Review Reports - Estimating the Effect of Deforestation on Runoff in Small Mountainous Basins in Slovakia

Round 1

Reviewer 1 Report

The article deals with an interesting and current topic. Climate change is accompanied, among other things, by an increase in the number and intensity of storms and the spread of the bark beetle epidemic. Both of these phenomena result in extensive deforestation, which has an impact on the hydrological cycle, among other things. In their study, the authors of the article focused on the first of the phenomena: they studied the effects of strong storms, causing massive forest breaks, on the outflow of water from the area. They deal with this topic in detail in their article. In addition to the immediate effects of severe storms, the impact of changes in precipitation intensity caused by advancing climate change is also addressed.

The study is prepared according to a clear methodology; however, I have comments on the implementation of some of its steps and also on the interpretation of the results.

When studying the effect of changes in the intensity of rainfall caused by climate change in the future, they work with the current land use (CLC for 2018); however, at the time (2070-2100) both river basins will look different, including different CNw values. How does this part of the study relate to the impact of deforestation?

In conclusion, the authors state that the impact of climate change increases with return periods (see Tables 10 and 11). However, it is interesting that this dependence does not manifest itself for the current precipitation intensities (see Tables 8 and 9). In both cases, the same terrain characteristics were used for modelling, but the results are different. Can the authors explain this difference in results?

Minor comments:

- manuscript should follow instructions to authors

- abbreviations must be explained before they are used (see ECM in the abstract, CSC-CN and RCM CLM on page 3, but also elsewhere)

- P1, the seventh line from the bottom: "land uses" -> land use

- P6, the last par .: three or four climatological stations?

- P7, equation (1) and others: all variables must be explained and supplemented by physical units

- P7, the first third of the page: "Simple scaling using the scaling of the statistical moments was applied in this paper." - Is it possible to explain this procedure in more detail?

- P7, middle of the page: beta_n = n_beta: Can you explain this equation?

- P8, equation 4-7: CN is not among the variables, on the contrary, the variable F is not used in the equations

- P8, tab. 2: table number is repeated, it should be 3 (other tables must be renumbered as well)

- P10, second paragraph: (Figure 5h) should be (Figure 5i)

- P14: "for rainfall durations from 60 minutes up to

1440 minutes ": explain why you used such a range

- P14: "In both of the analyzed, Liptovská Teplička and Boca, climatological stations" -> "In both of the analyzed climatological stations, i.e. Liptovská Teplička and Boca,"

- P14: Burn's diagram: What is it? What does it express?

- P15: Can you explain the trend analysis and its results in the tab. 3 and 4?

- P21: "In particular, land-use change and with it overall land use as well as climate change.": I do not understand this sentence.

Author Response

Comments and Suggestions for Authors

The study is prepared according to a clear methodology; however, I have comments on the implementation of some of its steps and also on the interpretation of the results.

Answer:Thanks for the question. Yes, it is assumed that the river basins will be in the period 2070-2100 in terms of land use. However, the authors wanted to point out only the impact of the development of climatic conditions for the scenario that it will be possible to maintain their current use, ie CLC 2018.

Note: CNs have changed (others are for the present and others for the future - we assumed with slightly worsened conditions), I do not know whether this should be commented on in the answer or not.

Answer: In the first step, we addressed the impact of land use, and the latter affects lower return periods, while in the second case, we addressed the impact of climate - climate affects higher return periods.

Minor comments:

- manuscript should follow instructions to authors

Answer: Thanks, we've reviewed the article template

- abbreviations must be explained before they are used (see ECM in the abstract, CSC-CN and RCM CLM on page 3, but also elsewhere)

Answer: Thank you for the comment. We added to the text.

- P1, the seventh line from the bottom: "land uses" -> land use

Answer: Thank you for your comment. The introduction has been modified and the term "land use" applied

- P6, the last par .: three or four climatological stations?

Answer: We corrected in the text, four climatological stations is the correct number.

- P7, equation (1) and others: all variables must be explained and supplemented by physical units ….. Answer: Thank you for the comment, the mistakes were corrected.

- P7, the first third of the page: "Simple scaling using the scaling of the statistical moments was applied in this paper." - Is it possible to explain this procedure in more detail?

Answer: It was supplemented and more detailed descripton and additional formulae was added in the text.

- P7, middle of the page: beta_n = n_beta: Can you explain this equation?

Answer: Thank you for the comment it was corrected.

- P8, equation 4-7: CN is not among the variables, on the contrary, the variable F is not used in the equations ………

Answer: Thank you for the comment, the mistakes were corrected.

- P8, tab. 2: table number is repeated, it should be 3 (other tables must be renumbered as well)… Answer: Thank you for the comment, the mistakes were corrected.

- P10, second paragraph: (Figure 5h) should be (Figure 5i)

Answer: Thank you for the comment, the mistakes were corrected.

- P14: "for rainfall durations from 60 minutes up to 1440 minutes ": explain why you used such a range Answer: Thank you for your comment, the range was replaced by exact selected durations of rainfall.

- P14: "In both of the analyzed, Liptovská Teplička and Boca, climatological stations" -> "In both of the analyzed climatological stations, i.e. Liptovská Teplička and Boca,"

Answer: Thank you for your comment, the suggestion was accepted.

- P14: Burn's diagram: What is it? What does it express?

Answer: More description on the methodology and explanation of the Burn's diagram was added into the text

- P15: Can you explain the trend analysis and its results in the tab. 3 and 4?

Answer: More detailed explanation about the presented results of the trend analysis was added to the text.

As the Burn's methodology and the Mann-Kendall trend test are well known and often applied methodologies, we have omitted in our paper a more detailed description with formula of these methodologies.

- P21: "In particular, land-use change and with it overall land use as well as climate change.": I do not understand this sentence.

Answer: Thank you for your comment, the sentence was rewritten.

Author Response File: Author Response.docx

Reviewer 2 Report

This is overall a good work. As a reader, I would prefer the authors could dig a little deeper into the hydrologic response due to the land use change and climate change. Please find my questions and comments as follows.

1. In Figure 1, it seems on the eastern end of IPLTICA catchment that there is an upstream channel that is cut by the watershed boundary. Can you confirm it? It is unclear to tell it from the map. If it does exist, how did you model the inflow? Did this inflow vary a lot and influence the discharge at the downstream gauge? Is there any other upstream tributary that feeds into the two catchments?

2. In Figure 2, why does the second DEM is presented in the gradient color but not in split color classes like the other figures in this panel?

3. In Equation 1, what are "T" and "d"?

4. I don't understand the logic behind equation 2.

5. In Equation 3, what is "E"?

6. Which RCP pathway is used to develop the CLM scenario?

7. Figure 5 and 6 need to be further polished to keep subtitles and border lines consistent. Also, I don't understand why period 2004-2016 needs to be shown twice in each pair of plots.

8. In Figure 7, what is the name of the along-radius axis? In the paragraph before the figure, you write "For the future scenario, a shift of the rainfall extremes to an earlier period of the month of July". However, it seems to be the opposite to me.

9. In Figure 8, the triangle symbols in the red color scheme are hard to distinguish from each other.

10. I am sorry if I miss it. Is there snow cover anywhere in the watershed or in the upstream area? Besides the precipitation change, how did the yearly snowmelt runoff and ET change? Will they contribute to the increase in runoff and affect your conclusions? It's better to have a simple water budget analysis just to make sure.

11. In my personal opinion, I would suggest to highlight your main conclusions in the abstract.

12. Please add line numbers to help with reviewing process.

Author Response

In Figure 1, it seems on the eastern end of IPOLTICA catchment that there is an upstream channel that is cut by the watershed boundary. Can you confirm it? It is unclear to tell it from the map. If it does exist, how did you model the inflow? Did this inflow vary a lot and influence the discharge at the downstream gauge? Is there any other upstream tributary that feeds into the two catchments?

Answer: Thank you, we have checked the watershed boundaries. White line is the boundary of the basin, which is correctly drawn. There are no other tributaries of the main river in the basin.

In Figure 2, why does the second DEM is presented in the gradient color but not in split color classes like the other figures in this panel?

Answer: Thank you for the comment, the Figures (DEM) have been corrected.

In Equation 1, what are "T" and "d"?

Answer: Thank you for the comment 3-5. More detailed description was added about T-return period and d-rainfall duration in the text (page 7).

I don't understand the logic behind equation 2.

Answer: Thank you for the comment. More detailed explanation about the duration function was inserted.

In Equation 3, what is "E"?

Answer: Thank you for the comment. More detailed description and formulae for deriving the scaling coefficients and the description of the variables was inserted in the text.

Which RCP pathway is used to develop the CLM scenario?

Answer: Thank you for the comment. More information about the CLM scenario was added to the text.

Figure 5 and 6 need to be further polished to keep subtitles and border lines consistent. Also, I don't understand why period 2004-2016 needs to be shown twice in each pair of plots.

Answer: Figures 5 and 6 have been modified. The period 2004 - 2016 classifies the period after deforestation (after the greatest forest calamity caused by wind), which was the evaluation period. And this period was compared with the whole observed period (1981 - 2016) and the period before deforestation (1981-2003).

In Figure 7, what is the name of the along-radius axis? In the paragraph before the figure, you write "For the future scenario, a shift of the rainfall extremes to an earlier period of the month of July". However, it seems to be the opposite to me.

Answer: Thank you for your comment, the mistake was corrected.

In Figure 8, the triangle symbols in the red color scheme are hard to distinguish from each other. Answer: Thank you for the comment, the color was changed for better distinguishing.

I am sorry if I miss it. Is there snow cover anywhere in the watershed or in the upstream area? Besides the precipitation change, how did the yearly snowmelt runoff and ET change? Will they contribute to the increase in runoff and affect your conclusions? It's better to have a simple water budget analysis just to make sure.

Answer: Thank you for the comment. The increased runoff is due to increased precipitation during the year. The effect of temperature was indicated in the spring months. Precipitation was lower, runoff higher, where the cause of this behaviour may be in the earlier snow-melting on the catchment. Earlier snowmelt was also demonstrated in the study by Hríbik et al. (2009) who analyzed a small river basin in central Slovakia.

Hríbik, A. Majlingova, J. Škvarenina and D. Kyselová, Winter Snow Supply in Small Mountain Watershed as a Potential Hazard of Spring Flood Formation, 2009, Bioclimatology and Natural Hazards pp 119-128

In my personal opinion, I would suggest to highlight your main conclusions in the abstract. ….. Answer: Thank you for the comment. The new abstract was inserted.

Please add line numbers to help with reviewing process. …..

Answer: We accept the comment, and we added the line numbers in the paper.

Author Response File: Author Response.docx

Reviewer 3 Report

General Comments

The introduction is rather disorganized. The motive behind the study and its novelty can be explained better. Also, the authors refer to climate model projections of future precipitation which seems out of context in this section. In Section 3.4, the authors state that they set the initial abstraction coefficient to 0 which will result in all precipitation becoming runoff. What is the basis for this assumption? The standard value for the coefficient is 0.2 but recent literature suggests it may be closer to 0.05 but I cannot find any basis for setting this as zero! Therefore, the entire discussion on design flows is rather suspect in my opinion. Section 4.3 which presents an analysis of the IDF characteristics for various time periods is difficult to understand – it is not clear how the authors are establishing rainfall durations and what the purpose of the future time period analysis is.

Specific (minor) comments

Page 7 – the authors mention hourly and daily precip data was available for the Vyšná Boca and the Liptovská Teplička stations. However, in the following page (8), they state “For the Vyšná Boca climatological station, only daily rainfall data for the 1981-2017 period were available.” The two statements contradict each other.

It is recommended that the authors split Figure 5 into smaller figures based on themes (flows, temp, precip separately). Also, it is recommended the authors use month names rather than I, II etc. as it can lead to confusion whether these are based on water or calendar years.

Author Response

Answer: Thank you for your comments. The introduction has been modified and the objectives written more clearly.

The standard value for the coefficient is 0.2. When using a constant value (λ = 0.2), the method tends to underestimate the direct runoff at smaller precipitation totals and to overestimate it at higher precipitation totals (Kozlovská and Toman, 2010; Malý, 2010; Spál et al. 2012).

In the study by Spál and Szolgay (2015)), the estimate of the initial loss coefficient was variable in the range of 0.05 - 0.5 from real precipitation-runoff events in small Slovak river basins. World studies, for example, Mocus (1972) have arrived at coefficients (λ) in the range of 0.013 to 2.1. Hawkins and Khojeini (2000) report values ranging from 0 to 0.0966, while Jiang (2001) reports that 90% of the values were less than value 0.2. Zhi-Hua Shi et al. (2009) on real data in China, they state a coefficient (λ) in the range of 0.010 to 0.154.

For design floods where the duration of the design precipitation (with the constant intensity and duration equals to time concentration) can be relatively short, a value of 0 (small river basins) or a value very close to zero is recommended. In this article, we have not dealt with its precise determination in more detail.

Kozlovská, S.–Toman, F., 2010, Význam indexu předchozích srážek při stanovení kulminačních průtoků metodou CN křivek, In: Konference s mezinárodní účastí: Rizika ve vodním hospodářství 2010. Sborník příspěvků. Brno. 45 – 52. ISBN 978-80-7204-703-1.

Spál, P.–Danáčová, M.–Szolgay, J.–Hlavčová, K., 2012, Stanovenie výšky priameho odtoku pomocou metódy SCS-CN v malokarpatskom regióne. In: Acta hydrologica Slovaca. Roč. 13, č. 1. 260 – 269. ISSN 1335-6291.

Malý, A., 2010, Porovnání výstupů metody odtokových křivek (SCS-CN) s pozorovanými daty z malých povodí. In: Hydrologické dny 2010. 7. konference českých a slovenských hydrologů a vodohospodářů. Hradec Králové, ČR, 25.- 27. 10. 2010

Mockus, V., 1972, Estimation of direct runoff from storm rainfall. In: SCS National Engineering Handbook, Sect. 4, Chapt. 10. 24 s.

Hawkins, R.H.–Khojeini, A.V., 2000, Initial abstraction and loss in the curve number method. (abstrakt), In: Proceedings Arizona Hydrologic Society

Jiang, R., 2001, Investigation of runoff curve number initial abstraction ratio. M. S. thesis, Watershed Management, University of Arizona, Tucson, AZ. S. 120 s

Spál, P., Szolgay, J. 2013. PARAMETRIZÁCIA ODHADU OBJEMU PRIAMEHO ODTOKU NA MALÝCH POVODIACH POMOCOU METÓDY ČÍSEL ODTOKOVÝCH KRIVIEK, Meteorological journal, 2013, Volume 16, Number 2. SLOVAK HYDROMETEOROLOGICAL INSTITUTE

http://www.shmu.sk/File/ExtraFiles/MET_CASOPIS/2013-2_MC.pdf

Section 4.3 - Thanks for your comment, the text has been edited.

Specific (minor) comments

Answer: Thank you for your comment. In our study the hourly rainfall data are available for CLM scenario for the both stations. Finally the results from the CLM scenario were compared to actual measured data. The actual measured rainfall intensities were available for the Liptovská Teplička st. in the hourly time step and for the Vyšná Boca station only daily rainfall data were available.

Answer: Thank you for your comment, the suggestion was accepted (month names that I, II etc). Dividing Figure 5 into smaller images would slightly lose the possibility of direct comparison, if necessary we can make adjustments according to themes (flows, temperature, precipitation separately).

Author Response File: Author Response.docx

Reviewer 4 Report

The study made several confused evaluations.

The objective of the study (1): identify land use change by windstorm. But from the result, only land use change including both natural disasters and deforestation. The study did not show the change as the objective (1).
the “SCS-CN” method is used with daily precipitation. Even few studies have been supposed some modified methods, the study did not used. It is extreme strange to estimate runoff with sort-term precipitation and calculate design floods. Furthermore, the snow issues seem be considered in this study because of low temperature.
The study used “SCS-CN” is lumped model not disturbed one.(3.4 distributed model)
The runoff and discharge are different definition in hydrology.
I believe the analysis of the hydrometeorological data in this study is good.
Two suggestions: (1) focus on identification of land use change or natural disasters using your data. (2) focus on the downscale and validation of short-term rainfall. Choose one of them.

Author Response

The study made several confused evaluations.

The objective of the study (1): identify land use change by windstorm. But from the result, only land use change including both natural disasters and deforestation. The study did not show the change as the objective (1).

Answer: Thank you for you comment. The text was edited directly in the article.

the “SCS-CN” method is used with daily precipitation. Even few studies have been supposed some modified methods, the study did not used. It is extreme strange to estimate runoff with sort-term precipitation and calculate design floods. Furthermore, the snow issues seem be considered in this study because of low temperature.

Answer: For estimation of design floods by SCS-CN method, we have considered extremes precipitation (in the form of short-term design rainfalls). These are dominant in the summer season in Slovakia.

The study used “SCS-CN” is lumped model not disturbed one.(3.4 distributed model)

Answer: Thank you for your comment.

The runoff and discharge are different definition in hydrology.

Answer: Thank you for you comment, we agree. The terms have been corrected in the text of the article

I believe the analysis of the hydrometeorological data in this study is good.

Answer: Thank you.

Two suggestions: (1) focus on identification of land use change or natural disasters using your data. (2) focus on the downscale and validation of short-term rainfall. Choose one of them.

Answer: These suggestions were mentioned in the main objectives study in the introduction. Thank you.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Thank you wery much for your effort with editing the manuscript. All my comments were addressed. I accept all reactions.

I have only two more comments:

Correct numbering of equations,
you use both SCS-CN and CN-SCS abbreviations. You shoud use the first one only - see l. 68.

Author Response

Thank you for raising professional advice for improving our paper.

Thank you very much for your effort with editing the manuscript. All my comments were addressed. I accept all my reactions.

I have only two more comments:

Correct numbering of equations.

Answer: Thank you for your comment. The numbering of equations edited directly in the article

you use both SCS-CN and CN-SCS abbreviations. You should use the first one only - see l. 68.

Answer: Thanks, it was edited.

Author Response File: Author Response.docx

Reviewer 2 Report

Thank you very much for the efforts of revising the manuscript. All my comments and questions have been carefully addressed. I only have two more follow-up short comments. After they are taken care of, I would suggest publishing this work.

1. Figures 4-6: the borderlines of subplots still seem to be broken. I am not sure if this is an error caused by the pdf conversion.

2. I would suggest adding this discussion and the reference into the paper to let the readers have a basic idea of the other components of the water budget:

Hríbik, A. Majlingova, J. Škvarenina and D. Kyselová, Winter Snow Supply in Small Mountain Watershed as a Potential Hazard of Spring Flood Formation, 2009, Bioclimatology and Natural Hazards pp 119-128

Author Response

Thank you for raising professional advice for improving our paper.

Figures 4-6: the borderlines of subplots still seem to be broken. I am not sure if this is an error caused by the pdf conversion.

Answer: It is a problem caused by the PDF conversion.

I would suggest adding this discussion and the reference into the paper to let the readers have a basic idea of the other components of the water budget:

Hríbik,M., Majlingova, A., Škvarenina J. and Kyselová, D. Winter Snow Supply in Small Mountain Watershed as a Potential Hazard of Spring Flood Formation, 2009, Bioclimatology and Natural Hazards pp 119-128

Answer: Thank you for your comment. It was added.

Author Response File: Author Response.docx

Reviewer 3 Report

The revisions made by the authors are adequate.

Author Response

Thank you for raising professional advice for improving our paper.

The revisions made by the authors are adequate.

Answer: Thank you

Reviewer 4 Report

Line 60~64, the author has modified the description of "event-based hydrological CN-SCS model",

But in the section 3.4, SCS-CN methodology, it seems no information about "event-based hydrological CN-SCS model". (no change at all)

Readers can not repeat the study especially in the CN-SCS method with hourly rainfall. Just changed the name is not good enough.

So two tasks should be done.

1.Cite a paper which use CN-SCS method(hourly) which the author used. 2.Validate the simulation (60mins) which can make sure the inferred frequency analysis more believable.

Author Response

Thank you for raising professional advice for improving our paper.

Line 60~64, the author has modified the description of "event-based hydrological CN-SCS model",

But in the section 3.4, SCS-CN methodology, it seems no information about "event-based hydrological CN-SCS model". (no change at all)

Readers can not repeat the study especially in the CN-SCS method with hourly rainfall. Just changed the name is not good enough.

So two tasks should be done.

1.Cite a paper which use CN-SCS method(hourly) which the author used. 2.Validate the simulation (60mins) which can make sure the inferred frequency analysis more believable.

Answer: Thank you for your comment. Yes, we are aware of the fact that the original method was developed for rainfall duration of 24 hours. The authors in the original paper mentioned that the duration of 24h was catchment-size dependent, and this was used to derive the R-R relationships.

see e.g. Engineering Field Manual, 1991, Chapter 2, Estimating runoff and peak discharges, SCS U.S. Government Printing Office,1991, p. 2-72:2-957

Lately, this methodology was adopted in other countries, without considering that the 24 h duration of rainfall overestimated the final direct runoff mostly in smaller catchments. Many authors pointed out the uncertainties in this method, not only concerning the rainfall duration but many others.

Meadows, M.E, 2016. Adjusting NRCS Curve Number for rainfall Durations Less Than 24 hours, Journal of South Carolina Water Resources, Volume 3, Issue 1, Pages 43-47

In our previous studies, we have modified the SCS-CN methodology for small watersheds in Slovakia, looking closer to the modification of parameter lambda. We have used both real and design rainfall-runoff data and altered the lambda values.

Kohnová, S., Karabová, B. and Hlavčová, K., 2015. On the possibilities of watershed parameterization for extreme flow estimation in ungauged basins, Proc. IAHS, 370, 171–175, doi:10.5194/piahs-370-171-2015

Karabová, B., Marková, R., 2013. Testing of regionalization of SCS-CN parameters in the Upper Hron River region, Slovakia. Adolf Patera Seminar 2013, Prague, Czech Republic.

Banasik, K., Rutkowska, A., Kohnová, S., 2014. Retention and curve number variability in a small agriculturals catchment: the probabilistic approach. Water, 6, 5, 1118–1133.

Karabová, B., 2014. Analýza parametrov metódy SCS – CN v podmienkach Slovenska. PhD thesis, SvF-13423-27852, SvF STU Bratislava, 2014, 172 (in Slovak).

Spál, P., Danáčová, M., Szolgay, J., and Hlavčová, K.: The estimation of direct runoff using the Curve Number Method (SCS CN) in the Vištucký creek catchment (in Slovak), Acta Hydrologica Slovaca, 12, 368–376, 2011.

Spál, P. Floods wave estimation for detention basin design on small catchments, 2013, Dissertation Thesis, Slovak University of Technology, Bratislava (in Slovak).

In this study, we proposed to use the duration of rainfall equal to the time of concentration on the catchments. In our case studies the estimated time of concentration was about 1- 1.5 hours so far less than 24 hours. The overestimation of design discharge using the daily rainfall would be too high and not corresponding the reality.

Author Response File: Author Response.docx