The Zonation of Mountain Frozen Ground under Aspect Adjustment Revealed by Ground-Penetrating Radar Survey—A Case Study of a Small Catchment in the Upper Reaches of the Yellow River, Northeastern Qinghai–Tibet Plateau

Round 1

Reviewer 1 Report

Comments on remotesensing-1687060

This manuscript has improved some, but is still problematic in various aspects.

1. Incorrect citations. Many citations are forged. For instance, the authors cite Neal’s (2004) GPR review paper several times, but most of those cited contents are not found in that paper. The even worse is that those statements have nothing to do with Neal’s work (Lines 93-95). Note that Neal’s (2004) GPR review paper focused on the use in sedimentology, not permafrost.
2. Confusing and unprofessional terminologies. An obvious example is terms like fixed antenna “distance”. In the GPR field, we call it “offset” rather than “distance”. Furthermore, the authors seem to mess up the directional angle, slope, aspect, and azimuth. Therefore, their reports mislead readers.
3. Unexplained figures and inconsistency between figure captions and text. A typical example is Fig. 6(c). The negative numbers shown in the figure are illogical, also, what does “a continuous function of slope aspect” indicate? I understand what the authors tried to say, but the use of the “continuous function” is not acceptable because it has a very rigid definition in mathematics. Don’t abuse a well-defined nomenclature. Since Fig. 6(c) is one of the highlights of the authors' findings. I suggest applying a more adequate method, cluster analysis, or similar algorithms, to gain an insight into the valuable result.  
4. The resolutions of Figs 2,4, and 5 are visually indiscernible, Figs 2 and 4 in particular, although I enlarged the figure size to 400%. This difficulty needs to be solved somehow.   

Author Response

Dear reviewer，

Thanks again for your professional comments，which are very helpful for the improvement of our manuscript. Based on them, we have made further modifications of the manuscript. The specific replies are as follows：

1. Incorrect citations. Many citations are forged. For instance, the authors cite Neal’s (2004) GPR review paper several times, but most of those cited contents are not found in that paper. The even worse is that those statements have nothing to do with Neal’s work (Lines 93-95). Note that Neal’s (2004) GPR review paper focused on the use in sedimentology, not permafrost.

Response: We check the manuscript and find that there are indeed ambiguous citations in some places, where we have improved the expression, or revised the citation. For the lines L93-95, we have removed the citation of Neal's paper (2004) and reorganized the sentences. In many cases, the GPR data processing method and principle reported by Neal (2004) for detecting sedimentary layers is universal for the detection of permafrost related features e.g. seasonal thawing depth. Therefore, when introducing the principle or method, we retain the reference to this paper.

2. Confusing and unprofessional terminologies. An obvious example is terms like fixed antenna “distance”. In the GPR field, we call it “offset” rather than “distance”. Furthermore, the authors seem to mess up the directional angle, slope, aspect, and azimuth. Therefore, their reports mislead readers.

Response: Many thanks for the guidance on the usage of terminology. The “antenna distance” has been replaced by “antenna offset” in the new version. We have unified the expression of “slope’s aspect”. In the revised manuscript, we used three words: aspect (as a universal expression, and abbreviated as “ASP” in the text), slope’s orientation (when describing aspect in degrees) and slope’s facing (when describing the aspect in terms of N, East, Northwest, etc.).

3. Unexplained figures and inconsistency between figure captions and text. A typical example is Fig. 6(c). The negative numbers shown in the figure are illogical, also, what does “a continuous function of slope aspect” indicate? I understand what the authors tried to say, but the use of the “continuous function” is not acceptable because it has a very rigid definition in mathematics. Don’t abuse a well-defined nomenclature. Since Fig. 6(c) is one of the highlights of the authors' findings. I suggest applying a more adequate method, cluster analysis, or similar algorithms, to gain an insight into the valuable result.  

Response: All the abbreviations used in figures have been spell out in the captions in the revised manuscript. The minus sign in Figure 6(c) has been deleted, and the scale is also explained in the caption of the figure. The expression of “continuous function” has been removed.

Thanks for the suggestions for our further work. Based on the interpreted results of GPR data currently obtained in this study area, we are doing some specific works on the characteristics of permafrost. For example, we are trying to apply machine learning methods to describe the detailed distribution of permafrost and the active layer thickness.  The main arm of this manuscript is to verify the effect of GPR in the extraction of permafrost boundaries in mountainous areas, and to obtain the characteristics of permafrost zonation. After the following works are completed, we will compare them with the results of this study. So this paper is like a seed for our study, and I think the expertise on GPR in your comments is also very valuable for our further work.

4. The resolutions of Figs 2,4, and 5 are visually indiscernible, Figs 2 and 4 in particular, although I enlarged the figure size to 400%. This difficulty needs to be solved somehow.   

Response: We have improved the resolution of all the images in the revised manuscript and will provided the original images with 600dpi to editors.

Please let me know if there is anything else need to be revised. Thanks again for all the comments and suggestions! 

Best Regards.

Reviewer 2 Report

Dear Authors,
many thanks for the work performed on your manuscript, which now I believe it has been significantly improved. My comments have been properly addressed.

I have just a very minor remark related to the low visibility of the legends within pictures (see for example Fig. 2 and 4). Please, do check this features for all the included graphics items.

Best Regards.

Author Response

Dear reviewer，

Thanks again for your comments，which are very helpful for the improvement of our manuscript. We hope to continue to receive your professional guidance in our future work.

Regarding the problem of images ‘resolution you mentioned, we have optimized the visualization of all images in the new manuscript and will provided all the original images with 600dpi to editors.

 

Thanks again for all the comments and suggestions!

Best Regards.

Round 2

Reviewer 1 Report

Comments on remotesensing-1687060_v2

Besides the English writing problem, there are still basic issues in this manuscript needed to be solved. Although the faults listed below may not challenge the results of this research, they reflect the insufficiency of the authors’ technical writing training and will degrade the journal’s prestige as published.  

1. Equations that are copied from other places (papers, reports, books, websites, etc.) are not cited. Explanations of some symbols and variables are missing. The formats of a few symbols adopted in equations are inconsistent with that in the main text.
2. Some abbreviations/acronyms lack explanations.
3. The scale legend doesn’t match the map scale, see Fig. 1 to find the error.

Since MDPI has excellent English editing service, and the faults listed above are obvious and easy to correct, even a high school student can handle them well, I will leave the decision or further review to the editorial office. There is no need to get back to me for approval.  

 

Author Response

Dear reviewer,

Thanks again for all your comments and suggestions. According to your recommendation, we adopted the language editing service of MDPI, and made an improvement to the entire manuscript in terms of language. In English writing, I will continue to strengthen training in future.

• In order to make readers to easily understand the formulas in the text, we have reedited the formula and supplemented the explanation of all the symbols.
• We have checked and revised the abbreviations used in the manuscript again, and removed the abbreviation "No." in the text.
• The scale bars in Figure 1 were automatically inserted in ArcGIS. We rechecked the scales but found no errors. We will ask the technical editor a favor to check the scale again.

Thanks again for your hard work on this manuscript, and I hope that I can receive your professional guidance in the future work.

Best Regards.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments on RS-1612765

The writing and figures of this manuscript are confusing because the authors don't adopt standard usages and terminologies. The way of presenting their research may be acceptable in local communities, but not standard. Besides, the authors ignore to provide the crucial points of their work probably they are not specializing in GPR and lack basic knowledge of statistics. These shortcomings should be seriously considered if they have the chance to revise the manuscript. Below are a few examples of my concerns.    

-- The terminologies used in this manuscript are inadequate, for instance, "unit degree", "unit meter", and the like are incorrect and confusing. Check dictionaries and the literature on their research subject to find the correct terms.

-- The unit expression in figures are completely wrong and misleading. Again, check dictionaries and the literature on their research subject to find the correct expressions of "percentage frequency distribution". In addition, the expressions of "Frequency/%", "Depth/m", "Elevation/m", and so on may be used in some local communities, but they are never acceptable usages worldwide.

-- Fig. 5 requires scales (both horizontal and vertical), it is meaningless otherwise.

-- The results and interpretations are in doubt because the authors fail to provide the details of their field investigations, and lack statistical knowledge. The authors carried out 130 profiles (survey lines?), but present only one profile, what about the rest 129 profiles?

-- The authors declare that both 100 MHz rough terrain antenna (RTA) and unshielded antenna (UA) were used in the survey, but no further details are given. The field parameters and resolutions of those two antennas are quite different, how the authors dealt with the two antennas in the field. If the authors used the UA for velocity analysis, and RTA for subsurface imaging, section 2.2.1 must be re-written. 

-- As with most modern GPR systems, the position of the RTA system can be tracked in real-time using DGPS or RTS (robotic total stations). The authors also have DEM (digital elevation model) for their survey, therefore, it is possible to display Fig. 3 in 3D which will be more realistic. If this suggestion is unfeasible for the authors, the errors and deviations of Fig. 3 should be clearly pointed out because no one believes that the survey line of Fig. 3 is straight in the rugged Tibet Plateau especially the RTA was used.    

Reviewer 2 Report

In this manuscript titled "The zonation of mountain frozen ground under slope aspect adjustment revealed by ground penetrating radar (GPR) survey -A case study of a small catchment in the upper reaches of the Yellow River, northeastern Qinghai Tibet Plateau", the Authors Guangyue Liu, Lin Zhao, Changwei Xie, Defu Zou, Tonghua Wu, Erji Du, Lingxiao Wang, Yu Sheng, Yong-hua Zhao, Yao Xiao, Chong Wang, and Yiwei Wang, focuses in the use of GPR to survey the permafrost boundary in the mountainous area of Qinghai Tibet Plateau.

The paper is well-written, presenting with the appropriate criteria the influence of slope direction on the mountain permafrost zoning model. It improves previous studies and the conclusions are well supported by the results. The manuscript merits a publication in Remote Sensing. The following comments may help the Authors to improve the presentation and readability of their work:

• use GPR or ground-penetrating radar in the title;
• line 23, out a "," before and;
• lines 58-59, explain better;
• line 82, add a set of sentences to describe already used methods;
• line 95, ...study ... study..., rewrite better;
• line 102, delete space before ".";
• line 121, add a space between usuallyhaving;
• line 157, delete a space;
• line 161, ...they were made... ?? rewrite better;
• lines 183-186, cut the long sentence in two parts and rewrite better;
• line 219, SPB already defined;
• line 272, advice that it is discussed below;
• line 285, define Lxx and describe;
• line 296, "- " ??;
• line 322, put a space before [36];
• line 439, radiation;
• line 466, put a space before (a);
• line 474, put spaces;
• you used many times "study", please use other terms like work,...;
• line 500, could you estimate the density of GPR profiles that are necessary in order to obtain a sufficient survey;
• line 507, delete space after "model";
• line 509 and 510, put a "," before and;
• line 511, (LLC) already defined;
• line 514, (LLD) already defined;
• line 520, del space before ".";

Reviewer 3 Report

I read the paper with interest as it is about joining  GPR data interpretation, punctual in situ temperature measurements  in boreholes and pits and regional spatial data. The idea is to spatially extend punctual permafrost detection (in boreholes), to kilometric profiles (using GPR) and to a whole mountaineous region in northeastern Qinghai Tibet Plateau. 

However, many typos are present in this version of the manuscript preventing the reader comprehension. In addition to these needed  corrections , as a radarist, I am disappointed by the only radargram shown in the paper, Figure 3. You should add some information from direct measurements done along this profile, or at least, an example of a wide angle reflection and refraction profile for velocity estimation. You must also indicate the position of this profile in Figure 1. In its current presentation, the reader has to trust your interpretration for the permafrost table determination. You have to advance more convincing arguments.And you should describe your GPR data processing.

In addition, you should add at least one or two  other GPR profiles to indicate the interest in using this geophysical method for extending punctual borehole and/or pits information laterally. In the present form, I feel that the statistic for permafrost distribution could be conducted only with  boreholes and pits data.

Equation 3 must be simplified. For doing this, you should introduce a math symbol for defining the threshold in Equation 2. 

 

 

 

Reviewer 4 Report

Dear Author,

Many thanks for your contribution. Though not novel in its application, the use of GPR technique for the development of a geographical model of permafrost distribution is interesting and worth of exploration.

On an overall basis, the article is properly structured and well developed, hence it is, in my opinion, easy to follow.

My only concern is related to the limited description of the GPR survey, which is critical considering that the model you developed is strongly based on the interpretation of the GPR profiles. In particular:

• The two kinds of antenna you used is not clear: what are the differences between the unshielded and the rough terrain?
• The acquisition parameters and geometry should be better detailed and justified. Sampling, profile location, as well as the WARR setting are just theoretically mentioned but not properly evaluated.
• You stated that 130 GPR profiles have been collected and among them, you selected 63 for the interpretation. It would be better to show why and where these profiles are located (given the wide are surveyed according to Fig. 1).
• The profile you show clearly results not only from the topographic correction, but I believe that several processing steps have been applied – in particular considering the achieved depth of approximately 10 metres and the highly heterogeneous nature of the subsurface.
• Time to depth conversion is not proved – how did you obtain the 7 cm/ns value for velocity? And how did you manage the highly heterogeneity of the subsurface for the computation of the actual depth? Given the importance that the subsurface layering has on the model, it is fundamental to properly address this reconstruction.
• As you also collected some borehole data, it would be useful to provide a direct comparison with the obtained GPR data, otherwise the interpretation remains somehow vague and difficult to ascertain.

As said, I think that without a strong experimental basis, the actual impact of an empirical model might be limited and its performance weak.  

Finally, previous applications and works on GPR for permafrost studies and model development should be deeper commented, highlighting limitations and weaknesses so that the value of the proposed contribution could better emerge. What are the challenges you are going to solve or overcome with respect to the state of the art? So far, I can only see a brief mention of previous works but no comments on them.

For these reasons I think that the paper in its current form requires some additional works. I know this might be disappointing, but I hope the provided comments could help the revision and improvements of the contribution.

Best Regards.