Review Reports - Application of SWAT in Hydrological Simulation of Complex Mountainous River Basin (Part I: Model Development)

Round 1

Reviewer 1 Report

The authors simulated hydrological processes in a mountainous river basin in central Nepal. The study is helpful to understanding the hydrological processes in the study basin. However, several points should be addressed before possible publication.

(1) As one of the series of two papers, the main (series) title should be exactly the same, while Part I/II should be placed between the series title and the specified manuscript sub-title.

(2) Line 12: What did you mean by snow-dominated? The fractions of land surface covered by snow or snow-melt contribution to total runoff are both not very high.

(3) Line 66: Artificial Neural Network (ANN) is a set of tools that can be used in data-driven hydrological modeling, but ANN itself is not a hydrological model.

(4) Line 95: Part B?

(5) Fig. 1: the inset should be limited to regions around the study basin to show the location clearer.

(6) Is there any glacier in the river basin? How was glacier-melt modelled if any?

(7) Fig. 2: explain land use types following the figure title or in the main text.

(8) Fig. 4 and some others: rainfall or precipitation?

(9) Table 2: SWAT parameters are generally HRU-dependent, i.e., they depend on land use, soil, and /or other factors. It is preferrable to consider the spatial variations of model parameters.

(10) Fig. 10: It seems water storage increases on an annual scale. Is there any evidence of such a result? More discussion should be added.

Author Response

Reviewers' comments on Paper I

Reviewer 1: The authors simulated hydrological processes in a mountainous river basin in central Nepal. The study is helpful to understanding the hydrological processes in the study basin. However, several points should be addressed before possible publication.

Response

Thank you for your appreciation on our paper. We would like to address the other comments and suggestions as follows:

SN	Comments	Response
1	As one of the series of two papers, the main (series) title should be exactly the same, while Part I/II should be placed between the series title and the specified manuscript sub-title.	We have revised the title of both the papers as recommended. Application of SWAT in Hydrological Simulation of Complex Mountainous River Basin (Part I: Model Development) Application of SWAT in Hydrological Simulation of Complex Mountainous River Basin (Part II: Climate Change Impact Assessment)
2	Line 12: What did you mean by snow-dominated? The fractions of land surface covered by snow or snow-melt contribution to total runoff are both not very high.	We realized that the term “snow-dominated” is not an appropriate terminology to describe the characteristics of the study basin. Therefore, we have maintained only “mountainous basin” throughout the manuscript.
3	Line 66: Artificial Neural Network (ANN) is a set of tools that can be used in data-driven hydrological modeling, but ANN itself is not a hydrological model.	We have re-written it as “Artificial Neural Network (ANN) based data-driven hydrological models” in the revised manuscript.
4	Line 95: Part B?	We have slightly rearranged this paragraph incorporating your suggestion.
5	Fig. 1: the inset should be limited to regions around the study basin to show the location clearer.	We believe that the inset map is required for the convenience of international readers to have a better idea of the study basin and its geographical location. Therefore, we have maintained Fig 1 in its original form.
6	Is there any glacier in the river basin? How was glacier-melt modelled if any?	. Yes, there is a small fraction covered by glaciers of this basin (~6% of total area). However, the glacier-melt model was not used in this study.
7	Fig. 2: explain land use types following the figure title or in the main text.	Thank you for your comment. These have been explained in Section 2.1.3 For your kind information, the same thing has been done for the soil types too.
8	Fig. 4 and some others: rainfall or precipitation?	It is “precipitation”. Please note that Figures 4, 5,6 and 7 are slightly modified to maintain consistency.
9	Table 2: SWAT parameters are generally HRU-dependent, i.e., they depend on land use, soil, and /or other factors. It is preferrable to consider the spatial variations of model parameters.	Yes, the SWAT parameters are HRU-dependent and different values have been adopted in the study basin accordingly and shown as a range (for eg., CN2, SOL_AWC and SOL_Z) in Table 2 in the revised version. However, we have applied the same value of some parameters to all the HRUs and sub-basins during calibration based on its sensitivity.
10	Fig. 10: It seems water storage increases on an annual scale. Is there any evidence of such a result? More discussion should be added.	Fig. 10: is the annual water balance of the basin. The following sentences have been added in Section 3.5 for clarity. Change in Storage (∆ storage) is defined as ∆ Storage = - [(Precipitation(P)-Water Yield(WY) -Evapotranspiration(ET)]. It implies that if P> (WY + ET), the excess water infiltrates and is stored, as soil moisture and GW storages, of the basin. On the other hand if (WY + ET)> P, the water deficit is met by soil and GW storages of the basin. For example in January, some water is released from the basin to meet the WY and ET while it is stored in the soil and GW storage in July. It is noted here that ∆ storage accounts for model error too.

Reviewer 2 Report

This paper describes a watershed modeling study of a mountainous area across China and Nepal. The method is technically sound. However, there is no significant contribution from the first paper. I actually found it exceedingly verbose. Much of the introduction is textbook information and can be removed. For example, the readers do not expect to learn what is lumped, semi-distributed and full distributed hydrologic models from the paper. And the equations of the models do not need to show here or a simple water balance equation is enough if there is no modification to the model. The entire paragraph ranging from Line 262 to 279 is just repetition of the figure and table. The two papers can be consolidated into one without any lost of important information and focus on the climate change study.

Author Response

Reviewers' comments on Paper I

Reviewer 2: This paper describes a watershed modeling study of a mountainous area across China and Nepal. The method is technically sound. However, there is no significant contribution from the first paper. I actually found it exceedingly verbose. Much of the introduction is textbook information and can be removed. For example, the readers do not expect to learn what is lumped, semi-distributed and full distributed hydrologic models from the paper. And the equations of the models do not need to show here or a simple water balance equation is enough if there is no modification to the model. The entire paragraph ranging from Line 262 to 279 is just repetition of the figure and table. The two papers can be consolidated into one without any lost of important information and focus on the climate change study.

Response

Thank you for your appreciation on the technical parts of our paper. We would like to respond to the other comments as follows:

We think that the paragraph (L63-73) is informative to existing modelers as well as new readers regarding the different softwares that are in use across the world.
Equation 1 (simple water balance equation of SWAT) has been maintained, however, equations 2 and 3 and their description have been deleted as per the suggestion of the reviewer.
As per the suggestion of the reviewer and without compromising the important results of the model calibration and validation, the paragraph (Line 266-277 of the revised manuscript) was shortened as follows:

Simulated and observed hydrograph for the calibration and the validation periods at Arughat on daily and monthly time steps are shown in Figure 5. The mean and standard deviation of the observed (simulated) flows are 168 (170) m³/s and 167 (168) m³/s, respectively for the calibration period and 154 (181) m³/s and 163 (180) m³/s, respectively for the validation period (Table 3). It can be seen from Figure 5 that the model simulates the flow pattern very well and the hydrographs are in good agreement with the rainfall pattern at daily and monthly timescales even for this length of time (20 years for calibration and 10 years for validation). Scatter plots of simulated vs. observed daily flows of these periods (Figures 6(a) and 6(b)) also show the good agreement between them. The difference in cumulative volume between the simulated and observed flow is very small for the calibration period (Figure 6 c), however, the model has over-estimated the flow in the validation period (Figure 6 d).

Our intention is not to increase the number of papers for publication. Rather, our decision to present the information in two complementary papers separately is based on the content and discussion we feel is needed. Our approach of calibrating and validating the hydrological model (in the first paper) considering a long duration of 30 years at daily timestep and extended validation using primary data collected by the study team in upstream and downstream of the calibration station is not a conventional practice in hydrological modelling. The focus of the second paper is concentrated completely on the impact of climate change on the flow of the study basin and therefore, we feel presenting ample results and discussion is necessary in this regard. With due respect to your suggestion, we still firmly believe that a single consolidated paper cannot provide the intended information with sufficient details.

Reviewer 3 Report

The study deals with application of SWAT hydrological model in Mountainous River Basin of central Nepal. It is a suitable article. Researchers very nicely presented the applicability of SWAT model. However, there are grammatical and typographical errors which needs to be rectified before its accepting the draft for publication.

Line 10……Please write……Soil and Water Assessment Tool (SWAT)

Line 12-13……Please write…… snow-dominated mountainous basin, the Budhigandaki River Basin (BRB) in central Nepal.

Line 31……Change the keywords with scientific appeal. Write SWAT model. Try to avoid using more numbers of place or location as keyword. Use one location instead of three.

Line 41-42……Check full stop and citation.

Line 64……Deletes “s” in For examples

Line 65……Please check spacing…..GR4J (Genie Rural a4parametres Journalier)

Line 74……Check spacing and use of citations [30,31]

Line 75-79……Revise the sentences. Several studies….., e.g. Nile Basin by…..Write the name of authors……Please check the mistakes.

Line 90-91……Write about the objective that has been presented in another study (Part II).

Line 101……Materials and methods

Line 113……Give space after full stop.

Line 114-115……Revise the sentence. Check grammar.

Line 129…..Tibet…..Check the font style

Line 259…..Unit should be presented uniformly throughout the manuscript, e.g., s/m3 in Table 2……3 should be in superscript form.

Line 302……Delete (e.g.,

Line 325-328…… Revise the sentence as it is very confusing. Results presented for Flow Duration Curve are not in readable form. Further, use comma before “and” and “respectively”.

Line 371-373………Spacing between the word should be properly checked.

Line 374………Give full stop before Similarly

Please avoid unnecessary use of bold in case of words/letters throughout the manuscript. Check the Journal’s format.

Discussion part may be elaborated considering the appropriate outcomes of the study along with proper justifications.

The ranking of the Model’s performance using term “very good” seems to be unscientific

Author Response

Reviewers' comments on Paper I

Reviewer 3: The study deals with application of SWAT hydrological model in Mountainous River Basin of central Nepal. It is a suitable article. Researchers very nicely presented the applicability of SWAT model. However, there are grammatical and typographical errors which needs to be rectified before its accepting the draft for publication.

Response

Thank you for your appreciation on the technical parts of our paper. We would like to respond to the other comments and suggestion as follows:

SN	Comments	Response
1	Line 10……Please write……Soil and Water Assessment Tool (SWAT)	The suggested change has been made in L10 of the revised paper.
2	Line 12-13……Please write…… snow-dominated mountainous basin, the Budhigandaki River Basin (BRB) in central Nepal.	Line 12-13 has been re-written as per the suggestion.
3	Line 31……Change the keywords with scientific appeal. Write SWAT model. Try to avoid using more numbers of place or location as keyword. Use one location instead of three.	We also realized this thing after reading your comment. We have revised the keywords as follows: Hydrological Simulation; SWAT; Water Balance; Complex Mountain; Budhigandaki
4	Line 41-42……Check full stop and citation.	It is typo and has been corrected.
5	Line 64……Deletes “s” in For examples	Comments incorporated.
6	Line 65……Please check spacing…..GR4J (Genie Rural a4parametres Journalier)	It is corrected as: Génie Rural à 4 paramètres Journalier
7	Line 74……Check spacing and use of citations [30,31]	The spacing has been adjusted in the revised paper.
8	Line 75-79……Revise the sentences. Several studies….., e.g. Nile Basin by…..Write the name of authors……Please check the mistakes.	We have revised the citation in L75-79 as per the suggestion: Several studies have been carried out to assess the water availability and impacts of climate change; land use and land cover changes around the world using SWAT model [21,32–34]. Nile Basin by Griensven et al. [35], Itapemirim River basin (Brazil) by Fukunaga et al.[36], Ganga Basin by Anand et al. [37], and in Mekong Basin by Tang et al. [38] are some of them.
9	Line 90-91……Write about the objective that has been presented in another study (Part II).	The objective of the second paper (Part II) has been mentioned in L100-101.
10	Line 101……Materials and methods	Comments incorporated.
11	Line 113……Give space after full stop.	Comments incorporated.
12	Line 114-115……Revise the sentence. Check grammar.	L115-116 (in the revised manuscript) has been rewritten as follows:. The temperature varies from -2.0 °C in winter to 33.0 °C in summer in the study basin [61].
13	Line 129…..Tibet…..Check the font style	Comments incorporated.
14	Line 259…..Unit should be presented uniformly throughout the manuscript, e.g., s/m3 in Table 2……3 should be in superscript form.	Consistency has been maintained in the units throughout the revised manuscript.
15	Line 302……Delete (e.g.,	Comments incorporated
16	Line 325-328…… Revise the sentence as it is very confusing. Results presented for Flow Duration Curve are not in readable form. Further, use comma before “and” and “respectively”.	L325-329 (of the revised manuscript) have been changed as below for clarity: Based on the flow duration curve, the magnitude of the observed flow at 10%, 40% and 90% exceedance probabilities are 431, 118 and 29 m³/s respectively. Similarly, the exceedance probabilities are respectively 453, 126 and 39 m³/s for simulated flow. This indicates that the fractional difference between the corresponding values of the two flow-series are 5%, 7%, and 33% respectively.
17	Line 371-373………Spacing between the word should be properly checked.	Comments incorporated.
18	Line 374………Give full stop before Similarly	Comments incorporated
19	Please avoid unnecessary use of bold in case of words/letters throughout the manuscript. Check the Journal’s format.	We have revised the manuscript as per the journal’s format requirement.
20	Discussion part may be elaborated considering the appropriate outcomes of the study along with proper justifications.	The Discussions section of this paper includes the following: We have calibrated and validated the SWAT model using 30 years historical data and compared the model performance with similar studies in this region. Our approach of calibration and validation of the hydrological model has considered extended validation using primary data collected by the study team in upstream and downstream of the calibration station. To the knowledge of the authors, adopting this non-conventional approach is seldom found in this type of study. We have discussed the probable underlying reasons for the differences between the observed and simulated values. We have presented the results of the long-term flow simulation as well as high and low-flows and also discussed their implications on the water availability of the study basin. We have discussed the spatial variation of precipitation within the study basin with appropriate reasons. We have explained the implication of the fractional difference of high and low flows depicted from the FDC. We have also presented how snowmelt and baseflow varies seasonally in the basin and its implications on the overall water resource management of the basin. A brief discussion has been presented regarding the water balance of the basin. We, therefore, feel that adding further elaborative sentences to the discussion will not make any value additions to the paper. With due respect to the reviewer’s comments, we have maintained the Discussion Section in its original form.
21	The ranking of the Model’s performance using term “very good” seems to be unscientific	We have ranked the model as “very good” based on the evaluation criteria defined by Moriasi et al., 2007 and Gupta et al., 2009.

Round 2

Reviewer 1 Report

(1) Fig.1: the inset map is not clear enough. It’s preferrable to show regions around the study basin instead of such a large area not related with the study basin.

(2) Why was glacier-melt not considered in the model? If possible, consider glacier-melt runoff because the basin is said to be “glacier-dominant” in the previous version. If impossible, its impact on simulation results should be discussed by referring to related studies.

(3) Line 308: Figure 6?

(4) Fig. 7: Peak floods were significantly overestimated in (a) and (b) while underestimated in (c), which should be discussed further.

Author Response

Dear Reviewer,

We would also like to thank for providing valuable comments and suggestions which have significantly improved the quality of the paper. As per the suggestion, we have prepared new inset map with clear study basin. Also checked the minor English language and grammar. Similarly, in the discussion part we added the two paragraphs describing the glacier melt runoff in the total hydrographs of the BRB (Line no. 424-434) and in the (Line no. 393-406) depicted comparison the upstream and downstream station with the simulated flow and historical flow series. Besides, we have rearranged the (line 344-346) per the advice of the reviewer. We have also changed the figure 8 to make consistency of the figures.

Author Response File: Author Response.pdf

Reviewer 2 Report

Thanks for the response to my comment.

The authors have addressed some of my comments. But I still considered the two papers can be consolidated into one without missing important information.

Author Response

Dear Reviewer,

We had tried to address this issue in our response in the previous version of the manuscript. With due respect to the reviewer’s views, we would further like to reiterate that the main focus of the second paper in this series is to assess the impact of CC on future flows of the study basin by using a well calibrated SWAT hydrological model which is discussed in detail in this paper. We still firmly believe that a single consolidated paper cannot do justice to our study without compromising on the details of the model setup as well as CC impact assessment. Therefore, we request your approval for presenting our study in the current form of two complementary papers.

Thank you.

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear Author,

The improvement of your manuscript is clearly visible so I am grateful for your cooperation. Although, some further cosmetic changes are still necessary.

Line no 109 it should be „ eighth highest peak in the world „ instead of „ eighth highest peak of the world“

Line no 109 it should be „ in the world, is situated „ instead of „ in the world is situated „

Line no 113 it should be „ Tibetan part of the basin receives „ instead of „ Tibetan parts of the basin receives „

Line no 255 it should be „ two different sets „ instead of „ two different set „

Line no 367 it should be „in the basins with a significant „ instead of „ in the basins with significant „

Author Response

Dear Reviewer,

We would also like to thank for providing valuable comments and suggestions which have significantly improved the quality of the paper. We have addressed all the five comments in the revised article.