Investigating Effect of Pumping Ratio on Effectiveness of Barrier Wells for Saltwater Intrusion: Lab-Scale Experiments and Numerical Modeling

Round 1

Reviewer 1 Report

Review for “Investigating effect of pumping ratio on effectiveness of barrier wells for saltwater intrusion: Lab-scale experiments and numerical modeling” by Ozaki  et al. (water-1223395).

The article deals with important issue – how to deal with sea water intrusion and investigates and proposes option – barrier well. The authors studied the ideal ratio between pump from a production well and from a barrier well. They used a lab experiment and a model.

I found the manuscript interesting since the experiments showed clearly how pumping ratio affect seawater intrusion, and their model described well the results from the experiments. However, I have some remarks, concerning the introduction (I think it is lack of references) and concerning their experiment and model. Therefore, I think that the article should be accepted with moderate revision. My remarks are listed below:

General remark – I think that when taking into consideration pumping ratio you need to think about the effect of distance of the wells from the seater intrusion toe, the seawater-freshwater interface (FSI), the geometry of the FSI and more parameters. I did not find any discussion about that in the paper and I think it is important to talk about issue.

In addition, you model describes the experiments’ results very well, but it does not clear what the modell contribute to the insights which discussed before.

• Lines 23-25 and 26-27 – I got the feeling that these two sentences are contradict each other. If so, “Moreover” is a word that should be replace.
• [1] (line 36) - Gaza is a good example for significant effect of seawater intrusion, but if you want to give only one example you should put a more serious situation, like the one in the Nile Delta aquifer in Egypt (100Km):

Shrif M, Sefelnasr A, Javadi A. Incorporating the concept of equivalent freshwater head in successive horizontal simulations of seawater intrusion in the Nile Delta aquifer, Egypt. 2012, J.Hudrol. Vol 464-465, pp. 186-198

• Lines 59-77 – you wrote only about research concerning pumping saltwater that used model. This issue is more complicated, and I think you should elaborate a bit more, or at list add more reference. For example:

Stein S, Yechieli Y, Shalev E, Kasher R, Sivan O. The effect of pumping saline groundwater for desalination on the fresh–saline water interface dynamics. 2019, Water research, Vol 156, pp. 46-57.‏

• 2 - I think it will help the readers to mention in the figure that well A is the barrier well and well B is the production well.
• 3 - I advise to add in the caption of the figure that this is result of experiment woth ratio of 2.6.
• Line 190 – I think that the symbol represents density of saltwater is wrong (not like the one written in Eq.14).
• 16 – there is no explanation in the text what is q’.
• Line 216 – there is no D_T in Eq. 16.
• Table 4- in table 1 there are only 6 experiments but in table 4 ther are seven experiments. I do not understand why. I think only one table showing 7 experiments are enough.
• Line 251 – why only experiments with ratio 1.2 and 1.9 are mentioned? What about experiment with ratio 1.5?
• Lines 253-255 – you should describe the time to achieve steady state to let the readers be able to compare it. We need to know those time to understand how significant is “the longest time”.
• Line 283-286 – I would have emphasized more that the situation in step 2 and step 4 is almost identical. It means that the process is reversible, and this is not trivial.
• Line 300 – in the figure 10 there is no image of experiment with ratio 2.3, only ratio 2.
• Lines 301-303 – You should discuss about the flow (Q_A) which describe (only !) in Fig.11. I believe that the fact that Q_A in experiment with ratio of 2.6 is lower than in the other experiments (except experiment with ratio 0.9) contribute to the phenomena described in this sentence. You should elaborate abot the different flow in the experiments (I advise also add to table 2) and explain why you did not conduct all the experiment with the same flow (Q_A).
• Lines 342-346 – I did not understand – the first sentence talked about how the barrier well is effective to eliminate seawater intrusion and the second sentence begins with “therefore” it is enough to pump groundwater that slightly concentrated with saltwater??
• Line 352 – I did not understand what it means “even after...production well”. Please elaborate.
• Lines 356-357 – this sentence summarize the whole paper it should not start with “Therefore”.

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 1 Comments

Point 1: The article deals with important issue – how to deal with sea water intrusion and investigates and proposes option – barrier well. The authors studied the ideal ratio between pump from a production well and from a barrier well. They used a lab experiment and a model.

 I found the manuscript interesting since the experiments showed clearly how pumping ratio affect seawater intrusion, and their model described well the results from the experiments. However, I have some remarks, concerning the introduction (I think it is lack of references) and concerning their experiment and model. Therefore, I think that the article should be accepted with moderate revision. My remarks are listed below:

General remark – I think that when taking into consideration pumping ratio you need to think about the effect of distance of the wells from the seater intrusion toe, the seawater-freshwater interface (FSI), the geometry of the FSI and more parameters. I did not find any discussion about that in the paper and I think it is important to talk about issue.

In addition, you model describes the experiments’ results very well, but it does not clear what the model contribute to the insights which discussed before.

Response 1: Please provide your response for Point 1. (in red)

Thank you very much for your peer review of our paper(water-1223395) with the productive comments.

We would like to thank you for your comments regarding the lack of references. We added the descriptions of some references about saltwater intrusion and a negative hydraulic barrier in the introduction.

We answered all your comments below and revised some parts according to your advice. Thank you so much again, all your comments helped us improve this paper and we appreciate them all.

Point 2:Lines 23-25 and 26-27 – I got the feeling that these two sentences are contradict each other. If so, “Moreover” is a word that should be replace.

Response 2: Please provide your response for Point 2. (in red)

Thank you for your suggestion. We replaced “. Moreover” with “, and ~~”. (Line24)

Point 3: [1] (line 36) - Gaza is a good example for significant effect of seawater intrusion, but if you want to give only one example you should put a more serious situation, like the one in the Nile Delta aquifer in Egypt (100Km):

Shrif M, Sefelnasr A, Javadi A. Incorporating the concept of equivalent freshwater head in successive horizontal simulations of seawater intrusion in the Nile Delta aquifer, Egypt. 2012, J.Hudrol. Vol 464-465, pp. 186-198

Response 3: Please provide your response for Point 3. (in red)

We added some descriptions of the paper, “Incorporating the concept of equivalent freshwater head in successive horizontal simulations of seawater intrusion in the Nile Delta aquifer, Egypt.”, introduced by the reviewer, in the introduction part. (Line35)

Point 4: Lines 59-77 – you wrote only about research concerning pumping saltwater that used model. This issue is more complicated, and I think you should elaborate a bit more, or at list add more reference. For example:

Stein S, Yechieli Y, Shalev E, Kasher R, Sivan O. The effect of pumping saline groundwater for desalination on the fresh–saline water interface dynamics. 2019, Water research, Vol 156, pp. 46-57.‏

Response 4: Please provide your response for Point 4. (in red)

Based on the reviewer's suggestions, we have added “The effect of pumping saline groundwater for desalination on the fresh–saline water interface dynamics.” as the reference. (Line62)

Point 5: 2 - I think it will help the readers to mention in the figure that well A is the barrier well and well B is the production well.

Response 5: Please provide your response for Point 5. (in red)

Thank you for your suggestion.

Since “A is the barrier well and well B is the production well” are related to step 3, we described it as “A is the barrier well and well B is the production well in step 3”. (Fig.2)

Point 6: 3 - I advise to add in the caption of the figure that this is result of experiment woth ratio of 2.6.

Response 6: Please provide your response for Point 6. (in red)

We added the Q_A/Q_B ratio of 2.6 to the title in Figure 3.(Fig.3)

Point 7: Line 190 – I think that the symbol represents density of saltwater is wrong (not like the one written in Eq.14).

16 – there is no explanation in the text what is q’.

Line 216 – there is no DT in Eq. 16.

Response 7: Please provide your response for Point 7. (in red)

I'm sorry we made a simple mistake. We modified the description of q' , and the DL in Equation 16 to DT. (Line232-233)

Point 8: Table 4- in table 1 there are only 6 experiments but in table 4 ther are seven experiments. I do not understand why. I think only one table showing 7 experiments are enough.

Response 8: Please provide your response for Point 8. (in red)

As the reviewer pointed out, we have combined Table 1 and Table 4 into one.

As explained in Line 257~261, R=2.3 was conducted as the secondary test to confirm R=1.9 is the critical pumping ratio, Table 1 just shows the main 6 experimental cases. (Table5, Line272,288)

Point 9: Line 251 – why only experiments with ratio 1.2 and 1.9 are mentioned? What about experiment with ratio 1.5?

Response 9: Please provide your response for Point 9. (in red)

We changed the expression as follows.

(Line251) When the ratio was within 1.2 to 1.9, the salt–freshwater interface curve moved right and upward without upconing to Well B. (Line270)

Point 10: Lines 253-255 – you should describe the time to achieve steady state to let the readers be able to compare it. We need to know those time to understand how significant is “the longest time”.

Response 10: Please provide your response for Point 10. (in red)

We have added a new table showing the time for the intrusion to steady state as Table.5.(Table5,Line288)

Point 11: Line 283-286 – I would have emphasized more that the situation in step 2 and step 4 is almost identical. It means that the process is reversible, and this is not trivial.

Response 11: Please provide your response for Point 11. (in red)

The findings pointed out by the reviewers have also been added to the Conclusions as important findings. (4. Conclusions)

Point 12: Line 300 – in the figure 10 there is no image of experiment with ratio 2.3, only ratio 2.

Lines 301-303 – You should discuss about the flow (QA) which describe (only !) in Fig.11. I believe that the fact that QA in experiment with ratio of 2.6 is lower than in the other experiments (except experiment with ratio 0.9) contribute to the phenomena described in this sentence. You should elaborate abot the different flow in the experiments (I advise also add to table 2) and explain why you did not conduct all the experiment with the same flow (QA).

Response 12: Please provide your response for Point 12. (in red)

It was commented that Figure 10 should be removed by the other reviewer as well. We have deleted Figure 10 according to the comments from both reviewers.

In addition, since it was difficult to keep the flow rates of Qa and Qb constant in this series of experiments, the experiment was conducted while checking the flow rates during the experiment.

Point 13: Lines 342-346 – I did not understand – the first sentence talked about how the barrier well is effective to eliminate seawater intrusion and the second sentence begins with “therefore” it is enough to pump groundwater that slightly concentrated with saltwater??

Response 13: Please provide your response for Point 13. (in red)

This part means it will be enough to pump groundwater with less salinity if the pumping ratio is even lower than the critical ratio, for example R=0.9 in this study since the critical ratio was R=1.9.

Therefore, we added “, as observed at pumping ratio 0.9.”. (Line364-367)

Point 14: Line 352 – I did not understand what it means “even after...production well”. Please elaborate.

Response 14: Please provide your response for Point 14. (in red)

Removed (Line372-374)

Point 15: Lines 356-357 – this sentence summarize the whole paper it should not start with “Therefore”.

Response 15: Please provide your response for Point 15. (in red)

We deleted this sentence. (Line378)

Author Response File: Author Response.pdf

Reviewer 2 Report

I liked the general idea of the paper, the experimental and numerical parts, and the potential application. However, I have a couple of suggestions:

* I feel the literature review a bit short (only 10 references in the introduction!). For example, in line 55, the authors say "many studies" but do not give any example in that line. And in the next paragraph, they cite only three studies. Moreover, those three studies are numerical works. They do not mention any laboratory study related to saltwater intrusion. It is unclear if it is the first laboratory experiment on the topic, but they could include references on laboratory experiments related to coastal aquifers, even addressing other issues. Overall, they should better explain what has been done and what is the novelty of their project.

* Line 45: "...sea level will rise more than 95%" They cite the IPCC, but I think something is missing (95% with respect to what?)

* Not sure how scientific is the term "pumping quantity" (e.g. line 71). Do the authors mean flow rate?

* Lines 106-108/Table 1: can the hydraulic conductivity change that much for that reason, or could it also be due to some experimental uncertainties (e.g. heads)?

• Lines 123-125/Table 1: If they wanted to compare six different pumping ratios, why didn't the authors keep QA or QB constant and only change one pumping rate?
• Line 130 I think it could be better explained how the pumping rates for QA/QB were determined

• Lines 217-234: The authors refer extensively to many tables and figures in a few lines. The paragraph is hard to follow and some tables containing the parameters at least could be merge in one.

• Lines 233-234: by results, do the authors mean a visual comparison of the interface? Or did they use something else?

• Is the experimental design based on optimal configurations from previous works? They found a critical pumping ratio of 1.9. Still, it may be interesting to provide in a short discussion if this critical pumping ratio is specific to these experiments or can be extrapolated to other conditions. The authors mention something in the conclusions (from line 352), but maybe they can expand a bit more in the previous section.
• In general the results and discussion part could be better organized and structured, in particular for what concerns the presentation and description of the figures. Moreover, the quantitative comparison between modeled results and experimental data could be improved.
• Either Figure 10 or 11  could be removed

Author Response

Response to Reviewer 2 Comments

Point 1: I liked the general idea of the paper, the experimental and numerical parts, and the potential application. However, I have a couple of suggestions:

 * I feel the literature review a bit short (only 10 references in the introduction!). For example, in line 55, the authors say "many studies" but do not give any example in that line. And in the next paragraph, they cite only three studies. Moreover, those three studies are numerical works. They do not mention any laboratory study related to saltwater intrusion. It is unclear if it is the first laboratory experiment on the topic, but they could include references on laboratory experiments related to coastal aquifers, even addressing other issues. Overall, they should better explain what has been done and what is the novelty of their project.

Response 1: Please provide your response for Point 1. (in red)

Thank you very much for your peer review of our paper(water-1223395) with the productive comments.

We would like to thank you for your comments regarding the lack of references. We added the descriptions of some references about saltwater intrusion and a negative hydraulic barrier in the introduction.

We answered all your comments below and revised some parts according to your advice. Thank you so much again, all your comments helped us improve this paper and we appreciate them all. (1.Introduction)

Point 2: * Line 45: "...sea level will rise more than 95%" They cite the IPCC, but I think something is missing (95% with respect to what?)

Response 2: Please provide your response for Point 2. (in red)

Revised as “in more than 95% of the ocean area on earth, ”. (Line55)

Point 3: * Not sure how scientific is the term "pumping quantity" (e.g. line 71). Do the authors mean flow rate?

Response 3: Please provide your response for Point 3. (in red)

Thank you for your suggestion. We replaced “pumping quantity” with “flow rates”. (Line83)

Point 4: * Lines 106-108/Table 1: can the hydraulic conductivity change that much for that reason, or could it also be due to some experimental uncertainties (e.g. heads)?

Response 4: Please provide your response for Point 4. (in red)

When the experimental equipment was made, the width of the permeation tank was 10.0 cm. In Fig. 2, it was described that the width of the permeation tank was 10.6 cm. The reason for this is the width of the permeation tank has expanded slightly by filling with glass beads. The width of the permeation tank was measured with vernier calipers at about 3 points during the experiment, and the average was 10.6 cm. For this reason, it is considered that the hydraulic conductivity was slightly different each time. (Line118-120)

Point 5: Lines 123-125/Table 1: If they wanted to compare six different pumping ratios, why didn't the authors keep QA or QB constant and only change one pumping rate?

Line 130 I think it could be better explained how the pumping rates for QA/QB were determined

Response 5: Please provide your response for Point 5. (in red)

Since it was difficult to keep the flow rates of Qa and Qb constant in this series of experiments, the experiment was conducted while checking the flow rates during the experiment.

Initially, we adjusted the discharge valve so that the QA / QB flow ratio was supposed to be 1.0, although it was actually 0.9 due to the difficulties to adjust the flow rates. After that, the flow rate ratio was changed by trial and error, and we confirmed at what flow rate the salt water would reach the production well. As a result, it was found that approximately 2.0 in the flow rate ratio is the critical value at the production well as for saltwater intrusion. (Table1, Line143)

Point 6: Lines 217-234: The authors refer extensively to many tables and figures in a few lines. The paragraph is hard to follow and some tables containing the parameters at least could be merge in one.

Response 6: Please provide your response for Point 6. (in red)

As the reviewer pointed out, we have combined Table 1 and Table 4 into one. (Table1, Line143)

Point 7: Lines 233-234: by results, do the authors mean a visual comparison of the interface? Or did they use something else?

Response 7: Please provide your response for Point 7. (in red)

We visually observed the saltwater colored by food red coloring.

Point 8: Is the experimental design based on optimal configurations from previous works? They found a critical pumping ratio of 1.9. Still, it may be interesting to provide in a short discussion if this critical pumping ratio is specific to these experiments or can be extrapolated to other conditions. The authors mention something in the conclusions (from line 352), but maybe they can expand a bit more in the previous section.

In general the results and discussion part could be better organized and structured, in particular for what concerns the presentation and description of the figures. Moreover, the quantitative comparison between modeled results and experimental data could be improved.

Response 8: Please provide your response for Point 8. (in red)

We could not find any previous studies on the experiment and the numerical analysis presented by the authors.

We found a critical pumping ratio of 1.9 and this critical pumping ratio is specific to this experiment, we suppose that it cannot be extrapolated to other conditions. However, this study found the possibility to manage saltwater intrusion by controlling the pumping ratio (by keeping it under the critical ratio).

(4. Conclusions)

Point 9: Either Figure 10 or 11 could be removed

Response 9: Please provide your response for Point 9. (in red)

We have deleted Figure 10 according to reviewer’s comments.

Fig.11 → Fig.10

Round 2

Reviewer 2 Report

I think the authors should perform a more careful literature review to insert their paper in a proper research context. See for example the works of Salandin, Camporese and co-workers as well as this reference https://link.springer.com/article/10.1007/s11269-020-02635-z just to mention some. Moreover, it seems that the experimental setup had several problems that could put the results into question. The authors should provide some comments about it. Finally, a quantitative and not only a qualitative comparison between experiment and model results should be given. I fear the manuscript still needs some major revisions prior publication.

Author Response

Response to Reviewer 2 Comments

Point 1: I think the authors should perform a more careful literature review to insert their paper in a proper research context. See for example the works of Salandin, Camporese and co-workers as well as this reference https://link.springer.com/article/10.1007/s11269-020-02635-z just to mention some. Moreover, it seems that the experimental setup had several problems that could put the results into question. The authors should provide some comments about it. Finally, a quantitative and not only a qualitative comparison between experiment and model results should be given. I fear the manuscript still needs some major revisions prior publication.

(Answer to the question about the experimental device in this study (the last question from the reviewer 2)

Response 1: Please provide your response for Point 1. (in red)

First of all, thank you very much for introducing the important reference.

Even though the experimental setup used in this study was designed without any reference in 2019, this setup has some similarities as the one used in the suggested reference paper by Vat et al. (2020), such as the general configuration of the device with 3 sections; saltwater chamber, freshwater chamber, and the central flow chamber, as shown in the suggested paper, glass beads simulating a coastal unconfined aquifer, saltwater prepared from tap water by adding commercial salt at a concentration of 35g/l. As mentioned in the suggested study, glass beads and saltwater colored by food color are commonly used in laboratory scale experiments. Therefore, it is difficult to clearly show the experimental setup is ‘optimal’ or the best design for the simulation in terms of the scale (for example the hydraulic heads and each length of the device do not represent any specific practical location, just designed in the lab-scale), but at least it seems to be reasonable considering there are some previous studies with the similar experimental setups. We put the reference in Line 86-90.

As for the reliability of the experimental setup, all hydraulic parameters, such as hydraulic conductivity, porosity, diameter of the porous media, and densities for freshwater and seawater, were designed and confirmed to be reasonable for a coastal sandy aquifer, and fresh/sea water.

Although there were some uncertainties as mentioned, such as the expansion of the device width and the unstable pumping rates, the findings of this study was regarded as reliable because the numerical and experimental results were examined based on the visual comparison and intrusion length at the steady-states. We added the explanation in Line 382-388 and Figure 9 and to avoid many figures, only case (4)R=1.9 (QA=2.7ml/s, QB=5.2ml/s), K=0.50 cm/s was described in the text.

Reliability confirmation

The experimental results and numerical simulation results were compared based on the general condition of the SWI and the intrusion length at each steady state. The error in the intrusion length was less than 2.5cm in any cases, and the general shape of the fresh/salt - water interface was calculated with high accuracy as shown below. The figures show the steady states in Step1, 2, and 3 for each experiment, as shown in Figure 9.

• R=0.9 (Q_A=1.8ml/s, Q_B=1.7ml/s), K=0.45 cm/s

• R=1.2 (Q_A=2.9ml/s, Q_B=3.5ml/s), K=0.40 cm/s

• R=1.5 (Q_A=2.5ml/s, Q_B=3.8ml/s), K=0.53 cm/s

• R=1.9 (Q_A=2.7ml/s, Q_B=5.2ml/s), K=0.50 cm/s

• R=2.6 (Q_A=2.2ml/s, Q_B=5.7ml/s), K=0.48 cm/s

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report

The authors tried their best to answer my comments. In my view, it is a pity that the comparison between model and experiment remains only qualitative and visual. This limits the relevance and the impact of the study, in particular considering that a model-based interpretation of the results would have allowed to investigate the impact of the (many) experimental uncertainties. My suggestion is to strongly encourage the authors considering a quantitative comparison between model and experiment based on optical imaging of the experiment. Considering that it is already the third round of review and the authors have not been able to provide an adequate answer to the comment, I believe it is not constructive to carry on the discussion any longer and the manuscript can be published in the present form.