the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 Aug 2024
Hydro-geomorphological modelling of leaky wooden dam efficacy from reach to catchment scale with CAESAR-Lisflood 1.9j
Abstract. Leaky wooden dams are woody structures installed in headwater streams that aim to reduce downstream flood risk through increasing in-channel roughness and decreasing river longitudinal connectivity in order to desynchronise flood peaks within catchments. Hydrological modelling of these structures omits sediment transport processes since the impact of these processes on flow routing is considered negligible in comparison to in-stream hydraulics. Such processes are also excluded on the grounds of computational expense. Here we present a study that advances our ability to model leaky wooden dams through a roughness-based representation in the landscape evolution model CAESAR-Lisflood, introducing a flexible and representative approach to simulating the impact of leaky wooden dams on reach and broader catchment-scale processes. The hydrological and geomorphological sensitivity of the model is tested against grid resolution as well as a variability in key parameters such as leaky dam gap size and roughness. The influence of these parameters are also tested in isolation from grid resolution, whilst evaluating the impact of simulating sediment transport on computational expense, model domain outputs and internal geomorphological evolution. The findings show that simulating sediment transport increased the volume of water stored in the test reach by up to an order of magnitude whilst reducing discharge by up to 31 % during a storm event. We demonstrate how this is due to the leaky dam acting to induce geomorphic change and thus increasing channel roughness. When considering larger grid resolutions, however, our results show that care must be due to overestimations of localised scour and deposition in the model and that behavioural approaches should be adopted when using CAESAR-Lisflood in the absence of robust empirical validation data.
-
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
(1657 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1657 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-2132', Anonymous Referee #1, 27 Sep 2024
General summary
This paper presents and tests a new extension of the CAESAR-Lisflood landscape evolution model, that enables hydro-geomorphological modelling of in-stream wooden leaky barriers. The paper is interesting and well presented. It is also novel by being the first model published that considers both geomorphic and hydrological processes and the interactions between them which is of great importance for practitioners of natural flood management (NFM). The topic is also relevant to the EGU Sphere Geoscientific Model Development journal.
The paper has potential as a worthy contribution of a new model tool with some interesting insights into processes and responses that could have real world implications. However, the paper would benefit from some improvements. Firstly, a more up to date review and comparison with the topic of the hydro-geomorphological functioning of leaky barriers is needed. Much progress has been made recently especially with regards to modelling the hydraulic and hydrological effects of leaky barriers but these studies are not included and may help to improve the insights given in the discussion. Secondly, a clearer research context to help justify the paper with better stated aims are needed. For example, at the moment there is little exploration of why it is important to consider geomorphic processes and what the specific geomorphic aims of this study are. Thirdly, more consideration of the validity of predictions and applicability of the tool in the real world is required. The authors state that the model is heuristic, and it is unvalidated which means its reliability for making worthwhile predictions is unknown. However, at the same time the authors advocate that the model is useful for practitioners but given the uncertainty of predictions perhaps this is not a valid viewpoint to take.
Specific points
L24 State recurrence interval or annual exceedance probability of storm event to give an idea of its magnitude. Also state for what catchment size.
L32-35 Low risk yes but potential for structure washout and displacement perhaps should be acknowledged.
L36 What is meant by ‘river engineering’? Seems like a vague term to use.
L46-48 More nuance and specific reference to sources that back up these claims is needed. These benefits are often perceived and have not been quantified comprehensively or assessed.
L50-62 A more up to date and accurate reflection on the knowledge gaps and recent advancements on understanding is needed. For example, see the work of Follett and Hankin (2022) and Geertsema et al., (2020) on approaches to model the hydraulic effects of in channel large wood interventions. Also, the work of Lo et al. (2022) gives field based observations on the geomorphic effects of leaky barriers and the work of Norbury et al. (2021) and Van Leeuwan et al. (2024) measure the hydrological effects of structures using field data. Flume based studies on hydro-geomorphic responses are also potentially useful to synthesise and compare with (e.g. Schalko et al., 2019; Muhawenimana et al., 2021).
L64-65 Clear statements on potential geomorphic processes, feedbacks and importance are needed. At the moment the importance of considering geomorphic processes in models isn’t coming through. For example, the aforementioned studies have given observations on the patterns of erosion and deposition in relation to structures that could have hydraulic effects.
L81-82 First part of this statement is not true. See references made above on progress made on representing the leakiness and lower gap effects of leaky barriers.
L82 What is meant by a ‘prototype real world location’?
L84 A new paragraph stating a clearer and more elaborated list of aims and, or hypotheses is needed. This would help to give the paper more structure and purpose.
L159-160 ‘Upwinding’ and ‘upwind’ are strange terms to use. Consider rewording?
L190 This approach of scaling n seems sensible but perhaps a caveat is needed here given that it isn’t based on an empirical relationship as it stands.
L210-214 More details on the prototype reach is needed in the paper rather than citing the thesis. The channel slope used seems quite low for a headwater stream where leaky barriers are typically used. Would it be worth testing the model over a range of slopes to see the effect? Why were different DEM resolutions tested? Certainly important but little context or purpose on this is given.
L221 What is meant by sediment types?
L238 Again like testing a range of slopes, it would be interesting to see the effects of a range of different flow events and perhaps would provide more insight into hydrogeomorphic effects of leaky barriers than testing a range of different DEM cell sizes.
L246 Why was an nmax value of 0.16 used?
Figure 6 To make it clear, mark on erosion and deposition labels. I.e. negative values show deposition and positive values show erosion which may be counter-intuitive at first glance. Similar remarks can be made for Figure 7 in relation to changes in water storage.
L343-344 The finding that increasing LD gap size resulted in less sediment being lost seems counter-intuitive.
L455-459 Perhaps a more important point is how valid is this model. Whilst there may be some value in using it in an heuristic manner for scenario testing, validation using empirical observations of changes in discharge and morphology would strengthen the value of this model and give practitioners more confidence in using it. This is a knowledge gap that should be stated. The word ‘behaviouralist’ seems a strange one to use.
L477-480 Yes this is a fair statement but there simply isn’t usually the data to calibrate or validate sediment transport in the real world unfortunately and processes are notoriously difficult to predict for different hydrological events. In contrast, observations of channel hydraulics and hydrology are more available o accessible to enable model assessment.
L491 Could the model be used to assess other types of NFM structure or in channel wood? Another aspect deserving further study in future?
Associated files with the paper. Appears that the model has been uploaded for both weblinks on Zenodo. Could not see the datasets.
Typo errors
L467-468 This sentence does not make sense.
References
Follett, E. and Hankin, B., 2022. Investigation of effect of logjam series for varying channel and barrier physical properties using a sparse input data 1D network model. Environmental Modelling & Software, 158: 105543.
Geertsema, T.J., Torfs, P.J., Eekhout, J.P., Teuling, A.J. and Hoitink, A.J., 2020. Wood‐induced backwater effects in lowland streams. River Research and Applications, 36(7): 1171-1182.
Lo, H.W., van Leeuwen, Z., Klaar, M., Woulds, C. and Smith, M., 2022. Geomorphic effects of natural flood management woody dams in upland streams. River Research and Applications, 38(10): 1787-1802.
Muhawenimana, V., Wilson, C.A., Nefjodova, J. and Cable, J., 2021. Flood attenuation hydraulics of channel-spanning leaky barriers. Journal of Hydrology, 596: 125731.
Norbury, M., Phillips, H., Macdonald, N., Brown, D., Boothroyd, R., Wilson, C., Quinn, P. and Shaw, D., 2021. Quantifying the hydrological implications of pre- and post-installation willowed engineered log jams in the Pennine Uplands, NW England. Journal of Hydrology, 603: 126855.
Schalko, I., Lageder, C., Schmocker, L., Weitbrecht, V. and Boes, R.M., 2019. Laboratory Flume Experiments on the Formation of Spanwise Large Wood Accumulations: Part II—Effect on local scour. Water Resources Research, 55(6): 4871-4885.
van Leeuwen, Z.R., Klaar, M.J., Smith, M.W. and Brown, L.E., 2024. Quantifying the natural flood management potential of leaky dams in upland catchments, Part II: Leaky dam impacts on flood peak magnitude. Journal of Hydrology, 628: 130449.
Citation: https://doi.org/10.5194/egusphere-2024-2132-RC1 -
RC2: 'Comment on egusphere-2024-2132', Paul Quinn, 06 Nov 2024
I would first like to congratulate Reviewer 1 for a very detailed analysis of the paper and that I fully agree with these recommendations. So, I will concentrate on general matters.
I think more papers are needed including more general reviews such as Lashford et al., 2022 and Quinn et al.,, 2022. As they pose important design concerns. Which is what my main discussion point is.
Overall, a very good paper, it is clear and it covers the scope (as stated) very well. The issue of Manning n having an impact on flow is the most important part of this paper. Perhaps plant some vegetation on the banks?
The most concerning issue, that I think can be easily solved, is the assumption that leaky dams with a rectangular lower slot is the only design of leaky dam of concern. There are many leaky dam designs. Designs that allow a dynamic alteration of discharge with flow magnitude/depth being the obvious example. What if the leaky dam has a vertical slot or multiple pipes at multiple depths? So, you must make it clear that you are restricting your analysis to just this design and it still tell us interesting things.
Your choice of storm and flows need clarification. O.1 RI rain storm or flow RI?
Where on the network is your reach? What is the catchment area entering the top end of the reach? This is important. You have said that you are testing sensitivity, BUT you have not tested sensitivity to changing inflow. LDs are drowned out quite quickly so if the inflow increases the effectiveness goes down quickly. Or are you stating another design restraint of your study that you assume LDs are only at the upper most part of headwater e.g. less than 1km2? In reality LDs can be effective at larger scales if designed correctly. So be clear.
If you have assumed no storage is created being the LD, then you have missed the key design concept that LDs work best if they can engage larger storage volumes. Being constrained within a channel is the worst design.
So just broaden your discussion and conclusion.
Citation: https://doi.org/10.5194/egusphere-2024-2132-RC2 -
RC3: 'Comment on egusphere-2024-2132', Anonymous Referee #3, 18 Nov 2024
The Landscape Evolution Model CAESARLisflood 1.9j is employed to simulate leaky wooden dams at reach scale. A roughness-based approach is followed, and the simulation of sediment transport is also included, which represent a significant improvement with respect to current simulation practice.
A theoretical case study is proposed, and the Author systematically analyse the results by combining the variation of several parameters, from cell size, to roughness, to lower gap size, observing the variation of hydraulic and geomorphological results. One LD is represented in a straight channel.
The contribution is certainly interesting, especially for its effort of understanding the dynamic behaviours of LD.
However, the approach is strongly theoretical, with no connection to field data, making it an interesting attempt, but still hardly applicable. At least, this limitation should be clearly discussed.
In addition, some inaccuracies are present in the paper, like the absence of the results for certain gap size (not clearly justified), the absence of figures related to the case study domain, and a very synthetized presentation of the results, which seems under “discussion” form rather than result analysis. For example, why not to show the actual discharge hydrographs, rather than showing immediately a discharge variation?
The methodology relies on the application of an evolution model, and, as said, the absence of a clear calibration is the main limitation of the selected approach.
Despite having employed different model resolution, the effect of the resolution is not discussed, possibly leading to results inaccuracies. For example, while erosion and sediment areas along the channel are similar, the effect on the variation of the discharge is different for different resolution. This aspect is only addressed in the final discussion, but I think it should be analysed with higher precision also in the results.
Please find below my detailed comments:
Lines 133-135: “This method is straightforward to apply within a model domain, as specific cells can be identified to place the LD in combination with other roughness variables such as in-channel or floodplain boundary roughness”. How does the model identify LD cells with respect to riverbed or floodplain cells? Is the cell dimension modified for rover or floodplain?
Figure 2: the figure shows a square with one LD, and either vertical (b) or horizontal (c) flow direction. However, I guess that the model allows also horizontal + vertical flow, in case of transversal connectivity (flooding the areas upstream of the LD). I think that the figure should highlight the possibility of such a behaviour. In addition, connected to the previous comment, the cell edge that represents the LD is representative of the physical dimension of the LD?
Line 190-191: what do the Authors mean with “If the LD is overtopped, there is less of the LD in contact with the water column, therefore 𝑛 is reduced.”? If the LD is overtopped, the water level is higher than htop. So, the entire LD will be in contact with the water column, which however will have additional area not in contact with LD, thus reducing the BR. Is this the reason why the roughness coefficient is reduced? Please adjust the sentence.
Lines 190-192: These sentences appear reasonable but highly qualitative. Was any calibration performed to justify the variation of that roughness behaviour?
Paragraph 3.1: A figure representing the DEM and the different resolution would help the reader in figuring out tests configuration.
Figure 3: Please also include the b-axis sketch (or a description in the text), to ensure the proper interpretation from readers that may be not confident with the grain-size analysis.
Paragraph 3.3: the time-step of the simulations is not reported.
Figure 4: To ease the readability of the figure, why not to use the SI units fo measurements also for time (seconds or hours, instead of minutes x 104)? I also believe that showing the output discharge hydrograph, for at least one of the tests, would help in clarifying the discussion,
Lines 271-273: the Authors say: “For all sediment transport enabled experiments, the falling limb of the storm and remainder of the simulation time shows up to 25% deviation from the baseline scenario due to 𝑄 becoming out of phase with the baseline.”. In Figure 4, C and D, a variation of the discharge is observed, the values od ΔQ being positive or negative depending on the cells size. Positive values reach 25% for DEMT4 and reached -10% for DEMT1. Which of the situations is reflected in the Authors’ sentence? I think that here you should further discuss the effect of the cells size in providing attenuation/incrementation of the simulated discharges.
Line 280: QS was not introduced before. Please, define it clearly.
Line 310: The Authros wrote: “the second most cell downstream of the LD”. I suggest rephrasing as follows: “the second cell most downstream of the LD”.
Figure 6: The amount of change appear strongly dependant on the cells size. Please add a comment about that.
Lines 318-319: “a 2 m DEM with a two-pixel wide channel (4 m)”. Do the Authors refer to DEMT2? If yes, please use the same acronym, to be consistent within the paper. If not, please explain better.
Paragraph 4.3.1: Is only no-gap simulation discussed? If yes, please include general comments also for other gaps. In addition, why is Figure 7 not showing 0.3 and 0.4 m gaps? They should be discussed too, since they are mentioned.
Line 338: the Authors say: “Sediment transport was found not to be influenced by the installation of the LD”. However, in Figure 8 a variation curve in the sediment yield is shown. Please consider modifying this sentence. Probably the variation is not significant nor clear, but the sediment yield appears to be at least slightly different from the baseline one. In addition, discuss also the results for 0.3 and 0.4 m gaps.
Figure 8: why are subfigures A’, B’ and C’ displayed? They are not commented in the text.
Lines 342-343: Please discuss why with a larger gap sediment storage is observed.
Figure 9: Also here, 0.3 and 0.4 m gaps are not shown.
Lines 374-376: the Authros suggest that: “CL can therefore be used to aid identification of the ideal locations for LDs throughout a given reach to achieve the desired behavioural response, such as increased flow attenuation or enhanced geomorphic diversity.” However, the approach here proposed only shows the potentiality of the model. In a real-case scenario, will the peak reduction, the erosion and deposition occur according to the model predictions. In my opinion, the approach is really interesting but, before being employed for NFM design, it needs to be calibrated and validate on different case-studies. This does not mean reduce the importance of the contribution but helps in identifying its limitations and stress the requirements of future efforts.
Lines 388-389: “In addition, as CL utilises a regular raster grid, the required resolution can be adapted based on user requirements as well as computational resource availability”.
Discussion: when discussing about computational burden, please consider including tables with the following data: cell resolution, number of cells, computational time. This will help the paper to be clearer on the effect of grid size on computational burden.
Line 469: I think it should be ”reduced” and no “reduce”; or “can reduce”?
Line 488: “which could further the understanding”. A verb is probably missing.
Citation: https://doi.org/10.5194/egusphere-2024-2132-RC3 -
AC1: 'Comment on egusphere-2024-2132', Josh Wolstenholme, 18 Dec 2024
Dear Editor and Reviewers,
Thank you for your time and effort in providing detailed comments and insightful suggestions, which have significantly improved this manuscript. Please find attached the detailed line-by-line responses for your consideration.
Best regards,
Josh Wolstenholme
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-2132', Anonymous Referee #1, 27 Sep 2024
General summary
This paper presents and tests a new extension of the CAESAR-Lisflood landscape evolution model, that enables hydro-geomorphological modelling of in-stream wooden leaky barriers. The paper is interesting and well presented. It is also novel by being the first model published that considers both geomorphic and hydrological processes and the interactions between them which is of great importance for practitioners of natural flood management (NFM). The topic is also relevant to the EGU Sphere Geoscientific Model Development journal.
The paper has potential as a worthy contribution of a new model tool with some interesting insights into processes and responses that could have real world implications. However, the paper would benefit from some improvements. Firstly, a more up to date review and comparison with the topic of the hydro-geomorphological functioning of leaky barriers is needed. Much progress has been made recently especially with regards to modelling the hydraulic and hydrological effects of leaky barriers but these studies are not included and may help to improve the insights given in the discussion. Secondly, a clearer research context to help justify the paper with better stated aims are needed. For example, at the moment there is little exploration of why it is important to consider geomorphic processes and what the specific geomorphic aims of this study are. Thirdly, more consideration of the validity of predictions and applicability of the tool in the real world is required. The authors state that the model is heuristic, and it is unvalidated which means its reliability for making worthwhile predictions is unknown. However, at the same time the authors advocate that the model is useful for practitioners but given the uncertainty of predictions perhaps this is not a valid viewpoint to take.
Specific points
L24 State recurrence interval or annual exceedance probability of storm event to give an idea of its magnitude. Also state for what catchment size.
L32-35 Low risk yes but potential for structure washout and displacement perhaps should be acknowledged.
L36 What is meant by ‘river engineering’? Seems like a vague term to use.
L46-48 More nuance and specific reference to sources that back up these claims is needed. These benefits are often perceived and have not been quantified comprehensively or assessed.
L50-62 A more up to date and accurate reflection on the knowledge gaps and recent advancements on understanding is needed. For example, see the work of Follett and Hankin (2022) and Geertsema et al., (2020) on approaches to model the hydraulic effects of in channel large wood interventions. Also, the work of Lo et al. (2022) gives field based observations on the geomorphic effects of leaky barriers and the work of Norbury et al. (2021) and Van Leeuwan et al. (2024) measure the hydrological effects of structures using field data. Flume based studies on hydro-geomorphic responses are also potentially useful to synthesise and compare with (e.g. Schalko et al., 2019; Muhawenimana et al., 2021).
L64-65 Clear statements on potential geomorphic processes, feedbacks and importance are needed. At the moment the importance of considering geomorphic processes in models isn’t coming through. For example, the aforementioned studies have given observations on the patterns of erosion and deposition in relation to structures that could have hydraulic effects.
L81-82 First part of this statement is not true. See references made above on progress made on representing the leakiness and lower gap effects of leaky barriers.
L82 What is meant by a ‘prototype real world location’?
L84 A new paragraph stating a clearer and more elaborated list of aims and, or hypotheses is needed. This would help to give the paper more structure and purpose.
L159-160 ‘Upwinding’ and ‘upwind’ are strange terms to use. Consider rewording?
L190 This approach of scaling n seems sensible but perhaps a caveat is needed here given that it isn’t based on an empirical relationship as it stands.
L210-214 More details on the prototype reach is needed in the paper rather than citing the thesis. The channel slope used seems quite low for a headwater stream where leaky barriers are typically used. Would it be worth testing the model over a range of slopes to see the effect? Why were different DEM resolutions tested? Certainly important but little context or purpose on this is given.
L221 What is meant by sediment types?
L238 Again like testing a range of slopes, it would be interesting to see the effects of a range of different flow events and perhaps would provide more insight into hydrogeomorphic effects of leaky barriers than testing a range of different DEM cell sizes.
L246 Why was an nmax value of 0.16 used?
Figure 6 To make it clear, mark on erosion and deposition labels. I.e. negative values show deposition and positive values show erosion which may be counter-intuitive at first glance. Similar remarks can be made for Figure 7 in relation to changes in water storage.
L343-344 The finding that increasing LD gap size resulted in less sediment being lost seems counter-intuitive.
L455-459 Perhaps a more important point is how valid is this model. Whilst there may be some value in using it in an heuristic manner for scenario testing, validation using empirical observations of changes in discharge and morphology would strengthen the value of this model and give practitioners more confidence in using it. This is a knowledge gap that should be stated. The word ‘behaviouralist’ seems a strange one to use.
L477-480 Yes this is a fair statement but there simply isn’t usually the data to calibrate or validate sediment transport in the real world unfortunately and processes are notoriously difficult to predict for different hydrological events. In contrast, observations of channel hydraulics and hydrology are more available o accessible to enable model assessment.
L491 Could the model be used to assess other types of NFM structure or in channel wood? Another aspect deserving further study in future?
Associated files with the paper. Appears that the model has been uploaded for both weblinks on Zenodo. Could not see the datasets.
Typo errors
L467-468 This sentence does not make sense.
References
Follett, E. and Hankin, B., 2022. Investigation of effect of logjam series for varying channel and barrier physical properties using a sparse input data 1D network model. Environmental Modelling & Software, 158: 105543.
Geertsema, T.J., Torfs, P.J., Eekhout, J.P., Teuling, A.J. and Hoitink, A.J., 2020. Wood‐induced backwater effects in lowland streams. River Research and Applications, 36(7): 1171-1182.
Lo, H.W., van Leeuwen, Z., Klaar, M., Woulds, C. and Smith, M., 2022. Geomorphic effects of natural flood management woody dams in upland streams. River Research and Applications, 38(10): 1787-1802.
Muhawenimana, V., Wilson, C.A., Nefjodova, J. and Cable, J., 2021. Flood attenuation hydraulics of channel-spanning leaky barriers. Journal of Hydrology, 596: 125731.
Norbury, M., Phillips, H., Macdonald, N., Brown, D., Boothroyd, R., Wilson, C., Quinn, P. and Shaw, D., 2021. Quantifying the hydrological implications of pre- and post-installation willowed engineered log jams in the Pennine Uplands, NW England. Journal of Hydrology, 603: 126855.
Schalko, I., Lageder, C., Schmocker, L., Weitbrecht, V. and Boes, R.M., 2019. Laboratory Flume Experiments on the Formation of Spanwise Large Wood Accumulations: Part II—Effect on local scour. Water Resources Research, 55(6): 4871-4885.
van Leeuwen, Z.R., Klaar, M.J., Smith, M.W. and Brown, L.E., 2024. Quantifying the natural flood management potential of leaky dams in upland catchments, Part II: Leaky dam impacts on flood peak magnitude. Journal of Hydrology, 628: 130449.
Citation: https://doi.org/10.5194/egusphere-2024-2132-RC1 -
RC2: 'Comment on egusphere-2024-2132', Paul Quinn, 06 Nov 2024
I would first like to congratulate Reviewer 1 for a very detailed analysis of the paper and that I fully agree with these recommendations. So, I will concentrate on general matters.
I think more papers are needed including more general reviews such as Lashford et al., 2022 and Quinn et al.,, 2022. As they pose important design concerns. Which is what my main discussion point is.
Overall, a very good paper, it is clear and it covers the scope (as stated) very well. The issue of Manning n having an impact on flow is the most important part of this paper. Perhaps plant some vegetation on the banks?
The most concerning issue, that I think can be easily solved, is the assumption that leaky dams with a rectangular lower slot is the only design of leaky dam of concern. There are many leaky dam designs. Designs that allow a dynamic alteration of discharge with flow magnitude/depth being the obvious example. What if the leaky dam has a vertical slot or multiple pipes at multiple depths? So, you must make it clear that you are restricting your analysis to just this design and it still tell us interesting things.
Your choice of storm and flows need clarification. O.1 RI rain storm or flow RI?
Where on the network is your reach? What is the catchment area entering the top end of the reach? This is important. You have said that you are testing sensitivity, BUT you have not tested sensitivity to changing inflow. LDs are drowned out quite quickly so if the inflow increases the effectiveness goes down quickly. Or are you stating another design restraint of your study that you assume LDs are only at the upper most part of headwater e.g. less than 1km2? In reality LDs can be effective at larger scales if designed correctly. So be clear.
If you have assumed no storage is created being the LD, then you have missed the key design concept that LDs work best if they can engage larger storage volumes. Being constrained within a channel is the worst design.
So just broaden your discussion and conclusion.
Citation: https://doi.org/10.5194/egusphere-2024-2132-RC2 -
RC3: 'Comment on egusphere-2024-2132', Anonymous Referee #3, 18 Nov 2024
The Landscape Evolution Model CAESARLisflood 1.9j is employed to simulate leaky wooden dams at reach scale. A roughness-based approach is followed, and the simulation of sediment transport is also included, which represent a significant improvement with respect to current simulation practice.
A theoretical case study is proposed, and the Author systematically analyse the results by combining the variation of several parameters, from cell size, to roughness, to lower gap size, observing the variation of hydraulic and geomorphological results. One LD is represented in a straight channel.
The contribution is certainly interesting, especially for its effort of understanding the dynamic behaviours of LD.
However, the approach is strongly theoretical, with no connection to field data, making it an interesting attempt, but still hardly applicable. At least, this limitation should be clearly discussed.
In addition, some inaccuracies are present in the paper, like the absence of the results for certain gap size (not clearly justified), the absence of figures related to the case study domain, and a very synthetized presentation of the results, which seems under “discussion” form rather than result analysis. For example, why not to show the actual discharge hydrographs, rather than showing immediately a discharge variation?
The methodology relies on the application of an evolution model, and, as said, the absence of a clear calibration is the main limitation of the selected approach.
Despite having employed different model resolution, the effect of the resolution is not discussed, possibly leading to results inaccuracies. For example, while erosion and sediment areas along the channel are similar, the effect on the variation of the discharge is different for different resolution. This aspect is only addressed in the final discussion, but I think it should be analysed with higher precision also in the results.
Please find below my detailed comments:
Lines 133-135: “This method is straightforward to apply within a model domain, as specific cells can be identified to place the LD in combination with other roughness variables such as in-channel or floodplain boundary roughness”. How does the model identify LD cells with respect to riverbed or floodplain cells? Is the cell dimension modified for rover or floodplain?
Figure 2: the figure shows a square with one LD, and either vertical (b) or horizontal (c) flow direction. However, I guess that the model allows also horizontal + vertical flow, in case of transversal connectivity (flooding the areas upstream of the LD). I think that the figure should highlight the possibility of such a behaviour. In addition, connected to the previous comment, the cell edge that represents the LD is representative of the physical dimension of the LD?
Line 190-191: what do the Authors mean with “If the LD is overtopped, there is less of the LD in contact with the water column, therefore 𝑛 is reduced.”? If the LD is overtopped, the water level is higher than htop. So, the entire LD will be in contact with the water column, which however will have additional area not in contact with LD, thus reducing the BR. Is this the reason why the roughness coefficient is reduced? Please adjust the sentence.
Lines 190-192: These sentences appear reasonable but highly qualitative. Was any calibration performed to justify the variation of that roughness behaviour?
Paragraph 3.1: A figure representing the DEM and the different resolution would help the reader in figuring out tests configuration.
Figure 3: Please also include the b-axis sketch (or a description in the text), to ensure the proper interpretation from readers that may be not confident with the grain-size analysis.
Paragraph 3.3: the time-step of the simulations is not reported.
Figure 4: To ease the readability of the figure, why not to use the SI units fo measurements also for time (seconds or hours, instead of minutes x 104)? I also believe that showing the output discharge hydrograph, for at least one of the tests, would help in clarifying the discussion,
Lines 271-273: the Authors say: “For all sediment transport enabled experiments, the falling limb of the storm and remainder of the simulation time shows up to 25% deviation from the baseline scenario due to 𝑄 becoming out of phase with the baseline.”. In Figure 4, C and D, a variation of the discharge is observed, the values od ΔQ being positive or negative depending on the cells size. Positive values reach 25% for DEMT4 and reached -10% for DEMT1. Which of the situations is reflected in the Authors’ sentence? I think that here you should further discuss the effect of the cells size in providing attenuation/incrementation of the simulated discharges.
Line 280: QS was not introduced before. Please, define it clearly.
Line 310: The Authros wrote: “the second most cell downstream of the LD”. I suggest rephrasing as follows: “the second cell most downstream of the LD”.
Figure 6: The amount of change appear strongly dependant on the cells size. Please add a comment about that.
Lines 318-319: “a 2 m DEM with a two-pixel wide channel (4 m)”. Do the Authors refer to DEMT2? If yes, please use the same acronym, to be consistent within the paper. If not, please explain better.
Paragraph 4.3.1: Is only no-gap simulation discussed? If yes, please include general comments also for other gaps. In addition, why is Figure 7 not showing 0.3 and 0.4 m gaps? They should be discussed too, since they are mentioned.
Line 338: the Authors say: “Sediment transport was found not to be influenced by the installation of the LD”. However, in Figure 8 a variation curve in the sediment yield is shown. Please consider modifying this sentence. Probably the variation is not significant nor clear, but the sediment yield appears to be at least slightly different from the baseline one. In addition, discuss also the results for 0.3 and 0.4 m gaps.
Figure 8: why are subfigures A’, B’ and C’ displayed? They are not commented in the text.
Lines 342-343: Please discuss why with a larger gap sediment storage is observed.
Figure 9: Also here, 0.3 and 0.4 m gaps are not shown.
Lines 374-376: the Authros suggest that: “CL can therefore be used to aid identification of the ideal locations for LDs throughout a given reach to achieve the desired behavioural response, such as increased flow attenuation or enhanced geomorphic diversity.” However, the approach here proposed only shows the potentiality of the model. In a real-case scenario, will the peak reduction, the erosion and deposition occur according to the model predictions. In my opinion, the approach is really interesting but, before being employed for NFM design, it needs to be calibrated and validate on different case-studies. This does not mean reduce the importance of the contribution but helps in identifying its limitations and stress the requirements of future efforts.
Lines 388-389: “In addition, as CL utilises a regular raster grid, the required resolution can be adapted based on user requirements as well as computational resource availability”.
Discussion: when discussing about computational burden, please consider including tables with the following data: cell resolution, number of cells, computational time. This will help the paper to be clearer on the effect of grid size on computational burden.
Line 469: I think it should be ”reduced” and no “reduce”; or “can reduce”?
Line 488: “which could further the understanding”. A verb is probably missing.
Citation: https://doi.org/10.5194/egusphere-2024-2132-RC3 -
AC1: 'Comment on egusphere-2024-2132', Josh Wolstenholme, 18 Dec 2024
Dear Editor and Reviewers,
Thank you for your time and effort in providing detailed comments and insightful suggestions, which have significantly improved this manuscript. Please find attached the detailed line-by-line responses for your consideration.
Best regards,
Josh Wolstenholme
Peer review completion
Journal article(s) based on this preprint
Data sets
Datasets associated with this paper Josh Wolstenholme, Chris Skinner, David Milan, Robert Thomas, and Daniel Parsons https://zenodo.org/doi/10.5281/zenodo.12795494
Model code and software
CAESAR-Lisflood 1.9j LD Josh Wolstenholme, Chris Skinner, David Milan, Robert Thomas, and Daniel Parsons https://zenodo.org/doi/10.5281/zenodo.12795494
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 275 | 83 | 182 | 540 | 11 | 10 |
- HTML: 275
- PDF: 83
- XML: 182
- Total: 540
- BibTeX: 11
- EndNote: 10
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United Kingdom | 1 | 134 | 24 |
| United States of America | 2 | 123 | 22 |
| Germany | 3 | 26 | 4 |
| Netherlands | 4 | 22 | 4 |
| Romania | 5 | 20 | 3 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 134
Joshua M. Wolstenholme
Christopher J. Skinner
David J. Milan
Robert E. Thomas
Daniel R. Parsons
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1657 KB) - Metadata XML