Influence of Check Dams on Flood and Erosion Dynamic Processes of a Small Watershed in the Loss Plateau
, Zhanbin Li 1,2, Peng Li 1, Guoce Xu 1,*, Haidong Gao 1, Lie Xiao 1, Feichao Wang 1 and Tian Wang 1Round 1
Reviewer 1 Report
This manuscript provides a study about the influence of the check dams on flood and erosion and how it affects the reduction of flow peak and therefore flooding and flow velocity. However, there are good efforts have been done in this work, this work can be considered more as a case study rather than a technical paper. Therefore, I recommend to change the manuscript type from article to case report.
Here are minor comments:
· Lines 42-45 more references are needed
· Lines 108-109 is there any other recent land use data? Probably this data is out of date
· In line 110 you mentioned the used model, Mike 11. Can you move this to the model set up section?
· Lines 116-118, why did you use 2D model to model the study domain? Is there any clarification for this? Or you did 2D overland flow with Mike. This section needs to be clearer.
Author Response
Response to Reviewer 1 Comments
Point 1: Lines 42-45 more references are needed
Response 1: References have been added to the manuscript.
Point 2:Lines 108-109 is there any other recent land use data? Probably this data is out of date
Response 2: Land use data needs to be consistent with the time of rainfall - runoff data. Therefore, recent land use data is not used.
Point 3: In line 110 you mentioned the used model, Mike 11. Can you move this to the model set up section?
Response 3: Line 110 mainly introduces the source of the cross-section data required by the Mike11 model, so it is more appropriate to put it in the data source part.
Point 4: Lines 116-118, why did you use 2D model to model the study domain? Is there any clarification for this? Or you did 2D overland flow with Mike. This section needs to be clearer.
Response 4: This study focuses on the influence of the check dams on the hydrodynamic process of the channel. However, the accuracy of the calculation results of the runoff from the slope is critical to the research results. In order to simulate the hydrological process in the watershed more accurately, the 2D model was selected to simulate the slope flow.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
General remarks:
The manuscript focuses on the issue of flood control and anti-erosion measures in the Chinese Loess Plateau. This issue is very topical in the context of the current climate change, which increases the frequency of heavy rainfall. The authors present well-documented results on the impact of check dams on flood characteristics and sediment transport in a small catchment. Information obtained from the field measurements is very valuable for further research on this issue and for the practical application of similar measures in other locations. I recommend the article for publication after minor adjustments.
Minor points:
Title, line 4: substitute "Loss" with "Loess"
Page 7, line 203: Modify the equation (3) to include the variable m, which is mentioned in the text below.
Page 7, lines 218-220: Check units in equation (5). Parameter g in the equation (5) is specific weight = density x gravity acceleration
Page 8, line 269: substitute “conditions was” with “conditions were”
Page 12, line 368: substitute “Figure 7 shows the correlation” with “Figure 7 shows relationships”
Page 12, lines 369-370: substitute “the correlation coefficient between” with “the coefficient of determination for the relationship between”
Page 12, line 371: remove “the correlation coefficient“
Page 12, line 371: substitute “There was a good correlation” with “There were close relationships”
Page 12, line 373: substitute “indicates” with “indicate”
Page 12, line 376: substitute “Figure 7. Correlation between” with “Figure 7. Relationships between”
Author Response
Response to Reviewer 2 Comments
Point 1: Title, line 4: substitute "Loss" with "Loess"
Response 1: It has been modified in the text.
Point 2: Page 7, line 203: Modify the equation (3) to include the variable m, which is mentioned in the text below.
Response 2: ‘m’ is the channel length unit and is not clearly stated in the manuscript, and has been modified.
Point 3: Page 7, lines 218-220: Check units in equation (5). Parameter g in the equation (5) is specific weight = density x gravity acceleration
Response 3: It has been modified in the text.
Point 4: Page 8, line 269: substitute “conditions was” with “conditions were”
Response 4: It has been modified in the text.
Point 5: Page 12, line 368: substitute “Figure 7 shows the correlation” with “Figure 7 shows relationships”
Response 5: It has been modified in the text.
Point 6: Page 12, lines 369-370: substitute “the correlation coefficient between” with “the coefficient of determination for the relationship between”
Response 6: It has been modified in the text.
Point 7:Page 12, line 371: remove “the correlation coefficient”
Response 7: It has been modified in the text.
Point 8:Page 12, line 371: substitute “There was a good correlation” with “There were close relationships”
Response 8: It has been modified in the text.
Point 9:Page 12, line 373: substitute “indicates” with “indicate”
Response 9: It has been modified in the text.
Point 10:Page 12, line 376: substitute “Figure 7. Correlation between” with “Figure 7. Relationships between”
Response 10: It has been modified in the text.
Author Response File: Author Response.pdf
Reviewer 3 Report
Check Eq 3
Author Response
Response to Reviewer 3 Comments
Point 1: Check Eq 3
Response 1: It has been modified in the text.
Author Response File: Author Response.pdf
Reviewer 4 Report
Dear Authors,
the paper presents a research of modeling of flood and sediment transport a case study of the check dams on the Wangmaogou watershed. The topic can be considered interesting and original - in research are use MIKE modeling. Modelling data was compared with field measurements to right calibration and validation. Research have a potential practical application on areas where high erosion dynamics. Also distribution of flood characteristics and sediments transport in different conditions (K, M, S dams in longitudial profile) have implementation on sediments management in the catchment. In my opinion presented in paper integrated modeling is a new part in scientific impact of state of knowledge.
I add some comments on different lines of the document.
--------------------------------------------
line 34: 650,000
I suggest change coma to „650 thous. km2”.
line 36: 472,000 … line 37: 91,200
I suggest change coma to „472 thous. km2” and „91.2 thous. km2”, and all parts of the paper.
line 81-82: 110°20′26″–110°22′46″E, 81 37°34′13″–37°36′03″N
I suggest add coordinates into Fig. 1 – as text line isn’t clear infomation.
line 93-95: were 23 normal check dams in Wangmaogou watershed, including two key dams, seven medium dams and 14 small dams
What mean „key, medium, small”? It’s size (heigh)?. It will be important to wrote some additional information.
line 96 (Fig. 1):
(b) I suggest add location of Wangmaogou watershed – I don’t see location?
(c) and (d) – text „250 500” on scale i stoo close – I suggest delete „250”.
line 101-102: The Suide Supervision Bureau established a watershed outlet hydrologic station in the Wangmaogou watershed
It’s missing data (water level, discharge, ect.) references?
line 104: content
I suggest write „contentration” and add information about indicator (eg. TSS in mg/l or turbidity)?
line 104-105: The terrain data was obtained from 104 the State Bureau of Surveying and Mapping.
It’s missing data references?
line 108: is then generated by the Hutchinson method.
It’s missing references to the method?
line 108: The land use data…
It’s missing data references?
line 124: evapotranspiration of the watershed is very small
I suggest add some references?
line 164: from 1 to 1
Please check, is missing „-„ (minus)?
line 189:
General question: You wrote that are 8 working condition designs (from N to KMS). It will be important to wrote that N means there is no dam in watershet, yes? In K are only key dams in watershet (2 dams) or all dams (23) are as key dam? That infomation is missing, but it should be clearly wrote.
line 213, 214, 251, 253 (tab.3): m.s-1
Why is „.” (dot) beetween „m” and „s”? In the paper you once use „/” once „-1” – please uniform.
line 219, 220
As above.
line 251, 284, 312 – figures 3-5:
In my opinion data (symbols: dot, rectangles…) on figure 3-5 are illegible – I suggest change to color figure or change symbols.
line 251, 284, 312 – figures 3-5:
I suggest in the top part of the figures 3-5 add information about dam location and types in longitudial profile. Becouse I see some changes of the parameters in longitudial profile but I can’t compre this with watershed (Fig. 1D). It should be beter to show increase and decrease of the indicators.
line 366: the connectivity index (Table 6) of eight (…) Figure 7 shows…
First you refer to Table 6 hen figure 7 – I suggest change location in paper.
line 388: [45].The
„Space” is missing.
line 410 (fig 8):
The type of colour scale in legend is not correct to the presented siltation modulus diagram – please change type of scale for interval (eg. 3, 5 – you decide).
Kind regards.
Author Response
Response to Reviewer 4 Comments
Point 1: line 34: 650,000
I suggest change coma to „650 thous. km2”.
Response 1: It has been modified in the text.
Point 2: line 36: 472,000 … line 37: 91,200
I suggest change coma to „472 thous. km2” and „91.2 thous. km2”, and all parts of the paper.
Response 2: It has been modified in the text.
Point 3: line 81-82: 110°20′26″–110°22′46″E, 81 37°34′13″–37°36′03″N I suggest add coordinates into Fig. 1 – as text line isn’t clear infomation.
Response 3: The coordinate system might not perform as well as what we expect, because the font is too small to show clearly.
Point 4: line 93-95: were 23 normal check dams in Wangmaogou watershed, including two key dams, seven medium dams and 14 small dams
What mean „key, medium, small”? It’s size (heigh)?. It will be important to wrote some additional information.
Response 4: The check dams were divided into three types: key dams, medium dams and small dams. Key dams are composed of the dam body, spillway, and discharge structure, while middle dams are composed of a dam body and a discharge structure and small check dams consists only of one dam body.
Point 5: line 96 (Fig. 1):
(b) I suggest add location of Wangmaogou watershed – I don’t see location?
(c) and (d) – text „250 500” on scale i stoo close – I suggest delete „250”.
Response 5: It has been modified in Figure 1.
Point 6: line 101-102: The Suide Supervision Bureau established a watershed outlet hydrologic station in the Wangmaogou watershed
It’s missing data (water level, discharge, ect.) references?
Response 6: It has been modified in the text.
Point 7: line 104: content
I suggest write ,contentration” and add information about indicator (eg. TSS in mg/l or turbidity)?
Response 7: It has been modified in the text.
Point 8: line 104-105: The terrain data was obtained from 104 the State Bureau of Surveying and Mapping. It’s missing data references?
Response 8: The terrain data was obtained from the Yellow River Water and soil Conservation Suide Supervision Bureau, and we purchased the terrain data directly from the Suthep Bureau.
Point 9: line 108: is then generated by the Hutchinson method.
It’s missing references to the method?
Response 9: References have been added.
Point 10: line 108: The land use data…
It’s missing data references?
Response 10: The land use data is from Yellow River Water and soil Conservation Suide Supervision Bureau.
Point 11: line 124: evapotranspiration of the watershed is very small
I suggest add some references?
Response 11: References have been added.
Point 12: line 164: from 1 to 1
Please check, is missing „-„ (minus)?
Response 12: NSE ranges from 0 to 1
Point 13: line 189:
General question: You wrote that are 8 working condition designs (from N to KMS). It will be important to wrote that N means there is no dam in watershet, yes? In K are only key dams in watershet (2 dams) or all dams (23) are as key dam? That infomation is missing, but it should be clearly wrote.
Response 13: N represents the case without dams; K represents the case with key dams only; M represents the case with medium dams only; S represents the case with small dams only; KM represents the case with key and medium dams; KS represents the case with key and small dams; KMZ represents the case with key, medium and small dams.
Point 14: line 213, 214, 251, 253 (tab.3): m.s-1
Why is „.” (dot) beetween „m” and „s”? In the paper you once use „/” once „-1” – please uniform.
Response 14: It has been modified in the text.
Point 15: line 219, 220
As above.
Response 15: It has been modified in the text.
Point 16: line 251, 284, 312 – figures 3-5:
In my opinion data (symbols: dot, rectangles…) on figure 3-5 are illegible – I suggest change to color figure or change symbols.
Response 16: It has been modified in the text.
Point 17: line 251, 284, 312 – figures 3-5:
I suggest in the top part of the figures 3-5 add information about dam location and types in longitudial profile. Becouse I see some changes of the parameters in longitudial profile but I can’t compre this with watershed (Fig. 1D). It should be beter to show increase and decrease of the indicators. Check dam
Response 17: It has been modified in the text.
Point 18: line 366: the connectivity index (Table 6) of eight (…) Figure 7 shows…
First you refer to Table 6 hen figure 7 – I suggest change location in paper.
Response 18: It has been modified in the text.
Point 19: line 388: [45].The„Space” is missing.
Response 19: It has been modified in the text.
Point 20: line 410 (fig 8):
The type of colour scale in legend is not correct to the presented siltation modulus diagram – please change type of scale for interval (eg. 3, 5 – you decide).
Response 20: It has been modified in the text.
Reviewer 5 Report
The Authors present a modeling study of how effective check dams are at reducing peak flood, or discharge and erosion in the Loss Plateau of China. In order to calculated erosion power, the study simulates runoff from precipitation events using MIKE SHE. The flow velocity along each stream reach is then used to estimate an erosion force within the channel. While the paper is concisely written, there are a few fundamental problems that need to be addressed in order for publication.
First: the model is not well described and it is unclear how MIKE SHE, which is used to calculate peak flow discharge and the erosive force is parameterized to have check dams. In other words how mathematically in the model does the presence of check damps affect flow? How is the additional storage capacity of check dams represented, and how is this simulated along the rising and falling portions of the hydrograph.
Second: It should be noted that the erosion in this analysis only accounts for channel scouring forces, when in reality, much of the sediment transported in the stream channels originates upslope outside of the channel from rain splash and sheet flow processes.
Third: While I think the use of MIKE SHE and the Saint-Venant equation to estimate watershed wide erosion is neat, I’m not convinced that the model is adequately simulating all the channel physics to make a complete estimate of how check dams may influence peak flow, flow velocity, and erosion power. We all acknowledge that models are simplifications of reality, and therefore do not account for all processes, but it is also worth stating that in the paper, and specifically adding a discussion of how the use of these simplifications of flow may deviate from reality. For example, in reality a channel has complex shapes and with a series of constrictions that speed flow up and wide areas that slow flow down. These will likely cause turbulent flow conditions that are neglected in the 1D Saint-Venant flow velocity approximation.
While these revisions and restructuring may seem quite large I am only recommend minor revision because I think the revisions are generally achievable and are mostly having to do with an improved methods description and an additional discussion.
Line by Line Comments:
The authors use the word ‘obviously’ too much, and in pretty much every case its use does not add meaning to the sentence. In almost every case where ‘obviously’ is used, I would omit it.
Line: 51-52: Omit the by in ‘…rainy season was blocked by up to 85.5%..’
Line 53-55: “Check dames are constructed across gullies to reduce the velocity of concentrated water flows, reduce erosion, control sediment and stabilize gullies [16-17].” I recommend leading this paragraph with this sentence. Seems out of place in the middle.
L67-68: Need citations to back this statement up. ‘Most studies conducted to date have focus on the effects of check dams on runoff and sediment …’
L123-125: Was evapotranspiration so small because the time scale of the simulation was just a storm, and therefore the flux of ET over the simulation time period did not add up?
L135-141: At this point it is not clear how MIKE SHE solved overland flow. It is later stated that the Saint-Venant equation. That should probably be made clear here, given how that sets the conditions that determine erosion force.
L142: Calibration and validation of the model: At this point the model has not been adequately described and therefore the calibration process does not make since to the reader. It is also unclear at this point in the manuscript how erosion is estimated and how it is connected to MIKE SHE. I recommend putting section 2.7 above section 2.4 and greatly expanding the model description to include how overland flow is calculated. Explicitly describe the Saint-Venant equation used in MIKE SHE is needed.
L153-154: It is not just that calibration works better with fewer parameters (which I agree with), but also what those parameters are and how they affect model behavior is very important. This is why a better model description is needed.
L155: What is the leakage coefficient? I know what the mannings number is, usually noted as ‘mannings n’. I also understand soil infiltration, though I’m not certain how MIKE SHE represents them. Again, having an adequate model description before the model calibration section would help.
L166: Do you mean rainfall that is associated with erosion, or rainfall that causes erosion?
Figures 2,3,4,4, and 8. Would all be better if they were in color.
Equation 3: It is not clear how channel connectivity affects runoff or erosion power. Rather it appears to be a simple metric of how many check dams are on a reach. Also, I don’t know how the presence of a check dam is represented in MIKE SHE.
L206: Change ‘better’ with ‘increased’
L216-217: The numbered reference should appear right after ‘Foster et al.,’
L263: My understanding of the Saint-Venant equations is that velocities in the vertical direction are simplified to zero. So it is confusing how a downward movement of water is accounted for. I think what you mean to say is ‘… into kinetic energy as water moves along the channel, …’
L265-267: Without the explicit explanation of how check dams are represented in MIKE SHE, it is hard to see how this statement is supported. Though I’m sure the model shows that velocity slows down, but how?
L354: Specify that your results are supported by [10], rather than just citing Castillo et al., 2007
L390-392: I don’t understand this sentence.
L394-369: Don’t lower reaches typically have lower erosion power compared to upper reaches because the channel slope is less? It is not explained here why lower reaches have lower erosion compared to upper reaches and it should be.
L516, Citation #31. Incomplete citation. Please include author names.
Comments for author File: Comments.pdf
Author Response
Response to Reviewer 5 Comments
Point 1: First: the model is not well described and it is unclear how MIKE SHE, which is used to calculate peak flow discharge and the erosive force is parameterized to have check dams. In other words how mathematically in the model does the presence of check damps affect flow? How is the additional storage capacity of check dams represented, and how is this simulated along the rising and falling portions of the hydrograph.
Response 1: The influence of the check dams on the hydrodynamic process of the channel was described by setting up a hydraulic structure such as a weir in the river network file of the Mike11 model.
Point 2:Second: It should be noted that the erosion in this analysis only accounts for channel scouring forces, when in reality, much of the sediment transported in the stream channels originates upslope outside of the channel from rain splash and sheet flow processes.
Response 2: This paper mainly considers the influence of the check dam system on the channel erosion dynamic factor, and does not consider the influence of the check dam on the slope erosion and sediment yield, which will be considered in the subsequent research.
Point 3:Third: While I think the use of MIKE SHE and the Saint-Venant equation to estimate watershed wide erosion is neat, I’m not convinced that the model is adequately simulating all the channel physics to make a complete estimate of how check dams may influence peak flow, flow velocity, and erosion power. We all acknowledge that models are simplifications of reality, and therefore do not account for all processes, but it is also worth stating that in the paper, and specifically adding a discussion of how the use of these simplifications of flow may deviate from reality. For example, in reality a channel has complex shapes and with a series of constrictions that speed flow up and wide areas that slow flow down. These will likely cause turbulent flow conditions that are neglected in the 1D Saint-Venant flow velocity approximation.
While these revisions and restructuring may seem quite large I am only recommend minor revision because I think the revisions are generally achievable and are mostly having to do with an improved methods description and an additional discussion.
Response 3: The 1D hydrodynamic model cannot accurately simulate complex channel hydrodynamic processes such as turbulent flow conditions. Therefore, the flow velocity and water flow shear force in the paper are only a description of the overall condition of the channel section. The influence of the check dam on the complex hydrodynamic process of the channel will be taken into account in subsequent research.
Point 4:The authors use the word ‘obviously’ too much, and in pretty much every case its use does not add meaning to the sentence. In almost every case where ‘obviously’ is used, I would omit it.
Response 4: According to the recommendations, deleted too many ‘obviously’.
Point 5:Line: 51-52: Omit the by in ‘…rainy season was blocked by up to 85.5%..’
Response 5: It has been modified in the text.
Point 6:Line 53-55: “Check dames are constructed across gullies to reduce the velocity of concentrated water flows, reduce erosion, control sediment and stabilize gullies [16-17].” I recommend leading this paragraph with this sentence. Seems out of place in the middle.
Response 6: It has been modified in the text.
Point 7: L67-68: Need citations to back this statement up. ‘Most studies conducted to date have focus on the effects of check dams on runoff and sediment …’
Response 7: It has been modified in the text.
Point 8:L123-125: Was evapotranspiration so small because the time scale of the simulation was just a storm, and therefore the flux of ET over the simulation time period did not add up?
Response 8: There are two main reasons for not considering evapotranspiration: First, the solar radiation during the rainstorm is small, Second, the duration of the rainstorm is very short.
Point 9:L135-141: At this point it is not clear how MIKE SHE solved overland flow. It is later stated that the Saint-Venant equation. That should probably be made clear here, given how that sets the conditions that determine erosion force.
Response 9: The MIKE SHE model calculates the slope flow by solving the diffusion wave form of Saint-Venant equation. This paper mainly simulates the hydrological process in the watershed, calculates the erosion factor in the channel, and does not simulate the erosion process in the model.
Point 10:L142: Calibration and validation of the model: At this point the model has not been adequately described and therefore the calibration process does not make since to the reader. It is also unclear at this point in the manuscript how erosion is estimated and how it is connected to MIKE SHE. I recommend putting section 2.7 above section 2.4 and greatly expanding the model description to include how overland flow is calculated. Explicitly describe the Saint-Venant equation used in MIKE SHE is needed.
Response 10: In this paper, the MIKE SHE model is used to simulate the process of production and convergence of the slope, and the hydrodynamic process of the channel is calculated by Mike 11. Based on the simulation results of Mike 11, the dynamic parameters affecting the erosion process, such as flow velocity and runoff shear force, are calculated by the formula of Section 7.2. The model does not simulate the erosion process. Therefore, it is better to put section2.7 after 2.4.
Point 11:L153-154: It is not just that calibration works better with fewer parameters (which I agree with), but also what those parameters are and how they affect model behavior is very important. This is why a better model description is needed.
Response 11: MIKE SHE and MIKE 11 are distributed physics models, each parameter of which has a physical meaning, so less parameter calibration is needed.
Point 12:L155: What is the leakage coefficient? I know what the mannings number is, usually noted as ‘mannings n’. I also understand soil infiltration, though I’m not certain how MIKE SHE represents them. Again, having an adequate model description before the model calibration section would help.
Response 12: Leakage coefficient refers to the leakage rate of the riverbed. Due to limited space, the model calibration process is not described in detail in this paper.
Point 13:L166: Do you mean rainfall that is associated with erosion, or rainfall that causes erosion?
Response 13: I am referring to the rainfall associated with erosion.
Point 14:Figures 2,3,4,4, and 8. Would all be better if they were in color.
Response 14: It has been modified in the text.
Point 15:Equation 3: It is not clear how channel connectivity affects runoff or erosion power. Rather it appears to be a simple metric of how many check dams are on a reach. Also, I don’t know how the presence of a check dam is represented in MIKE SHE.
Response 15: The channel connectivity index represents the influence of the number and location of check dams on the hydrodynamic process of the channel. The check dams are described by setting up a hydraulic structure in Mike 11.
Point 16:L206: Change ‘better’ with ‘increased’
Response 16: It has been modified in the text.
Point 17:L216-217: The numbered reference should appear right after ‘Foster et al.,’
Response 17: It has been modified in the text.
Point 18:L263: My understanding of the Saint-Venant equations is that velocities in the vertical direction are simplified to zero. So it is confusing how a downward movement of water is accounted for. I think what you mean to say is ‘… into kinetic energy as water moves along the channel, …’
Response 18: I mean from upstream to downstream, the potential energy of the flow is constantly being converted into kinetic energy, and there is a constant inflow.
Point 19:L265-267: Without the explicit explanation of how check dams are represented in MIKE SHE, it is hard to see how this statement is supported. Though I’m sure the model shows that velocity slows down, but how?
Response 19: The influence of check dam on hydrodynamic process of channel is described by setting hydraulic structure in Mike11.
Point 20:L354: Specify that your results are supported by [10], rather than just citing Castillo et al., 2007
Response 20: Literature 10 supports my results.
Point 21:L390-392: I don’t understand this sentence.
Response 21: The runoff depth and peak discharge reflect some characteristics of the individual rainstorm flood, and cannot comprehensively reflect the effect on erosion and sediment yield.
Point 22:L394-369: Don’t lower reaches typically have lower erosion power compared to upper reaches because the channel slope is less? It is not explained here why lower reaches have lower erosion compared to upper reaches and it should be.
Response 22: It has been modified in the text.
Point 23:L516, Citation #31. Incomplete citation. Please include author names.
Response 23: It has been modified in the text.
Author Response File: Author Response.pdf
