Towards Coupling of 1D and 2D Models for Flood Simulation—A Case Study of Nilwala River Basin, Sri Lanka

Round 1

Reviewer 1 Report

In this work the authors show an interesting work of coupling of 1D and 2D Models for flood simulation. 1D model is HEC-HMS and 2D model is the hydraulic model iRIC. HEC-HMS is used to obtain the streamflow used as input in the hydraulic model starting from hourly rainfall. The study area, shown in figure 1, is the Nilwala River Basin (Sri Lanka). This figure has to be improved: colors refer to…??? I‘d include the location of Sri Lanka in a larger map. The authors calibrated HEC-HMS model using data from two flood events (May 2018 and September 2019) and then validate the model using data from the flood event registered in May 2017. The authors have to show more details about the calibration steps. Did they used the calibration tools of HEC-HMS or did they use another methodology? I didn’t find streamflow data corresponding to the events used for calibration… figure 2 shows numerical and observed streamflows for these events but both (panels a and b) are identical each other… asumming that one of these panels is ok (for example panel a) peak value for event registered in May 2018 is seven times less than the peak value of the event registered in May 2017. I think the authors have to use similar streamflows to calibration and validation stages. Later, in figure 3, the authors show the numerical and the observed streamflows for the validation event (May 2017) but, as the authors say “Due to the severity, the gauge station at Pitabeddara was completely flooded during the event's peak.”... if the gauge station was completely flooded, the registered streamflow can’t be used as reference. Also, the authors say that the peak discharge was obtained from a water mark placed on a building… how the authors calculate the peak discharge from a water mark? Maybe, the peak value of water depth can be accurately estimated from a water mark but the uncertainty associated to the peak value of discharge obtained from a water mark can be high… Regarding figure 4, where are the inlets located in the map? Please include a map to locate these inlets…Figure 5 show the surveyed flood extent, what is the table included in the figure? Maybe the numerical flood extent can be included in this figure. Where are located the control points defined by the authors to compute the producer accuracy, user accuracy and overall accuracy? Why the authors use a value of water depth equal to 0.5 meters to define “flooded area”? Why not use 0.25 m? How the extent of the flood change with this value? There are two figure 6… in the second one I can’t read the labels of the colour bar (please, use white). Figure 8 has to be improved… labels are too small. Regarding the computation times… the authors use a grid value equal to 80 m and the associated computational time was equal to 5 h, did the authors run simulation with other values of grid size to check the accuracy of the results and the different computational times associated to these values of grid size. The authors say that they coupled HEC-HMS and iRIC but no details of this coupling are shown… did they use Python scripting to automatically link these models??? I think that this is a key point of the paper and it is not detailed at all… In my opinion, the paper is very interesting but it has to be improved according to these comments in order to be reconsidered.

Author Response

Thank you very much for your comments to improve this manuscript. We have addressed the comments made by four reviewers and yourself.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper presents a hydrologic-hydraulic modeling applied to the analysis of floodings for a specific case study. It is a useful study for practitioners and engineers in charge of flood management. The use of free-licensed software is particularly interesting.

Nevertheless, there are some aspects that would be opportune to deepen.

It would be interesting to include in the Introduction Section a comparisson between the proposed 1D-2D coupled model with other models (such as HEC-RAS 2D). Which are the advantage of using this coupled model instead of others? 

The proposed model was applied to a few cases, including calibration/validation events. It would be useful to have some context of the magnitude of the flood events. This information would help to determine if the model provides good results for a wide range of floods. 

The limitation of the research derived from applying the analysis to one case study and only three events should be explained. It would be opportune to mention how these limitations may be addressed.

Author Response

Thank you very much for your comments to improve this manuscript. We have addressed the comments made by four reviewers and yourself.

Author Response File: Author Response.pdf

Reviewer 3 Report

Title: Coupling of 1D and 2D Models for Flood Simulation- A Case Study of Nilwala River Basin, Sri Lanka

 

Abstract: Please provide a brief description of the models used in the simulation

I recommend the authors to include the resulting Flood hazard map in the Abstract and Conclusion, as it is one of the major results of the research.

 

1. Introduction: 

Please include a statement explaining the reason why the particular models (HEC-HMS and iRIC) are used in this study. 

Please clarify the meaning of the statement:

L50 "It has been identified that precipitation within the basin increases from south to north with increasing elevation"

L52 "There is considerable period in between extreme precipitation in upper Nilwala and inundation."

L77 I would like to recommend the authors to use "we coupled a 1D and 2D models.." instead of "we adopted a 1D-2D coupled model" since HEC-HMS and iRIC are two separate models that are intended to be used independently. 

L93 Include the complete name of the acronyms (SWAT, SWMM, iRIC) when first mentioned

L100 "Nays2DFlood solver in iRIC, a 2D hydraulic model used for flood 100 inundation mapping [19, 26–29]." does not qualify as a complete sentence, please consider revision.

L116 Please include the elevation data in the figure

2. Materials and Methods

L119, 162 The tables should be introduced in the statement first. 

L185 What is does this mean? "These basic equations, which are in a rectangular coordinate system, are mapped to a general coordinate system"

L189 What is "Background images"?

L220 misspelling of "model"

L221,225 The equations should be introduced in the discussion

3. Results and Discussions

Figures and tables must be introduced in the discussion first before appearing.

L257,265 Please include the rainfall data together with the discharge data, to visualize the lag effect of increased precipitation in the streamflow discharge

L288 Show the inlet points in the map

L293 What is the downstream water level boundary condition?

L324 If the depth is recorded on a river gauge station, I believe it should be noted as Water level instead of Flood depth.

L336 Figure 7 instead of Figure 6

L336 Please make the legend readable. Include the units used as well

L367-369 "This section may be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn." What is this for?

 

 

Author Response

Thank you very much for your comments to improve this manuscript. We have addressed the comments made by four reviewers and yourself.

Author Response File: Author Response.pdf

Reviewer 4 Report

The title of this paper is “Coupling of 1D and 2D Models for Flood Simulation – A Case Study of Nilwala River Basin, Sri Lanka”. The title of this paper is interesting. However, the contents of this paper needs a lot of revision for publication. Therefore, I recommend a rejection in this paper.

1. In introduction, there is no necessity for this study. The necessity of the content carried out in this study should be written.
2. The coupling process of 1D and 2D models should be shown as a flowchart.
3. From Equation (2) to Equation (8), the meaning of each equation should be mentioned.
4. Sections 2.3.1, 2.3.2 and 2.3.3 should be merged into one.
5. What did “cumecs” mean in graphs?
6. In Figure 5, the legend is not visible. The scale and traverse table are required.
7. In Figure 6, the legend is not visible.
8. It needs to be clearly indicated that this study could improve something.
9. It is necessary to mention the limitations of this study and follow-up studies.

 

Finally, I hope that my comments will help you complement your study.

Author Response

Thank you very much for your comments to improve this manuscript. We have addressed the comments made by four reviewers and yourself.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Manuscript Number - hydrology-1526877

 

18/01/2022

Some of the questions have been clearly answered by the authors but I have doubts about some points:

1. lines 278-282: “The peak discharge of the flood was estimated by flood marks. Thus, there is a considerable uncertainty in the peak discharge of this event. However, the simulated flow using the averaged parameters showed the goodness of fit with the observed flow with a NSE of 0.927, a PBIAS of -8.33%, and an RMSE of 0.3. Like calibration, the PBIAS indicates a model overestimation bias.” How did the authors estimate the peak discharge from flood marks? Did they use this estimated value to compute NSE, RMSE or PBAS? What it is the value of the estimated peak discharge?

2. Figure 4 shows the numerical domain defined for the iRIC simulations. I think that this area can be defined in Figure 1 just clarify the area under study for the iRIC model. Also the area analysed with HEC-RAS could be included in figure 1. I send you an example attached to my comments.

 

3. In figure 5: what is the difference between “Inundated Area” and “Flood Extent”?
4. Figure 7: there is a label in panel (a) “Time: 777600 sec”. This label doesn’t appear in panel (b), why?
5. In line 204 “The discharge hydrographs obtained from the HEC-HMS model were then imported into iRIC manually and were set as boundary conditions.” Therefore, in my opinion, there is no coupling between HEC and iRIC. I think the authors should consider changing the title of the paper. For example: “Towards a coupling of 1D and 2D Models for flood simulation – A Case Study of Nilwala River Basin, Sri Lanka”
6. In line 175 there is a reference to table 2 “For simulation, three flood events were selected; May 2017, May 2018, and September 2019 (Table 2).” I think this reference is not ok.
7. The difference between producer’s accuracy and user’s accuracy is not clear to me. In fact, the results seem to be contradictory. On one hand the producer's accuracy indicates that the model overestimates the flood extent to a certain degree and, on the other hand, user’s accuracy indicates that a majority of the flood extent was accurately simulated. I think the authors have to clarify this point.

Comments for author File: Comments.pdf

Author Response

We are extremely grateful for your constructive suggestions and comments which will go a long way in the development of this manuscript. We hope that the changes we have made address your doubts about the paper

Author Response File: Author Response.pdf

Reviewer 4 Report

The title of this paper is “Coupling of 1D and 2D Models for Flood Simulation – A Case Study of Nilwala River Basin, Sri Lanka”. The paper has been well revised in consideration of the contents pointed out. Therefore, I recommend an acceptance in this paper.

Comments for author File: Comments.docx

Author Response

Thank you