Water Quality Indices: Challenges and Application Limits in the Literature
, Fabienne Trolard 1,2,*, Mohamed Kefi 5, Sihem Jebari 4 and Guilhem Bourrié 6Round 1
Reviewer 1 Report
The manuscript is not original research article but could be considered as a review paper on Water Quality Index. The authors describe WQI evolution from historical point of view and provide in appendices the description of various methods for WQI calculations. The authors provide evidence that applying different methods on the same database could lead to contradictions. The authors point out fuzzy logics as the best approach but they do not provide any application nor comparison with the methods mentioned before. Therefore I suggest that this flaw should be removed prior to publication.
Other specific comments and suggestions are:
L226 Replace ‘scheme’ with ‘Figure’
L292 I did not find the AND operator in the Icaga’s method.
Figure 2 was placed in wrong section. It is referred for the first time in paragraph 4.4. So please move it after L289 or even into relevant appendix. Function mf1 is inappropriately described on Figure 2. Compare it with relevant equation after L526 in which variables a and b1 are involved. Why function mf5 presented on Figure 2 if it is not considered in Appendix E?
Replace DBO with BOD in Tables A2 and A3.
There is inconsistency of Equation A7 which has upper limit equal 11 with L418 where 12 variables are mentioned.
The Appendix E is unclear even if it was copied from Icaga (2007) work. The authors tried to replace the variable QP with QV but didn’t do it completely (L528 and 530).
L528 The rules are incorrect. If QP=IV then Output=I? In Icaga it was IV. What are the meanings of sub-indices i, j, k, l anyway?
Both tables A8 and A9 are taken from Icaga’s paper but only one is referred to it.
In my opinion the review presented in this manuscript falls within the scope of Water journal but requires major revisions regarding the fuzzy logics paragraph.
Author Response
Reviewer 1:
Comment 1 : L226 Replace ‘scheme’ with ‘Figure’
- Thank you for the suggestion, the modification was done.
Comment 2 : L292 I did not find the AND operator in the Icaga’s method.
- Thank you for your comment. Based on Icaga’s method explanation, he applied OR operator to insure the maximum of membership to a quality class. However, the operator AND can be used in the approach to insure the minimum of membership. We agree with the reviewer comment. For this reason, we have revised this section to explain more clearly AND Operator. In this context, we highlighted Line 296 – line 302.
Comment 3 : Figure 2 was placed in wrong section. It is referred for the first time in paragraph 4.4. So please move it after L289 or even into relevant appendix. Function mf1 is inappropriately described on Figure 2. Compare it with relevant equation after L526 in which variables a and b1 are involved. Why function mf5 presented on Figure 2 if it is not considered in Appendix E?
- Thank you for the comments.
o Figure has been moved following the correction section
o For mf1 as the other 3 membership functions mf2, mf3 and mf4 in the figure 2 are only the graphical resolution of the equations described in the 3rd step of Icaga WQI (L525). In his paper, Icaga explained that if there are not 2 upper limits in the function 2 and 3 , b=b1=b2 and c=c1=c2.
o The Fuzzy logic method has been carefully reviewed and used in the validated PhD of Moez Kachroud. It was reported that mf5 can be proposed in the case of the adequate temperature for the fecal coliform reproduction.
Comment 4 : Replace DBO with BOD in Tables A2 and A3.
- Thank you for your recommendations. Modifications were done.
Comment 5 : There is inconsistency of Equation A7 which has upper limit equal 11 with L418 where 12 variables are mentioned.
- Correction was made. Equation A9.
Comment 6 : The Appendix E is unclear even if it was copied from Icaga (2007) work. The authors tried to replace the variable QP with QV but didn’t do it completely (L528 and 530).
- Thank you for your suggestion. The corrections were made. In fact, we just mixed “P” and “V”. We use the term water quality variable since they vary in the time and the space,
Comment 7 : L528 The rules are incorrect. If QP=IV then Output=I? In Icaga it was IV. What are the meanings of sub-indices i, j, k, l anyway?
- Agreed and Corrected.
- Sub-indices i, j, k and l are counters used to make the distinction of the number of variables in each rules.
Comment 8 : Both tables A8 and A9 are taken from Icaga’s paper but only one is referred to it.
- Reference added in A10
Reviewer 2 Report
The article needs to be carefully proofread for English because there are many phrases that are a little off. Some examples:
•“water quality monitoring has one of the highest priorities” (vs. “is among the highest-priority areas” or similar)
•“the evaluation of rivers water quality has been intensified in the developing countries recently” (vs. “developing countries have intensified efforts to evaluate the quality of rivers” or similar)
•“they made the state of the art of the water quality indices” (not sure what you mean… that they evaluated the latest science? To “make the state of the art” seems to imply that they made an advance.)
“Alves et al. 2014 [19] in their statistical analysis review, found 554 articles dealing with the use of WQI’s between 1974 and 2011, of which only 38% are applied in India and 9,5% in China. This is due to the scarcity of water resources in these two countries, which is connected directly to the food scarcity”
•I’m confused. What do you mean “applied in” and why were you expecting more than 38% and 9.5% respectively in India and China? Why are we zooming in on India and China at all here?
“Nb of variables” (Table 1)
•“No.” or “#” are the common abbreviations for “number”.
“Thus, an index is basically a comparative tool to evaluate the efforts made to ameliorate the water quality and not really a tool to evaluate water quality absolutely”
•Is this always true? Plenty of indices report water quality scores (for example, beaches often list water quality as “A”, “B”, etc.) for the purposes of informing decisions.
“Toxicity is explicitely excluded by Horton, on the basis that ‘under no circumstances should
streams contain substances that are injurious to humans, animals and aquatic life. Water containing such substances, therefore, is considered not eligible for index rating.’”
•Explicitly* (typo)
•This is interesting in historical context. Obviously today we understand that all water contains trace amounts of substances that are “injurious to humans” (in some amount) and so this rule would disqualify all waters from being scored. I think it would make sense to comment here that this was before many of the breakthroughs in our understanding of the prevalence of low levels of organic contaminants in the water (for example, the NORS and NOMS investigations by the EPA in the early 1970s) and how this standard would not be realistic today.
“Disagreements appear in three cases: i) the same WQI is used, but the limits of classes differ; ii) different WQIs are used, on the basis of the same variables, and lead to different classifications; iii) different WQIs are used, on the basis of different types or numbers of variables.”
•I thought you were arguing for considering WQIs only in the context of changes in quality? Above you say: “Thus, an index is basically a comparative tool to evaluate the efforts made to ameliorate the water quality and not really a tool to evaluate water quality absolutely”. But here you are concerned that the absolute water quality has different scores according to different scales? I’m having trouble following the argument.
Fuzzy logic is a form of many-valued logic in which the truth values of variables may be any real number between 0 and 1. It is employed to handle the concept of partial truth, where the truth value may range between completely true and completely false. By contrast, in Boolean logic, the truth values of variables may only be the integer values 0 or 1.”
•This is copied verbatim from Wikipedia… https://en.wikipedia.org/wiki/Fuzzy_logic
•I stopped reviewing the article after discovering this copying. I do not know how much of the rest of the article is copied verbatim from other sources. I therefore stopped reviewing the article at this point and referred the issue to the editor for review.
Author Response
Reviewer 2:
Comment 1 : The article needs to be carefully proofread for English because there are many phrases that are a little off. Some examples:
• “water quality monitoring has one of the highest priorities” (vs. “is among the highest-priority areas” or similar)
• “the evaluation of rivers water quality has been intensified in the developing countries recently” (vs. “developing countries have intensified efforts to evaluate the quality of rivers” or similar)
• “they made the state of the art of the water quality indices” (not sure what you mean… that they evaluated the latest science? To “make the state of the art” seems to imply that they made an advance.)
“Alves et al. 2014 [19] in their statistical analysis review, found 554 articles dealing with the use of WQI’s between 1974 and 2011, of which only 38% are applied in India and 9,5% in China. This is due to the scarcity of water resources in these two countries, which is connected directly to the food scarcity”
Thank you for your suggestion. We revised the English style and we improved the sentences suggested.
Comment 2 : I’m confused. What do you mean “applied in” and why were you expecting more than 38% and 9.5% respectively in India and China? Why are we zooming in on India and China at all here?
Based on the reference, it was mentioned that WQI was used in India and China. To clarify the sentence, we replace applied in by used in as highlighted L40-L41.
Comment 3 : “Nb of variables” (Table 1)
• “No.” or “#” are the common abbreviations for “number”.
- Correction was made in table 1
Comment 4 : “Thus, an index is basically a comparative tool to evaluate the efforts made to ameliorate the water quality and not really a tool to evaluate water quality absolutely”
• Is this always true? Plenty of indices report water quality scores (for example, beaches often list water quality as “A”, “B”, etc.) for the purposes of informing decisions.
- In this sentence, we just want to point out idea related to Horton’s method
Comment 5 : “Toxicity is explicitely excluded by Horton, on the basis that ‘under no circumstances should streams contain substances that are injurious to humans, animals and aquatic life. Water containing such substances, therefore, is considered not eligible for index rating.’”
• Explicitly* (typo)
- Line 74. Correction was made
Comment 6 : This is interesting in historical context. Obviously today we understand that all water contains trace amounts of substances that are “injurious to humans” (in some amount) and so this rule would disqualify all waters from being scored. I think it would make sense to comment here that this was before many of the breakthroughs in our understanding of the prevalence of low levels of organic contaminants in the water (for example, the NORS and NOMS investigations by the EPA in the early 1970s) and how this standard would not be realistic today.
“Disagreements appear in three cases: i) the same WQI is used, but the limits of classes differ; ii) different WQIs are used, on the basis of the same variables, and lead to different classifications; iii) different WQIs are used, on the basis of different types or numbers of variables.”
• I thought you were arguing for considering WQIs only in the context of changes in quality? Above you say: “Thus, an index is basically a comparative tool to evaluate the efforts made to ameliorate the water quality and not really a tool to evaluate water quality absolutely”. But here you are concerned that the absolute water quality has different scores according to different scales? I’m having trouble following the argument.
Thank you for your comments. As the paper is a review, we tried to identify several applications of WQI. Literature review analysis help us to identify the approach to quantify and to classify the WQI. “Starting from Horton to more advanced index”. In addition, we gave, as examples, some disagreement evidences to illustrate the contradiction obtained in the literature when using some well-used WQIs.
Comment 7 : Fuzzy logic is a form of many-valued logic in which the truth values of variables may be any real number between 0 and 1. It is employed to handle the concept of partial truth, where the truth value may range between completely true and completely false. By contrast, in Boolean logic, the truth values of variables may only be the integer values 0 or 1.”
• This is copied verbatim from Wikipedia… https://en.wikipedia.org/wiki/Fuzzy_logic
We are sorry about this matter. In fact, Wikipedia was used in this case just to introduce the concept of Fuzzy logic, which will be helpful for the reader to understand the approach with an easy definition. Based on the comment of the reviewer, we have reformulated the sentence.
I stopped reviewing the article after discovering this copying. I do not know how much of the rest of the article is copied verbatim from other sources. I therefore stopped reviewing the article at this point and referred the issue to the editor for review.
Thank you for you relevant comment to improve the quality of the paper. We would like ensure that this work is original and no copyright violation were made. We will pay attention in the future to use only published papers as a reference.
Reviewer 3 Report
It needs a major revision to update the references as most of them are very old.
The manuscript looks like a review rather than an article.
So it is suggested to make a major revision.
Author Response
Reviewer 3:
Comment 1 : It needs a major revision to update the references as most of them are very old.
The manuscript looks like a review rather than an article.
So it is suggested to make a major revision.
Thank you for your comment and we totally agree. As the paper is a review, we tried to emphasize on the research related to WQI from old to more recent applications. Our target is mainly to track the evolution of WQI concept in the space and in the time. Since 1965, the context was used in water quality assessment, then, WQI took different forms with the development of the technology.
Round 2
Reviewer 1 Report
In general the authors answered satisfactory all my specific comments. However my main concern about implementation of fuzzy logics in WQI was only partially addressed. Therefore please consider the following comments:
1/ The new section 4.3 which presents a test of 4 WQIs on the same dataset improved the paper considerably but it lacks in depth analysis of results.
2/ It appears from table 8 that FWQI is the least restrictive method. Is it good?
3/ I presume that the authors used Fuzzy logics for calculation of FWQI. If this is the case please state this clearly in the text.
4/ Please refer each selected method for testing with relevant appendix for calculation details. Also all abbreviations (L217-218) should be explained when they appear for the first time. I do not see FWQI in appendices.
Author Response
Response to Reviewers
First of all on behalf of all the co-authors, I would like the editors and 3 reviewers for their relevant comments to improve this paper. I would like to give some clarifications about the comments suggested by reviewer 3
Reviewer 1:
In general the authors answered satisfactory all my specific comments. However my main concern about implementation of fuzzy logics in WQI was only partially addressed. Therefore please consider the following comments:
1/ The new section 4.3, which presents a test of 4 WQIs on the same dataset improved the paper considerably but it lacks in depth analysis of results
- Thank you for your comment, additional interpretations were added to improve the analysis in section 4.3 (From Line 227-240)
2/ It appears from table 8 that FWQI is the least restrictive method. Is it good?
- Thank you again. Explications about FWQI characteristics and accuracy were added in the same section of comment 1.
3/ I presume that the authors used Fuzzy logics for calculation of FWQI. If this is the case please state this clearly in the text.
Thank you for your comment. Somme modifications were added in L221-222 and L553.
4/ Please refer each selected method for testing with relevant appendix for calculation details. Also all abbreviations (L217-218) should be explained when they appear for the first time. I do not see FWQI in appendices.
- Thank you for the suggestion. We referred all methods (L216-226).
Reviewer 3 Report
It was improved and revised as suggested before, therefore, it is suggested to be accepted.
Author Response
Thanks for your comments to improve this paper.
