Next Article in Journal
The Use of Hydrodynamic Models in the Determination of the Chart Datum Shape in a Tropical Estuary
Next Article in Special Issue
Temporal Effects of Groundwater on Physical and Biotic Components of a Karst Stream
Previous Article in Journal
Allocating Water Environmental Capacity to Meet Water Quality Control by Considering Both Point and Non-Point Source Pollution Using a Mathematical Model: Tidal River Network Case Study
Previous Article in Special Issue
Determining the Discharge and Recharge Relationships between Lake and Groundwater in Lake Hulun Using Hydrogen and Oxygen Isotopes and Chloride Ions
Article Menu
Issue 5 (May) cover image

Export Article

Open AccessArticle

Peer-Review Record

Quantification of Temporal Variations in Base Flow Index Using Sporadic River Data: Application to the Bua Catchment, Malawi

Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Reviewer 4: Anonymous
Water 2019, 11(5), 901; https://doi.org/10.3390/w11050901
Received: 30 March 2019 / Revised: 25 April 2019 / Accepted: 26 April 2019 / Published: 29 April 2019
(This article belongs to the Special Issue Groundwater-Surface Water Interactions)

Round 1

Reviewer 1 Report

Dear authors. Please see the attached document for my comments.

Comments for author File: Comments.pdf

Author Response

Please see the attached word document providing our responses to your comments.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper you want to published can be improved. Firstly take care about the language and the written text (some missing letters I found in your text). I Have some suggestions to improve your research:

row 34...season...you refer to climatic season?

Rows 44-484 here we can find just  general definitions. It is not necessary to put this kind of definitions in this research.

Row 57-59... BFI equal with 0.0...in fact it is just theoretical without real application

entire chapter 2.3. refers also to general definitions and it is not connected to the paper and also you have some missing letter in the title

row 296 at the station 5D2 you find annual values for BFI  equal with 1...another theoretical value in fact every river has small contribution from underground. Look over the data maybe you will find some errors

figure 2 and 3 show what you describe in the table 4 and do not add anything to the analysis.

Chapter 3.3.2. Comments on the source of base flow...a lot of speculations without scientific support

Rows 432-440 must be rewritten because it has nothing to do with the results

Conclusion also must be rewritten  and synthetically presented

Author Response

Please see the attached word document providing our responses to your comments.

Author Response File: Author Response.docx

Reviewer 3 Report

Dear Authors,

the scope of the paper is to perfom a baseflow separation by  using an automated , easy to use, easly accessible and free filter method to exchange knowledge with the Government of Malawi. The paper is well organized and respects the format of the journal. The introduction of the paper must be improved  adding some references about others methods such as the mass balance filtering, specifing the preferences about the filtering method. In the paragraph 2.4 it is not clear what decision procedure  supported the choice of using the BFI Programm.

For all these considerations, the paper will be accepted after minor revision suggested in the following:

Line 47: modify "for example;"in  "for example:" 

Line 49: Introduce others references on the hydrograph separation methods, such as the filtering and the mass balance

Line 137: delete "which"

Figure 1: Please to Include the figures 1b and 1c in the 1a, in the upper left corner. The symbol of the weir must be enlarge because it is not visible in the picture. Include the sub-catchments units.

Line 183: What types of "direct analysis" are perfomed? Please, Specify it

Line 185: Change "desription" in to "description"

Lines 188-190: THis sentences is not clear "Second, the BFI is calculated as the ratio of the flow 188 under the baseflow hydrograph; the baseflow volume, to the flow under the river hydrograph; the 189 total flow volume [9]". Include the formula of the BFI in text to improve the readability of the text.

Lines 215-216:How did you compare the diverse filtering tools? Which  parameters did you use as input of the filtering method? Why did not use another method too, to compare the results and to better substain  your findings? 

 Best regards

Author Response

Please see the attached word document providing our responses to your comments.



Author Response File: Author Response.docx

Reviewer 4 Report

Major comments:

The authors need to pay attention to the interpretation of results. High BFI in dry season does not necessarily mean that "groundwater sustains flow". You can have nearly no flow in the river but BFI equals to 1 based on the method the authors used to separate baseflow. Also, dry season BFI derived using UKIH should not have any trend because it is supposed to be very close to 1. So, I do not believe the premise of the BFI trend tests for dry season is correct. Actually, I think the BFI analysis should be conducted only for the wet season, not to lump different flow regimes. Additionally, the analysis should be accompanied with trend analysis of rainfall (and possibly ET or other water balance components).

On experimental design, baseflow separation is subjective. I strongly recommend the authors to compare BFI derived from different baseflow separation methods to confirm consensus on the trends. UKIH is a filtering approach, other baseflow separation methods may include field-based method (Saraiva Okello et al. 2018), digital filter (Eckhardt 2005), recession analysis (Wittenberg & Aksoy 2010), and combinations (Mei & Anagnostou 2015). Also, those different method should be reviewed in the introduction.

Substantial revisions on the manuscript are needed and I suggest a major revision at this time. The authors can find some specific editing in the annotated manuscript. Hope these help.

Comments for author File: Comments.pdf

Author Response

Please see below for our responses to your major comments. In addition, please see the attached PDF for individual responses to your annotations.


*Reviewers comments have been copied and pasted here and are in black. Comments comprise overview remarks and further comments by annotated manuscript. The author's response comments have added in red. Where we have made changes to the paper, tracked changes have been used.

 

Overview remarks:

Firstly, thank you kindly for your time and consideration of our paper. We appreciate all your comments and feel by addressing them the paper has been significantly improved. As such, we are extremely grateful for your input.

 

Major comments: The authors need to pay attention to the interpretation of results. High BFI in dry season does not necessarily mean that "groundwater sustains flow". You can have nearly no flow in the river, but BFI equals to 1 based on the method the authors used to separate baseflow.

Response: Revised. A high BFI in the dry season, based on the methods used, indicates a high proportion of baseflow in the river. Many rivers in Malawi flow in the dry season. However, you are correct, in many cases, the flow is significantly reduced and, in some rivers, to nearly zero flow (even though the BFI appears close to 1.0). It's very important to highlight in the paper as we do not mean to state ‘groundwater water sustains flow’, more ‘baseflow sustains any dry season flows which exist. As such we have made changes to the sections which mention ‘baseflow sustaining flows’; i.e Line 21, Line 37 and the conclusion.


Also, dry season BFI derived using UKIH should not have any trend because it is supposed to be very close to 1. So, I do not believe the premise of the BFI trend tests for the dry season is correct.

Response: Noted. In this study, we expected the dry BFI to be close to 1 due to the limited amounts of rainfall which fall in Malawi. As you can see in the results, dry season BFI for all gauges are >0.80 (this includes gauge 5C1 presented in Table 2, and the other gauges 5D1, 5D2, 5D3, 5E6 and 5F1 as presented in the supplementary material).

 

One of the study’s objectives was to identify trends in dry season BFI. Identifying trends in BFI is important for sustainable catchment management as changes in baseflow can indicate an imbalance in the water system which is crucial information for policymakers and those managing the catchment. These imbalances may be caused by natural factors (i.e. natural variations in BFI, climate and weather variations etc) or human activities (i.e. deforestation, over-abstraction of water, artificial recharge to groundwater).

 

As mentioned above, dry season BFI was found to be >0.8 for all gauges and no trends were found in the dry season. Although one might ‘assume’ that there will be no trend, scientific evidence is required to support this assumption. Further, an increasing or decreasing trend could have been found.  For example, a statistically significant ‘increasing’ trend in BFI could have been found (e.g. a BFI changing from 0.8 to 1). Alternatively, a statistically significant ‘decreasing’ trend in BFI could have been found (e.g. a BFI changing from 0.8 to say 0.6). Although these changes may appear minimal, baseflow behavioural change is indicated and the potential drivers for such changes should be investigated and the magnitude of the trend quantified.

 

Actually, I think the BFI analysis should be conducted only for the wet season, not to lump different flow regimes.

Response: Noted. Many countries, for example, Malawi, do not have a full understanding or appreciation of seasonal variations in baseflow. For example, the fact dry season BFI generally increases significantly from the wet season and annual BFI. The introduction highlights studies from Malawi which focus on annual values of baseflow and BFI and identifies the need to assess seasonal variations. Identification of the differences in seasonal BFI is very important for sustainable catchment management, and policy makers and water managers should be as informed as possible. Further, as discussed in the final paragraph of section 3.3, the BFI has many applications in hydrology and hydrogeology and use of an incorrect BFI could lead to inaccurate assessment on water resources (for example, using an annual value of 0.5 in a dry season assessment, where the dry season BFI is actually 0.9 could result in inaccurate results which could be elemental to water resources allocation in the area). As such, the authors feel presenting dry season BFI is extremely important and is required to draw attention to seasonal variations.

 

Additionally, the analysis should be accompanied with trend analysis of rainfall (and possibly ET or other water balance components).

*Response: Noted. Trend analysis of rainfall was conducted as referred to in section 3.4, second paragraph (of the revised manuscript) and the supplementary information, Table 5.  There was not enough groundwater data for trend analysis.

 

On experimental design, baseflow separation is subjective. I strongly recommend the authors to compare BFI derived from different baseflow separation methods to confirm consensus on the trends.

*Response: Noted. We agree baseflow separation is subjective and recognise the merit in using different baseflow separation methods to confirm BFI values. However, we have decided not to address this recommendation for the following reason.

 

This paper is an output of Laura Kelly’s (corresponding author) PhD research on ‘Developing a framework to assess temporal variations in a baseflow-A case study of Malawi.  Although comparing BFI values from different methods was initially considered, it was decided against due to time constraints. As research progressed, it became apparent that the crucial objective was the selection of an appropriate implementation tool based on specific criteria, not the filtering method. The selected tool, BFI Programme, utilises only one filtering method (some tools provide the option to use different filtering methods). We did consider using another implementation tool, which used the same filtering technique to compare values, but this then deviated from the PhD objectives and time has become a critical issue. We are hopeful that future research will use another method to compare our findings. We hope you understand and appreciate our reasons.

 

UKIH is a filtering approach, other baseflow separation methods may include field-based method (Saraiva Okello et al. 2018), digital filter (Eckhardt 2005), recession analysis (Wittenberg & Aksoy 2010), and combinations (Mei & Anagnostou 2015). Also, those different method should be reviewed in the introduction.

Response: Revised. A new paragraph has been included in the introduction to describe and reference other methods.

 

Additional information of relevance to this comment: As mentioned above, this paper is an output of Laura Kelly’s PhD research.  A key objective of the PhD was the selection of an appropriate method for a recommendation in the final framework. A chapter in Laura’s thesis will address this objective and reviews all methods for investigation of baseflow including desktop methods, field surveys and site investigation (and associated submethods).  The method was required to meet a strict set of criteria and resulted in the selection of baseflow separation (with BFI). As such, the starting point for this paper is the use of baseflow separation.

 

This paper did not compare the diverse filtering tools, instead, it compared the implementation tools available. The selection of the tool was very important in this study to allow the exchange of knowledge with the Government of Malawi. As such, its selection took precedence over the comparison of filtering methods and was required to meet certain criteria. New text has been added to provide more clarity and reinforce this point and is included as section 2.3 ‘decision procedure for selection of baseflow separation method and implementation tool’. A comparison table listing the considered tools against the criteria has also been included at the request of another reviewer.

 

Further comments by annotated manuscript:

Substantial revisions on the manuscript are needed and I suggest a major revision at this time. The authors can find some specific editing in the annotated manuscript. Hope these help.

Response: Responses have been individually added to the annotated manuscript.

 

 

Thank you again for your valuable comments, we hope that you are happy with our responses.




Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Your manuscrip was improved with my suggestions and I agree to be published în this form

Author Response

Comment - Your manuscript was improved with my suggestions and I agree to be published în this form


Response - Thank you kindly. Again, we appreciate all your comments.

Reviewer 4 Report

Please see the attached.

Comments for author File: Comments.zip

Author Response

*Reviewers comments have been copied and pasted from the annotated PDF (on the main paper) to here and are in black. The author's response comments have added in red. Where we have made changes to the paper, tracked changes have been used.

Thank you kindly for your time and consideration of our paper. We appreciate all your comments and feel by addressing them the paper has been significantly improved. As such, we are extremely grateful for your input.

 

Annotated manuscript (main paper):

Unfortunately, we could not upload two PDF response files. Hence, the responses to the annotation on the main manuscript are included here.

 

Line 12.

Response: Revised. ‘The’ removed.


Line 16.

Response: Revised. ‘MK’ removed.


Line 28.

Response: Revised. ‘BFI’ removed and replaced by Base Flow Index.


Line 33.

Response: Revised. ‘metling’ added.


Line 59. ‘The baseflow is commonly expressed in numerical fashion as the BFI’ I don't think so. Baseflow and BFI are two different concepts. Baseflow typically refers to the volume or the baseflow rate.

Response: Revised. Incorrect wording used here. We have removed this sentence and added a more appropriate sentence.

Line 62.

Response: Revised. ‘The’ removed.


Line 120. ‘BFI approaches’. What is this referring to?

Response: Revised.  Unclear wording. Removed.


Line 120. ‘Observations’. It is just an estimation.

Response: Revised. Used the word estimation instead.


Line 197. ‘Data was’ "Data" is always in plural form in scientific writing.

Response: Revised.

 

Line 213. You can perform graphical methods manually or using a computer program. It is just the user's choice when it comes to the method implementation.

Response: Revised. Sentence removed and the previous line revised.


Line 232/Table 1. How do you define "Easy to use"? What is "easy"? I feel like this is too subjective.

Response: Revised. Easy to use was defined as 'requiring minimal time and training to use'. For example, Flow Screen for R requires a lot of time to learn how to use due to its programming like nature. RAP requires initial training sessions. We have removed 'easy to use' (from the table and text, and rephrased as 'requires minimal training to use'.


Line 232/Table 1. I don't understand why unknown is presented here? The authors claimed that they have performed an evaluation on different tools. But if the tool is not accessibly how do the authors perform their evaluations?

Response: Revised. You are correct. Many of the tools were not easily accessible or free to obtain, hence we did not have access to those tool to evaluation them against the other criteria (i.e. easy to use and selection of seasonal periods). We wanted to indicate this in the table by using 'unknown/unk'. However, perhaps this is not appropriate. We have thus removed unk and replaced with a hyphen.


Line 219/Table 3. The authors should report the actual number of year with the available data record.

Response: Revised. This is the number of years of available data. We have added the number of periods of data (i.e number of annual/wet season/dry season periods which had with no missing values).


Line 359/Table 4. Helping the reader not to use a calculator during their reading is not a valid reason of repeating information in scientific publication. I still recommend deleting this repeated information.


Response: Revised. Thank you for restating this. Removed CV row.

Line 460/Table 5. ‘MK p-value’ No need to show this. And ‘Hypotheses accepted’ No need to show this too. The authors can simply use bold or italic form to highlight the items under "Trend" to indicate significance.

Response: Revised. p-value row removed. Also removed from supplementary material Table 6. Row removed and italic used to highlight the statistically significant trends.

 

Annotated manuscript on the supplementary material:

Response: Responses have been individually added to the annotated supplementary material.

 

 

Thank you again for your valuable comments, we hope that you are happy with our responses.

 


Author Response File: Author Response.pdf

Water EISSN 2073-4441 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert
Back to Top