The Named Storms Catalogue: unlocking learnings from past events

Gonzalez, Paula L. M.; Ackerley, Duncan; Matthews, Eloise; Suri, Dan; Wallace, Emily; Wheeler, Matthew; Lowe, Jason A.; Harley-Nyang, Daisy

doi:10.5194/essd-2026-181

Preprints

https://doi.org/10.5194/essd-2026-181

Preprints

17 Apr 2026

| 17 Apr 2026

Status: a revised version of this preprint was accepted for the journal ESSD and is expected to appear here in due course.

The Named Storms Catalogue: unlocking learnings from past events

Paula L. M. Gonzalez, Duncan Ackerley, Eloise Matthews, Dan Suri, Emily Wallace, Matthew Wheeler, Jason A. Lowe, and Daisy Harley-Nyang

Abstract. Extratropical cyclones are one of the leading causes of widespread damage from extreme weather over the United Kingdom and western Europe due to their strong winds and heavy precipitation. Storm naming has been used as a strategy to communicate risks and enhance preparedness of the general public and key affected sectors. Given the first ten seasons of storm naming, the opportunity arises to collect and characterise for the first time the set of named storms identified by the UK Storm Centre. Collecting information about these events enables to objectively assess what makes them different from other storms and from one another; and supports preparedness for future events.

This paper introduces the Named Storms Catalogue as an open-source dataset accessible from https://doi.org/10.5281/zenodo.18877013. The catalogue contains a set of storm tracks, storm development and hazard metrics that allow to objectively describe, rank and compare past named storm events, unlocking learning opportunities by the preparedness and resilience as well as the research communities.

This dataset introduction describes the content and metrics and the methodologies used to develop them, while a set of case studies highlights how the catalogue can be used to assess and compare past events. The paper presents a preliminary climatology for UK named storms and discusses some ongoing lines of research that the authors are exploring, such as using the metrics included in the catalogue to distinguish named events from others, and to identify different types of storms within the set.

Received: 06 Mar 2026 – Discussion started: 17 Apr 2026

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.

Download & links

Paula L. M. Gonzalez, Duncan Ackerley, Eloise Matthews, Dan Suri, Emily Wallace, Matthew Wheeler, Jason A. Lowe, and Daisy Harley-Nyang

Status: closed

RC1:
'Comment on essd-2026-181', Mika Rantanen, 28 May 2026
Review of Earth System Science Data Discussion manuscript “The Named Storms Catalogue: unlocking learnings from past events” by Gonzalez et al.
This data description manuscript presents a new named storms dataset for the UK and western Europe. The dataset contains the tracks and development and intensity metrics of all named extratropical cyclones during the ten-year period from 2015/2016 to 2025/2026.
Several catalogues and databases of European windstorms already exist. However, these datasets often include only the most intense storms or do not systematically identify storms by their operational names, making it difficult to link individual events across different data sources and reports. Thus, the main novelty of the present dataset is that it provides a comprehensive collection of all officially named UK storms during the study period, together with consistent track and intensity information. As far as I am aware, such a dataset does not yet exist.
The manuscript was really clearly written and I enjoyed reading it. The structure of the paper was good, however, the authors could pay attention to the figures, especially their font sizes and the use of abbreviations (see comments below). Overall, I can still recommend publication of the paper. I have some relatively minor comments which I hope the authors can address before the publication.
Minor comments:
At L128, it is said that operational meteorologists confirmed that the TRACK-derived tracks corresponded to the named events. However, it was not explicitly stated whether the opposite was also true. Were all named storms associated with a “trackable” extratropical cyclone? Is storm naming inherently tied to the presence of a well-defined low-pressure system, or can situations occur where the Met Office names a storm that is later not captured by TRACK? More generally, how did you ensure that TRACK identified all named storms?

The manuscript is not mentioning if post-tropical storms are included in the dataset. I have the understanding that the Met Office or other met institutes do not rename post-tropical cyclones but retain the original name if the storm is forecast to cause damage in Europe. There seems to be “exOphelia” in the dataset in 2017, but are there other post-tropical cyclones? This could be mentioned in the text. In Figure 4, one track is originating from the tropics. Was that a tropical cyclone?

The whole dataset and thus the manuscript is based on ERA5 reanalysis. Although ERA5 is the state-of-the-art reanalysis and arguably the best possible choice for this analysis, it still has some biases, for example in wind speeds. It was mentioned at L389 that ERA5 did not capture the full severity of Storm Eunice. Do you think that the use of ERA5 may systematically underestimate (or in some cases overestimate?) some of the development or hazard-related variables analysed in the manuscript? This aspect might deserve a short discussion paragraph in the last section for the end-users, especially since ERA5 biases associated with windstorms have likely been documented in previous literature.

Tables and Figures. I find the extensive use of abbreviations in tables and figures (e.g. MxAcP, Mxppr, etc.) somewhat difficult to follow. While some abbreviations (such as Mxgust) are intuitive, many others are not immediately clear to the reader. For clarity and accessibility, I would recommend either spelling out the variables in full or providing a consistent legend/caption where all abbreviations are clearly defined. Especially in Figures 6-8.

Also, especially the Figures 6-8 (and others too, such Fig. 9 subplots) have lots of small text which I was not able to read from a printed copy. Can you increase the font size of the figures?

Line-to-line comments:
L46: Are all windstorms in the Western naming group named in advance based on forecasts, or are some named after impacts became evident? Have some storms been named retrospectively? A brief clarification of the naming procedure (and timing relative to forecast vs. observed impacts) would be useful here. For example, in Finland, a windstorm may sometimes be named once it is clear that it is causing significant impacts, sometimes on the same or even the next day it affects the country.
Section 2.1. I guess you used ERA5 in its native resolution (0.25°)? This is not mentioned.
Section 2.2. It was not fully clear from the text why there are two ocean datasets. Which one do you use e.g. in Fig. 12, and the figures in the database?
L71: Why did you track the systems with 6-hourly fields and not e.g. 3-hourly? 6-hourly data may be somewhat too coarse for representing local impacts, especially for fast-moving systems where the location and intensity of the strongest winds and precipitation can change substantially within six hours. Using 3-hourly fields could improve the representation of storm evolution by better resolving rapid deepening and the intensity peaks.
L98: Here you write that you calculated vorticity from wind components. But vorticity is available directly at pressure levels from ERA5. Why did you calculate the vorticity fields by yourselves?
L121: Vorticity maxima may not always be associated with MSLP minima, particularly during the early development stage of cyclones. Were such situations included in the tracking procedure, and how did you deal with them?
Table 1, bearing. I wasn’t sure about how this is calculated. Here you state “6h positions of maxima”. Which maxima?
L184: Surge residual is a new term for me. Can you explain it here a bit more?
L186: Section 0?
L216: “adding 6 hours on - 3 hours for each end of the track.” I can’t follow here, did you add 6 hours or 3 hours to the durations?
Table 2, Maximum wind gust. Is this the single highest ERA5 grid-point value within the storm footprint? Since peak gusts are often associated with mountainous / high-elevation regions, it would be useful to clarify whether these maxima are strongly influenced by high-elevation grid points.
Table 2, maximum wind gust. Your tracks are every 6 hours. Did you get the ERA5 wind gust data every 1 one hour to obtain the absolute maximum wind gust associated with the tracks, or are these just the instantaneous wind gusts from the 6-hourly wind gust fields? It was not entirely sure from L73 whether the wind gust data is hourly or not (it could be stated more explicitly).
L264: Full range may be misleading as it’s definitely not the physically possible “full range” due to small(ish) sample size. Maybe just the widest range?
Figure 6, 7 and others: There is no unit in the y-axis. Also, could the titles be explicitly written, without the abbreviations, in order to ease the interpretation (and with bigger size). The abbreviations may be clear for the authors that have used them for a long time, but not for a regular reader who reads the paper for the first time.
L281: What do you mean by “corrections to account for changes in the background climate”? Do you mean that the storms in 2015 occurred in a colder climate than the storms in 2025, and this may have an impact on the hazard metrics? The effect is probably very small, given the short 10-year period involved (~0.2 °C of global warming) and the relatively limited response of extratropical cyclones to warming.
L282: I think it’s interesting that DJF season has lots of named storms with close to zero hazard metrics (Fig. 7a–c), while this is clearly not the case for JJA. Does this imply that named JJA storms have a higher probability to cause damage? One possible explanation is that winter storms are more often named based on forecast intensity and anticipated impacts, even if their hazard metrics remain modest because strong winds are more common in winter climatology. This also seems somewhat inconsistent with the Introduction statement that naming is primarily based on impacts (L46-47), so a short discussion of this could be useful.
Figure 8. As commented earlier, the abbreviations could be written open. Particularly in the caption of the figure.
L375: Citation comes before the dot.
L387-388: Volonte et al. 2024 is missing from the reference list. How about the ERA5 resolution? Can that be one reason for the underestimation of the severity?
Conclusions. Are you planning to update the dataset in the future? If yes, this could be mentioned here.
Citation: https://doi.org/10.5194/essd-2026-181-RC1
- AC1: 'Reply on RC1', Paula LM Gonzalez, 06 Aug 2026
  
  We thank the reviewer for their comments. We have included detailed replies in the attached document.
  
  Citation: https://doi.org/10.5194/essd-2026-181-AC1
RC2:
'Comment on essd-2026-181', Anonymous Referee #2, 30 Jun 2026

Summary
The authors present a new open-source storm catalogue of named storms and showed how this catalogue can be used in risk assessment.
General
The manuscript is well written. Also, the newly derived storm catalogue is of interest for the scientific but also the stakeholder community, e.g. insurance companies. Thus, I recommend minor revision prior to possible publication in ESSD.
Major comment
I think the literature review in the introduction is a bit short. H. Dacre initiated a storm atlas (Dacre et al. 2012, BAMS). What is the difference between your catalogue and the storm atlas by Dacre? In Line 37 you mention several naming partnerships, but the author team stays rather vague. You need to explain this in more details, what are the differences, which partnerships exist and what are their aims?

Minor comments
L20 and elsewhere: Please use “Method” instead of “Methodology” throughout the manuscript.
L33 and L55: Compound extreme events associated with cyclone activity are presented in Messmer and Simmonds (2021, WACE) and could be cited here.
L72 Please cite the tracking method here. How does this tracking method compare to others, e.g. Neu et al. (2013, BAMS)
L87: please change to “ However, we consider them”
L149: Comma after “In this paper”
L199-202: How sensitive is the choice of 98^th and 99^th percentile to your results presented later, why are different thresholds used? How is the sensitivity testing performed? You need to be explicit wrt the criteria used.
Figures: The labels are too small in all figures, so please revise all figures.
Discussion: Please discuss also other storm catalogues/ storm atlas and the storm naming partnerships in more details.

Citation: https://doi.org/10.5194/essd-2026-181-RC2
- AC2: 'Reply on RC2', Paula LM Gonzalez, 06 Aug 2026
  
  We thank the reviewer for their comments. We have included detailed replies in the attached document.
  
  Citation: https://doi.org/10.5194/essd-2026-181-AC2
EC1:
'Comment on essd-2026-181', Graciela Raga, 25 Jul 2026

Dear Authors,
A third reviewer has made comments available for you, after the deadline for upload, and I am uploading them here.
Please address them when you revise your manuscript. Best regards, Graciela Raga

Citation: https://doi.org/10.5194/essd-2026-181-EC1
- AC3: 'Reply on RC3', Paula LM Gonzalez, 06 Aug 2026
  
  We thank the reviewer for their comments. We have included detailed replies in the attached document.
  
  Citation: https://doi.org/10.5194/essd-2026-181-AC3

Status: closed

RC1:
'Comment on essd-2026-181', Mika Rantanen, 28 May 2026
Review of Earth System Science Data Discussion manuscript “The Named Storms Catalogue: unlocking learnings from past events” by Gonzalez et al.
This data description manuscript presents a new named storms dataset for the UK and western Europe. The dataset contains the tracks and development and intensity metrics of all named extratropical cyclones during the ten-year period from 2015/2016 to 2025/2026.
Several catalogues and databases of European windstorms already exist. However, these datasets often include only the most intense storms or do not systematically identify storms by their operational names, making it difficult to link individual events across different data sources and reports. Thus, the main novelty of the present dataset is that it provides a comprehensive collection of all officially named UK storms during the study period, together with consistent track and intensity information. As far as I am aware, such a dataset does not yet exist.
The manuscript was really clearly written and I enjoyed reading it. The structure of the paper was good, however, the authors could pay attention to the figures, especially their font sizes and the use of abbreviations (see comments below). Overall, I can still recommend publication of the paper. I have some relatively minor comments which I hope the authors can address before the publication.
Minor comments:
At L128, it is said that operational meteorologists confirmed that the TRACK-derived tracks corresponded to the named events. However, it was not explicitly stated whether the opposite was also true. Were all named storms associated with a “trackable” extratropical cyclone? Is storm naming inherently tied to the presence of a well-defined low-pressure system, or can situations occur where the Met Office names a storm that is later not captured by TRACK? More generally, how did you ensure that TRACK identified all named storms?

The manuscript is not mentioning if post-tropical storms are included in the dataset. I have the understanding that the Met Office or other met institutes do not rename post-tropical cyclones but retain the original name if the storm is forecast to cause damage in Europe. There seems to be “exOphelia” in the dataset in 2017, but are there other post-tropical cyclones? This could be mentioned in the text. In Figure 4, one track is originating from the tropics. Was that a tropical cyclone?

The whole dataset and thus the manuscript is based on ERA5 reanalysis. Although ERA5 is the state-of-the-art reanalysis and arguably the best possible choice for this analysis, it still has some biases, for example in wind speeds. It was mentioned at L389 that ERA5 did not capture the full severity of Storm Eunice. Do you think that the use of ERA5 may systematically underestimate (or in some cases overestimate?) some of the development or hazard-related variables analysed in the manuscript? This aspect might deserve a short discussion paragraph in the last section for the end-users, especially since ERA5 biases associated with windstorms have likely been documented in previous literature.

Tables and Figures. I find the extensive use of abbreviations in tables and figures (e.g. MxAcP, Mxppr, etc.) somewhat difficult to follow. While some abbreviations (such as Mxgust) are intuitive, many others are not immediately clear to the reader. For clarity and accessibility, I would recommend either spelling out the variables in full or providing a consistent legend/caption where all abbreviations are clearly defined. Especially in Figures 6-8.

Also, especially the Figures 6-8 (and others too, such Fig. 9 subplots) have lots of small text which I was not able to read from a printed copy. Can you increase the font size of the figures?

Line-to-line comments:
L46: Are all windstorms in the Western naming group named in advance based on forecasts, or are some named after impacts became evident? Have some storms been named retrospectively? A brief clarification of the naming procedure (and timing relative to forecast vs. observed impacts) would be useful here. For example, in Finland, a windstorm may sometimes be named once it is clear that it is causing significant impacts, sometimes on the same or even the next day it affects the country.
Section 2.1. I guess you used ERA5 in its native resolution (0.25°)? This is not mentioned.
Section 2.2. It was not fully clear from the text why there are two ocean datasets. Which one do you use e.g. in Fig. 12, and the figures in the database?
L71: Why did you track the systems with 6-hourly fields and not e.g. 3-hourly? 6-hourly data may be somewhat too coarse for representing local impacts, especially for fast-moving systems where the location and intensity of the strongest winds and precipitation can change substantially within six hours. Using 3-hourly fields could improve the representation of storm evolution by better resolving rapid deepening and the intensity peaks.
L98: Here you write that you calculated vorticity from wind components. But vorticity is available directly at pressure levels from ERA5. Why did you calculate the vorticity fields by yourselves?
L121: Vorticity maxima may not always be associated with MSLP minima, particularly during the early development stage of cyclones. Were such situations included in the tracking procedure, and how did you deal with them?
Table 1, bearing. I wasn’t sure about how this is calculated. Here you state “6h positions of maxima”. Which maxima?
L184: Surge residual is a new term for me. Can you explain it here a bit more?
L186: Section 0?
L216: “adding 6 hours on - 3 hours for each end of the track.” I can’t follow here, did you add 6 hours or 3 hours to the durations?
Table 2, Maximum wind gust. Is this the single highest ERA5 grid-point value within the storm footprint? Since peak gusts are often associated with mountainous / high-elevation regions, it would be useful to clarify whether these maxima are strongly influenced by high-elevation grid points.
Table 2, maximum wind gust. Your tracks are every 6 hours. Did you get the ERA5 wind gust data every 1 one hour to obtain the absolute maximum wind gust associated with the tracks, or are these just the instantaneous wind gusts from the 6-hourly wind gust fields? It was not entirely sure from L73 whether the wind gust data is hourly or not (it could be stated more explicitly).
L264: Full range may be misleading as it’s definitely not the physically possible “full range” due to small(ish) sample size. Maybe just the widest range?
Figure 6, 7 and others: There is no unit in the y-axis. Also, could the titles be explicitly written, without the abbreviations, in order to ease the interpretation (and with bigger size). The abbreviations may be clear for the authors that have used them for a long time, but not for a regular reader who reads the paper for the first time.
L281: What do you mean by “corrections to account for changes in the background climate”? Do you mean that the storms in 2015 occurred in a colder climate than the storms in 2025, and this may have an impact on the hazard metrics? The effect is probably very small, given the short 10-year period involved (~0.2 °C of global warming) and the relatively limited response of extratropical cyclones to warming.
L282: I think it’s interesting that DJF season has lots of named storms with close to zero hazard metrics (Fig. 7a–c), while this is clearly not the case for JJA. Does this imply that named JJA storms have a higher probability to cause damage? One possible explanation is that winter storms are more often named based on forecast intensity and anticipated impacts, even if their hazard metrics remain modest because strong winds are more common in winter climatology. This also seems somewhat inconsistent with the Introduction statement that naming is primarily based on impacts (L46-47), so a short discussion of this could be useful.
Figure 8. As commented earlier, the abbreviations could be written open. Particularly in the caption of the figure.
L375: Citation comes before the dot.
L387-388: Volonte et al. 2024 is missing from the reference list. How about the ERA5 resolution? Can that be one reason for the underestimation of the severity?
Conclusions. Are you planning to update the dataset in the future? If yes, this could be mentioned here.
Citation: https://doi.org/10.5194/essd-2026-181-RC1
- AC1: 'Reply on RC1', Paula LM Gonzalez, 06 Aug 2026
  
  We thank the reviewer for their comments. We have included detailed replies in the attached document.
  
  Citation: https://doi.org/10.5194/essd-2026-181-AC1
RC2:
'Comment on essd-2026-181', Anonymous Referee #2, 30 Jun 2026

Summary
The authors present a new open-source storm catalogue of named storms and showed how this catalogue can be used in risk assessment.
General
The manuscript is well written. Also, the newly derived storm catalogue is of interest for the scientific but also the stakeholder community, e.g. insurance companies. Thus, I recommend minor revision prior to possible publication in ESSD.
Major comment
I think the literature review in the introduction is a bit short. H. Dacre initiated a storm atlas (Dacre et al. 2012, BAMS). What is the difference between your catalogue and the storm atlas by Dacre? In Line 37 you mention several naming partnerships, but the author team stays rather vague. You need to explain this in more details, what are the differences, which partnerships exist and what are their aims?

Minor comments
L20 and elsewhere: Please use “Method” instead of “Methodology” throughout the manuscript.
L33 and L55: Compound extreme events associated with cyclone activity are presented in Messmer and Simmonds (2021, WACE) and could be cited here.
L72 Please cite the tracking method here. How does this tracking method compare to others, e.g. Neu et al. (2013, BAMS)
L87: please change to “ However, we consider them”
L149: Comma after “In this paper”
L199-202: How sensitive is the choice of 98^th and 99^th percentile to your results presented later, why are different thresholds used? How is the sensitivity testing performed? You need to be explicit wrt the criteria used.
Figures: The labels are too small in all figures, so please revise all figures.
Discussion: Please discuss also other storm catalogues/ storm atlas and the storm naming partnerships in more details.

Citation: https://doi.org/10.5194/essd-2026-181-RC2
- AC2: 'Reply on RC2', Paula LM Gonzalez, 06 Aug 2026
  
  We thank the reviewer for their comments. We have included detailed replies in the attached document.
  
  Citation: https://doi.org/10.5194/essd-2026-181-AC2
EC1:
'Comment on essd-2026-181', Graciela Raga, 25 Jul 2026

Dear Authors,
A third reviewer has made comments available for you, after the deadline for upload, and I am uploading them here.
Please address them when you revise your manuscript. Best regards, Graciela Raga

Citation: https://doi.org/10.5194/essd-2026-181-EC1
- AC3: 'Reply on RC3', Paula LM Gonzalez, 06 Aug 2026
  
  We thank the reviewer for their comments. We have included detailed replies in the attached document.
  
  Citation: https://doi.org/10.5194/essd-2026-181-AC3

Paula L. M. Gonzalez, Duncan Ackerley, Eloise Matthews, Dan Suri, Emily Wallace, Matthew Wheeler, Jason A. Lowe, and Daisy Harley-Nyang

Data sets

UK Named Storms Catalogue (2015-2025) Paula L. M. Gonzalez et al. https://doi.org/10.5281/zenodo.18877013

Paula L. M. Gonzalez, Duncan Ackerley, Eloise Matthews, Dan Suri, Emily Wallace, Matthew Wheeler, Jason A. Lowe, and Daisy Harley-Nyang

Viewed

Total article views: 654 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
444	163	47	654	39	48

HTML: 444
PDF: 163
XML: 47
Total: 654
BibTeX: 39
EndNote: 48

Views and downloads (calculated since 17 Apr 2026)

Month	HTML	PDF	XML	Total
Apr 2026	174	57	15	246
May 2026	145	13	1	159
Jun 2026	16	6	2	24
Jul 2026	36	16	2	54
Aug 2026	45	45	11	101
Sep 2026	28	26	16	70

Cumulative views and downloads (calculated since 17 Apr 2026)

Month	HTML	PDF	XML	Total
Apr 2026	174	57	15	246
May 2026	145	13	1	159
Jun 2026	16	6	2	24
Jul 2026	36	16	2	54
Aug 2026	45	45	11	101
Sep 2026	28	26	16	70

Viewed (geographical distribution)

Total article views: 605 (including HTML, PDF, and XML) Thereof 605 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 16 Sep 2026

Short summary

The Named Storms Catalogue is the first dataset to systematically evaluate ten years of UK storm naming using carefully designed metrics developed with Met Office meteorologists. It reveals how storms evolved and the hazards they created, enabling unprecedented comparison, ranking, and insight. The work introduces the dataset, showcases three case studies, and explores current and future uses with a view to helping experts and responders better understand and manage storm impacts.


Total:	0
HTML:	0
PDF:	0
XML:	0