Mapping Alteration Mineralogy in Eastern Tsogttsetsii, Mongolia, Based on the WorldView-3 and Field Shortwave-Infrared Spectroscopy Analyses

Round 1

Reviewer 1 Report

Data are presented in a confusing way. Figure 6 is difficult to read, and suggests wrong identification of fresh surface and weathered surface spectra.

Good general outline. Suggestions to supress unnecessary detailed information are made.

Please see the attached pdf file.

Comments for author File: Comments.pdf

Author Response

We would like to thank your review of the paper. In accordance with the comments and suggestions, we revised the manuscript.

"Please see the attachment."

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors provided a manuscript on  “Mapping Alteration Mineralogy in Eastern Tsogttsetsii, Mongolia, Based on the WorldView-3 and Field Shortwave-Infrared Spectroscopy Analyses”, but the overall accuracy of their mapping method is too low to guarantee an adequate abundance map of alteration minerals (Porphyry Cu deposits).

In my opinion, in the present form the manuscript is not ready for publication and needs a major revision.

Some additional comments are to be considered by the authors before the manuscript can be considered for publication.

Abstract

Line 17: good is not adequate as a K of 0.47 is bordering on sufficiency. Moreover, why don't you also show the accuracy results of the band math?

Introduction:

This section needs to be completely revised, some parts should be moved from here and put into a separate paragraph (e.g. add a new section2. with the Study area). For example, Lines 44-45 and Table 1, Lines 77-86 and Figure 1. Moreover, without this parts the Introduction is too poor and has to be improved following the standards of the Remote Sensing journal.

Lines 66-68: provide a reference for this sentence.

3. Methods:

Line 179: editing error "There are".

As mentioned in the line 183, the authors conducted a study on the same area using ASTER data, it would be nice the comparison of the results of these two studies but with a real atmospheric correction applied to both datasets.

Lines 193-194: add a separate sub-section with a description of the XRD, XRF and EPMA used for this study and add also the acronyms' meaning.

Section 3.2: the IAR atmospheric correction method used is too simple and doesn't guarantee adequate reflectances, I suggest to the authors to use other atmospheric correction methods that implement the MODTRAN (or similar) radiative transfer code.

Lines 223-230: this paragraph is completely wrong, remove it. This is not a correct way to verify the quality of the applied atmospheric correction, in particular the concept of "reflectance calibration" is wrong as we must speak of atmospheric compensation or correction of the satellite data and not calibration!

Lines 416-417: why applying to WV3 data a DS? It is not clear to me. It is only a qualitative way to operate and old method used by [44] for another kind of images and application.

Lines 434-435: DS and BM combination bands were used to highlight or to map? This is not clear to me and if you used them to map how can you be confident what are you mapping? In those SWIR spectral ranges with the WV3 spectral resolution and FWHM there are several minerals absorption features.

Results and discussion sections are well written and presented with the relative figures, but the methods used to obtain the results are not adequate to produce reliable quali/quantitative mineral maps. 

Conclusions:

Line 678: with such a low accuracy with a K of 0.47, obtained for a single WV3 SWIR imagery, atmospherically corrected with the IAR method, how can you affirm that the WV3 SWIR data are effective for mapping alteration minerals? 

Author Response

We would like to thank your review of the paper. In accordance with the comments and suggestions, we revised the manuscript.

"Please see the attachment."

Author Response File: Author Response.pdf

Reviewer 3 Report

Overall comments

This paper tackles then extraction of mineral alteration information associated with a porphyry system in Mongolia using WV3 multi-spectral satellite data and a variety of traditional image processing methods, namely decorrelation-stretch, band ratios and MTMF.   The results are validated against ground spectral data and laboratory analyses.   

This is a good piece of work in terms of its design, content and expression and thus should be considered for publication. 

Specific comments

One suggestion for consideration/change by the authors is their use and appraisal of band ratios versus MTMF (from Lines 498 onwards).   The ammonium-bearing minerals absorption near 1550 nm is arguably better targeted using WV3 bands 2, 3 (and 4) - not bands 5 and 6 (and 2).   What do the weightings for the buddingtonite (and NH4-illite) MTMF eigenvector show?  As expected for the limited WV-3 spectral resolution, there will be complicating overlap of the long-wavelength absorption shoulder to the 1400 nm water absorption by minerals like illite and montmorillonite with this narrower ammonium 1550 nm feature.    But this is problem a problem no matter what type of information extraction is used.  

The challenge for methods like MTMF is that they rely on scene statistics to generate a “background” in order to then calculate a given target's "mixing" chord.   In contrast, band ratios should be designed to use a given pixel's mineral absorption spectral-continuum as its “background”.  That is, continuum-band ratios are not reliant on scene-derived statistics and thus are more transferable to other scenes. 

You have generated 8 MTMF spectral endmembers in Figures 14 and 15, which of course is the maximum given the 8 available WV3 bands (data dimensionality = N bands +1 with one of these dimensions being albedo/topographic shading).   The problem is whether your 8 “endmember” spectra uniquely span the available 8 dimensions.   Clearly they do not given your reported commission problems and the large amount of unmapped areas in Figure 15.  Ultimately, the success (accuracy) of any information extraction method for the wider user community is how well such methods can be “seamlessly” transferred to other image scenes.  

Though outside the wavelength region scope covered by your paper, you do repeatedly mention hematite as an alteration product, which WV3 has more VNIR bands than ASTER for better mapping of  iron oxides.     Maybe an opportunity for the authors to write an associated paper.  

Ln 168-174: Delete

Ln 204-216:  How did you assess the accuracy of your offset correction in the image data?   For example, was topographic shading variation nulled over higher relief areas? 

Ln 614: higher Na and

Ln 622: Pyrophyllite!   A key advanced argillic alteration mineral and with pronounced 2165 nm absorption.    Why no significant consideration given in the paper?

Lns 631-632:  Not true – well ordered kaolinite forms in weathering environments.

Lns 633-635:  What is the difference between: “forms in a zone of advanced argillic alterations or during hydrothermal activity by high sulfidation”?

Author Response

We would like to thank your review of the paper. In accordance with the comments and suggestions, we revised the manuscript.

"Please see the attachment."

Author Response File: Author Response.pdf

Reviewer 4 Report

Mapping Alteration Mineralogy in Eastern Tsogttsetsii, Mongolia, Based on the WorldView-3 and Field Shortwave-Infrared 3

Spectroscopy Analyses

Authors proposed a method in alteration mineral maps using WorldView-3 (WV-3) data and 10

field shortwave-infrared (SWIR) spectroscopy. They used X-ray diffraction, X-ray fluorescence, and electron probe X-ray micro analyzer and mixture-tuned mated filter.

Herewith, these are some highlighted points of this paper:

• The research subject is interesting as it describes a thermographic approach for measuring which might absorb many readers;
• The analysis seems adequate and represents considerably;
• The paper is clear for presenting the approach;

Besides these points, there are some suggestions that can improve the quality of your work:

• The major comment concerns with this manuscript would be to ignore state-of-the-art methods particularly [1-5]. Particularly using matched filtering authors need more literature connection.
• The novelty of this approach is not clear.
• Please make your figure bigger and more readable than the current one. There are many valuable information that need to be described better.

In general, this work is almost adequate but before that authors need to really work on it to be ready. Therefore, I recommend major revision.

Thank you,

References

1. Assessing the reliability of an automated system for mineral identification using LWIR Hyperspectral Infrared imagery
2. Mineral mapping using simulated Worldview-3 short-wave-infrared imagery
3. Comparison assessment of low rank sparse-PCA based-clustering/classification for automatic mineral identification in long wave infrared hyperspectral imagery
4. Mineral identification using artificial neural networks and the rotating polarizer stage
5. Alteration mineralogy at the Cerro La Mina epithermal prospect, Patagonia, Argentina: field mapping, short-wave infrared spectroscopy, and ASTER images

Author Response

We would like to thank your review of the paper. In accordance with the comments and suggestions, we revised the manuscript.

"Please see the attachment."

Author Response File: Author Response.pdf

Round 2

Reviewer 4 Report

Authors responded to my comments well. I have no further questions. Congratulations!