Human Health Impact Analysis of Contaminant in IoT-Enabled Water Distributed Networks

Round 1

Reviewer 1 Report

This study analyzed the impact of contaminate piped water on human health in a water distribution system in the case with/without water quality sensors. Early detection of toxic chemical and pathogenic microbial contamination in water at a distribution network and suspension of the water would reduce the impact on human health.  

However, more clear and detailed explanation about the materials and methods is necessary for readers’ understanding and interest. The following is should be revised to understand the approach in this study.

1)Insufficient explanations especially about materials and methods (parameter value setting etc.) .

e.g.:

What is the value of “the water ingestion rate”?

What contaminant was applied to this study, and what is its LD50 value?

How to classify “infection”, “disease”, and “fatalities“?

There is no reference for “our proposed approach presented in (Shahra and Wu 2020)”. Even if the reference is quoted, the authors should explain the approach briefly.

The explanation of EPANET, TEVA-SPOT, and the methods of “Health Impact Analysis” should be described in the Materials and Methods, not in the Results and Discussion section, to understand the approach of this study.

The information of the sensor is also necessary (e.g. accuracy, monitoring parameters). What is the criteria

2) Inconsistency of terms

It seems there are several inconsistent terms, which also makes it difficult to understand the approach and results. Please be sure to use the same terms throughout the paper including the figures and tables.

3)Small legends and low resolution of Fig.1-9.

The reviewer also suggests that the authors should mention the advantage, importance, and novelty of this study by quoting the reference in “Relate Work” section rather than crating the section.

Author Response

The authors' response is in attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

In general, the manuscript is well documented and is enriched with pictures and tables, however several aspects need to be clarified. Moreover, the results and discussion are a quite confuse and the statistical analysis should be more informative.

It's hard to understand the methods for evaluation the human health impact analysis of contamination in water distributed networks. In my opinion the description of the contaminants in this study should be more descriptive, indicating each parameter and each concentration susceptible to cause disease. The title seems be also excessive.

Line 157 and 158: At the Introduction part you referred the water contaminants, such as nitrate and fluoride, while in the study design these are not included. In my opinion these chemicals and also microbiological contaminants should be included in this study.

At the “Methods” part, you should refer the bibliographic references supported to apply the formulas described. You should also describe the equations 4, 5 and 6 at the "Methods" part.

At the Discussion part (lines 199-200) you write "In this work, you used the demand based method to define the number of water Gallons used by person per day (GPD) to define the distribution of the population”. Is a quite confuse the number of water Gallons. Shouldn't this information be included in the Methods?

Between the lines 204 and 205 there are a lot of text without identified lines. In this part you referred that "simulated the effect of water contaminants in different locations". However, are not described which those contaminants are, neither mention how was determined the degree of toxicity of the contaminants. Are also do not explained what are the contaminants to which the lethal dose (LD) estimate is applied.

At the lines 219 and 220, it is not clear how was calculated the number of different diseases.

At lines 233 and 234: How can you estimated the people affected by contamination in term of fatalities, infection, and disease?

At the table 1: “maximum concentration” and “maximum individual dose” refers to what type of contaminant?

Other detailed comments in the pdf document.

Comments for author File: Comments.pdf

Author Response

The authors' response is in attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper evaluates the impact on human health of contamination which occurred in water distribution system. The impact on the population is computed as function of ingested contaminant mass, which is a common practice in the literature. Then, the effects of contaminations were computed for three scenarios: 1) the first one without quality sensors in water distribution network, 2) the second one with 10 quality sensors placed minimizing the total volume of contaminant and 3) last one with 10 quality sensors placed minimizing the detection time. The paper is well structured. In the related works section, it is suggested to add the following references:

• Ung, H.; Piller, O.; Gilbert, D.; Mortazavi, I. Accurate and Optimal Sensor Placement for Source Identification of Water Distribution Networks. J. Water Resour. Plan. Manag. 2017, 143, 04017032.
• Hooshmand, F.; Amerehi, F.; MirHassani, S. Risk-Based Models for Optimal Sensor Location Problems in Water Networks. J. Water Resour. Plan. Manag. 2020, 146, 04020086.
• Santonastaso, G.F.; Di Nardo, A.; Creaco, E.; Musmarra, D.; Greco, R. Comparison of topological, empirical and optimization-based approaches for locating quality detection points in water distribution networks. Environ. Sci. Pollut. Res. 2020, 1–10.
• Hu, C.; Yan, X.; Gong, W.; Liu, X.; Wang, L.; Gao, L. Multi-objective based scheduling algorithm for sudden drinking water contamination incident. Swarm Evol. Comput. 2020, 55, 100674.

The case study is clear but it would be necessary to specify the studied contamination events, the number of the events, and the starting time of contamination and demand pattern. The obtained results are sufficiently discussed but it is advisable examine in deep a comparison between the last two scenarios. The conclusions are adequate. Nevertheless, some minor revisions are kindly required to improve the manuscript

1) Lines 158-159: “Through this, the effect of water contamination on human health will be calculated and analyzed in all the previous cases that have been explained” It is not clear which are the previous cases.

2) Lines 269, in references section misses the following citation: “Shahra and Wu 2020”

3) It is advisable provide some information about the applied procedure to place quality sensors in water distribution network.

4) The captions should explain better the various figures and table.

5) The quality of images is very poor. It is suggested improve it.

6) The English language should be improved.

Author Response

The authors' response is in attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The reviewer appreciates the effort to revise the manuscript. However, the reviewer would like to ask the authors more clear explanation about the data applied to this study.

Table 1:

Please add a reference(s) for “Water Ingestion Rate (Liters per day) 1.41”. Is this applicable in this study area?

“Gallons per Person per Day (GPD) 200” is high. Is this because the GPD include commercial, industrial, and other water uses as well as domestic use in this case study area? Since the authors mentioned that the target network is for a drinking water purpose in the case study area (subsection 3.1.1), a reference and explanation more about the GPD in this study area is necessary.

A LD50 value influences on the health impact analysis. Please explain the decision criteria of the LD50 value (0.001) in this study. The reviewer would also like to ask the following.

1) The LD50 value applied in this study is small, which means “Chemical/toxic” has very high toxic. What chemical/toxic as the source of deliberate and accidental contaminations was assumed in this case study? It would be better to show an example case.

2) Can the contaminations (answer of 1)) be detected by the sensor based on the change of the water quality parameters (i.e. Chlorine, TOC, Turbidity, and Conductivity)? It would be better to mention a limitation(s) of the approach and analysis in this study if any.

The above information would be useful and interesting for the readers including water utilities.

There are two “Duration”. Please delete one of them.

The authors set “Start time” at 24:00. Since customers generally do not consume much water during the nighttime, the water flow in the nighttime is lower than that in daytime. The contaminant spread time of 8 hours was calculated based on water flow in the nighttime?

Please also explain the reason why the authors decided that time as “Start time”. The result of health impact (e.g. Figure 7) could be changed if “Start time” is not 24:00.

Abstract (line 16-17) and Conclusion (line 414)

The authors mentioned “The results show that the health was reduced by up to 98.37 % by using water quality sensors.” in the abstract, while “the infections cases were reduced by 98.3” in the conclusion. Please correct the number of reduction rate (98.37 or 98.3). In addition, “the health was reduced” sounds strange.

Line 335-336

It needs to be a sentence.

Please delete the following.

Figure 7: Please delete “taminant”

3.2.2: There are unnecessary “s”s. 

Author Response

Thank you for your comments.

Please see attached file  (reveiwer1.pdf)  for our response

Author Response File: Author Response.pdf

Reviewer 2 Report

In general, the manuscript is well documented and is enriched with pictures and tables. The authors were concerned with reviewing all aspects identified in the first review. The introduction and methods have been improved considerably. The contents of the article are presented in a clearer and more organized way.
It is worth noting some comments in the new text and inconsistency in figures and tables:

- Line 138: You should explain the s designation on the first time this appear in the text.
- Line 157: In my opinion it is preferable to place the Author et al [32], as they did in the ptevious reference [31].
- Lines 164 and 172: Again, is preferable to place the Author et al, as you did previously. 
- Line 214: Please correct “dose-level”.
- Line 317: "Chlorine, TOC, Turbidity, and Conductivity" - be consistent along the manuscript using the large/small print.
- The legends of figures 10, 11, 12 and 13 are very close between the different graphs. Try to distance them a little further.
- Tables 3, 4 and 5 appear only at the end of the text and overlapped with references. Can't you place them in supplementary material?

Author Response

Thank you for your comments. Please see the attachment for our response

Author Response File: Author Response.pdf

Reviewer 3 Report

Authors have satisfied all requests of reviewer. Nevertheless some minor changes are required: 

Section 3.2.1: have authors considered only one contaminant event (injection node at the water source (Reservoir)”? In this case they should explain the purpose of this choose.

Lines: 186-190: It is advisable include the information reported in the response to reviewer: “The three cases were divided based on the water demand and It is labelled in the case study Figure 1, by circle, square, and node_3654."

Author Response

Thank you for your comments, please see the attachment for our response.

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

The reviewer appreciates the effort to revise the manuscript.