Groundwater Flow-Modeling and Sensitivity Analysis in a Hyper Arid Region

Round 1

Reviewer 1 Report

This manuscript deals with groundwater flow modeling for the limestone and dolomitic limestone in a hyper arid region in Iraq. Overall, this manuscript needs to be resubmitted after more detail analysis with considering climate change and more hydrogeological information. For the groundwater modeling of the study area, one needs to obtain accurate input data such as hydraulic conductivity, groundwater level, aquifer geometries, boundary conditions, etc. Among these data, hydraulic conductivity (K) is highly important in limestone areas depending on the distribution of caverns and rock matrix. In this study, the K values are between 1x10^-3 (or 86.4 m/d) to 1x10^-4 m/s (8.64 m/d) for the aquifers and between 1x10^-7 (or 8.64x10^-3 m/d) to 1x10^-8 m/s (or 8.64x10^-4 m/d) for the aquicludes and is 3.3x10^-5 m/s (or 2.85 m/d) (line 94-95). Hence, how did you estimate the K values of the aquifers and aquicludes as well as the fault zones? In addition, the role of the two fault zones is important in the study area. Please indicate the strike, dip angle, and thickness of the fault zones. I don’t understand an assumed thickness of 5 mm/year (line 16-17). In addition, is the faults partly distributed in the model area like in Figure 3?

Figure 1 should include the line for the cross section of Figure 2. In Figure 2, discharge zone should be displayed. The names of aquifers in Figure 2 should correspond to the aquifers in Table 1. Also, you need to add the geological map showing rocks distribution with fault 1 and 2 lines.

You mentioned only the groundwater levels of the Dammam aquifer (line 42-43). What are the groundwater levels of the Um Er Radumma and Tayarat aquifers? I can’t see the effect of groundwater flow by the control of the fault zones (Figure 5). Also, the section ‘3.2 Model sensitivity’ is too shortly described but the analysis of model sensitivity is one of key subjects of this study. You should explain the result of sensitivity analysis for hydraulic conductivity, groundwater recharge, and boundary conditions in more detail. According to the calibration in Table 5, Kx and Ky of the aquifers are different: Kx>Ky in Dammam Aquafer, Kx=Ky in Um Er Radumma Aquifer, and Kx<Ky in Tayarat Aquafer. I don’t understand the difference depending on the aquifers since groundwater flows from west to east.

‘4. Conclusion’ is too short. You have to mention all the key result of your study including sensitivity analysis and fault zone effect on the groundwater flow with stating some uncertainty of the modeling.

Additionally, you should check ‘References’ by the instruction of ‘Water’.

Some duplicate and other comments are indicated in the review file.

Author Response

Dear Reviewer,

Thank you very much for reviewing our humble thoughts and manuscript. We also greatly appreciate the reviewers for their time , complimentary comments and suggestions. We have revised the manuscript accordingly. Please find enclosed a point-by-point response to reviewer’s concerns also all have been reflected in red at the revised manuscript in addition to a clean version to have easier orientation . We hope that you find our responses satisfactory.

Sincerely,  

Reviewer # 1 Comments and Suggestions for Authors

This manuscript deals with groundwater flow modeling for the limestone and dolomitic limestone in a hyper arid region in Iraq. Overall, this manuscript needs to be resubmitted after

• More detail analysis with considering climate change and more hydrogeological information.

Answer: Thanks for the comment, a paragraph on this context (Climate change and hydrogeology) has been added to the section of 2. Materials and Methods.

• For the groundwater modeling of the study area, one needs to obtain accurate input data such as hydraulic conductivity, groundwater level, aquifer geometries, boundary conditions, etc. Among these data, hydraulic conductivity (K) is highly important in limestone areas depending on the distribution of caverns and rock matrix. In this study, theK values are between 1x10^-3 (or 86.4 m/d) to 1x10^-4 m/s (8.64 m/d) for the aquifers and between 1x10^-7 (or 8.64x10^-3 m/d) to 1x10^-8m/s (or 8.64x10^-4 m/d) for the aquicludes and is 3.3x10^-5 m/s (or 2.85 m/d) (line 94-95). Hence, how did you estimate the K values of the aquifers and aquicludes as well as the fault zones?

Answer: Thanks, Due to the lack of detailed hydrogeological data in the region we used a uniform Kf value for each stratigraphic formation and then change it in reasonable range by manual and followed automatic calibration. Transmissivties (Table 1) were transfored into initial hydraulic conductivities. The Kf of faulty zone was obtained from previours conducted pumping test close to the fault zone. A paragraph has been added to the section to reflect this along with the reference of the previous study.

• In addition, the role of the two fault zones is important in the study area. Please indicate the strike, dip angle, and thickness of the fault zones. I don’t understand an assumed thickness of 5 mm/year (line 16-17). In addition, is the faults partly distributed in the model area like in Figure 3?

Answer: Thanks for the heads up, the typo “year” word has been removed based on your review, The thickness has been assumed as referred in the paper . Since there is no package to imitate a real fault in MODFLOW; there are generally two options: First: assign the expected or analysis obtained Kf value to the model cells where faulty zone is distributed. Second: using a Wall boundary condition, or so called Horizontal Flow Barrier (HFB) package included with MODFLOW. The package was developed to simulate thin, vertical, low permeability features that could impede the horizontal groundwater flow. The boundary hast to be assigned to the fault and adjacent cells in the finite difference grid (width of the barrier). The hydraulic characteristic of the barrier then calculated by dividing the hydraulic conductivity of the barrier cells by the width of that barrier.  A paragraph has been added to the section 2.2.2 boundary conditions to clarify this case.

• Figure 1 should include the line for the cross section of Figure 2.

Answer: Thanks, the Fiogure1 has been modified and cross section has been added.

• In Figure 2, discharge zone should be displayed.

Answer: The discharge zone has been displayed.

• The names of aquifers in Figure 2 should correspond to the aquifers in Table 1.

Answer: The aquifer names for Figure 2 has been corresponded to the aquifer Tavle.1

• you need to add the geological map showing rocks distribution with fault 1 and 2 lines.

Answer: Geological map has been added.

 

 

• You mentioned only the groundwater levels of the Dammam aquifer (line 42-43). What are the groundwater levels of the Um Er Radumma and Tayarat aquifers? I can’t see the effect of groundwater flow by the control of the fault zones (Figure 5).

Answer: Thanks for the comments, that represents the static groundwater table. As shown in the plane view the fault zone plays in some parts to the north as barrier with limited impact on groundwater flow, particularly the fault in southern edge of the model. This can be explained the cross direction of the fault parallel to the major groundwater flow direction, so hydraulic barrier effects on the regional scale isn’t significantly remarkable.

 

• Also, the section ‘3.2 Model sensitivity’ is too shortly described but the analysis of model sensitivity is one of key subjects of this study. You should explain the result of sensitivity analysis for hydraulic conductivity, groundwater recharge, and boundary conditions in more detail.

Answer: Thanks for the comment, the paragraph has been improved and more details have been given considering the reviewer comment, The model isn’t sensitive to the transfer component of the hydraulic conductivity which could be explained by the general groundwater flow direction from the East to the west of the model with an average velocities much high than those in transfer or vertical direction. That movement also limited the hydraulic impact of the fault zone, where groundwater flow obliquely to barrier. On the other hand, the model is sensitive to the groundwater recharge close the southwester edge of the model, higher value than 17 mm/a caused model inconvergance and unreasonable groundwater flow pattern.  

• According to the calibration in Table 5, Kx and Ky of the aquifers are different: Kx>Ky in Dammam Aquafer, Kx=Ky in Um Er Radumma Aquifer, and Kx<Ky in Tayarat Aquafer. I don’t understand the difference depending on the aquifers since groundwater flows from west to east.

Answer: Thanks for your comment, the paragraph has been improved since the automatic parameter optimization shows that, he Dammam and Tayarat Aquifer isn’t sensitive to the change in Ky value, so after 4 iteration a homogenous K distribution was observed suggesting same K value.

• Conclusion’ is too short. You have to mention all the key result of your study including sensitivity analysis and fault zone effect on the groundwater flow with stating some uncertainty of the modeling.

Answer: Thanks for your comment, the conclusion has been modified to reflect your kind suggestion, providing the key results of the study.

• Additionally, you should check ‘References’ by the instruction of ‘Water’.

Answer: References have been reviewed and developed based on Water journal policy.

• Some duplicate and other comments are indicated in the review file.

Answer: Thank you very much for your  remarks, the comments have been considered and modified as reviewer recommended.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors,

I find the article very interesting, especially since the model investigates these “strategic” aquifers (i.e. Um Er Radumma and Dammam) for the region. Please consider the following queries from my side, that in my opinion will improve the paper in some minor details.

Query 1: more detailed objectives of the paper are need in the last paragraph of introduction.

Query 2: although Figure 2 illustrates the recharge and groundwater flow processes very well, it should contain a vertical axis or any kind of illustration that provides information about the thickness of the aquifer layers (although mentioned in Table 1).

Query 3: the paper lacks information about the time discretization of the model (stress periods, time steps), the spatial discretization (grid dimensions etc), model layers and the type of flow during each calibration step (steady/transient state)

Query 4: how does the model considers the presence of wadis in the study area? 

Query 5: The discussion section should be improved with additional information (E.g. summary of the model and the study area) main conclusions about the critical parts of the model (boundary conditions and calibration process). I would also be interested in listening the author’s view on the effectiveness of MODFLOW for simulating limestone aquifers (such as in this case) as well as on the use of UZF package in your model.

Author Response

Dear Reviewer,

Thank you very much for reviewing our humble thoughts and manuscript. We also greatly appreciate the reviewers for their time , complimentary comments and suggestions. We have revised the manuscript accordingly. Please find enclosed a point-by-point response to reviewer’s concerns also all have been reflected in red at the revised manuscript in addition to a clean version to have easier orientation . We hope that you find our responses satisfactory.

Sincerely,  

Reviewer # 2 Comments and Suggestions for Authors

I find the article very interesting, especially since the model investigates these “strategic” aquifers (i.e. Um Er Radumma and Dammam) for the region. Please consider the following queries from my side, that in my opinion will improve the paper in some minor details.

• Query 1: more detailed objectives of the paper are need in the last paragraph of introduction.

Answer: Thanks for your comment, the objective has been improved.

 

• Query 2: although Figure 2 illustrates the recharge and groundwater flow processes very well, it should contain a vertical axis or any kind of illustration that provides information about the thickness of the aquifer layers (although mentioned in Table 1).

Answer: Thanks for the heads up, Figure1 has been modified.

• Query 3: the paper lacks information about the time discretization of the model (stress periods, time steps), the spatial discretization (grid dimensions etc), model layers and the type of flow during each calibration step (steady/transient state)

Answer: The current model is limited to steady state simulation, providing static conditions of the hydraulic system, modeling of transient condition would need more data and we agree it would be good option for future project. We added a sentence on this context in the2.2.1. Properties and initial data and conclusion.

 

• Query 4: how does the model considers the presence of wadis in the study area? 

Answer: Thanks for your comment, The main Wadi has been studied in details in our previous work , please have a look on the reference below:

Al-Muqdadi, S.W. and Merkel, B.J., 2011. Automated watershed evaluation of flat terrain. Journal of Water Resource and Protection, 3(12), p.892.

 however simulating the Wadi using drain boundary condition is beyond the current project.

• Query 5: The discussion section should be improved with additional information (E.g. summary of the model and the study area) main conclusions about the critical parts of the model (boundary conditions and calibration process). I would also be interested in listening the author’s view on the effectiveness of MODFLOW for simulating limestone aquifers (such as in this case) as well as on the use of UZF package in your model.

Answer: Thanks for your comment, the conclusion has been improved to reflect your kind suggestion, providing the key results of the model and the debate of using the MODFLOW and the UFZ has been highlighted as well in both the conclusion and the methods section.

Author Response File: Author Response.pdf