Methods of In Situ Assessment of Infiltration Rate Reduction in Groundwater Recharge Basins

Round 1

Reviewer 1 Report

Line 51: delete “was performed”

Line 85: define “m/a”

Line 90: Is “FDR” a commercial name?

Fig 1; make text eligible

Equation 1 is inconsistent in terms of units.

Figure 3: the values in the y-axis use point (0.5) not comma (0,5). The latter is incorrect.

Figure 5: same as figure 3

Figure 6: same

Figure 7: same

Figure 8: same

Line 291: What is “R”

Line 294: Change “adquire” to “acquire”

Author Response

Point 1: Line 51: delete “was performed”

Response 1: Sentence was reformulated deleting the desired words.

Point 2: Line 85: define “m/a”

Response 2: The units of hydraulic loading rate (m/a) was explained: “This designed hydraulic loading rate represents the recharge of a 300 m high water column with an area of 1 m² within one year.”

Point 3: Line 90: Is “FDR” a commercial name?

Response 3: FDR is a water content measurement technique. It corresponds to frequency domain reflectometry. This was added into the manuscript.

Point 4:  Fig 1; make text eligible

Response 4: Text font was increased.

Point 5:  Equation 1 is inconsistent in terms of units.

Response 5: The fraction t₅₀/z was corrected into z/t₅₀

Point 6:  Figure 3: the values in the y-axis use point (0.5) not comma (0,5). The latter is incorrect.

Figure 5: same as figure 3

Figure 6: same

Figure 7: same

Figure 8: same

Response 6: The decimal separator was changed in all figures

Point 7:  Line 291: What is “R”

Response 7: This was a typo. It is now deleted.

Point 8:  Line 294: Change “adquire” to “acquire”

Response 8: Done

Reviewer 2 Report

Overall comments

The paper is entitled “Methods of in situ assessment of infiltration rate reduction in groundwater recharge basins.” This paper’s aims is to evaluate a variety of methods to calculate decrease in infiltration rates.

I think this paper is well-written and the some included information is beneficial for readers, however, the importance of the study is somewhat weak. The weakness is that the most methods in the study are empirical, i.e., site specific and boundary condition dependent parameters are necessary.  If the authors can develop some equation or way to estimate the empirical parameters from site information, boundary conditions, and so on, the manuscript becomes much stronger.

Specific comments

L82                  Figure 1 is not readable. Authors need to show it with better resolution. Also There is no dimension in the figure.

L90                  FDR must be abbreviation of frequency domain reflectometry.

L114                Please explain what v50 and v50o are.

L117                Is this equation valid? Please give a citation for the equation.

L122                Explain what is theta_o.

 L185              Is there any standard for managers to take an action? What kinds of action are made?

L201                Why the initial water content decreased and water content at the dry end increased? Seems like bulk density of soil changed by the cycles.

L205                In figure 4, is this data from field or lab? How deep the sensor was located for this data?

L272                What is the wall effect. Please explain in detail. Lab experiment usually is performed to obtain better dataset due to the simplicity of the boundary/initial conditions. Why the lab data fitting is always worse than field data fitting. This should be mentioned precisely.

L284-286        This is the critical point of the manuscript. Not only the boundary condition, but also sensor arrangement, soil properties affect the empirical parameters. Thus I questioned how these models can be utilized.

L291                What is “R”?

Author Response

Point 1: L82 - Figure 1 is not readable. Authors need to show it with better resolution. Also There is no dimension in the figure.

Response 1: Text font was increased. Both sketches are not scaled; the depth of the layers were included.

Point 2: L90 - FDR must be abbreviation of frequency domain reflectometry.

Response 2: The abbreviation meaning was added into the manuscript.

Point 3: L114 - Please explain what v50 and v50o are.

Response 3: An explanation of the linear velocity based on t₅₀ – v₅₀ was added. v_50o represents the linear velocity at the first infiltration cycle.

Point 4:  L117 - Is this equation valid? Please give a citation for the equation.

Response 4: The formula was slightly incorrect. The fraction t₅₀/z was corrected into z/t₅₀ (linear velocity)_. A citation for the determination of t₅₀ was included.

Point 5:  L122 - Explain what is theta_o.

Response 5: The initial water content (theta_o) explanation was included.

Point 6:   L185 - Is there any standard for managers to take an action? What kinds of action are made?

Response 6: Until now there are no specific guidelines for assessing clogging in MAR infiltration basins. In every basin all over the planet they proceed differently. For instance in Orange County (USA) they remove completely the clogged layer in the upper centimetres when the infiltration rate is only 30% of the initial infiltration capacity, whereas in Israel they don’t remove the clogged layer, they plough the basin floor. Here comes the relevance of these methods to give the managers a solid indicator on how advanced is the clogging process of their basins.

Point 7:  L201 - Why the initial water content decreased and water content at the dry end increased? Seems like bulk density of soil changed by the cycles.

Response 7: This is exactly because of the development of a clogging layer in the floor basin when using river water. This algae/colloids formed layer reduces the hydraulic conductivity of the soil, and therefore let less water flow through. This ends up in less filled pores with water in the unsaturated zone. During the drying phase the clogging layer blocks as well the air entrance delaying also the draining of the water filled pores. These changes of water content were proven to follow also the changes in the hydraulic conductivity.

Point 8:  Line 294: Change “adquire” to “acquire”

Response 8: Done

Point 9: L205 - In figure 4, is this data from field or lab? How deep the sensor was located for this data?

Response 9: This data is from the field. Information was included now in the caption of the figure. – Information about the depth of the sensor was added in the first paragraph of materials and methods chapter.

Point 10: L272 - What is the wall effect. Please explain in detail. Lab experiment usually is performed to obtain better dataset due to the simplicity of the boundary/initial conditions. Why the lab data fitting is always worse than field data fitting. This should be mentioned precisely.

Response 10: “Wall effect” was renamed into “sidewall flow”, which is a preferential flow path that might appear in the inner edge between packing material and the tanks wall.

The trigger time method is very sensible to this issue in experiments with walls as boundaries because it tracks the time of the first appearance of a change in the water. The influence is greater when it comes to unsaturated experiments with drying and wetting cycles happening constantly.

The water plume vertical flow is influenced by the wall – the vertical water flow is forced if the water reaches the wall, overestimation of the flow velocity in opposite to the field where the water can still flow lateral.

Point 11: L284-286 - This is the critical point of the manuscript. Not only the boundary condition, but also sensor arrangement, soil properties affect the empirical parameters. Thus I questioned how these models can be utilized.

Response 11: Yes, it is assumed that the sensor should stay at the same depth. The idea is that the MAR sites are going to require a “calibration round”. This is a first infiltration event with tracer and the installed sensors at a given depth. From this data a regression analysis can be executed as it was done in this research and with the obtained empirical equation the managers could track the infiltration capacity of their basin for the upcoming infiltration cycles in time and decide/program the best moment for maintenance.

Point 12:  L291 - What is “R”?

Response 12: This was a typo in the text and was deleted.

Reviewer 3 Report

The submitted manuscript "Methods of in situ assessment of infiltration rate reduction in groundwater recharge basins" compares different methods of estimating the infiltration rate reduction through time using in situ water content sensors. Overall, I think that the justification for these types of methods could benefit the water community. However, there are some significant shortcomings of the current approach. For this reason I am unable to recommend publication at this time.

My biggest concern is the methods of the approach that is being implemented. The regression models are developed using tracer experimental data to determine the parameter values. These values are then tested directly against the same tracer experimental data. So, the good fit of these regression models should not be surprising since the parameters are being tested against the same data that was used to develop them. This is the biggest shortcoming of the study.

Furthermore, an article of this nature needs much more discussion as to the application and shortcomings of the methods. Currently, the article has no discussion section.

I also think that the english writing needs significant improvement and proof-reading. I understand that English may not be the authors first language, so perhaps a colleague or native english speaking friend would be willing to assist in this regard. This could greatly increase the impact of any final product.

Below are some more specific comments by line number:

43-44: Hydraulic conductivity, by definition, is for water and permeability is for a fluid. Please correct for consistency in your definition.

78-79: This statement seems somewhat out of place. Perhaps a further site description could be written in the methods section with this type of information.

89-90: So what depth or depths are the sensors? This is confusing here. Also FDR or TDR?

Fig. 1: Please further describe the setup, instrument depths, etc. I understand it is described in another paper, but a brief summary would go a long way here.

Eq. 1: V50 and V50o are never defined.

119: please cite Richards and Darcy-Buckingham original papers.

125: This inferring needs to be better spelled out for the reader. Jumping from equations for hydraulic conductivity to an equation for flux should be better supported.

134: Why is theta a subscript?

137: this equation, and the one prior, are long enough to be it's own numbered equation.

Eqs. 4-6: using tracer data to develop the parameters and then comparing to the same tracer data is a major issue as I mention earlier.

186-187: This sentence is poorly written and an example of the necessary proof-reading for english writing that I noticed throughout the manuscript.

204: So the most important part of the drying data occur in the first 60 minutes, the same time that is used as justification to not use the wetting period data. 

258-259: Please correct equation numbers.

Conclusion section: Much of the conclusions, especially the last 2 paragraphs, are not actually conclusions but more of a discussion. Please write a discussion section to discuss the implications of the methods that expands on these statements. Conclusions should be more definitive, based on the data.

Author Response

Point 1: 43-44: Hydraulic conductivity, by definition, is for water and permeability is for a fluid. Please correct for consistency in your definition.

Response 1: This was edited.

Point 2: 78-79: This statement seems somewhat out of place. Perhaps a further site description could be written in the methods section with this type of information.

Response 2: The statement was reformulated.

Point 3: 89-90: So what depth or depths are the sensors? This is confusing here. Also FDR or TDR?

Response 3: The sensor is 28 cm below the infiltration basin. The infiltration basin is 6 cm deep. The latter information was deleted. The water content sensors are FDR-based.

Point 4:  Fig. 1: Please further describe the setup, instrument depths, etc. I understand it is described in another paper, but a brief summary would go a long way here.

Response 4: A more detailed description of the setup was included.

Point 5: Eq. 1: V50 and V50o are never defined.

Response 5: Definition was incorporated.

Point 6:  119: please cite Richards and Darcy-Buckingham original papers.

Response 6: Citations were included.

Point 7:  125: This inferring needs to be better spelled out for the reader. Jumping from equations for hydraulic conductivity to an equation for flux should be better supported.

Response 7: A more detailed explanation of the equation was incorporated.

Point 8:  134: Why is theta a subscript?

Response 8: This was edited now.

Point 9:  137: this equation, and the one prior, are long enough to be it's own numbered equation.

Response 9: Both equations were numbered.

Point 10:  Eqs. 4-6: using tracer data to develop the parameters and then comparing to the same tracer data is a major issue as I mention earlier.

Response 10: The purpose of comparing the developed equation with the same data that was used for the regression is to demonstrate that the different explained methods, especially water content, can follow up the changes of the hydraulic conductivity. Water content could also from the beginning show a completely different trend than cannot explain hydraulic conductivity changes. We are aware the equations need to be validated with the data coming from other experiments and for this, upcoming test are being programmed.

Point 11:  186-187: This sentence is poorly written and an example of the necessary proof-reading for english writing that I noticed throughout the manuscript.

Response 11: The sentence was reformulated and the paper was once again given to proof-reading to native speakers and experts in the area of science and engineering.

Point 12:  204: So the most important part of the drying data occur in the first 60 minutes, the same time that is used as justification to not use the wetting period data. 

Response 12: We apologize for this inaccurate description of the data. The saturation conditions are reached within the first 10 min of infiltration, whereas a slower draining gradient starts just after 70 min but is still changing after several hours. We include a graphical presentation (to be found in the attached file) of the water content in the first infiltration cycle in the field experiment. The infiltration runs until minute 1440 (1 day). We also included in the annex a table with the first 150 min of wetting and drying for this cycle, so you can confirm the changes of water content.

Point 13:  258-259: Please correct equation numbers.

Response 13: This was edited.

Point 14: Conclusion section: Much of the conclusions, especially the last 2 paragraphs, are not actually conclusions but more of a discussion. Please write a discussion section to discuss the implications of the methods that expands on these statements. Conclusions should be more definitive, based on the data.

Response 14: The conclusion section was reduced in order to be more definitive. The last paragraph aims exactly to explain that a validation process of the defined equations has to be made with a new set of data.

Annex

Author Response File: Author Response.docx

Reviewer 4 Report

This manuscript is well written

Author Response

Point 1: This manuscript is well written

Response 1: Thank you! We appreciate it!