Evaluation of the Main Function of Low Impact Development Based on Rainfall Events

Round 1

Reviewer 1 Report

This is an interesting modelling evaluation of the potential benefits of implementing LID in the Munji and Gwanpyeong regions.

An important part of model preparation is the calibration. Without knowing the two regions very well it is difficult to determine whether 5 events are sufficient to accurately calibrate the SWMM model to the observed records. Statistically, I would expect that at least 10 events would be required. Further, the 5 events have been selected across 2015, 2016 and 2017. Is it correct that there have only been 5 events in 3 years?

It is interesting that the various LID scenarios have been modelled, if understood correctly, by increasing the infiltration area and/or rate. It would be equally interesting to evaluate some of the other input parameters to the PCSWMM model and their influence on the infiltration/runoff volume/peak flows, and compare these with the model results for LID.

I would be cautious of ranking the LIDs based only on the SWMM model results without any sensitivity analysis of the input parameters.

There are a few spelling and grammatical errors that require correction;

• Line 340, "...while it becomes sane serious as Gwanpyeong...."
• Line 347, "...regulation ability is getting better with the rainfall regression period." Not sure what the meaning is here.
• Line 354, "...destroyed..." Suggest replace with a less emotive word.
• Line 358, "...is better that before development." Should this be "than"?
• Line 439 "...show LIDs could can lessen..." Delete "can".
• Line 452 "...unbalance..." replace with "imbalance"

Author Response

Dear anonymous reviewer,

First of all, we would like to express our deepest gratitude to the distinguished reviewer for spending your precious time to review our manuscript. Your comments and advice are highly appreciated, and they are useful for us to revise our manuscript for producing a high-quality journal. We made major revisions based on the comments, the corrected portions at the revised manuscript revision are highlighted in red color for the benefits of the reviewers to identify and verify the corrections. And the followings summarize the corrections have been made corresponding to the comments given by the distinguished reviewer. The following are point-by-point the responses to the comments of respected reviewers.

Point 1: This is an interesting modelling evaluation of the potential benefits of implementing LID in the Munji and Gwanpyeong regions. An important part of model preparation is the calibration. Without knowing the two regions very well it is difficult to determine whether 5 events are sufficient to accurately calibrate the SWMM model to the observed records. Statistically, I would expect that at least 10 events would be required. Further, the 5 events have been selected across 2015, 2016 and 2017. Is it correct that there have only been 5 events in 3 years?

Response 1: In fact, we collected and utilized more than 5 rainfall runoff event data for this study. But, unfortunately, some of them gave unreasonable results during our calibration procedures, in that some SWMM parameters from those event data are not in the appropriate range suggested by the SWMM manuals and the previous research papers. Therefore, we needed to discard them from our analysis. In future studies including LID for Chinese cities as well as Korean cities, we will select the more event data and perform the preprocess for the sake of data correction before main analysis in order to get the large enough number of events.

Point 2: It is interesting that the various LID scenarios have been modelled, if understood correctly, by increasing the infiltration area and/or rate. It would be equally interesting to evaluate some of the other input parameters to the PCSWMM model and their influence on the infiltration/runoff volume/peak flows, and compare these with the model results for LID.

Point 3: I would be cautious of ranking the LIDs based only on the SWMM model results without any sensitivity analysis of the input parameters.

Response 2, 3: We felt that we might be able to answer to the comment 2 and 3 from the reviewer by addressing our sensitivity analysis. We have added the sensitivity analysis of the input parameters into section 3.2 (line 274 to 303). The sensitivity of seven major parameters in SWMM and their influence on the infiltration, runoff volume, peak flows are evaluated. It is concluded that conductivity is the most sensitive parameter with regard to infiltration and runoff volume, which followed by N-pervious and suction head, meanwhile, the peak flow is greatly affected by N-pipe and N-pervious. Then, it is interesting that the results indicate that the N-pipe (the manning roughness coefficient of the conduits) play a significant role in Gwanpyeong than that in Munji in terms of peak flow and runoff volume. That is to say, the roughness of the conduits might be the reason for difference in the flow reduction effect of LIDs in two regions. The flow velocity is greatly affected by the roughness of pipe, if the flow velocity is too fast, the peak flow will be too large to lead to waterlogging; otherwise, if the velocity is too slow, the pollutant will be deposited in the pipe network. Therefore, additional pipe network analysis, for example the combined Hardy Cross method with the optimal control, is needed to estimate the appropriate peak flow in the further study.

Point 4: There are a few spelling and grammatical errors that require correction;

• Line 340, "...while it becomes sane serious as Gwanpyeong...."
• Line 347, "...regulation ability is getting better with the rainfall regression period." Not sure what the meaning is here.
• Line 354, "...destroyed..." Suggest replace with a less emotive word.
• Line 358, "...is better that before development." Should this be "than"?
• Line 439 "...show LIDs could can lessen..." Delete "can".
• Line 452 "...unbalance..." replace with "imbalance"

Response 4: Thank you for pointing out the spelling errors, and we have corrected them in the manuscript. And our manuscript has been read and re-edited again by a professionally native English-speaking colleague.

Reviewer 2 Report

- there is no reference in the discussion of the results to analyses made by other authors,  - The authors independently analyse two catchment areas without attempting to determine which factors (catchment characteristics) determine the determined relationships,  - The authors should analyse the influence of the catchment characteristics on the obtained relationships (fig. 7) and discuss the results obtained in the light of other authors' research,  - It is advisable for the authors to perform a sensitivity analysis, for two precipitation events independently, to analyse the influence of SWMM model parameters on the modeled hydrogram volume, or maximum flow,  - on the basis of fig. 3 it results that the values of rainfall in 60 minutes increase, it is a little bit impossible, I suggest that the authors should determine total rainfall curves; empirical models should be adjusted to the course obtained this way. The results obtained in Fig. 3 confirm the defect of the polynomial model, therefore the correction described above should be made.     

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Figure 3 should be corrected, i.e. sum curves should be determined and the empirical model should be adapted to them. After making changes, the article can be published. Polynomials are good models, but they have their disadvantages and in order to eliminate them and develop a good precipitation model it is necessary to modify the curves in Fig.3.

Author Response

Dear anonymous reviewer,

First of all, we would like to express our deepest gratitude to the distinguished reviewer for spending your precious time to review our manuscript. Your comments and advice are highly appreciated, and they are useful for us to revise our manuscript for producing a high-quality journal. We made major revisions based on the comments, the corrected portions at the revised manuscript revision are highlighted in red color for the benefits of the reviewers to identify and verify the corrections. And the followings summarize the corrections have been made corresponding to the comments.

Point 1: Figure 3 should be corrected, i.e. sum curves should be determined and the empirical model should be adapted to them. After making changes, the article can be published. Polynomials are good models, but they have their disadvantages and in order to eliminate them and develop a good precipitation model it is necessary to modify the curves in Fig.3.

Response 1: Firstly, we have corrected Figure 3 in the revised manuscript (line 230 to line 233), the dimensionless cumulative distribution of Huff method (3rd quartile storm) is shown as Figure 3(a), and the empirical model is adjusted to form the Figure 3(b). Figure 3(b) shows the temporal rainfall distribution of 60-min duration under different return period. The general procedure of the design rainfall is summarized in the line 198 to line 233. In addition, the main objective of this study is to identify the primary hydrological function of LIDs under different rainfall events. And the relevant results in the paper are updated with the changes of precipitation distribution in Figure 3(b).

What should be highlighted here is all rainfall-runoff simulations in the target watersheds were conducted using the rainfall data on a 6-minute scale. And we compared those results with 1-minute interval in the previous study process, and found only a small difference. That is to say, although the time interval of temporal rainfall distribution changed, it has no significant influence on the hydrological function results of LIDs under the same total amount of rainfall.

Round 3

Reviewer 2 Report

The currently executed figure 3 is correct and the course of the curves is determined.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The manuscript is an interesting evaluation of the contribution that LID might make to some regions of South Korea, utilising a software model for the hydrological changes. It is a commendable effort for non-native English speakers, however, it is recommended that the manuscript be edited by a professional English-speaking writer to improve the quality of the paper prior to publication.

The modelled area of the proposed LIDs is described in Table 2, but there does not appear to be any discussion around whether there is enough physical area available to practically implement these solutions.

I'm unsure that % variation is the best way to represent the change in runoff and infiltration for Figures 8 and 9, since 200-300% variation seems significant, but the overall quantum may be small. Is there an alternate metric to use here?

Overall, a good manuscript that needs a few areas of improvement.

Author Response

Dear anonymous reviewer,

  First of all, we would like to express our deepest gratitude to the distinguished reviewer for spending your precious time to review our manuscript. Your comments and advice are highly appreciated, and they are useful for us to revise our manuscript for producing a high-quality journal. We made major revisions based on the comments, while the corrected portions at the revised manuscript revision ( please see the attachment) are highlighted in blue color for the benefits of the reviewers to identify and verify the corrections. The followings summarize the response have been made corresponding to the comments given by the distinguished reviewer.

Point 1: The manuscript is an interesting evaluation of the contribution that LID might make to some regions of South Korea, utilising a software model for the hydrological changes. It is a commendable effort for non-native English speakers, however, it is recommended that the manuscript be edited by a professional English-speaking writer to improve the quality of the paper prior to publication.

Response 1: Our revised manuscript has been read and re-edited by a professionally native English-speaking colleague.

Point 2: The modelled area of the proposed LIDs is described in Table 2, but there does not appear to be any discussion around whether there is enough physical area available to practically implement these solutions.

Response 2: We have added the descriptions for the area of the proposed LIDs into line 253 to 273. The land use map of the study sites in Figure 2 were obtained from the Munji district development report and previous study in Gwanpyeong region, and the descriptions were presented in line 253 to 259. Line 260-273 showed the available area to install LIDs to replace existing impervious surfaces in urban region. The design areas of each LIDs were obtained by reference to the actual land use data and specific measurements at Google Earth.

Point 3: I'm unsure that % variation is the best way to represent the change in runoff and infiltration for Figures 8 and 9, since 200-300% variation seems significant, but the overall quantum may be small. Is there an alternate metric to use here?

Response 3: Figures 8 and 9 in the original manuscript, were drawn based on the % variation of the infiltration and surface runoff between post-development without LIDs scenario (be taken as a baseline) and the designed LID scenario, and there didn’t use any alternate metric. In order to understand the change of hydrological conditions more intuitively, the variation of infiltration and surface runoff under different scenarios were showed by histograms in Figure 10 and Figure 11 at the revised manuscript revision.

Author Response File: Author Response.pdf

Reviewer 2 Report

Rainfall = f(duration rain)? What do you mean, after 60 minutes the amount of precipitation in a rain episode increases? Unclear ?
What is innovative? Nowadays it is only an engineering task, not a scientific one, because there is a lot of work to be done. Why did the authors use fractal geometry for this purpose? What were the results used for?  What were the results of fractal analysis compared to? It is not clear what is the purpose of fractal analysis in this paper? Return period of 100 years? It is rather a fictional calculation, because who is able to predict even a change in rainfall dynamics? What is the relationship between the results of fractal analysis and LID? I do not see in this approach the need for fractal analysis.  

Author Response

Dear anonymous reviewer,

  First of all, we would like to express our deepest gratitude to the distinguished reviewer for spending your precious time to review our manuscript. Your comments and advice are highly appreciated, and they are useful for us to revise our manuscript for producing a high-quality journal. We made major revisions based on the comments, while the corrected portions at  the revised manuscript revision (please see the attachment) are highlighted in blue color for the benefits of the reviewers to identify and verify the corrections. The followings summarize the responses have been made corresponding to the comments given by the distinguished reviewer.

Point 1: Rainfall = f (duration rain)? What do you mean, after 60 minutes the amount of precipitation in a rain episode increases? Unclear?

Response 1: In order to cope with extreme rainstorm events, designed rainfall was adopted for hydrological simulation in this study. The descriptions for design rainfall were presented in line 213 to 232 at the new revised revision.

Point 2: What is innovative? Nowadays it is only an engineering task, not a scientific one, because there is a lot of work to be done.

Response 2: Although, many studies have been conducted to evaluate the effectiveness of LIDs by experiment and modeling, the studies about performance of LIDs based on urban morphology are still sparse. Therefore, this study evaluated the hydrological effectiveness from runoff reduction or hydrologic balance aspect in two different urban form regions using SWMM. The results can provide the urban stormwater managers with decision-making references in future urban planning.

The innovations of this paper are:

- LIDs could make a better role in infiltration increase than runoff reduction, so that the mainly function of LIDs should be focused on restoration of hydrologic balance rather than flooding reduction.

- LIDs would take a relatively better role in the urban area having loose pattern of buildings than the one having compact pattern

Point 3: Why did the authors use fractal geometry for this purpose? What were the results used for? What were the results of fractal analysis compared to? It is not clear what is the purpose of fractal analysis in this paper?

Response 3:

- The fractal geometry was computed to quantify the urban morphology of the study areas.

- The results of fractal analysis are the basis of this paper. Urban morphology relies heavily on the calculation of fractal dimension of the drainage pipe network. Because the drainage pipe networks are often laid under urban roads or green belts, is closely related to the distribution density of buildings.

- The hydrological benefits of LIDs were evaluated based on urban morphology in this paper, the purpose of fractal analysis is to quantify the parameter that characterizes the morphometric variability of an irregular shape for accessing the region morphology of the two study areas.

Point 4: Return period of 100 years? It is rather a fictional calculation, because who is able to predict even a change in rainfall dynamics?

Response 4:

- The probability that storm events will occur is often expressed as a return period. Return period of 100 years means that, in any given year, there is a 1% chance that it will happen the storm event, regardless of when the last similar event was. With the high frequency of extreme weather events occur in the future, design rainfall is widely used for hydrological simulations.

- For this purpose, the frequency analysis software FARD 2006 is used to calculate the probability rainfall. This software can be used to estimate the quantiles for the normal, lognormal, gamma, log-Pearson type III, generalized extreme value (GEV), Gumbel (GUM), log-Gumbel, Weibull, Wakeby, generalized logistic, generalized Pareto, and kappa probability distributions. Although predictions are not guaranteed to be precise, it is at least could predict the possible storm events that occur in the future.

Point 5: What is the relationship between the results of fractal analysis and LID? I do not see in this approach the need for fractal analysis.

Response 5:

- The results of fractal analysis are the important part and basis to evaluate the performance of LIDs under different urban morphology. The region morphology of study areas need be estimated by the calculation of fractal dimension of the drainage pipe network.

- It is necessary to do fractal analysis in this study, we can evaluate what are the hydrological performance of LIDs based on the density of the drainage pipe network (urban morphology) and explore the impact factor of LIDs hydrological performance for providing the urban stormwater managers with decision-making references in future urban planning.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

- the answers in relation to fractal geometry do not justify the need to apply in the paper; however, if the authors perform such calculations it is for the catchment of the LID,
- still the distribution adopted for the calculations is not clear (Fig. 4),
- the calculations are unitary in nature (case study) and how to transfer the results to others on their basis?
- I do not see any further innovation approach, it is interesting to combine the LID solution with the measure of variability of the drainage basin seal,  

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report

Dear Editorial Office,
The corrections made by the authors and the answers are not satisfactory. I proposed additional analyses, but the authors did not do them. The proposed combination of fractal analysis as a separate element, which is not closely related to the calculation methodology (LID). The authors did not address all the remarks in a sufficient way.