the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Jul 2022
The three-dimensional structure of fronts in mid-latitude weather systems as represented by numerical weather prediction models
Abstract. Atmospheric fronts are a widely used conceptual model in meteorology, most encountered as two-dimensional (2-D) front lines on surface analysis charts. The three-dimensional (3-D) dynamical structure of fronts has been studied in the literature by means of “standard” 2-D maps and cross-sections and is commonly sketched in 3-D illustrations of idealized weather systems in atmospheric science textbooks. However, only recently the feasibility of objective detection and visual analysis of 3-D frontal structures and their dynamics within numerical weather prediction (NWP) data has been proposed, and such approaches are not yet widely known in the atmospheric science community. In this article, we investigate the benefit of objective 3-D front detection for case studies of extratropical cyclones and for comparison of frontal structures between different NWP models. We build on a recent gradient-based detection approach, combined with modern 3-D interactive visual analysis techniques, and adapt it to handle data from state-of-the-art NWP models including those run at convection-permitting kilometer-scale resolution. The parameters of the detection method (including data smoothing and threshold parameters) are evaluated to yield physically meaningful structures. We illustrate the benefit of the method by presenting two case studies of frontal dynamics within mid-latitude cyclones. Examples include joint interactive visual analysis of 3-D fronts and warm conveyor belt (WCB) trajectories, and identification of the 3-D frontal structures characterising the different stages of a Shapiro-Keyser cyclogenesis event. The 3-D frontal structures show agreement with 2-D fronts from surface analysis charts and augment the surface charts by providing additional pertinent information in the vertical dimension. A second application illustrates the effect of convection on 3-D cold front structure by comparing data from simulations with parameterised and explicit convection and shows that convection could strengthen the cold front. Finally, we consider “secondary fronts” that commonly appear in UK Met Office surface analysis charts. Examination of a case study shows that for this event the secondary front is not a temperature-based but purely a humidity-based feature. We argue that the presented approach has great potential to be beneficial for more complex studies of atmospheric dynamics and for operational weather forecasting.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(19717 KB)
Interactive discussion
Status: closed
-
RC1: 'Comment on wcd-2022-36', Anonymous Referee #1, 18 Aug 2022
In the manuscript "The three-dimensional structure of fronts in mid-latitude weather systems as represented by numerical weather prediction models" document an implementation and slight adaptation of an existing 3-dimensional front surface detection algorithm in the context of the interactive analysis software Met.3D. The authors then showcase their implementation for two case studies of autumn/winter storms over Europe and relate their detected fronts to other (mostly very well established) meterological concepts.The manuscript is quite comprehensive and touches upon many aspects around mid-latitude cyclones and their fronts. With the wide variety of aspects covered, it however remains somewhat unclear what, in its core, this manuscript is about. Discussions generally remain superficial and don't add much new to the literature except further anecdotal support for otherwise already very well established meteorological concepts (i.e., WCBs, Shapiro-Keyser cyclone model). Because of this overall lack of direction and novelty in the discussions, the manuscript appears in its present form to be mainly an advertisement for Met.3D. I recommend the editor to reject this manuscript, and the authors to submit more focused and targeted analyses/discussions on any or all of the mentioned topics individually.
More specifically:
(1) The front surface detection seems to be only a minor modification/optimisation of the algorithm introduced and implemented in Kern et al. (2019). Judging from the illustrations in Kern et al. (2019), the algorithm was already then implemented in Met.3D. Yet, the algorithm is introduced and discussed here in as much detail as if it was new. Further, the authors "validate" well-established meteorological concepts such as the Shapiro-Keyser cyclone model and the spatial relation between WCBs and fronts using their visualisation and front detection algorithm. Given how successful these concepts have been over decades, I find this quite assuming. If these concepts had failed to show up in their analysis, I would much rather doubt the implementation and visualisation in question rather than these meteorological concepts. Now, given that everything looks as expected, I am unsure what to take away from the "validation" beyond that the algorithm and visualisation is working fine---and so much that had already been shown by Kern et al. (2019).
(2) The authors discuss briefly the best choice of thermodynamic variable for the front detection. This choice remains an subject of debate, and a new perspective on this choice could warrant another publication. This would however require considerable additonal analyses; based on only two case studies, the authors are not in a position to give general recommendations (as presently done in the summary and discussion section).
(3) Similarly, a front classification into humidity and temperature-dominanted fronts would most likely be worthwhile and well warrant a publication. But this aspect is discussed by far too superficially to justify the publication of the present manuscript.
(4) Similarly, the comparison of WCBs and frontal structures in parameterised-convection versus convection-resolving models is both timely and interesting. It would certainly warrant a publication of its own. But again, this aspect is discussed by far too superficially here.
I would very much encourage the authors to extend their analyses in particular on the topics in issues (3) and (4) and to submit manuscripts with in-depth analyses on these. The 3D front surface detections will surely certainly be helpful for either.
Finally a comment on the frequent use of 3D visualisations in the present manuscript. Besides the 3D visuals, I find the visuals and text to be clear. Met.3D is undoubtedly a powerful tool for the interactive and exploratory analysis of a meteorological dataset--the interactive demonstration in video supplement 3 shows that clearly. At the same time, as static images on (digital) paper, I don't find the shown 3D visualisations useful. Figure 3 in the manuscript is a good example: Everyone who has looked at weather charts before will be able to grasp the chart in panel (a) within fractions of a second and have a clear impression of the synoptic situation. This would still be true even if additional lines were plotted to depict the intersection of the frontal surface with different vertical levels. But it is not true for panel (b), where I remain unsure about the frontal structure and synoptic situation even after looking at the panel for minutes. The main cyclone core is not visible behind the front surfaces, so I need quite some mental energy to disentangle the occluded from the cold front (which some ambiguity remaining) to then infer with my meteorological intuition that there should be another cyclone core hidden somewhere---an inference that I couldn't be sure about without the verification in panel (a). Because, given only panel (b), may be the occluded front is actually a secondary cold front and the only cyclone core in the chart is the one close to Iceland, the one that is peaking out behind the occluded front? Essential aspects of the 3D structure can still be conveyed in 2D maps; I would find such maps much less ambiguous, easier to parse, and thus more suitable for a static publication.
Citation: https://doi.org/10.5194/wcd-2022-36-RC1 -
CC1: 'Reply on RC1', Marc Rautenhaus, 23 Aug 2022
Dear Reviewer,
Thank you for providing your comments. We recognize your concern that our manuscript provides too little new insight on dynamic meteorology to the literature, and we would like to use the possibility of this discussion to briefly clarify the goals of our paper. We will reply in more detail after the discussion period has ended.
There were two major objectives that we wanted to achieve with our work: (a) a reimplementation and generalization of the Kern et al. (2019) method such that it is robust when used with current model data and can handle additional filter options, and to document the tool and make it available to the community as an open-source release, alongside the paper (which the original Kern et al. method was not); and (b) provide guidance to researchers on how to use 3-D front detection and visualization by investigating suitable method parameters and by showing the potential of the method for meteorological analyses with selected examples.
Since many people in the meteorological community are not used to working with 3-D visualization, we consider it valuable to demonstrate its potential. We tried to do this using familiar examples that are easily related to conventional 2-D depictions (for example in Fig. 3), but show features that would not be easily seen in the conventional views. It is true that static 3-D images can require significant effort to interpret, but our aim was to help and encourage the reader to make this effort, and eventually to explore the interactive tool. It is helpful in this regard that it is now possible to publish animations along with a paper (as we have done).
We did consider submitting the article to GMD, which would be the usual place to publish descriptions of models and analysis tools, but we chose to submit to WCD in order to reach the audience that is most concerned with analyses of atmospheric dynamics and that would thus benefit from learning how to use 3-D visualization for their work. We agree that we picked up several aspects that would (and should!) require much more in-depth research and could be published on their own – that, however, was not the goal of this paper, and is something we see as future work that could be approached by using our method.
Kind regards,
Marc Rautenhaus, on behalf of the author team
Citation: https://doi.org/10.5194/wcd-2022-36-CC1
-
CC1: 'Reply on RC1', Marc Rautenhaus, 23 Aug 2022
-
RC2: 'Comment on wcd-2022-36', Anonymous Referee #2, 05 Oct 2022
Summary:
The paper by Beckert et al. documents the method, capabilities and calibration of a frontal identification algorithm that can be used to then produce 3D plots of those frontal structures. I found the paper to be written well with many interesting plots presented. Nonetheless, the paper does not provide significant advancement of an already existing method. Furthermore, while the case studies described in the paper are interesting and certainly do have the potential to be published in WCD (and I strongly encourage the authors to do this), their brief descriptions in this paper are not sufficient for WCD. I feel this paper would be better suited to GMD, for example, given the brief analyses of the case studies. I therefore do not recommend publication in WCD and that the paper should be passed on to GMD for consideration instead. I would happily re-review this paper from a GMD context if it were submitted there. I have provided some more major considerations for the authors below along with a list of minor points.
Major points:
The paper feels disjointed in its current setup. I feel that Section 3 should be re-distributed into Section 5 so that 3.1 gets mixed in with the introductory paragraph of section 5, 3.2 goes at the start of section 5.1 (5.1.1 and 5.1.2?) and section 3.3 does likewise in section 5.2 (5.2.1 and 5.2.2). There would then be a nice flow from introducing the case study and the plots would flow logically from “concept” to “analysis”. Currently there are methods in section 2, an introduction of the case studies in section 3, another methods section is given in section 4 and we then return to the case studies in section 5. As you can see, the paper “jumps” around a bit. Putting section 3 into section 5 and re-ordering the figures will make it flow much better instead of having to refer-back to section 3 from section 5. Furthermore, section 5 jumps around too – better to stick with the one case cyclone and focus in on several features associated with it i.e. make 5.3 part of 5.1 e.g. 5.1.1, 5.1.2 and 5.1.3 then push section 5.4 into 5.2 (5.2.1, 5.2.2 and 5.2.3).
Section 2.3 and Figure 2: The wording in the numbered list should match the plots and the plots should then be referred to in each of the points of the numbered list e.g. point 1 goes with Fig 2a, point 2 with Fig 2b etc. It seems that this Figure-numbered list relation does not hold true in all cases so the authors should either adjust the list or adjust the figure to make the two complimentary. There are also no descriptions of the panels in the Figure 2 caption so you should say “see Section 2.3 for a description of the panels a-h” in the figure caption.
Figure 3b: This is difficult to interpret and would probably be better if the fronts were colour coded to show the difference between warm, cold and occluded (red, blue and purple?). I just find the figure to have a lot of “green swirls” that really need to be separated visually to make the features stand out. Furthermore, the scale is too smooth to really show the location of the fronts in the vertical. Using different colours for the type of front or markedly different colours at each pressure level might make these (and all the other figures that use the green-white colour scaling for the pressure heights) clearer.
Figure 9 and Lines 475 – 490: This whole description is very difficult to see as both the paper and online plots are far too “busy”. Surely you can distinguish between the “fast ascending” and “slow ascending” trajectories and plot them separately. I would discard figures 9e and 9f and replace them with figures showing “fast” and “slow” ascending trajectories. You would then only need to slightly re-word lines 475-490 to account for this change. I feel that the whole paragraph would then read much better with the adjusted figure.
Figure 10: I’m not convinced this figure shows anything a 2D figure wouldn’t show. It is trying to do too much by overlaying everything on the same plot and so the details are lost. I would like to see this separated out into ~6 panels that show the 3D structure clearly from the best angle. I would also suggest removing the land and just focussing on the cyclone itself (maybe adding in isobars for cyclone-centric orientation). In its current format, it shows less than what a 2D plot shows – but has clear potential to be excellent if it were better focussed (I can see why you would want to see all of this in 3D, it just does not show up well). Figure A3 for cyclone Egon is actually clearer; however, A3 would also benefit from having a slightly more upright angle, removing the land and adding isobars.
Lines 554-568 and Figures 11, A4 and A5: I found this passage very difficult to read and follow. The 3D plots make things difficult to reconcile. I would suggest circling exactly where you want the reader to look instead of trying to describe it (lines 554-556). You could then say something simpler like, “There is a gap in the frontal surface between 700-600 hPa in the ECMWF data whereas the frontal surface is present in the COSMO simulation (circled in Figs 12a, b).” It just focuses the reader on the point you want them to look at. I also think it might be worth including Figure A5 in Figure 12 and even plotting the difference between the THETA_W fields between COSMO and ECMWF (and possibly likewise for THETA). The reason for that it that I’m not convinced by your “convection drives differences in the temperature gradient” argument. It is possible that the opposite is true e.g. the temperature gradient around 700 hPa is stronger in COSMO (i.e. simulated better), which then leads to the development of convection along the frontal zone. The front may have been going through frontogenesis and the convection is just the result of that. I therefore do not believe your description of this process is convincing enough to be certain of the process you describe. The analysis does not contain enough detail.
Minor points:
Figure A1: I cannot find any reference to this figure in the text. Can it be removed?
Figure A2: I think this can be removed if you adjust Figure 9 (you could even include the jet in figure 9).
L221-224: Sentence starting “The method uses…” is very long. Please split this in two.
L287: Change “a decay stages” to “a decay stage”.
L300: Change “UTC5.3” to “UTC”.
Figure 3 caption: remove the extra “)5.3” near the end of the third line.
Line 321: Add “on” before “17 January 2018”.
Line 324: Change “As a result of the cyclone, high wind speeds were registered…” to “ The cyclone caused high wind speeds…” to be more concise.
Line 328: Change “… this is a Shapiro… “ to “… this was a Shapiro…” as it happened in the past.
Figure 7: Please include the time and date for these plots. It helps for stopping the video in the relevant place (I can see this information is in Figure 2 but should also be here). Also, the caption is unnecessarily detailed as you say most of it in the text. The caption only needs the description of the figures, not the explanation about what each step does (as you explain in the text). Please trim the caption down to make it easier to read.
Figure 7c: Maybe I’m missing something, but it looks like the feature is still in the plot under the blue circle (unlike in 7e where the northern feature disappears). Is this figure correct? Additional – Line 429 (related to Fig 7c comment) – OK I see this more now, but it is very subtle. I would focus that blue circle in to EXACTLY where you want the reader to look.
Figures 7d: it is very hard to work out where this cross section is taken without looking at 6f as there’s too much shading. If you put a line on Figure 6c to show the location of the cross-section then that would help make it clearer (then refer to it in the caption).
Line 427: should it be “of the filter” instead of “of filter”?
Figure 9: The plots get very ‘busy’ with time, especially figure (f). If you could show where the viewpoint 2 and viewpoint 3 cross sections are located in figure (a) for example, then it’d help. If the land masses weren’t blocked out so much in / at the very periphery of figure (f) then that would help.
Line 475: “north-easterly direction” should really be “north-eastward direction”.
Fig. 11: For clarity, it might be worth making the THETA_W scale blue-red so that the grey shading shows up better. The grey-blue end could be confused with the horizontal gradient shading.
L585: Change “The most easterly front, ranging from…” to “The most eastward front, extending from…”.
L587: Do you mean blue tubes for Fig 13b not green?
Lines 594-595: I do not see the need to describe the feature between 700 hPa and 500 hPa as it has no relevance to what you are focusing on (i.e. the low-level THETA_W feature). Please remove.
L647-648: “We find that cold frontal…”, I disagree with this sentence because you have not shown this. The description of the case study is not detailed enough to be certain of this reasoning (as mentioned in the major points).
Citation: https://doi.org/10.5194/wcd-2022-36-RC2 -
EC1: 'Comment on wcd-2022-36', Irina Rudeva, 08 Oct 2022
Dear Authors,
Thank you for submitting your manuscript to WCD. Unfortunately, based on the evaluation of two reviewers, I believe that the submitted version of the paper is not well suited for publishing in WCD. Your willingness to reach out to a wider audience is very much appreciated; however, both reviewers noted that the submitted manuscript has not provided significant advancement of our understanding of weather and climate dynamics, which is the scope of WCD.
Therefore, I recommend withdrawing the paper and re-evaluating it. If the main goal remains documenting the 3D tool and providing guidance on how it can be used by other researchers then I suggest following the reviewers’ advice on submitting the manuscript to GMD. If you choose to focus more on atmospheric dynamics facilitated by the 3D tool, then such study can be considered for WCD as a new submission. I believe that such paper may still achieve your goal to introduce the 3D frontal algorithm to the meteorological community.
If you decide to submit the paper to GMD, you are welcome to mention that the paper has been previously discussed in WCD and was recommended for submission to GMD. I’ll be happy to assist the new GMD editor should they contact me.
Please feel free to contact me if you need any further information.
Kind regards,
IrinaCitation: https://doi.org/10.5194/wcd-2022-36-EC1 -
AC1: 'Comment on wcd-2022-36', Andreas Beckert, 16 Nov 2022
Dear Irina Rudeva and dear Reviewers,
Thank you for providing your comments and feedback on our manuscript about the three-dimensional structure of fronts in mid-latitude weather systems. We recognize your concerns that our manuscript in the submitted version is not well suited for publishing within the scope of WCD. Our main goal remains documenting our generalized reimplementation of the Kern et al. (2019) method, to provide guidance to researchers on how to use 3-D front detection and visualization, and to demonstrate the potential of working with interactive 3-D visualization for analysis of atmospheric dynamics. We have hence decided to follow your recommendation to withdraw the manuscript from WCD and to resubmit to GMD. For this resubmission, we have revised the article according to the reviewer’s suggestions. We reply to the reviewer comments in detail in the attached PDF, so that future GMD reviewers can refer to the revisions already made.
Thank you very much for your efforts and kind regards,
Andreas Beckert and Marc Rautenhaus, on behalf of the author team
Interactive discussion
Status: closed
-
RC1: 'Comment on wcd-2022-36', Anonymous Referee #1, 18 Aug 2022
In the manuscript "The three-dimensional structure of fronts in mid-latitude weather systems as represented by numerical weather prediction models" document an implementation and slight adaptation of an existing 3-dimensional front surface detection algorithm in the context of the interactive analysis software Met.3D. The authors then showcase their implementation for two case studies of autumn/winter storms over Europe and relate their detected fronts to other (mostly very well established) meterological concepts.The manuscript is quite comprehensive and touches upon many aspects around mid-latitude cyclones and their fronts. With the wide variety of aspects covered, it however remains somewhat unclear what, in its core, this manuscript is about. Discussions generally remain superficial and don't add much new to the literature except further anecdotal support for otherwise already very well established meteorological concepts (i.e., WCBs, Shapiro-Keyser cyclone model). Because of this overall lack of direction and novelty in the discussions, the manuscript appears in its present form to be mainly an advertisement for Met.3D. I recommend the editor to reject this manuscript, and the authors to submit more focused and targeted analyses/discussions on any or all of the mentioned topics individually.
More specifically:
(1) The front surface detection seems to be only a minor modification/optimisation of the algorithm introduced and implemented in Kern et al. (2019). Judging from the illustrations in Kern et al. (2019), the algorithm was already then implemented in Met.3D. Yet, the algorithm is introduced and discussed here in as much detail as if it was new. Further, the authors "validate" well-established meteorological concepts such as the Shapiro-Keyser cyclone model and the spatial relation between WCBs and fronts using their visualisation and front detection algorithm. Given how successful these concepts have been over decades, I find this quite assuming. If these concepts had failed to show up in their analysis, I would much rather doubt the implementation and visualisation in question rather than these meteorological concepts. Now, given that everything looks as expected, I am unsure what to take away from the "validation" beyond that the algorithm and visualisation is working fine---and so much that had already been shown by Kern et al. (2019).
(2) The authors discuss briefly the best choice of thermodynamic variable for the front detection. This choice remains an subject of debate, and a new perspective on this choice could warrant another publication. This would however require considerable additonal analyses; based on only two case studies, the authors are not in a position to give general recommendations (as presently done in the summary and discussion section).
(3) Similarly, a front classification into humidity and temperature-dominanted fronts would most likely be worthwhile and well warrant a publication. But this aspect is discussed by far too superficially to justify the publication of the present manuscript.
(4) Similarly, the comparison of WCBs and frontal structures in parameterised-convection versus convection-resolving models is both timely and interesting. It would certainly warrant a publication of its own. But again, this aspect is discussed by far too superficially here.
I would very much encourage the authors to extend their analyses in particular on the topics in issues (3) and (4) and to submit manuscripts with in-depth analyses on these. The 3D front surface detections will surely certainly be helpful for either.
Finally a comment on the frequent use of 3D visualisations in the present manuscript. Besides the 3D visuals, I find the visuals and text to be clear. Met.3D is undoubtedly a powerful tool for the interactive and exploratory analysis of a meteorological dataset--the interactive demonstration in video supplement 3 shows that clearly. At the same time, as static images on (digital) paper, I don't find the shown 3D visualisations useful. Figure 3 in the manuscript is a good example: Everyone who has looked at weather charts before will be able to grasp the chart in panel (a) within fractions of a second and have a clear impression of the synoptic situation. This would still be true even if additional lines were plotted to depict the intersection of the frontal surface with different vertical levels. But it is not true for panel (b), where I remain unsure about the frontal structure and synoptic situation even after looking at the panel for minutes. The main cyclone core is not visible behind the front surfaces, so I need quite some mental energy to disentangle the occluded from the cold front (which some ambiguity remaining) to then infer with my meteorological intuition that there should be another cyclone core hidden somewhere---an inference that I couldn't be sure about without the verification in panel (a). Because, given only panel (b), may be the occluded front is actually a secondary cold front and the only cyclone core in the chart is the one close to Iceland, the one that is peaking out behind the occluded front? Essential aspects of the 3D structure can still be conveyed in 2D maps; I would find such maps much less ambiguous, easier to parse, and thus more suitable for a static publication.
Citation: https://doi.org/10.5194/wcd-2022-36-RC1 -
CC1: 'Reply on RC1', Marc Rautenhaus, 23 Aug 2022
Dear Reviewer,
Thank you for providing your comments. We recognize your concern that our manuscript provides too little new insight on dynamic meteorology to the literature, and we would like to use the possibility of this discussion to briefly clarify the goals of our paper. We will reply in more detail after the discussion period has ended.
There were two major objectives that we wanted to achieve with our work: (a) a reimplementation and generalization of the Kern et al. (2019) method such that it is robust when used with current model data and can handle additional filter options, and to document the tool and make it available to the community as an open-source release, alongside the paper (which the original Kern et al. method was not); and (b) provide guidance to researchers on how to use 3-D front detection and visualization by investigating suitable method parameters and by showing the potential of the method for meteorological analyses with selected examples.
Since many people in the meteorological community are not used to working with 3-D visualization, we consider it valuable to demonstrate its potential. We tried to do this using familiar examples that are easily related to conventional 2-D depictions (for example in Fig. 3), but show features that would not be easily seen in the conventional views. It is true that static 3-D images can require significant effort to interpret, but our aim was to help and encourage the reader to make this effort, and eventually to explore the interactive tool. It is helpful in this regard that it is now possible to publish animations along with a paper (as we have done).
We did consider submitting the article to GMD, which would be the usual place to publish descriptions of models and analysis tools, but we chose to submit to WCD in order to reach the audience that is most concerned with analyses of atmospheric dynamics and that would thus benefit from learning how to use 3-D visualization for their work. We agree that we picked up several aspects that would (and should!) require much more in-depth research and could be published on their own – that, however, was not the goal of this paper, and is something we see as future work that could be approached by using our method.
Kind regards,
Marc Rautenhaus, on behalf of the author team
Citation: https://doi.org/10.5194/wcd-2022-36-CC1
-
CC1: 'Reply on RC1', Marc Rautenhaus, 23 Aug 2022
-
RC2: 'Comment on wcd-2022-36', Anonymous Referee #2, 05 Oct 2022
Summary:
The paper by Beckert et al. documents the method, capabilities and calibration of a frontal identification algorithm that can be used to then produce 3D plots of those frontal structures. I found the paper to be written well with many interesting plots presented. Nonetheless, the paper does not provide significant advancement of an already existing method. Furthermore, while the case studies described in the paper are interesting and certainly do have the potential to be published in WCD (and I strongly encourage the authors to do this), their brief descriptions in this paper are not sufficient for WCD. I feel this paper would be better suited to GMD, for example, given the brief analyses of the case studies. I therefore do not recommend publication in WCD and that the paper should be passed on to GMD for consideration instead. I would happily re-review this paper from a GMD context if it were submitted there. I have provided some more major considerations for the authors below along with a list of minor points.
Major points:
The paper feels disjointed in its current setup. I feel that Section 3 should be re-distributed into Section 5 so that 3.1 gets mixed in with the introductory paragraph of section 5, 3.2 goes at the start of section 5.1 (5.1.1 and 5.1.2?) and section 3.3 does likewise in section 5.2 (5.2.1 and 5.2.2). There would then be a nice flow from introducing the case study and the plots would flow logically from “concept” to “analysis”. Currently there are methods in section 2, an introduction of the case studies in section 3, another methods section is given in section 4 and we then return to the case studies in section 5. As you can see, the paper “jumps” around a bit. Putting section 3 into section 5 and re-ordering the figures will make it flow much better instead of having to refer-back to section 3 from section 5. Furthermore, section 5 jumps around too – better to stick with the one case cyclone and focus in on several features associated with it i.e. make 5.3 part of 5.1 e.g. 5.1.1, 5.1.2 and 5.1.3 then push section 5.4 into 5.2 (5.2.1, 5.2.2 and 5.2.3).
Section 2.3 and Figure 2: The wording in the numbered list should match the plots and the plots should then be referred to in each of the points of the numbered list e.g. point 1 goes with Fig 2a, point 2 with Fig 2b etc. It seems that this Figure-numbered list relation does not hold true in all cases so the authors should either adjust the list or adjust the figure to make the two complimentary. There are also no descriptions of the panels in the Figure 2 caption so you should say “see Section 2.3 for a description of the panels a-h” in the figure caption.
Figure 3b: This is difficult to interpret and would probably be better if the fronts were colour coded to show the difference between warm, cold and occluded (red, blue and purple?). I just find the figure to have a lot of “green swirls” that really need to be separated visually to make the features stand out. Furthermore, the scale is too smooth to really show the location of the fronts in the vertical. Using different colours for the type of front or markedly different colours at each pressure level might make these (and all the other figures that use the green-white colour scaling for the pressure heights) clearer.
Figure 9 and Lines 475 – 490: This whole description is very difficult to see as both the paper and online plots are far too “busy”. Surely you can distinguish between the “fast ascending” and “slow ascending” trajectories and plot them separately. I would discard figures 9e and 9f and replace them with figures showing “fast” and “slow” ascending trajectories. You would then only need to slightly re-word lines 475-490 to account for this change. I feel that the whole paragraph would then read much better with the adjusted figure.
Figure 10: I’m not convinced this figure shows anything a 2D figure wouldn’t show. It is trying to do too much by overlaying everything on the same plot and so the details are lost. I would like to see this separated out into ~6 panels that show the 3D structure clearly from the best angle. I would also suggest removing the land and just focussing on the cyclone itself (maybe adding in isobars for cyclone-centric orientation). In its current format, it shows less than what a 2D plot shows – but has clear potential to be excellent if it were better focussed (I can see why you would want to see all of this in 3D, it just does not show up well). Figure A3 for cyclone Egon is actually clearer; however, A3 would also benefit from having a slightly more upright angle, removing the land and adding isobars.
Lines 554-568 and Figures 11, A4 and A5: I found this passage very difficult to read and follow. The 3D plots make things difficult to reconcile. I would suggest circling exactly where you want the reader to look instead of trying to describe it (lines 554-556). You could then say something simpler like, “There is a gap in the frontal surface between 700-600 hPa in the ECMWF data whereas the frontal surface is present in the COSMO simulation (circled in Figs 12a, b).” It just focuses the reader on the point you want them to look at. I also think it might be worth including Figure A5 in Figure 12 and even plotting the difference between the THETA_W fields between COSMO and ECMWF (and possibly likewise for THETA). The reason for that it that I’m not convinced by your “convection drives differences in the temperature gradient” argument. It is possible that the opposite is true e.g. the temperature gradient around 700 hPa is stronger in COSMO (i.e. simulated better), which then leads to the development of convection along the frontal zone. The front may have been going through frontogenesis and the convection is just the result of that. I therefore do not believe your description of this process is convincing enough to be certain of the process you describe. The analysis does not contain enough detail.
Minor points:
Figure A1: I cannot find any reference to this figure in the text. Can it be removed?
Figure A2: I think this can be removed if you adjust Figure 9 (you could even include the jet in figure 9).
L221-224: Sentence starting “The method uses…” is very long. Please split this in two.
L287: Change “a decay stages” to “a decay stage”.
L300: Change “UTC5.3” to “UTC”.
Figure 3 caption: remove the extra “)5.3” near the end of the third line.
Line 321: Add “on” before “17 January 2018”.
Line 324: Change “As a result of the cyclone, high wind speeds were registered…” to “ The cyclone caused high wind speeds…” to be more concise.
Line 328: Change “… this is a Shapiro… “ to “… this was a Shapiro…” as it happened in the past.
Figure 7: Please include the time and date for these plots. It helps for stopping the video in the relevant place (I can see this information is in Figure 2 but should also be here). Also, the caption is unnecessarily detailed as you say most of it in the text. The caption only needs the description of the figures, not the explanation about what each step does (as you explain in the text). Please trim the caption down to make it easier to read.
Figure 7c: Maybe I’m missing something, but it looks like the feature is still in the plot under the blue circle (unlike in 7e where the northern feature disappears). Is this figure correct? Additional – Line 429 (related to Fig 7c comment) – OK I see this more now, but it is very subtle. I would focus that blue circle in to EXACTLY where you want the reader to look.
Figures 7d: it is very hard to work out where this cross section is taken without looking at 6f as there’s too much shading. If you put a line on Figure 6c to show the location of the cross-section then that would help make it clearer (then refer to it in the caption).
Line 427: should it be “of the filter” instead of “of filter”?
Figure 9: The plots get very ‘busy’ with time, especially figure (f). If you could show where the viewpoint 2 and viewpoint 3 cross sections are located in figure (a) for example, then it’d help. If the land masses weren’t blocked out so much in / at the very periphery of figure (f) then that would help.
Line 475: “north-easterly direction” should really be “north-eastward direction”.
Fig. 11: For clarity, it might be worth making the THETA_W scale blue-red so that the grey shading shows up better. The grey-blue end could be confused with the horizontal gradient shading.
L585: Change “The most easterly front, ranging from…” to “The most eastward front, extending from…”.
L587: Do you mean blue tubes for Fig 13b not green?
Lines 594-595: I do not see the need to describe the feature between 700 hPa and 500 hPa as it has no relevance to what you are focusing on (i.e. the low-level THETA_W feature). Please remove.
L647-648: “We find that cold frontal…”, I disagree with this sentence because you have not shown this. The description of the case study is not detailed enough to be certain of this reasoning (as mentioned in the major points).
Citation: https://doi.org/10.5194/wcd-2022-36-RC2 -
EC1: 'Comment on wcd-2022-36', Irina Rudeva, 08 Oct 2022
Dear Authors,
Thank you for submitting your manuscript to WCD. Unfortunately, based on the evaluation of two reviewers, I believe that the submitted version of the paper is not well suited for publishing in WCD. Your willingness to reach out to a wider audience is very much appreciated; however, both reviewers noted that the submitted manuscript has not provided significant advancement of our understanding of weather and climate dynamics, which is the scope of WCD.
Therefore, I recommend withdrawing the paper and re-evaluating it. If the main goal remains documenting the 3D tool and providing guidance on how it can be used by other researchers then I suggest following the reviewers’ advice on submitting the manuscript to GMD. If you choose to focus more on atmospheric dynamics facilitated by the 3D tool, then such study can be considered for WCD as a new submission. I believe that such paper may still achieve your goal to introduce the 3D frontal algorithm to the meteorological community.
If you decide to submit the paper to GMD, you are welcome to mention that the paper has been previously discussed in WCD and was recommended for submission to GMD. I’ll be happy to assist the new GMD editor should they contact me.
Please feel free to contact me if you need any further information.
Kind regards,
IrinaCitation: https://doi.org/10.5194/wcd-2022-36-EC1 -
AC1: 'Comment on wcd-2022-36', Andreas Beckert, 16 Nov 2022
Dear Irina Rudeva and dear Reviewers,
Thank you for providing your comments and feedback on our manuscript about the three-dimensional structure of fronts in mid-latitude weather systems. We recognize your concerns that our manuscript in the submitted version is not well suited for publishing within the scope of WCD. Our main goal remains documenting our generalized reimplementation of the Kern et al. (2019) method, to provide guidance to researchers on how to use 3-D front detection and visualization, and to demonstrate the potential of working with interactive 3-D visualization for analysis of atmospheric dynamics. We have hence decided to follow your recommendation to withdraw the manuscript from WCD and to resubmit to GMD. For this resubmission, we have revised the article according to the reviewer’s suggestions. We reply to the reviewer comments in detail in the attached PDF, so that future GMD reviewers can refer to the revisions already made.
Thank you very much for your efforts and kind regards,
Andreas Beckert and Marc Rautenhaus, on behalf of the author team
Video supplement
Development of 3-D frontal structures, jet stream and WCB trajectories of Vladiana Andreas A. Beckert, Lea Eisenstein, Annika Oertel, Tim Hewson, George C. Craig, Marc Rautenhaus https://doi.org/https://doi.org/10.5446/57570
Comparison of objectively 725 detected 3-D fronts in wet-bulb potential temperature and potential temperature Andreas A. Beckert, Lea Eisenstein, Annika Oertel, Tim Hewson, George C. Craig, Marc Rautenhaus https://doi.org/https://doi.org/10.5446/57600
Interactive front analysis of storm Friederike using the open-source meteorological 3-D visualization framework "Met. 3D" Andreas A. Beckert, Lea Eisenstein, Annika Oertel, Tim Hewson, George C. Craig, Marc Rautenhaus https://doi.org/https://doi.org/10.5446/57944
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 816 | 312 | 42 | 1,170 | 27 | 27 |
- HTML: 816
- PDF: 312
- XML: 42
- Total: 1,170
- BibTeX: 27
- EndNote: 27
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
