectory. I assume these dots represent every 2 days, please clarify? On the topic of clarity, can the dots be colored as the line color, in (b) there are several overlapping lines/dots that make the plot cluttered and hard to interpret.
Supplemental Fig. 7: The SOM numbers are very hard to read.
Citation: https://doi.org/10.5194/egusphere-2023-889-RC1 -
RC2: 'Comment on egusphere-2023-889', Anonymous Referee #2, 06 Jul 2023
Review of the paper “Cold wintertime air masses over Europe: Where do they come from and how do they form?” By Nygård et al. This is an interesting paper that provides a very thorough analysis of cold air outbreaks in Europe. However, I do think that the paper is maybe too broad and all-encompassing, at the penalty of becoming slightly overwhelming and thus perhaps not as readable as it could have been.
I’ll note first that I have read the comments of one reviewer who submitted their comments before me.
I don’t have many comments on the scientific content, which seems sound. It’s more a matter of some choices that you’ve made.
I agree with Reviewer #1 that it’s not completely clear why you subselect such a small subset of the SOMs. Just by eyeballing, it certainly looks like e.g. map 41 is very cold in Scandinavia, but you write that you only select maps that satisfy the cold criteria in continental Europe or the British Isles. Why do you make this choice? Scandinavia is part of Europe… Furthermore, your subdivision of the maps into groups seems to be subjective. Is group VI really that different from group V? This needs more motivation.
When it comes to the description of the Cold anomaly types in Section 3.2, it’s long, and quite frankly a bit boring to read. I think the reason for this is that you seem to follow a template for each of the groups. This makes it rigorous and useful as a reference, but it does feel quite repetitive.
I think you could also reduce the number of panels. In Fig. 3 for example, it’s not clear why you would show three MSLP panels for each group. Some of them are very similar. It’s also not clear, if one does not wish to read the SI, how the selection of the most common and second most common groups was done. Maybe this could be motivated more clearly, or maybe it would be better to reduce the number of panels.
Perhaps you could try to reduce the length of Section 3 substantially by, instead of going through a fixed template for each group, try to extract the most interesting features, i.e. the ones that stand out? The Discussion is an attempt at that.
This is meant as well-intentioned advice. I think the topic and the approach are interesting, but I think the paper will be more widely read if you drastically reduce the number of figures and text and focus more on the results that stand out.
Citation: https://doi.org/10.5194/egusphere-2023-889-RC2 - AC1: 'Comment on egusphere-2023-889', Tiina Nygård, 09 Aug 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-889', Anonymous Referee #1, 21 Jun 2023
Summary of the work: The analysis considers cold air outbreaks from 1979-2020 in Europe using SOMs, to investigate the thermodynamic pathways for the extreme cold events. Despite the general warming, these extreme cold events have not weakened, so the authors partition the 1979-2020 period into two halves to investigate the changes to the thermodynamic contributions to cold events. The analysis uses kinematic back trajectories to investigate the formation of the cold air and makes some assumptions about the sources of advective, adiabatic, and diabatic changes in airmass temperature. The analysis shows that the general warming is compensated by diabatic cooling at decadal scales.
Review discussion: Overall, this analysis is solid, and the methods are sound. The analysis of the thermodynamic changes to parcel/airmass trajectories follows the methods of Röthlisberger and Papritz (2023), which is very appropriate for the goals of the research. However, the motivation for using separate SOM analyses for the interpretation of European cold events is not as well motivated/explained as perhaps it could be. Currently, it is fair but with more clairty and detail, it could be excellent. The SOM analysis for temperature/cold, makes sense, but then attributing the cold clusters to the individual nodes on a separate MSLP SOM that don’t necessarily align with the notes in the temperature SOM is a bit confusing. Some of the confusion comes from the fact that these additional SOMs are only shown as supplemental figs. I see that you get more information about the flow configurations contributing to each type of cold type with the additional SOM analysis, but when the common MSLP or GPH modes represent less than 20% of the cold nodes/cases (e.g., type 1 in Fig. 3 & 4), is this the best way to present the result? Perhaps an analysis more like the chicklet plot in Supplemental Fig. 7 would be better suited for the discussion in the main text. Such a change would allow for broader explanation & description of the variety of flow configurations in each cold type than the current Fig. 3/4 discussion. If you take such an approach, the MSLP and GPH SOM figures would need to be moved from supplemental figures to full figures. With all that being said,
L194-195: Typo – The years for your latter and earlier periods need to be swapped.
L209-212: When you consider the events in which the SOM composite cold anomalies are less than the 10th percentile, how many of the days (or what fraction of days) in each Type/SOM group reach the cold threshold vs have the right pattern but don’t meet the threshold?
L345: The assessment here about wind is based on the composite fields, are the weak gradients a result of the composite smoothing or are the gradients actually weak in most/all the cases?
L553: “The more efficient radiative cooling is most probably related to the generally higher radiative cooling rate of warmer and moister air.” This statement seems important in this discussion and could be examined in more detail, “most probably” is a bit hand-wavy. In general, this paragraph has several locations where suggestive wording is used but could be
Fig. 8: The text on this figure is hard to reach, please make the text bigger or bolder. What do the dots on the trajectories represent. The caption says these are 10 d trajectories, but there are not 10 dots, there are 6 dots per trajectory. I assume these dots represent every 2 days, please clarify? On the topic of clarity, can the dots be colored as the line color, in (b) there are several overlapping lines/dots that make the plot cluttered and hard to interpret.
Supplemental Fig. 7: The SOM numbers are very hard to read.
Citation: https://doi.org/10.5194/egusphere-2023-889-RC1 -
RC2: 'Comment on egusphere-2023-889', Anonymous Referee #2, 06 Jul 2023
Review of the paper “Cold wintertime air masses over Europe: Where do they come from and how do they form?” By Nygård et al. This is an interesting paper that provides a very thorough analysis of cold air outbreaks in Europe. However, I do think that the paper is maybe too broad and all-encompassing, at the penalty of becoming slightly overwhelming and thus perhaps not as readable as it could have been.
I’ll note first that I have read the comments of one reviewer who submitted their comments before me.
I don’t have many comments on the scientific content, which seems sound. It’s more a matter of some choices that you’ve made.
I agree with Reviewer #1 that it’s not completely clear why you subselect such a small subset of the SOMs. Just by eyeballing, it certainly looks like e.g. map 41 is very cold in Scandinavia, but you write that you only select maps that satisfy the cold criteria in continental Europe or the British Isles. Why do you make this choice? Scandinavia is part of Europe… Furthermore, your subdivision of the maps into groups seems to be subjective. Is group VI really that different from group V? This needs more motivation.
When it comes to the description of the Cold anomaly types in Section 3.2, it’s long, and quite frankly a bit boring to read. I think the reason for this is that you seem to follow a template for each of the groups. This makes it rigorous and useful as a reference, but it does feel quite repetitive.
I think you could also reduce the number of panels. In Fig. 3 for example, it’s not clear why you would show three MSLP panels for each group. Some of them are very similar. It’s also not clear, if one does not wish to read the SI, how the selection of the most common and second most common groups was done. Maybe this could be motivated more clearly, or maybe it would be better to reduce the number of panels.
Perhaps you could try to reduce the length of Section 3 substantially by, instead of going through a fixed template for each group, try to extract the most interesting features, i.e. the ones that stand out? The Discussion is an attempt at that.
This is meant as well-intentioned advice. I think the topic and the approach are interesting, but I think the paper will be more widely read if you drastically reduce the number of figures and text and focus more on the results that stand out.
Citation: https://doi.org/10.5194/egusphere-2023-889-RC2 - AC1: 'Comment on egusphere-2023-889', Tiina Nygård, 09 Aug 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 240 | 129 | 17 | 386 | 42 | 8 | 9 |
- HTML: 240
- PDF: 129
- XML: 17
- Total: 386
- Supplement: 42
- BibTeX: 8
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Lukas Papritz
Tuomas Naakka
Timo Vihma
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(5117 KB) - Metadata XML
-
Supplement
(3813 KB) - BibTeX
- EndNote
- Final revised paper