Low-Flow Similarities between the Transboundary Lauter River and Rhine River at Maxau from 1956 to 2022 (France/Germany)

Round 1

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

Dear authors,

Thank you for your extensive work. However, I still have some questions:

1) According to the previous comments (perhaps the version was changed after the last authors' responses, which were done for the second application, and now I can't find the reference mentioned):
1.1) “Added in first paragraph of section 5.2, new citation of Hannaford (2011).” – I can’t find this citation.

 

1.2) ”In line 376 which includes citation 30, it is clearly indicated which probability distribution method we used for our study.” - Citation of the toolbox is not equal to the citation of the methodology, if you want to cite user guide, you should add reference on this user guide.

1.3) “SPI is for rainfall, and SSI is based on streamflow. It is not possible to mix the two.” - Of course, input data different, but methodology the same as the process of normalization and output data distribution. maybe you have some citation with explanation why another type of drought classification is better?

2) Citation 24 in section 3.2 – not represent WMO citation, it is another paper.

3) The Standardized Drought Analysis Toolbox (SDAT), as marked in the methodology, does not include the Standardized Streamflow Index (SSI). Instead, this toolbox offers the Standardized Soil Moisture Index (SSI), Standardized Runoff Index (SRI), or Standardized Streamflow Index (SSFI). Please provide some explanation in the methodology (or in the introduction part).

 

4) Figure 3. explanation of all data/colors should be added.

5) It is noteworthy that both the available data of Bobenthal station and Salmbach Pas-sage station (which only had data available until the end of 2020) identified gaps within the low flow events of the Maxau station. - could you explain this sentence.

Best regards

 

 

Author Response

Dear reviewer,

Thank you so much and please find below our reply.

1) According to the previous comments (perhaps the version was changed after the last authors' responses, which were done for the second application, and now I can't find the reference mentioned):
1.1) “Added in first paragraph of section 5.2, new citation of Hannaford (2011).” – I can’t find this citation.

Please find the citation 42.

1.2) ”In line 376 which includes citation 30, it is clearly indicated which probability distribution method we used for our study.” - Citation of the toolbox is not equal to the citation of the methodology, if you want to cite user guide, you should add reference on this user guide.

Please find in the latest version line 300, citation 31, which is the corresponding paper explaining the methodology used by the toolbox. This paper stated the probability distribution which was used in our study.

1.3) “SPI is for rainfall, and SSI is based on streamflow. It is not possible to mix the two.” - Of course, input data different, but methodology the same as the process of normalization and output data distribution. maybe you have some citation with explanation why another type of drought classification is better?

This paper aiming at comparing different approaches in defining hydrological drought period. SSI is the only index recommended by the WMO for streamflow drought analysis.

2) Citation 24 in section 3.2 – not represent WMO citation, it is another paper.

This citation was changed to the Handbook of Drought Indicators and Indices by WMO.

3) The Standardized Drought Analysis Toolbox (SDAT), as marked in the methodology, does not include the Standardized Streamflow Index (SSI). Instead, this toolbox offers the Standardized Soil Moisture Index (SSI), Standardized Runoff Index (SRI), or Standardized Streamflow Index (SSFI). Please provide some explanation in the methodology (or in the introduction part).

Please refer to our explanation in previous round that Standardized Streamflow Index (SSFI) is the same as Standardized Streamflow Index (SSI).

4) Figure 3. explanation of all data/colors should be added.

Please find the explanations in the legend at the bottom of Figure 3.

5) It is noteworthy that both the available data of Bobenthal station and Salmbach Passage station (which only had data available until the end of 2020) identified gaps within the low flow events of the Maxau station. - could you explain this sentence.

Thank you so much for pointing this out. This sentence was rephrased and further explained in lines between 694 to 698.

Best regards

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

This article identifies low flow using three methods: variable threshold method, fixed threshold method, and standardized flow index method, and compares them with the nearby Maxau station in the Rhine River during the same period. Explain the differences between various low flow identification methods and reveal the differences in low flow characteristics between the Laut River Basin and the Rhine River Basin. The research results can promote hydrological drought warning and drought management for water users in the Laut Basin and some areas of the lower Rhine River. But there are still some issues as follows:

1. Lack of quantitative conclusions

2. FR, DE, and CH in Figure 1 can be explained in the comments.

3. The data sources in the study were not clearly explained, such as altitude data, where they came from, and what the resolution was. Further clarification can be provided.

4. There are only three methods for comparison. The original text mentioned that the variable threshold method has stronger robustness, but the accuracy test of each method does not seem to be presented in the results.

5. In the conclusion, it seems that there is no relevant quantitative research on whether "FT responds faster to direct climate or artificial impacts (such as irrigation or dams), while VT responds faster to indirect factors (such as reduced groundwater flow and increased evapotranspiration)" can be quantified.

6. Why choose these three methods for low traffic identification? Can be explained appropriately.

7. It is recommended to cite the latest relevant articles in the literature review section, such as

Compound and successive events of extreme precipitation and extreme runoff under heatwaves based on CMIP6 models；Quantitative Evaluation of Runoff Simulation and Its Driving Forces Based on Hydrological Model and Multisource Precipitation Fusion

Author Response

1. Lack of quantitative conclusions

Quantitative conclusions were added.

2. FR, DE, and CH in Figure 1 can be explained in the comments.

Added in the description of Figure 1.

3. The data sources in the study were not clearly explained, such as altitude data, where they came from, and what the resolution was. Further clarification can be provided.

Data source for Table 1 and Table 2 was added.

4. There are only three methods for comparison. The original text mentioned that the variable threshold method has stronger robustness, but the accuracy test of each method does not seem to be presented in the results.

The paper stated that the VT and FT showed high coherence in defining low flow period over the years. FT is useful defining low flow period for navigation purpose and VT is more appropriate for defining the low flow period for ecological service. There is no right and wrong for each approach regarding each one’s own serving purpose. This paper is aiming at comparing the differences among the three low flow identification methods, hence assisting local integrated water resources management.

5. In the conclusion, it seems that there is no relevant quantitative research on whether "FT responds faster to direct climate or artificial impacts (such as irrigation or dams), while VT responds faster to indirect factors (such as reduced groundwater flow and increased evapotranspiration)" can be quantified.

The conclusion was drawn from the comparison between VT and FT over the Bobenthal station and Salmbacher Passage station from line 564 to line 578. Given the fact that the Salmbacher Passage station was more vulnerable to excessive evapotranspiration and anthropogenic water abstraction, and Bobenthal station was buffered by the forest area and gave more straightforward indication for the headwater storage condition. And FT of the Salmbacher Passage station gave out low flow signals three months earlier than the Bobenthal station. We argue that the fixed threshold reacts faster to direct climatic or artificial impacts such as water abstraction whereas the variable threshold reacts more strongly to more complex signals issued from indirect factors such as decreasing subsurface flow. This is typical for small headwater catchments that are more sensitive to flow disturbances. Other research based on European catchments agrees that the VT approach is more flexible when dealing with anomalies in low flow conditions (citations 39 and 40).

6. Why choose these three methods for low flow identification? Can be explained appropriately.

Defining low flow periods using predefined threshold approach has been long employed by hydrologist. And recent studies showed a paradigm shift by including anomalies into drought period identification, which is the Standardized Streamflow Index (SSI). Based on our current knowledge, these three methods covered the commonly used low flow identification methodologies of both academic research and government management approach.

7. It is recommended to cite the latest relevant articles in the literature review section, such as

Compound and successive events of extreme precipitation and extreme runoff under heatwaves based on CMIP6 models; Quantitative Evaluation of Runoff Simulation and Its Driving Forces Based on Hydrological Model and Multisource Precipitation Fusion.

New citation added.

 

Best regards

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

Comments and Suggestions for Authors

1. Title of the manuscript is too general; it should be more specific.

2. In the Introduction there is a statement: "the objective of this research is to provide comparisons among different low flow identification approaches and their applicability to different water use sectors". No specific methods are proposed to address the applicability of the approach to different water use sectors. Just an "analysis" does not cover the problem.

3. Section 3. Identification Methodology is too long; methods developed by other authors should be cited, and only modifications of commonly used approaches should be described.

4.The aim of the study mentioned in the abstract as : "revealing particularities of the Lauter catchment compared to the Rhine  River" is too general and does not provide clear purpose of the study.

5. Novelty of the study is missing.

6. Conclusions of the study do not refer to the possibility of use of low flow indicators applied to different water use sectors.

7. Final Remark: In my opinion, the manuscript requires major revisions by re-formulating the title, abstract, main objectives, methodology tuned to objectives and final conclusions.

Author Response

1. Title of the manuscript is too general; it should be more specific.

We changed the title.

 

2. In the Introduction there is a statement: "the objective of this research is to provide comparisons among different low flow identification approaches and their applicability to different water use sectors". No specific methods are proposed to address the applicability of the approach to different water use sectors. Just an "analysis" does not cover the problem.

We modified the sentence. This paper aims to perform fundamental research on the different behaviors of different low flow indices in identifying low flow events. However, in the discussion we propose and recommend the most appropriate methods that could be applied for a small basin like the Lauter and a regional scale basin at Maxau on the Rhine. 

 

3. Section 3. Identification Methodology is too long; methods developed by other authors should be cited, and only modifications of commonly used approaches should be described.

We are in a dilemma here since the other reviewer asked for more details in the methodology. We included the calculation steps in the methodology which are essential for understanding the methods used. According to the demand by the second reviewer to further clarify some of these steps, we have had to include more detail. We prefer if the editor takes the final decision.

 

4. The aim of the study mentioned in the abstract as : "revealing particularities of the Lauter catchment compared to the Rhine  River" is too general and does not provide clear purpose of the study.

This has been rephrased to “revealing the differences in low flow characteristics of the Lauter catchment compared to the Rhine River” to clarify our objectives.

 

5. Novelty of the study is missing.

We have pointed out in several parts of the article why the article is novel. There are four major novel elements in our paper:

• No comprehensive comparison among different low flow indices has been made so far for river basins in this region. This has been specifically demanded in a recent article by Stahl et al. in reference number 12 “The Challenges of Hydrological Drought Definition, Quantification and Communication: An Interdisciplinary Perspective”. This reference has been added.
• No low flow or hydrological studies exist so far for the Lauter River catchment.
• No comparison has ever been made between a small, local tributary basin of the Rhine (i.e. the Lauster) and a close-by station on the Rhine River itself.
• No long-term low flow studies exist for a small mid-European catchment putting the extreme 2022 drought into a long-term perspective since the 1950s.

6. Conclusions of the study do not refer to the possibility of use of low flow indicators applied to different water use sectors.

This paper is not comparing different water management strategies of different water use sectors. It therefore does not aim at comparing different low flow indices for individual water sectors, but rather for water resource management in general. It is the first article that compares different low flow indices in this region and the conclusion can therefore support local water managers. We did indicate the general usefulness of certain indicators for navigation e.g. in the Rhine or ecosystem services in the Lauter. 

 

7. Final Remark: In my opinion, the manuscript requires major revisions by re-formulating the title, abstract, main objectives, methodology tuned to objectives and final conclusions.

The reviewer gives suggestions from a water resource management perspective, but this paper is focused on low flow indices analysis and hydrological drought identification. We have re-written major parts of the article in response to all 3 reviewer comments.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have done a great job and presented interesting material in the article. This paper presents the analysis of different approaches for drought identification. The investigation centered around a small river catchment (a tributary of the Rhine). The article is generally well-prepared but requires serious revisions and changes:

1. Authors should consider a shorter article title.

2. Why did you choose the Lauter catchment?

3. Please provide a clearer explanation of low flow and hydrological drought, maybe their differences and interconnection because they are different events (low flow can be observed every year - it is part of the river regime, but hydrological drought is an event that can occur in different years). After analyzing your paper, I propose using only hydrological drought, as you used low flow indexes to identify hydrological drought, but this is only a suggestion.

4. Is it suitable to compare a small river with the Rhine?

5. Why did you choose an 80% exceedance rate? Zimbabwe and France/Germany have different climate conditions and water consumption. Also, the article you refer to has other thresholds (lines 243-244).

6. It will be good to see the equation for QMNA 5 index.

7. Could you describe more clearly the calculation of the Fixed Threshold?

8. Could you provide a reference on SSI and its methodology?

9. You noticed that SSI is based on SPI, so why did you use another type of drought classification?

10. Reference under number 33 describes the Standardized Soil Moisture Index; are you sure that it is similar to your Standardized Streamflow Index (SSI)? Maybe you used Standardized Streamflow Index (SSFI), which was proposed by Modarres in 2007?

11. Please check lines 398-399; this text does not correspond to Table 4.

12. Authors should depict periods with no data on Figures 4 and 5; at the moment, it looks like the periods without data didn't have drought events.

13. Please check page numeration.

14. How can you explain the big difference in the average duration FTD between Bobenthal and Salmbacher Passage stations (table 5)?

15. The trendlines are not displayed correctly; also, please add them to the legend in Figure 6.

16. In table 6, we can notice the big difference for the Exceptional class between Lauter River and Rhine River; maybe you have some assumptions?

17. It will be better to have some figures with threshold examples for one WGS during one year.

18. Could authors explain why SSI-1 is not suitable? It is not noise; it is just months' classification (line 630). When you use a bigger accumulation period, it works like smoothing, so an accumulation period of 12 months looks more suitable. But it is only the result of a specific classification of months according to the river's discharge.

19. Is it suitable to compare two threshold methods, which use monthly data, with the index approach, which uses data with a duration of 1 year? In such a case, the most extreme drought always will be after a few months after the real extreme drought. Maybe you should consider SSI with a shorter accumulation period?

20. Please add more references to the discussion; it would be great to see some comparison between your results and other authors.

21. In subdivision 5.2., again, is it suitable to compare a small river with the Rhine (which has a watershed area more than 100 times bigger)?

22. In lines 919-220, across the paper, you described only one tributary of the Rhine River, so how can you be sure that other tributaries will show the same results? For example, in the Lauter catchment, you noticed a drought event, also from the other side of the Rhine River, you will find another small catchment that will show a wet event (for Rhine, it is normal because its watershed covers a huge area with different climate regimes and anthropogenic impact). In such a case, can you say that these tributaries predict drought in the Rhine?

Also, the authors should better emphasize the novelty and relevance of the work, what makes it unique and what problems it solves

Best regards

Comments on the Quality of English Language

Authors should revise the work for minor errors and in some cases use a more scientific description.

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions highlighted in red in the re-submitted files.

 

1. Authors should consider a shorter article title.

We changed the title.

 

2. Why did you choose the Lauter catchment?

My PhD is financed by the Interreg project RiverDiv which focuses particularly on the rural Lauter catchment. As a transboundary catchment between France and Germany, this project is a pioneering example for cross-national water resource management.

 

3. Please provide a clearer explanation of low flow and hydrological drought, maybe their differences and interconnection because they are different events (low flow can be observed every year - it is part of the river regime, but hydrological drought is an event that can occur in different years). After analyzing your paper, I propose using only hydrological drought, as you used low flow indexes to identify hydrological drought, but this is only a suggestion.

A clearer explanation was added in the introduction section. The terminology was unified to low flow.

 

4. Is it suitable to compare a small river with the Rhine?

We are comparing the Lauter river with the nearest station on the Rhine River. Hence, focusing on the Upper Rhine valley particularly. Although the Rhine is rectified and has many sluices, flow is not regulated by dams in the navigated section between Germany and France that we are focusing on. Thus there is no drought control possible.

 

5. Why did you choose an 80% exceedance rate? Zimbabwe and France/Germany have different climate conditions and water consumption. Also, the article you refer to has other thresholds (lines 243-244).

The citation was employed for referring to a generic statement “For perennial streams, the exceedance rate is normally set within the range of 75% to 90%”, the reference number 21 “Revisiting Hydrological Drought Propagation and Recovery Considering Water Quantity and Quality” also echoed in this statement. 80% was chosen here to remain consistent with the QMNA 5 index in the fixed threshold method, as QMNA 5 index is commonly used in the local water resource management.

 

6. It will be good to see the equation for QMNA 5 index.

More descriptions were added regarding the QMNA 5 calculation.

 

7. Could you describe more clearly the calculation of the Fixed Threshold?

More descriptions were added and new Figure 2 was added to better describe the calculation.

 

8. Could you provide a reference on SSI and its methodology?

Citation 30 “A Generalized Framework for Deriving Nonparametric Standardized Drought Indicators” provides a methodology for calculating SSI. All standardized drought indices follow the same calculation procedure.

 

9. You noticed that SSI is based on SPI, so why did you use another type of drought classification?

SSI is independent of SPI. Both are commonly used tools for extracting deviations from normal in the time series. The important difference between the two is that SSI uses the streamflow series as its only input, whereas SPI uses precipitation time series as its only input.

 

10. Reference under number 33 describes the Standardized Soil Moisture Index; are you sure that it is similar to your Standardized Streamflow Index (SSI)? Maybe you used Standardized Streamflow Index (SSFI), which was proposed by Modarres in 2007?

Reference 33 was a bit misleading for the abbreviation. SSI is commonly used for denoting the Standardized Streamflow Index, which means exactly the same as the SSFI. And the abbreviation SSI is more prevailing nowadays than SSFI. See reference number 13 “The Challenges of Hydrological Drought Definition, Quantification and Communication: An Interdisciplinary Perspective”.

 

11. Please check lines 398-399; this text does not correspond to Table 4.

These lines stated that VTM generates higher maximum deficit volume than VTD which corresponds to Table 4.

 

12. Authors should depict periods with no data on Figures 4 and 5; at the moment, it looks like the periods without data didn't have drought events.

Starting time was added in the labels for each station.

 

13. Please check page numeration.

Modified.

 

14. How can you explain the big difference in the average duration FTD between Bobenthal and Salmbacher Passage stations (table 5)?

Based on FTD index, Bobenthal has longer average duration but less total events counts, whereas Salmbacher Passage station has shorter duration but more total events counts. Our hypothesis is that the Bobenthal station receives stable groundwater recharge as it is located at the exit of the forested zone. However, the Salmbacher Passage station is located downstream of a major town and agricultural area which introduces more fluctuations into its streamflow. Therefore, when employing the fixed threshold method, the Salmbacher Passage station gives more separate drought events but of shorter duration. We are in the process of building the hydrological model of the Lauter catchment to further investigate the water balance and provide more comprehensive explanations in our next publication.

 

15. The trendlines are not displayed correctly; also, please add them to the legend in Figure 6.

Legend modified.

 

16. In table 6, we can notice the big difference for the Exceptional class between Lauter River and Rhine River; maybe you have some assumptions?

Thanks for pointing that out. It is noticeable that the Lauter river has a relatively higher percentage for the Extreme class but a lower percentage for the Exceptional class compared with the Rhine. The difference is mainly caused by the time period between 1971 and 1978. The streamflow of the Lauter river was around 1 m³/s during that time period. We presume that these low values are due to the limitations of the measuring equipment that was unable to capture the entire picture of low flow dynamics during drought events. We initially assumed that the difference was due to the statistical analysis but couldn’t reach the conclusion that the Rhine river is more prone to drought conditions.

 

17. It will be better to have some figures with threshold examples for one WGS during one year.

Some new, clearer graphs (Fig. 2) have been added.

 

18. Could authors explain why SSI-1 is not suitable? It is not noise; it is just months' classification (line 630). When you use a bigger accumulation period, it works like smoothing, so an accumulation period of 12 months looks more suitable. But it is only the result of a specific classification of months according to the river's discharge.

Thanks for the suggestion. The paragraphs regarding SSI-12 index was modified. We could not observe any significant percentage change (above 0.1%) in the extreme or exceptional drought class when increasing the accumulation period from 1 month to 12 months, neither for the Rhine River nor for the Lauter river. Citation 27 mentions that an appropriate time scale should be chosen to better represent the catchment drought characteristics. We considered that by combining the SSI-12 index with the threshold approach we could better support local water managers in drought severity classification by including antecedent drought conditions.

 

19. Is it suitable to compare two threshold methods, which use monthly data, with the index approach, which uses data with a duration of 1 year? In such a case, the most extreme drought always will be after a few months after the real extreme drought. Maybe you should consider SSI with a shorter accumulation period?

Please refer to our answer to the previous question. We could not observe any significant percentage change (above 0.1%) in the extreme or exceptional drought class when increasing the accumulation period from 1 month to 12 months, neither for the Rhine River nor for the Lauter river. We suggest using the threshold approach for faster anomalies detection and the SSI-12 index for better drought severity classification.

 

20. Please add more references to the discussion; it would be great to see some comparison between your results and other authors.

There are very few articles available comparing different low flow indices. We added another three relevant publications, reference number 37, 40, 41 in the discussion.

 

21. In subdivision 5.2., again, is it suitable to compare a small river with the Rhine (which has a watershed area more than 100 times bigger)?

We are fully aware of the differences, however drought phenomena are much more wide spread than floods and thus are easier to compare over large regions over the Upper Rhine region. Very few structures control the river flow of the Rhine along the navigable reach between France and Germany above the confluence with the Lauter. We are suggesting here that observing the drought signals of the headwater catchment of this region can help providing early drought diagnosis for the Rhine River since the reactions in the confluences set on earlier than in the Rhine.

 

22. In lines 919-220, across the paper, you described only one tributary of the Rhine River, so how can you be sure that other tributaries will show the same results? For example, in the Lauter catchment, you noticed a drought event, also from the other side of the Rhine River, you will find another small catchment that will show a wet event (for Rhine, it is normal because its watershed covers a huge area with different climate regimes and anthropogenic impact). In such a case, can you say that these tributaries predict drought in the Rhine?

Our hydrological analyses of all the main tributaries bordering the Rhine in Alsace (France) and Baden Württemberg (Germany) show that hydrological drought events are extremely similar for the past 20 years. This is because droughts are large-scale, long-lasting phenomena that have a direct influence on the streams, whether draining the Vosges or Black Forest.

Reviewer 3 Report

Comments and Suggestions for Authors

This article conducts an analysis of hydrological drought periods in the Lauter River and Upper Rhine, employing three methods: the variable threshold method, the fixed threshold method, and the Standardized Streamflow Index.

While the article features a well-organized structure and is generally well presented (though some figures -figure 2- and sections require editing), it falls short in terms of novelty, limiting its appeal to a broader audience. All methods utilized are established methodologies, lacking innovative implementations. Methodologically, it fails to introduce anything beyond what is available in the cited references.

On the other hand, the presentation of the results is, from my perspective, overly lengthy and, at times, less rigorous, making it tough to extract practical conclusions applicable in new contexts. Furthermore, conclusions often rely on conjecture (for instance, lines 760, 777, 805, etc) undermining the overall analytical strength.

In conclusion, the article lacks a substantial scientific contribution, and the discussions need to be more concise and precise to facilitate the extraction of novel insights.

Author Response

This article conducts an analysis of hydrological drought periods in the Lauter River and Upper Rhine, employing three methods: the variable threshold method, the fixed threshold method, and the Standardized Streamflow Index.

While the article features a well-organized structure and is generally well presented (though some figures -figure 2- and sections require editing), it falls short in terms of novelty, limiting its appeal to a broader audience. All methods utilized are established methodologies, lacking innovative implementations. Methodologically, it fails to introduce anything beyond what is available in the cited references.

 

We have redone Fig. 2 and added several new figures to facilitate interpretation for a larger audience.

We have pointed out in several parts of the article why the article is novel. There are five major novel elements in our paper:

• No comprehensive comparison among different low flow indices has been made so far for river basins in this region. This has been specifically demanded in a recent article by Stahl et al. in reference number 12 “The Challenges of Hydrological Drought Definition, Quantification and Communication: An Interdisciplinary Perspective”. This reference has been added.
• There are no methodological studies whatsoever available on low flow or hydrological drought analyses in the Lauter or any comparable catchment nearby in France or Germany.
• No low flow or hydrological studies exist so far for the Lauter River catchment.
• No comparison has ever been made between a small, local tributary basin of the Rhine (i.e. the Lauster) and a close-by station on the Rhine River itself.
• No long-term low flow studies exist for a small mid-European catchment putting the extreme 2022 drought into a long-term perspective since the 1950s.

 

On the other hand, the presentation of the results is, from my perspective, overly lengthy and, at times, less rigorous, making it tough to extract practical conclusions applicable in new contexts. Furthermore, conclusions often rely on conjecture (for instance, lines 760, 777, 805, etc) undermining the overall analytical strength.

In conclusion, the article lacks a substantial scientific contribution, and the discussions need to be more concise and precise to facilitate the extraction of novel insights.

 

We are in a dilemma here since the other reviewer asked for more details in the methodology and results. We included the calculation steps in the methodology and clarified the results. We hope that these modifications help in extracting more practical conclusions. However, the aim, as the title suggest, is a more technical, methodological approach, not a water manager approach.

This paper is not comparing different water management strategies of different water use sectors. It therefore does not aim at comparing different low flow indices for individual water sectors, but rather for water resource management in general. It is the first article that compares different low flow indices in this region and the conclusion can therefore support local water managers. We did indicate the general usefulness of certain indicators for navigation e.g. in the Rhine or ecosystem services in the Lauter. 

We prefer if the editor takes the final decision.

The five points listed above should resolve the issues of how our paper provides a substantial scientific contribution. We hope that the modifications allow for extracting novel insights, for example the role of the Lauter in its low flow predictive for the Rhine.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

1. The manuscript includes confusing statements, e.g.: in Abstract: "The Lauter catchment is a mostly natural, forested catchment but is strongly influenced by past and present cultural human activities." Comment: Is it natural or strongly impacted catchment??

2. Title is not informative enough; the sentence is wrongly constructed.

3. In Abstract: "It appears that the FT reacts faster to direct climate or anthropogenic impacts whereas VT is more sensitive to indirect factors such as decreasing subsurface flow which is typical for small headwater catchment, such as the Lauter where flow dynamics react faster to flow disturbances." Comment: Differentiation between climate and anthropogenic factors was not investigated in this contribution, neither the subsurface flow. Conclusions cannot be drawn based on the suppositions.

4. "Therefore, we recommend that a thorough comparison among different low flow indices should be performed before giving recommendation for local water resources regulations." This does not sound as a recommendation.

5. In Discussion: "Since more “flash droughts” and “flood-drought cascading events” are disturbing the flow rate time series due to climate change, VT is more robust than FT in defining the beginning and the end of low flow periods." Flash droughts or cascading events were not investigated in this study. Thus, conclusions cannot be drawn based on the suppositions.

In my opinion, the main defficiency of this manuscript is lack of the synthetic form of presentation, and lack of applicability of results. The revised form of this manuscript did not improve the quality of the contribution.

 

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you to the authors for responding to the comments and taking them into account. At this point, the work looks significantly better. However, there are still a few comments that need clarification or explanation:

4. You are right, but is it suitable to compare a WGS with a catchment area of 345 km2 with a WGS catchment area of 50196 km2? This is the main question; the difference is too big, leading to different processes and impacts of hydrological drought on the rivers.

8. Great, but in the text, it is unclear that the authors used the calculation methodology from this citation. It only mentions some researchers "assessing the accuracy of SSI."

9. Is there any significant change in the methodology of SSI calculation compared to SPI (besides using runoff data instead of rainfall)? If not, what is the reason for using another type of drought classification?

10. You are correct; a similar index is SDI, and sometimes the same calculations are found under the abbreviation RDI. But the main question is: why did you use drought categories from reference 33, which describes only indexes based on precipitation and soil moisture without river runoff data? Your work concerns hydrological drought. In such a case, a more common drought classification for SSI would be better (e.g., Mild drought 0.00 to −0.99; Moderate drought −1.00 to −1.49; Severe drought −1.50 to −1.99... similar to what the WMO uses for SPI). If you disagree, please provide an explanation why you chose the drought classification from article number 33 in your reference list.

22. I fully agree with your statement if we talk about a huge area with the same physical and geographical factors, but the upper part of the Rhine is located in different climate conditions. Drought in different catchments can be similar in general, but their initiation can be different. For example, WGSs on the Lauter river may show hydrological drought one month later than data from the Tamina river, or vice versa. This is especially true if the catchments of these rivers have different climate conditions and flow regimes, which can offset the impact of both catchments and give average discharge for the Rhine (especially if we contrast the catchments from the lowland/low mountain catchment to the mountain catchment).

Of course, I don't know this region as well as you do, but like any other researcher, I understand that river discharge in the Maxau WGS is influenced not only by the Lauter river and its nearest catchments but also by rivers from Switzerland that flow into the Rhine, with different climate conditions and flow regimes. Could you provide some facts to confirm your statements that the Lauter river in general can show similar drought periods with other rivers from the upper part of the Rhine, especially from the mountain region, or provide some data showing the small impact of river discharge from the Swiss part of the Rhine (or the mountain part, for example, near Basel) compared to discharge at the Maxau WGS? (It could be a previous paper or a paper by another researcher).

And based on your answer, I can now be sure that data from Lauter River WGSs can indicate hydrological drought in this region, especially in the Rhine tributaries. However, once again, the Rhine is much bigger and has another flow regime.

Best regards,

Reviewer 3 Report

Comments and Suggestions for Authors

After carefully reviewing the revised manuscript, I must acknowledge that it represents an improvement over the original submission. The revisions have enhanced the clarity of figures and clarified certain results. 

However, I must emphasize that the manuscript still suffers from a significant flaw identified in the original version. It fails to effectively synthesize its results.I strongly recommend a more focused effort to synthesize the results and discussion, emphasizing the most relevant findings supported by the methodology. From my perspective, streamlining the content in this manner will significantly enhance the clarity and impact of the manuscript.