Integration of Electrical Resistivity Tomography and Seismic Refraction Tomography to Investigate Subsiding Sinkholes in Karst Areas

Round 1

Reviewer 1 Report

Dear Authors,

Thank you very much for submiting your work. It is my opinion that your work needs a very deep reorganization before I can accept it. 

I am afraid as it is I recommend to reject it. If however you decide to reorganize it I can have another go at it.

Please read the following notes:

This work intends to locate sinkholes and subsidence using the combination of Resistivity –ERT – and seismic refraction tomography -SRT.

Regarding the resistivity survey it applies the dipole –dipole array and then produces models using a 2D imaging commercial software and then an open source software pyGIMLi. As a first point I fail to see the importance of inverting resistivity data with both software systems. If the open source has proved its application before, as it apparently has done, I don’t understand the need to use both programs. Authors claim that the Python software can use arrays with arbitrary shape, can detect noisy points and outliers but commercial software derived from the res2Dinv can also do that easily. So if authors wish to compare software that would be another work, and should not distract readers from the main aims of this work, that is, resistivity plus seismic refraction in the location of sinkholes and subsidence. The scheme in fig. 4 is similar to any 2D inversion scheme I am afraid.

I feel the work needs strong reorganization. First, different chapters should be in bold letters to be highlighted. The way they are presented one can hardly notice them.

Then authors should organize the work as Introduction, containing a general overview of the problem, the problem and the aims of this work. Second the geology, location and characteristics of the area under investigation. Then the location of profiles with regard to the targets and the justification of the profile location. At the moment this information is in figure 1. This figure must, and this is a must, the surface limits in dashed lines of the sinkholes. Then the methodology, a brief description of the techniques, as they are widely known, but I precise description of the geometry and characteristics of this survey, as well as the justification of the parameters undertaken.

Then the processing. In my view if authors wish to compare different programs they distract the reader from the aims of the work. Choose one and justify. Figure 4 in my opinion is just a general scheme for the 2D inversion of resistivity data and should be deleted. The explanations of the python routine are tedious and I am sure that a reference to manual of the program would do.

Now the results. Authors have performed 7 resistivity profiles. Figure 5 and 6 provide the models ( I assume they are models not field data) obtained from the use of commercial software. First these figures must be three dimension. Authors have parallel profiles and profiles that intersect each other. A three D view is mandatory. The North and South, the East and West, etc must be marked at the respective end of the models. Both axis must have unities. The legend below the section must be explained (iterations, RMS previously it was RMSE –line 411 – if I am right, L2 ??) and interpretation must be in dashed lines, never a full line.

Authors should add that these models are the outcome of the commercial software they used.

I am not going into the discussion of figures  5 and 6 as they  must be 3D showing the intersections of the profiles and the fitting of the models in the different profiles. Any interpretation should be based in the geology. I must say that it is not easy to understand that sinkholes are sometimes resistive and other times they seem to be conductive. What is happening? They are adjacent sinkholes aren’t they? Are there any orientational effects? Anisotropy?

The discussion on outliers, misfit, errors can also be done with commercial software. Res2Dinv does that and accepts arbitrarily shaped arrays.

Sections in Fig. 11 are not identified, no North, South, etc. Once again it is a must to present them in a three D view. Colours must be the same as the ones in Figures 5 and 6 for a full comparison. Colours mislead people easily. Interpretation must be in dashed lines always and as I pointed out before.

Regarding the seismic survey authors mention they used 5 shoots. There is an urgent need to know their location, far-end, end, central, end, far-end I suppose if they use the generalized reciprocal method. Authors must also tell the reader the distances to the next geophone, that is the geometry of the spread.

I don’t understand why the elevation axis are different from the resistivity ones. The values in the curves cannot be read and the numbers in the colour scale are very small.  Authors must explain what is Wp/P) in the right of figures 12 and 13.

I cannot find the 2D WET tomographic model. What is it?

I am not going into details. I could have a go at the reorganized paper if authors are interested in producing it. As it is, it is my opinion that this work should be rejected. The problem is well known, these methods have been widely applied to these studies, and conclusions are trivial and authors keep saying “… it appears…” which is not scientific at all. In the last paragraph they even use “..which might lead…”.

The English must have some revision.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Minor comments

p.3, l. 124: Fig. 1 – indicate the surface outline of the two sinkholes

 p.13: Fig. 5. - Label N and S etc. properly, i.e. N at left and S at right (?) and NOT N-S in the right corner. Same for all other figures.

Also use the spatial same scale (e.g. in meters) – L1 is ca. twice as long as L2, and L3, etc.

 p. 15: Fig. 6 -same as for Fig. 5.

 p.16: Fig. 7 – what is a relative misfit of -96.05 % - a good fit? Please explain.

 p.17: Fig. 8 – use identical color scale in both figures (L1 and L3)

 p. 21: Fig. 11 – use the same horizontal and vertical scale and construct a 3 D grid plot.

 p.22, L.565 - Figure 12 shows

Major comments

 ERT:

Do the values shown agree at crossing points of profiles (e.g. profiles in Fig. 11)

Produce a 3D grid of the ERT profiles to show (or not show) the internal consistency of the models derived.

 Is a 3D inversion possible?

 SRT:

p. 22/23. Fig. 12, 13 & 14:

L2 and L3 cross L1. – do the values derived for each figure agree at their crossing points?

Show a 3D plot with the 3 lines intersecting.

I suggest to furthermore discard (blank out) areas in the left part of these figures, that are covered by less than 10 rays in the right part of each figure. These areas are not reliably resolved, e.g. all areas deeper than 10 m in Fig. 12. Keeping them in the velocity images falsely suggest that these areas are reliably resolved. This holds even so the RMS values are small.

 Comparison of results of ERT and SRT:

The findings of the two methods employed should be compared and re-discussed in the light of the above comments.

--

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Authors,

Thank you very much for your efforts and for the improvement you carried ou tin your work.

Please read the attached file or the following lines

2nd review on the paper “Integration of Electrical Resistivity Tomography (ERT) and Seismic Refraction Tomography (SRT) Methods to Investigate the Effects of Sinkhole Subsidence in Sierra de Gádor, Spain.”

Authors: Oussama Jabrane * , Pedro Martínez-Pagán , Marcos A. Martínez-Segura , Francisco Javier Alcalá , Driss El Azzab , Marco D. Vásconez-Maza , Mohammed Charroud

Line 20 “This work is aimed at proving the feasibility of” change to “This works aims to use the Electrical ….” NOTE: These methods have long been used with these aims.

Line 49 change “weakness” to “weakened” or to “areas of weakness”

Line 54 remove “involved”

Line 63 remove “-“

Line 64 change “ which show  well marked contrast regarding the bedrock” to “ whose physical properties can be very different from those of the surrounding rocks”.

line 71 change “ because “ to “and”

line 73 do you mean “reading” or “readings” that is field measurements?

Line 80 “bedrock” change to “rocks”. Bedrock has a different meaning. Here carbonate rocks are not bedrock but rocks that surround cavities.

Line 83 change “layers’ ” to “layers”

Line 151 “ in a vertical 2D plane in line with the “. This could be correct if there were no lateral and 3D changes. I would prefer to write it as “ in the region bellow the acquisition line”

Line 158 the English is awful. “This array is most sensitive to horizontal ER changes than to vertical ones, thus it is good at mapping vertical structures better than horizontal ones” Perhaps like this “ Tis array is  more sensitivity to lateral changes of ER than to vertical changes. Therefore, its response to vertical structures is better than that for horizontal structures”.

Line 159 “apparent ER pseudo sections” is repeating the concept twice. You can write “apparent ER sections” or “ER pseudo sections”. Please be careful with the terms you use.

I am sorry but figure 3 depicts pictures so small that details of field operations cannot be seen  properly. For example, one cannot hardly the electrodes, the impact plate and its coupling to the ground is impossible to see.

Line 171 “roll-along” must have a reference. It was not invented by the authors.

If you used a constant electrode distance of 2.5 metres did you have problems at the longest distances? Don’t answer but add comment or explain in the paper.

Line 177 I believe I noticed it before. Shaking is usually  related with S waves. You cause impact into the ground to produce P waves. So rewrite please.

Line 179 the technique doesnot analyze P waves. It uses P waves. Please rewrite.

Line 180/81 can be delected. Practitioners in the XXI century know how to calculate P wave velocities.  Furthermore it is causes confusion “beyond direct waves”?

Line 191 “conventional reciprocal mode” needs a reference.

Authors must include a sketch of the geophones positions. I asked for this in my previous review. If five shots were used readers need to know their positioning if the reciprocal method is to be used.

Line 193 needs a reference.

Line 202 “sensors of vertical movements”? delete this.

Line 204/205 As I wrote before a sketch of the geophones positioning is a MUST.

Line 205/206 ??? please delete and draw a sketch.

Line 206 How did you monitor noise level? Please explain in the text.

Line 207 if you are staking the records as noise is not in phase the more strikes you have the less noise is added as it cancels out. It should be like this.

So the question remains: how did you monitor the noise level?

Line 213 change “achieved” to “ calculated”

Line 222. Change “; they ” to “and”

Line 222 change “exhibited” to “show”

Libe 228 you refer “three resistive regions” but the first layer has resistivity values between 14 and 40 ohm.s, whilst the others depict values more than 10 times higher. I don’t understand. Please rephrase. In my view one is conductive and two are resistive.

Line 237 I mention it before: when you use the word bedrock usually you mean a layered earth. Here I presume you refer to the rocks that surround the cavities. Isn’t it?  Please rephrase.

Line 239 by “quite” you mean “some” or “several” or “many”? Change accordingly.

Line 244 again “resistive” see previous note and change.

Line 250 again “bedrock”. See previous note.

Be careful with the use of the word “bedrock” in this context.

Line 253 you mention a “fracture”. Did you see in the field? Is it marked in the geological maps? Or is ti inferred by the geophysics interpretation? If is after geophysics you must write “interpreted from geophysical data”

In figure 4 what are the black dashed lines and the white dashed lines? What are the numbers 1, 2 etc. You must mention these features in the discussion between lines 255 and 274.

Line 287 change “allows visualization” to “provides an image”.

Line 279 change “allowed” to “contributed”.

Figure 5 again the white dashed lines. What us their meaning?

In the legend change ohm-m to ohm.m. The same in figure 4.

Figure 6 again white dashed lines. What is their meaning? What is the horizontal scale of figure 6? You show the vertical scale but not the horizontal. The vertical scale is interpret depth? You must say it or if it is not say what it is.

The axis in figure 7 need unities (horizontal axis). In all figures you show “Elevation,m” change to “Elevation (m)”

Line 367/369 this is trivial.

Be careful with the use of the word bedrock See the notes above.

Libe 372 change “identified” to “interpreted”

Line 375 what do you mean by “cartography”?. Change or explain in the text.

Line 377/378 again this is trivial. Otherwise you would not have refractions to interpret. Remove.

OK this version of the work is vastly improved from the previous version. It can be published since the above notes are completely observed. This work is not a development of the techniques, even of the combined techniques, but it is a case study of local interest. So I regard it as a simple case study.

 Comments for author File: Comments.pdf

Dear Sirs,

The present version of the work is an considerable improvement from the previous version. In the review.pdf file I send you my remarks. I feel the observations I make must be adressed for a better clarification and improvement of the work.

The work is a simple case study with local application and does not give any improvement on the techniques used.

I feel that some English editing is still desirable.

Regards

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Minor comment:

L3: sinkhole -> sinkholes

 Mayor comment:

Fig. 7: Overlaying the ERT result with the SRT result makes the underlying ERT almost impossible to see.

 I strongly suggest to show the 4 profiles separately and individually (!) above each other (i.e. S2 located above E2) and (S3 located above E3) – i.e. 4 Figs. Each with its color scale, and no overlay (!).

It seems also as if the S2 and S3 profiles were truncated and not shown over their whole length, as given in Fig. 6.

Author Response

Please see the attachment.

Author Response File: Author Response.docx