Still normal? Near-real-time evaluation of storm surge events in the context of climate change

Liu, Xin; Meinke, Insa; Weisse, Ralf

doi:https://doi.org/10.5194/nhess-22-97-2022

Articles | Volume 22, issue 1

https://doi.org/10.5194/nhess-22-97-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/nhess-22-97-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 22, issue 1

Research article

|

21 Jan 2022

Research article |

| 21 Jan 2022

Still normal? Near-real-time evaluation of storm surge events in the context of climate change

Xin Liu, Insa Meinke, and Ralf Weisse

Download

Final revised paper (published on 21 Jan 2022)
Preprint (discussion started on 23 Apr 2021)

Interactive discussion

Status: closed

RC1:
'Comment on nhess-2021-75', Andreas Sterl, 11 May 2021
Synopsis

The paper describes a website presenting sea level observations along the German coasts. The intention is to put actual observations (today!) into a historical perspective in a form that is easily accessible for the non-scientist. The site does so by providing a set of easy-to-understand statistics and compares the actual measurements with that statistic. For instance, the actual water level measurement is placed on a return-time plot, which allows to say 'this measurement is expected to occur very n years'. The relevant statistics are re-calculated after each event, keeping them up-to-date. Besides water levels, the site also provides the annual number of events exceeding a given threshold, their length and their severity.

Discussion

The website allows an easy and quick assessment of how (un)usual the present situation is in the perspective of historical observations. By recalculating the relevant statistics after each event, it is always up-to-date. I can imagine that the site meets the needs of the general public as well as of managers of coastal-defense systems for easy and quick information. Extending the geographical scope of the website, as suggested by the authors, is certainly a good idea that should be perused.

I wonder, however, whether a scientific journal is the right place to present this service. The bulk of the paper is formed by lengthy descriptions of the time-evolution of the four parameters (water level, number, length and severity of events) at five of the ten stations available, without making any effort to explain or interpret the measurements. If the paper is accepted for publication, this part (essentially section 3) should be much shortened.

Detailed comments

p 3, l 94 : unsuable -> unusual (?)

p 4, l 121: putting -> put

p 9, l 223: de-clustering: how robust are the results to other choices of the de-clustering interval?

p 9, l 227: Start a new paragraph after MLE. Start it with something like "This choice is motivated by ...", or another way to say that what follows is an explanation of the choices (GEV + MLE) made.

p 11, l 279/80: "... and as a solid line when it is significant at the level of 95 ": The line is solid everywhere! You shouldn't place a general description of a figure in the caption of a specific figure.

This remark applies to all figures!

p 11, l 289: has -> have

p 11, l 299: The 'blue' event (2 Dec) has a return time longer than 5 years.

p 12, l 308: could -> can (?)

p 12, l 313: add 'number of' before 'historical monthly maximum'

p 13, l 328: particular -> the

p 14, l 371: spatially -> space

p 15, l 384: increasement -> increase

p 15, l 386-388: "... not fully explored ..." - you mean "... not fully explored here, ..."? I mean, there must be a lot of literature about the reasons for increasing water levels in Hamburg, e.g., von Storch & Woth (2008) (doi: 10.1007/s11625-008-0044-2) and references therein, but certainly there are more recent ones.

p 19, l 486-88: But heights were very modest. Sow hat does this long series of events mean in terms of threat for coastal defences?

p 20, l 523-27: This seems inconsistent: max water levels have no trend, which is consistent with results based on annual data. Presumably, this means that annual data show no trend either. However, in the next sentence you state that sea level in the Baltic did increase. So what is true? Or what do you want to say, but what I do not understand? Besides, isn't it remarkable that the mean water level increased, but the maxima did not? What does this say about other drivers of water level variability?

p 22, l 560-62: However ... sea level rise was excluded in those studies: At first sight, this looks contradictory, but after some thinking I find that it can happen: If the frequency of exceeding a threshold has no trend while the mean sea level increases, the strength of other (meteorological?) drivers of sea level variations must have decreased. That is interesting and would deserve more investigation. - This is an example of what I meant by "without making any effort to explain or interpret the measurements" above.

p 22, l 567: "notably increase at Warnemünde" - notably? The increase is not significant!

p 22, l 571-73: This is a repetition. Refer back to earlier figures. Applies to the following figures, too.

p 22/23, l 580-83: Strange. You find no trend in surge duration, but Weidemann, who excludes sea level rise from is analysis, finds an increase. I would expect sea level rise to increase the duration of events above a threshold to increase if the mean sea level rises, so if Weidemann finds an increase, you should find an even larger one.
Citation: https://doi.org/10.5194/nhess-2021-75-RC1
- AC1:
  'Reply on RC1', Ralf Weisse, 12 Aug 2021
  We appreciate the constructive comments of the reviewer. They helped us to clarify and improve our manuscript. In the following, we show how we will address the points raised by the reviewer in the revised manuscript.
  In the general discussion of the manuscript, the reviewer wonders whether a scientific journal would be the right place to present and discuss the developed service. We argue, that our manuscript describes and establishes a new method, which is well within the scope of a scientific journal. More specifically, we chose Natural Hazards and Earth System Sciences as journal, as its scope includes
  “the detection, monitoring, and modelling of natural phenomena, and the integration of measurements and models for the understanding and forecasting of the behaviour and the spatial and temporal evolution of hazardous natural events as well as their consequences;”
  
  and
  “the design, development, experimentation, and validation of new techniques, methods, and tools for the detection, mapping, monitoring, and modelling of natural hazards and their human, environmental, and societal consequences;”.
  
  From our perspective, the storm surge monitor and our manuscript are fully in-line with these points and the scope of the journal.
  However, we agree with the reviewer that our points were not made clear enough to be easily taken up. We will re-write the manuscript to make these points clearer and more explicit. Specifically, we will follow the suggestion of the reviewer and shorten the suggested parts (section 3) and extend our efforts to explain the approach and to interpret the data.
  The specific points of the reviewer will be addressed as follows:
  Lines 94, 121: The typos will be corrected.
  
  Line 223: We agree that the choice is somewhat arbitrary. We checked and found that no events occurred for which this de-clustering had an effect. We will remove this part from the text and the analysis.
  
  Line 227: Will be modified as suggested.
  
  Lines 279/280: We will follow the suggestion and adopt captions throughout the manuscript and shift general discussions and descriptions into the text.
  
  Line 289: The typo will be corrected.
  
  Line 299: The text will be changed accordingly.
  
  Lines 308, 313 (318), 328, 371, 384 (386): Will be corrected as suggested.
  
  Lines 386-388: We agree with the reviewer. We will add references and extend the discussion in the revised manuscript.
  
  Lines 486-488: The reviewer is right that heights were modest. But the results from the cited literature showed that particularly a series of events even of modest height may challenge coastal protection. We will be more explicit about this in the revised manuscript.
  
  Lines 523-527: Mean sea level is rising in the Baltic Sea but does show a strong spatial gradient with smaller values along the German Baltic Sea coast (see e.g., https://doi.org/10.3389/fmars.2021.647607). Moreover, there is strong decadal variability in sea level trends and large interannual and decadal variability. We will discuss this in the revised manuscript and put the statements into context.
  
  Lines 560-562: We will take up this point raised by the reviewer and extend the discussion accordingly.
  
  Line 567: The text will be revised accordingly.
  
  Lines 571-573: We will carefully check the manuscript for repetitions and remove them where not required within the context.
  
  Lines 580-583: As stated in the text, the increase described in cited literature was not significant. The same line of reasoning as above (see reply Lines 523-527) should hold. We will introduce a corresponding discussion in the text.
  
  Citation: https://doi.org/10.5194/nhess-2021-75-AC1
RC2:
'Comment on nhess-2021-75', Klaus Grosfeld, 17 Jun 2021
Review of Xin Liu, Insa Meinke, and Ralf Weisse “Still normal? Contextualizing real-time data with long-term statistics to monitor anomalies and systematic changes in storm surge activity – Introduction of a prototype web tool storm surge monitor for the German coasts”

General comments:

This manuscript describes the concept and implementation of a new online platform, the “Storm Surge Monitor” (www.stormsurge-monitor.eu / www.sturmflut-monitor.de), aiming to provide contextualized near-real-time data of storm surge activities along the German North Sea and Baltic Sea coasts. While the knowledge of storm surge activities, their occurrence rate, long-term evolution of maximum water levels as well as the return period of storm surge heights play a crucial role for coastal planners and authorities’, general information of potential threats under climate change with respect to sea level rise and changed wind / storm activities for the general public and dedicated stakeholders with critical infrastructures (e.g. big companies, harbors, container terminals etc.) is needed. The confrontation with theses kinds of risks in our daily life is important to rise awareness and acceptance and provoke support for mitigation and adaptation measures of climate change impacts.

A necessary precondition for this is a basic knowledge and process understanding on different levels which sets actual observations and events in a historical context. This is given by the Monitor, which combines near-real-time data of ten tide gauges (six of German North Sea and four of Baltic Sea) with long-term statistics of the available observational period (back to 70 – 120 years ago) providing contextualized information and daily updated graphs on height, frequency, intensity and duration of extreme events. This includes maximum water levels per season, the return period of an event at a local tide gauge, annual numbers of events exceeding a given threshold, and intensity and duration of a storm surge event allowing a critical consideration of its severity and threat.

In times of rising importance of climate change risks and induced political and societal measures on mitigation and adaptation such kind of tools are very helpful for the communication and exchange of knowledge between science and society. Therefore, I suggest publication of this manuscript with minor changes.

Specific comments:

The manuscript is well written and includes a clear description of the Monitor and used methods and statistics. Depending on the temporal resolution of the data not all graphs (duration and intensity) can be shown. This is mentioned in the manuscript but I couldn’t find it in the Monitor. Chapter 3 describes five stations (two North Sea, three Baltic Sea) in detail. While this is in general interesting, it occurs a bit lengthy and redundant, especially where information / comparison to other stations is given but not shown. I wonder, if some information could also be summarized in a table (e.g. trends) which would make the description a bit more compact. Instead, I would like to sea some emphasis on different evolutions and signals!

On page 16, line 407-409 the authors discuss the occurrents of events at stations Helgoland, Cuxhaven, Norderney (fewer) and more frequently at Husum, Hamburg and Bremen. I wonder, why Cuxhaven has fewer events while located at the estuarian tip between Weser and Elbe?

The authors emphasize that such online platforms “may provide substantial added value for professional stakeholders or the public discussion in general” (p.3, l. 96-99; p. 30-31, l. 775-781). I was wondering, if the authors have assessed the demand and need for such a Monitor and if they got into contact with key stakeholders to discuss the usefulness, the design and demand for the Monitor and the provided information? In times of modern knowledge exchange, we do know how important co-design of such processes is and the early involvement of potential users and stakeholders. A short additional paragraph on this aspect would be worthy for the readers and other scientist who plan similar services.

The title of the manuscript is really long? Is this really needed or could it be shortened to make it handier?

Technical corrections:

P 3, l 94: unusable -> unusual

P 4, l 121: putting -> put

Figure 2 and p 6, l 223: What is the reason for “four hours” de-cluster interval?

Table 2: NHN explain already here the meaning of NHN (German reference level, Normalhöhenull, NHN)

Figure 3 b: There is missing information on the right axis, which is also Water level in m relative to MThw

P 16, l 420-427 and p 19 l 490-500 and p 23, l 591-599: there is a mixture between abbreviated “hours” as h and a fully written hours when describing intensity and duration. I would suggest to unify this.

P 18, l 460-461: shift reference to Figure at correct position, since not all stations are shown there: A total of one or two events was observed at Helgoland, Cuxhaven (Fig. 4a), and Norderney, whereas six or eight events were observed at Husum, Hamburg (Fig. 7), and Bremen.

P 29, l 739: distinguisch -> distinguish

Figure captions: Some information is valid for many graphs, so it might be helpful to refer only to an extended description in a first graph (e. g.: The color of the bars denotes the degree of severity (green – minor; blue – severe; red – very severe). The trend (gray line) is shown with the 95 % confidence interval (light gray band) and as a solid line when it is significant at the level of 95 %.)

References (check bold marks):

Gönnert, G. and Buß, T.: Sturmfluten zur Bemessung von Hochwasserschutzanlagen, Landesbetrieb Straßen, Brücken und Gewässer 2/2009, ISSN 1867–7959, 2009.

Kodeih, S., et al.,: Climate information needs from multi-sector stakeholders, Deliverable 1.B, Work Package 1, Project ECLISEA, http://www.ecliseaproject.eu/wpcontent/uploads/2019/2008/d2011.b_report_stakeholder-needs_final_20180629.pdf, 2018.

Check weblink: page couldn’t be found and add all co-authors

Schaper, J., Ulm, M., Arns, A., Jensen, J., Ratter, B., and Weisse, R.: Transdisziplinäres Risikomanagement im Umgang mit extremen Nordsee-Sturmfluten – Vom Modell zur Wissenschafts-Praxis-Kooperation, Die Küste, 87, doi: 10.18171/1.087112, 2020.
Citation: https://doi.org/10.5194/nhess-2021-75-RC2
- AC2:
  'Reply on RC2', Ralf Weisse, 12 Aug 2021
  We appreciate the constructive comments of the reviewer. They helped us to clarify and improve our manuscript. In the following, we show how we will address the points raised by the reviewer in the revised manuscript.
  The reviewer makes the following specific comments (in bold) that we will address as follows (in italic).
  “Depending on the temporal resolution of the data, not all graphs (duration and intensity) can be shown. This is mentioned in the manuscript but I couldn’t find it in the Monitor."
  We will add a corresponding statement to the Monitor.
  “Chapter 3 describes five stations … in detail. While this is in general interesting, it occurs a bit lengthy and redundant, especially where information/comparison to other stations is given but not shown. I wonder if some information could also be summarized in a table (e.g. trends) which would make the description a bit more compact. Instead, I would like to see some emphasis on different evolutions and signals!”
  We will shorten this description and focus on the main evolutions and signals following also a similar suggestion by reviewer #1. We will also summarize the results as suggested here by reviewer #2.
  “On page 16, line 407-409 the authors discuss the occurrents of events at stations Helgoland, Cuxhaven, Norderney (fewer) and more frequently at Husum, Hamburg and Bremen. I wonder, why Cuxhaven has fewer events while located at the estuarian tip between Weser and Elbe?“
  There are primarily two reasons. First, the specific configuration of the coastline and bathymetry makes a tide-gauge more/less susceptible to storm surges. Also, the wind direction that most effectively generates storm surges differs between the tide gauges. Second, the Monitor uses a common threshold for all tide-gauges that is used by the Federal Maritime Agency (BSH) to issue storm surge warnings. In other statistics, local thresholds are used that are defined according to the DIN 4049. This will lead to differences in the number of detected surges. The issue is taken up and described in the FAQ section of the monitor. To address this comment, we will also discuss this more clearly in the revised manuscript. Taking feedbacks from stakeholders into account, we plan to include both statistics in the next version of the Monitor.
  “I was wondering, if the authors have assessed the demand and need for such a Monitor and if they got into contact with key stakeholders to discuss the usefulness, the design, and demand for the Monitor and the provided information? In times of modern knowledge exchange, we do know how important co-design of such processes is and the early involvement of potential users and stakeholders. A short additional paragraph on this aspect would be worthy for the readers and other scientists who plan similar services.”
  The Monitor was developed in close contact with representatives of authorities responsible for coastal protection. We constantly receive feedback (see the reply on the comment above) and plan to include such feedback in new releases. Presently a survey is performed among stakeholders in the region the feedback of which will be evaluated and taken up to improve the Monitor. As suggested by the reviewer we will add a paragraph briefly discussing these issues.
  “The title of the manuscript is really long? Is this really needed or could it be shortened to make it handier?”
  We agree with the reviewer and we will replace the title with a shorter more suited one.
  The reviewer further suggests several technical corrections that we will address as follows:
  Lines 94, 121: The typos will be corrected.
  
  Figure 2, Line 223: We agree that the choice is somewhat arbitrary. We checked and found that no events occurred for which this de-clustering had an effect. We will remove this part from the text and the analysis.
  
  Table 2: Explanation will be added.
  
  Figure 3b: Explanation will be added.
  
  Lines 420-427, 490-500, 591-599: We will unify as suggested.
  
  Lines 460-461: Will be modified as suggested.
  
  Line 739: The typo will be corrected.
  
  Figure captions will be revised taking also the comments from reviewer #1 into account.
  
  References will be checked and corrected.
  
  Citation: https://doi.org/10.5194/nhess-2021-75-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Reconsider after major revisions (further review by editor and referees) (01 Sep 2021) by Paolo Tarolli

AR by Ralf Weisse on behalf of the Authors (06 Oct 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (14 Oct 2021) by Paolo Tarolli

RR by Andreas Sterl (21 Oct 2021)

RR by Klaus Grosfeld (30 Oct 2021)

ED: Publish subject to minor revisions (review by editor) (10 Nov 2021) by Paolo Tarolli

AR by Ralf Weisse on behalf of the Authors (16 Nov 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (04 Dec 2021) by Paolo Tarolli

AR by Ralf Weisse on behalf of the Authors (13 Dec 2021)

Short summary

Storm surges represent a threat to low-lying coastal areas. In the aftermath of severe events, it is often discussed whether the events were unusual. Such information is not readily available from observations but needs contextualization with long-term statistics. An approach that provides such information in near real time was developed and implemented for the German coast. It is shown that information useful for public and scientific debates can be provided in near real time.