Conversion of Induced Polarization Data and Their Uncertainty from Time Domain to Frequency Domain Using Debye Decomposition

Round 1

Reviewer 1 Report

The paper describes in detail a general purpose approach for the conversion of tomographic TD IP data sets into the FD based on the Debye decomposition, of which the Authors  describe the theoretical basis before formulating the inverse problem regarding the TD to FD conversion. Though the analytical parts are well described, I find the field examples not sufficienty described. I find that little emphasis has been given in the comments on figures 11 and 12 with respect to the geological-mining context of the area characterized by the presence of an important porphyry copper deposits.

Does the survey area have any boreholes that could be used for data calibration? It could be interesting for the readers of the Mineral journal to see the results of the tomographic inversion of data from the Maletoyvaemskoie field site (Kamchatka, Russia) with respect from the borehole data if available. If no borehole data were available, considering the importance of the area from a mining point of view, a more accurate interpretation of the tomographic inversion in geomining terms could be of great interest.

 Paragraph INTRODUCTION:

Row 31 of the original text (8-13): the scientific literature on these topics is very rich. The Authors can supplement it. 

Paragraph 5. Application to tomographic field measurements

Row 288: Please insert the location of the profiles on a geological skech map.

Row 388: The obtained images of the complex resistivity are realistic, given the expected geological context.

The Authors could more accurately analyze and describe the correlation between the results of the field data inversion (Figures 11 and 12) with the specific mining – geological conditions.

 

 

 

 

- Some sentences could be expressed more clearly by improving the style of the English language.

- Row 302 of the original text: argillic instead of argilic

Author Response

Please see the attachement.

Author Response File: Author Response.pdf

Reviewer 2 Report

Hase and others present a new workflow for conversion of induced polarization (IP) from the time to frequency domain. The contribution is highly analytical and mostly deals with the mathematics of geophysical inverse theory, and I found this to be unexpected for the journal Minerals given the limited amount of application to field data presented in this paper. I suppose that this is a decision for the authors and editors. Accepting that caveat, my opinion is that this is a well written paper describing an interesting theoretical method that may improve mineral exploration. I found only minor issues with the manuscript. However, I am an applied geophysicist and not a practitioner of theoretical arts of geophysical inverse theory. I did not double-check the derivation of the inverse problem presented in section 2.3. It looks correct. The description and formulation of the Debye equations look good, and represent a new way to utilize Debye decon in the frequency domain inversion.

The use of synthetic data to check the accuracy of the TD-FD transform using the newly developed method shows that this is a valid method but with some concerns at low (rarely used) relaxation times (cf. Figure 1). However, it seems to work well enough as illustrated in the figure 3 with error propagation considered.

The choice of the validation field data in Section 5 is a bit unusual. There are a couple of aspects to this dataset that might seem to make it less than ideal for the validation purposes. The dataset was a pole-dipole survey without a reciprocal. It seems that the ground control from boreholes or other a priori information is sparse. Also, there were a number of transients that required manual filters. All of these issues point raise questions as to why this was the applied dataset. Further justification for this dataset, or the choice of another clearer dataset as an example, should be given in the opening of Section 5. 

Another minor issue that requires some further explanation is the benefit of exploiting the FD signal in time-domain IP data. This is simply asserted at on lines 32-33. Some laboratory scale IP analysis is done in frequency domain, but I'm not aware of any field data at larger scale that uses anything but time-domain IP with geoelectric resistivity. What is the actual benefit? How is it worth the extra processing steps? Both a presentation of this justification and what the conclusion is, given the error models so carefully considered in the paper, should be included in both the paragraph and conclusion.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf