the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Lava flow hazard modelling during the 2021 Fagradalsfjall eruption, Iceland: Applications of MrLavaLoba
Melissa A. Pfeffer
Sara Barsotti
Simone Tarquini
Mattia de' Michieli Vitturi
Bergrún Óladóttir
Ragnar Heiðar Þrasstarson
Abstract. On March 19, 2021, the first eruption in ca. 800 years took place in Fagradalsfjall on the Reykjanes Peninsula, in the backyard of the capital Reykjavík. This effusive eruption was the most visited eruption in Iceland to date and needed intense lava flow hazard assessment and became a test case for hazard assessment for future eruptions on the Peninsula, which can issue lava into inhabited areas or inundate essential infrastructure.
In this study we documented how lava flow modelling strategies were implemented using the stochastic code MrLavaLoba, evaluating hazards during the 6-month long effusive event. Overall, the purposes were three-fold; (a) Pre-eruption simulation to investigate potential infrastructure at danger for lava flow inundation (b) Syn-eruptive simulations for short-term (two weeks’ time frame) lava flow hazard assessment and (c) Syn-eruptive simulations for long-term hazard assessments (months to years). Furthermore, strategies for lava barrier testing were developed and incorporation of near-real time syn-eruptive topographic models were implemented.
During the crisis the code was updated to increase functionalites such as considering multiple active vents as well as code optimization that led to a substantial decrease in the computational time required for the simulations, speeding up the delivery of final products.
Gro B. M. Pedersen et al.
Status: final response (author comments only)
-
RC1: 'Comment on nhess-2022-166', Anonymous Referee #1, 07 Dec 2022
Revision
The paper is a chronological narration of how the software MrLavaLobe was used and modified during the 2021 Fagradalsfjall eruption occurred in the Reykjanes peninsula, Iceland. The manuscript is really interesting, and some sections are also compelling. But the text has some problems. In the manuscript, the authors initially describe the eruption, referring to a work in press, and this description seems a summary of the cited work. Then they describe the software without focusing on input parameters and procedures but describing the functionality with a qualitative approach. Then narrate the eruptive crisis and the results of the software used in real time. Finally, they describe the modification and the improvements to the software that were necessary to communicate with the stakeholders and predict lava inundation in a future eruption. The manuscript is too long and, while reading, the scientific motivations of the work are lost. The manuscript also lacks a clear introduction to the software and this lack impedes a full understanding of the reported improvements if not by reading the original paper and making comparison with this one.
I found the manuscript having a journalistic approach, not suitable for a scientific journal. Otherwise, if the authors intended to publish a technical report about the software improvements due to a real time application, then they should remove the chronological description of the happenings and concentrate on the technical improvements of the software and its application to the 2021 Fagradalsfjall eruption.
I appreciated the paper that is well organized in sections but it is too long and the text gets lost in long explanations that could be summarized and made simpler. Moreover, in the manuscript I found some repeated sentences that authors should eliminate. The English is somewhere not fluent and I requested to rewrote some sentences. Sentences are often too long and dispersive; verbs are somewhere used in the wrong way.
I also found that figures are not correctly cited and in the section 2 a figure or a citation to a figure is missing.
Fig.1a and 1b are never cited in the text, while the paragraphs 2 and 2.1 need reference to figures to understand the geography and the eruptive history. The same occurs for other figures. I suggest inserting in the text the right citations of all the figures by indicating also the figure boxes useful in the text.
I attached the pdf of the manuscript where I put my comments with suggestions and critical points, but the list is not exhaustive. I think that the authors should do an effort to re-reading the manuscript and re-writing the longer and twisted sentences and eliminate repetitions. The authors should also rethink the qualitative setting given to their paper substituting the long descriptive part with short quantitative sentences. For these reasons I suggest a major revision.
-
AC1: 'Reply on RC1', Gro Pedersen, 14 Mar 2023
First the authors want to thank RS1 for taking on the task of reviewing this paper. The authors have responded to the general comments from RS1 section by section. The bold text is general comment made by RS1 and the regular text is our response to the comment above. Response to the detailed comments can be found in the annotated pdf.
The paper is a chronological narration of how the software MrLavaLobe was used and modified during the 2021 Fagradalsfjall eruption occurred in the Reykjanes peninsula, Iceland. The manuscript is really interesting, and some sections are also compelling.
Thanks.
But the text has some problems. In the manuscript, the authors initially describe the eruption, referring to a work in press, and this description seems a summary of the cited work.
We agree that the manuscript can be improved. It is correct that we do only reference the Pedersen et al. (in press) paper for the eruption evolution. Upon submission this was the only paper “in press” describing this eruption and is therefore the only paper cited. This can now be improved because more papers have been published.
Then they describe the software without focusing on input parameters and procedures but describing the functionality with a qualitative approach.
We choose to cite the paper (de’ Michieli Vitturi and Tarquini, 2018) which was published with the release of the software instead of describing the technical details in this paper, because we want much more emphasis in this work on the usage of this code during the Fagradalsfjall 2021 eruption. In table 1 we do provide an overview of procedures applied, their goal, the approach and key input parameters. The purpose of having these together in a table is exactly to describe the functionality in a qualitative way that enables the reader to both understand the motivation and process and to be able to duplicate our work. Table 1a (in the appendices) provide all input parameters (or input parameter ranges). We do agree that the code changes made during this eruption have to be described better, and not only in the qualitative fashion as done in this manuscript. We will therefore add an appendix describing the exact changes in the code. We will also do a better job to refer to the tables, so this is much clearer for the reader.
Then narrate the eruptive crisis and the results of the software used in real time. Finally, they describe the modification and the improvements to the software that were necessary to communicate with the stakeholders and predict lava inundation in a future eruption. The manuscript is too long and, while reading, the scientific motivations of the work are lost.
We do agree that the manuscript is too long and should be shortened, repetitions should be deleted, and we must do a better job being concise and making sure that the motivation for the different modelling approaches is conveyed better. In the detailed comments we have made suggestions on how to improve this, partly by adding a new figure that show the evolution of the eruption, how the different activity phases and styles required different approaches to the modelling and what we did to achieve our objectives (code changes or changes in how the code was implemented and/or results were displayed). We will also make sure that both figures and tables are better referenced since lots of information that RS1 requests can be found there (especially in the tables).
The manuscript also lacks a clear introduction to the software and this lack impedes a full understanding of the reported improvements if not by reading the original paper and making comparison with this one.
We will improve the introduction to the software in section 3 and provide more technical information in an appendix. However, the goal is to apply an already published code to the Fagradalsfjall 2021 eruption and we therefore think it is logical that the reader may need to read the de’ Michieli Vitturi and Tarquini (2018) to get the full overview of what the code does. Otherwise, this already too long paper gets even longer, and we think that our goal of the paper to show the usage of the code gets further lost in describing the code itself. However, we will make sure that all changes made in the code during this volcanic crisis (and thus changes in the code from when it was published in 2018) will be described better, both in the paper itself and with substantial detail in an appendix.
I found the manuscript having a journalistic approach, not suitable for a scientific journal. Otherwise, if the authors intended to publish a technical report about the software improvements due to a real time application, then they should remove the chronological description of the happenings and concentrate on the technical improvements of the software and its application to the 2021 Fagradalsfjall eruption.
We think this critique derives from our goals for the paper not coming through well enough. This paper is not intended to be a technical report focusing on the code changes, - but rather it should be addressing how evolving lava hazards of the Fagradalsfjall 2021 eruption led to changing demands to the team providing the model results, which were addressed through changes in the MrLavaLoba code and how it was run. We want to highlight how flexible the MrLavaLoba code is and how this can be used in an ongoing effusive eruption, to show all the novel aspects of this interplay between the needs of end users and the potential of the scientific tool that has not been described before. Furthermore, we wish to highlight the caveats as of now and improvements that should be emphasized in the future. We will make sure that these goals are clearer in the abstract, introduction and conclusion of this paper, so the reader is clear on the goals of the paper.
The authors think the chronical narrative is necessary since the changing in activity led to different ways of implementing the code and to code changes. The way this was done, was very much a consequence of the eruption evolution and doing this during this eruption. Many of the tasks we were trying to address to provide information to the hazard assessment would have been done differently post-eruption and many of them we could not have imagined in the pre-eruption phase prior to the Fagradalsfjall 2021 eruption. Since the eruption lasted half a year and the activity changed multiple times during this eruption, it is obviously a complex task and it is clear that we need to do better to convey this message.
I appreciated the paper that is well organized in sections but it is too long and the text gets lost in long explanations that could be summarized and made simpler. Moreover, in the manuscript I found some repeated sentences that authors should eliminate. The English is somewhere not fluent and I requested to rewrote some sentences. Sentences are often too long and dispersive; verbs are somewhere used in the wrong way.
We fully agree with this statement. It is clear that our message is a bit lost in this lengthy manuscript and we need to improve the text. We have described numerous ways of how we will do this in the detailed comments to RS1, that we also want to thank for taking the time to point out these weaknesses.
I also found that figures are not correctly cited and in the section 2 a figure or a citation to a figure is missing. Fig.1a and 1b are never cited in the text, while the paragraphs 2 and 2.1 need reference to figures to understand the geography and the eruptive history. The same occurs for other figures. I suggest inserting in the text the right citations of all the figures by indicating also the figure boxes useful in the text.
We fully agree, we will work on improving how all figures and tables are referenced.
I attached the pdf of the manuscript where I put my comments with suggestions and critical points, but the list is not exhaustive. I think that the authors should do an effort to re-reading the manuscript and re-writing the longer and twisted sentences and eliminate repetitions. The authors should also rethink the qualitative setting given to their paper substituting the long descriptive part with short quantitative sentences. For these reasons I suggest a major revision.
Thanks again for these comments, we have gone through them all agree with many of them. Those where we disagree, we have explained why. We will shorten the text, delete repetitions, and improve the language. We will focus on improving the text so that our message is brought to the forefront and we will include more important technical details in the appendix.
-
AC1: 'Reply on RC1', Gro Pedersen, 14 Mar 2023
-
RC2: 'Comment on nhess-2022-166', Anonymous Referee #2, 12 Jan 2023
This paper describes the use of a computer code to simulate lava flows during an eruption. However, the paper is badly written and missing all scientific information that would be expected. The authors should completely rewrite the paper with a focus on what they are doing. They need to describe in a scientific manner what they are doing and how they change the code in different scenarios. In present form the paper cannot be accepted, there are several flaws in their references, where they are used to insinuate that they have information on the eruption, but that information does not exist in those papers. In general, the paper should be turned down in its present form. That said I do also encourage the authors to rewrite the paper and be more scientific in such a way that the reader can reproduce their calculation wishes him to replicate what they have done.
Current version is written in such a way that the scientific community can not be proud of.
The authors should also be more careful how they use references, keeping in mind that a reference does need to have that information that you claim to get from it.
-
AC2: 'Reply on RC2', Gro Pedersen, 14 Mar 2023
The authors have responded to the general comments from RS2 section by section. The bold text is general comment made by RS2 and the regular text is our response to the comment above. Response to the detailed comments can be found in the annotated pdf.
This paper describes the use of a computer code to simulate lava flows during an eruption. However, the paper is badly written and missing all scientific information that would be expected.
We disagree. Whilst the paper may be too lengthy, and language at places can be improved, it is incorrect that this paper is missing all scientific information expected. Honestly this is a very confusing claim, that just gets more confusing through the review, below are some examples based on the detailed review:
- For example, RS2 requests an overview of the purpose of the simulations, which is directly stated in Section 4 and is provided in Table 1 giving a systematic overview of all simulations conducted.
- RS2 requests all input parameters which are provided in Table 1a.
- RS2 requests coordinates for simulations which are again all provided in Table 1a.
- RS2 requests a metric comparison in Figure 3 between real time volume and simulation volume (see L325-330): Figure 3 shows exactly that.
- RS2 is surprised that syn-eruptive DEMs are made (L372-373) when this is described the data section 3.1.2 "Syn-eruptive photogrammetric surveys" and when section 3.2.1 in the methods section describe how this was done ( L213-215)
The authors should completely rewrite the paper with a focus on what they are doing. They need to describe in a scientific manner what they are doing and how they change the code in different scenarios.
This is exactly what we do. In Section 4, Table 1 and Table 1a we explicitly describe why and how we changed the code and input parameters. We will shorten and make Section 4 more concise, but RS2 is simply wrong when stating that we do not focus on what we are doing.
In present form the paper cannot be accepted, there are several flaws in their references, where they are used to insinuate that they have information on the eruption, but that information does not exist in those papers.
Again, this comment is very confusing for the following reasons:
- Throughout the paper RS2 wants the authors to delete the citation Pedersen et al. (in press) and at the same time claim that we do not cite where we have various information that stems from this paper (see e.g., comments to line 130-153). RS2 wants us to delete the very citation that provides the information whilst commenting that we have to cite the work that provides the information. Those are mutually exclusive. Possibly RS2 is misunderstanding what paper the “Pedersen et al., in press” is, but this is abundantly clear as we used proper citation of it and it is referenced properly in the References. It is clear that it is a paper in GRL led by the same author that led this manuscript. This frequently cited paper was “in press” upon submission of this manuscript, and was publicly available for everyone, including RS2. It is a key paper for this manuscript since the data provided in the paper and the repository published along with the paper have been used for our simulations. It obviously must be cited, and it is wrong when RS2 wants us to delete this citation.
-Based on comment L184 to the reference Pedersen et al. (in press) stating “If published ok, otherwise delete” it seems that RS2 simply doesn´t know what paper we are referring to. A paper “in press” will obviously be published since it is in press. We have been wondering if RS2 thought we are citing this submitted paper that RS2 reviewed, which obviously isn´t the case, and this has led to their insulting comment that we “Insinuate we have information that we don´t have in those papers” which is an extremely serious personal attack. This could potentially explain why the reviewer thought that data we had available for our runs wasn´t available (e.g. regarding comments 4.2.3). These comments simply seem unqualified and undermines our belief that RS2 has read this paper with their responsibilities as a reviewer in mind.
Another place RS2 criticizes is our reference to one conference abstract. This is done five times from L264-274. This submitted manuscript describes the lava simulation during the 2021 Fagradalsfjall eruption, and was submitted in June 2022, fairly quickly after the eruption ended. Therefore, multiple papers describing the eruption had not been published, namely the Sigmundsson et al 2022 Nature paper, which was under strict ban when we submitted this paper. Instead, at the request of the authors of that paper we cited an abstract from an EGU presentation (Geirsson et al., 2021). We know that this was not the ideal citation, but it was basically the only citation available upon submission, and this is common practice when working with recent geologic events. Now, papers have been published and the Geirsson abstract will be exchanged by the Sigmundsson et al. (2022) paper.
In general, the paper should be turned down in its present form. That said I do also encourage the authors to rewrite the paper and be more scientific in such a way that the reader can reproduce their calculation wishes him to replicate what they have done.
It seems that RS2 has not realized that all information necessary to reproduce our calculations is already contained within the paper or referenced within the paper and available to the reviewer. As stated in the Code and Data availability statement the code is available on github, the data we use is available on the Zenodo repository already published by Pedersen et al. in press (now Pedersen et al., 2022) and Table 1A provides the full overview of all input parameters. Thus, already in the current state RS2 can access the code used, the necessary data and input parameters to replicate our results.
Current version is written in such a way that the scientific community can not be proud of. The authors should also be more careful how they use references, keeping in mind that a reference does need to have that information that you claim to get from it.
Please see the comment above regarding citations.
On the note of scientific pride: Professional feedback should focus on how to improve the paper and according to the principals of this journal is required to include reasoning and justification so that the authors submitting the paper can understand where the comment is coming from. Opinions, expressed in condescending and patronizing language are irrelevant and unprofessional. This includes insinuations about the authors’ scientific capability to understand the very field they are writing a paper on, and suggestions of dishonest scientific practice. We therefore find that this review is clearly in violation of the journal´s policy on Publication Ethics:
“The reviews of submitted manuscripts must be done objectively, and the referees should express their views clearly with supporting arguments.”
This reviewer is not objective, their views are not expressed clearly, and they do not provide supporting arguments.
Further, the reviewer violated their responsibility as a reviewer.
“A referee of a manuscript should judge objectively the quality of the manuscript and respect the intellectual independence of the authors. In no case is personal criticism appropriate.“
This intelligence of the authors was not respected and personal criticism was rampant throughout the review.
Below are examples where RS2 made personal attacks on the authors rather than providing a scientific review of the work (See authors response to these in annotated pdf):
“the other does not fully understand the complexity of lava flow modelling” L 45
“The authors are here presenting a computer exercise with MrLavaloba, “ L48, L235, L242, L258.
“Otherwise this has to be seen as invented data” L269 & L275
“Line 264-274. This is serious, you are giving some conditions here and referring to an abstract that is dealing with an unrest in 2020 in Svartsengi, that is in completely different area to the one that actually erupted. At worst this is giving the reader wrong information on purpose.”L275.
Explain what you are doing” : L255, L290, L300,L340 in sections where we are explaining what we are doing
“This paper is not written in emergency mode, thus you need to tell the reader what you are doing, you can not make any shortcuts in this paper. Also it would be good if you could explain the complexity involved in a appropriate manner.” L320
“All this simulation needs much more thorough explanation, one of the fundamentals of research papers are that what you do should be possible to replicate. Her all information are missing and you are clearly not taking into account the complexity of the lava flow by taking into account internal pathways.” L379. Lava flow models are approximations of the actual process, none of them describe every aspect of lava flow emplacement, very few if any lava flow simulation models manage to take internal lava flow paths into accounts. This comment is therefore very pretentious criticism that can be directed to any existing lava flow models.
“This paper is starting to be more and more futuristic in an fantasy stile. Here you are changing completely the strategy for the simulation. Instead of using vents you are now using lava fronts. You really need to introduce this in the methods section and tell us if the code was changed again to be able to address this. Your paper is missing all scientific arguments and approches.” L397-400.
“Explain to us how you did find that out
This is getting more fantastic by every line...” L402-405
“You have been showing us the advance and thickness so far in one map, so what is the difference here? do the thickness map not show realistic values or what? You can not present your work this way, in a novel yeas or an autobiography yeas, but not in scientific paper” L409
“Again you need to explain all changes to the code and what effect they have. It is like you are trying to convince the reader that MrLavaLoba is the solution to all our problems, you are in a seller mode now.” L450
The authors doubt that RS2 has the ability to objectively judge this manuscript, either because RS2 has something personal at stake or because RS2 has not put in the appropriate time to provide a professional review.
That said, we do think that the paper is lengthy, that language can be improved, and we will do a better job being concise and making sure that the motivation for the different modelling approaches is conveyed better. We will add a new figure that shows the evolution of the eruption, how the evolving activity required different approaches to the modelling and what was done in order to tackle this (code changes and/or changes in how code was implemented). We will also make sure Figures and Tables are better referenced since lots of information that RS1 requests can be found there.
-
AC2: 'Reply on RC2', Gro Pedersen, 14 Mar 2023
Gro B. M. Pedersen et al.
Gro B. M. Pedersen et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 498 | 231 | 21 | 750 | 9 | 8 |
- HTML: 498
- PDF: 231
- XML: 21
- Total: 750
- BibTeX: 9
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1