the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 23 Feb 2024
Storm damage beyond wind speed – Impacts of wind characteristics and other meteorological factors on tree fall along railway lines
Abstract. Strong winter wind storms can lead to billions in forestry losses, disrupt train services and amount to millions of Euro spend on vegetation management alongside the German railway system. Therefore, understanding the link between tree fall and wind is crucial.
Existing tree fall studies often emphasize tree and soil factors more than meteorology. Using a dataset from Deutsche Bahn (2017–2021) and meteorological data from ERA5 reanalysis and RADOLAN radar, we employed stepwise model selection to build a logistic regression model predicting the risk of a tree falling on a railway line in a 31 km grid cell.
While daily maximum gust speed is the strongest risk factor, we also found that daily duration of strong wind speeds, precipitation, soil water volume, air density and the precipitation sum of the previous year increase tree fall risk. A high daily gust factor decreases the risk. Using interaction terms between maximum gust speed and duration of strong wind speeds as well as gust factor improves the model performance. Therefore, our findings suggest that high and prolonged wind speeds, especially in combination with wet conditions (high precipitation and high soil moisture) and a high air density, increase tree fall risk. Incorporating meteorological parameters linked to local climatological conditions (through anomalies or in relation to local percentiles) improved the model accuracy. This indicates the importance of taking tree adaptation to the environment into account.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Natural Hazards and Earth System Sciences.
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 preprint. The responsibility to include appropriate place names lies with the authors.-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(747 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(747 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-120', Anonymous Referee #1, 11 Apr 2024
This study assess the impact of several meteorological factors on tree fall risk along the German railway network. The results contribute to the understanding of the relative importance of different factors on the tree fall risk in Germany and they can be extrapolated by some extent to other countries as well, especially to those having similar climatic and environmental conditions. The manuscript has a clear structure and it is well written. However, I have a couple of general and a few more specific minor comments.
General comments
1. From line 70 onwards it is explained how much Deutche Bahn has recently spend money on vegetation management, but the number of tree fall events causing disruption in the railway service has remained high. The paragraph ends with justification that this kind of study can add value to the management of vegetation along the transportation routes. Can you comment your results from this viewpoint in the Discussion section? In which extent do you expect, that these results could be implemented to mitigate the wind-related damage?
2. I understand that Figure 4 presents the most relevant results of this study. Here, one model parameter is varied in each plot while the others are fixed to a certain value. How these fixed values were defined? Can you also perhaps slightly elaborate, how the interaction of different variables has been taken into account here? In general, the results presented in Fig. 4 seem as expected with the exception of the impact of wind direction on tree fall risk. As noted on lines 286-288, southeasterly winds seem to produce the smallest and northwesterly the highest risk. The impact of wind direction is furthermore discussed on lines 343-348 where the authors note that trees tend to adapt to local wind direction. As the predominant wind direction in Germany has probably some western component, it is interesting and unexpected for me that the northwesterly winds would cause the highest risk. Do you have any idea what could be the reason for this result? Could it even be just random noise due to the fact that very local features might dominate here. To return to interaction terms, a notable feature in Fig. 5 is that the model with no interaction terms for the gust factor yields generally higher tree fall risk than the model with interaction terms included. Can you comment on that?
Specific comments
1. Line 51. The term "gust factor" is probably familiar for the readers with background from storm damage studies but not for all other readers. You could shortly explain it by a few words here, for example like this "A high daily gust factor (i.e., the ratio of gust wind speed to mean wind speed) decreases the risk" etc
2. Line 64. I assume that here should read "losses" instead of "loses"
3. Line 80. Please check this sentence, as it seems like some words are accidentally missing.
4. Line 138. Should it read "Monthly percentages" instead of "Yearly percentage" in the caption of figure 2?
5. Line 145. I would estimate from the figure that approximately 28% of tree fall events occurred in December, January and February in total, so the claim that "the majority" of the events occurred within these months is clearly incorrect. Moreover, December is in fact the month with the third smallest number of tree fall events in the data, and the three months with the largest number of tree fall events are January, February and March. However, even these three months are far from producing "the majority" of the events.
6. Line 149. Is there a reason why you use ERA5 data instead of a more finer resolution ERA5-Land data?
7. Line 172. I assume that here should read "where" instead of "were"
8. Line 242. Did you use a two-tailed z-test or t-test?
9. Line 250. It is stated here that explanations for the different predictor abbreviations are given in Table Fehler: Verweis nicht gefunden. This apparently German table is referred several times thereafter. What table is this and where it can be found?
10. Line 305. I assume that here should read "limitations" instead of "limiation"
11. Line 313. I assume that the word "and" is missing between the words "risk" and "improve"
12. Line 415. I assume that here should read "The" instead of "Teh"
13. Line 436. I would suggest to rephrase the caption of Figure 7 as follows: "Comparison of interaction effect. Gusty day: dur90 = 2 and gf = 5; sustained day: dur90 = 12 and gf = 2..."
Citation: https://doi.org/10.5194/egusphere-2024-120-RC1 - AC1: 'Reply on RC1', Rike Lorenz, 28 Jun 2024
-
RC2: 'Comments on egusphere-2024-120', Anonymous Referee #2, 01 May 2024
- AC2: 'Reply on RC2', Rike Lorenz, 28 Jun 2024
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-120', Anonymous Referee #1, 11 Apr 2024
This study assess the impact of several meteorological factors on tree fall risk along the German railway network. The results contribute to the understanding of the relative importance of different factors on the tree fall risk in Germany and they can be extrapolated by some extent to other countries as well, especially to those having similar climatic and environmental conditions. The manuscript has a clear structure and it is well written. However, I have a couple of general and a few more specific minor comments.
General comments
1. From line 70 onwards it is explained how much Deutche Bahn has recently spend money on vegetation management, but the number of tree fall events causing disruption in the railway service has remained high. The paragraph ends with justification that this kind of study can add value to the management of vegetation along the transportation routes. Can you comment your results from this viewpoint in the Discussion section? In which extent do you expect, that these results could be implemented to mitigate the wind-related damage?
2. I understand that Figure 4 presents the most relevant results of this study. Here, one model parameter is varied in each plot while the others are fixed to a certain value. How these fixed values were defined? Can you also perhaps slightly elaborate, how the interaction of different variables has been taken into account here? In general, the results presented in Fig. 4 seem as expected with the exception of the impact of wind direction on tree fall risk. As noted on lines 286-288, southeasterly winds seem to produce the smallest and northwesterly the highest risk. The impact of wind direction is furthermore discussed on lines 343-348 where the authors note that trees tend to adapt to local wind direction. As the predominant wind direction in Germany has probably some western component, it is interesting and unexpected for me that the northwesterly winds would cause the highest risk. Do you have any idea what could be the reason for this result? Could it even be just random noise due to the fact that very local features might dominate here. To return to interaction terms, a notable feature in Fig. 5 is that the model with no interaction terms for the gust factor yields generally higher tree fall risk than the model with interaction terms included. Can you comment on that?
Specific comments
1. Line 51. The term "gust factor" is probably familiar for the readers with background from storm damage studies but not for all other readers. You could shortly explain it by a few words here, for example like this "A high daily gust factor (i.e., the ratio of gust wind speed to mean wind speed) decreases the risk" etc
2. Line 64. I assume that here should read "losses" instead of "loses"
3. Line 80. Please check this sentence, as it seems like some words are accidentally missing.
4. Line 138. Should it read "Monthly percentages" instead of "Yearly percentage" in the caption of figure 2?
5. Line 145. I would estimate from the figure that approximately 28% of tree fall events occurred in December, January and February in total, so the claim that "the majority" of the events occurred within these months is clearly incorrect. Moreover, December is in fact the month with the third smallest number of tree fall events in the data, and the three months with the largest number of tree fall events are January, February and March. However, even these three months are far from producing "the majority" of the events.
6. Line 149. Is there a reason why you use ERA5 data instead of a more finer resolution ERA5-Land data?
7. Line 172. I assume that here should read "where" instead of "were"
8. Line 242. Did you use a two-tailed z-test or t-test?
9. Line 250. It is stated here that explanations for the different predictor abbreviations are given in Table Fehler: Verweis nicht gefunden. This apparently German table is referred several times thereafter. What table is this and where it can be found?
10. Line 305. I assume that here should read "limitations" instead of "limiation"
11. Line 313. I assume that the word "and" is missing between the words "risk" and "improve"
12. Line 415. I assume that here should read "The" instead of "Teh"
13. Line 436. I would suggest to rephrase the caption of Figure 7 as follows: "Comparison of interaction effect. Gusty day: dur90 = 2 and gf = 5; sustained day: dur90 = 12 and gf = 2..."
Citation: https://doi.org/10.5194/egusphere-2024-120-RC1 - AC1: 'Reply on RC1', Rike Lorenz, 28 Jun 2024
-
RC2: 'Comments on egusphere-2024-120', Anonymous Referee #2, 01 May 2024
- AC2: 'Reply on RC2', Rike Lorenz, 28 Jun 2024
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 595 | 195 | 38 | 828 | 38 | 42 |
- HTML: 595
- PDF: 195
- XML: 38
- Total: 828
- BibTeX: 38
- EndNote: 42
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Rike Lorenz
Nico Becker
Barry Gardiner
Uwe Ulbrich
Marc Hanewinkel
Schmitz Benjamin
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(747 KB) - Metadata XML