Statistical Modelling for the Source Rock Parameters of the Montney Formation, NE British Columbia, Canada

Round 1

Reviewer 1 Report

Understanding the factors affecting hydrocarbon potential is critical to reducing the risk of unconventional resource exploration and development. The authors' research using statistical model analysis has some significance, but there are still some problems in the text, so a minor revision is recommended for acceptance.

1. Line 51. There is a problem with reference citation.
2. Line 85. You may mean Fig. 1 here.
3. Line 92. You may mean Fig. 2 here.
4. Authors should double-check the entire text to ensure that the specialized vocabulary is presented consistently in context, for example, lines 171-184. Formatting issues should also be checked, for example, line 192.
5. There are many very, very long sentences in the text that affect the reading experience, for example lines 219-223.
6. It seems that the authors simply introduced a clustering method into this analytical study.  Compared to the traditional method, from which reflects the characteristics of both easy and fast.

Author Response

Reviewer 1

I want to thank you very much for your valuable comments, which are all taken into consideration. In response to your comments, please review the amendments attached to the manuscript.

• The entire manuscript has undergone rigorous linguistic revision.
• The methodology section has been reconstructed.
• Results and discussion has been rearranged in a proper way.
• References problems have been fixed (comment 2,3).
• The specialized vocabulary has been checked and presented consistently in the context.
• The long sentences have been rephrased and shortened.
• A brief section about the TwoStep method is added, and a comparison between it and the HCA was conducted.

Reviewer 2 Report

The present study employed statistical modeling of source rock parameters for a case study from Canada. Overall, almost all the sections of the paper are poorly presented. In addition, there are some serious organizational and technical issues. Hence, in my opinion, the current work does not qualify for a journal article in its present format. In addition, I have the following main concerns.

 

Beginning with abstract, which is very confusing and poorly written throughout. Further, several acronyms in the abstract have been used without defining/introducing them before.

 

In the introduction, there is no coherency and smooth flow of information. Instead, various concepts are randomly presented. A review of existing relevant literature is very inadequate as well.

 

Study purpose and motivations are not explicitly presented. The authors should have highlighted the scientific novelty and key contributions to state of the art. Also, it would have been appealing if the study's usefulness to practitioners and concerned authorities was provided.

 

It is also recommended to highlight the rationale/motivation behind selecting the Montney Formation as a case study.

 

The materials and methods section should be split into sub-sections as “sample collection,” experimental program,” to facilitate readers' understanding. I can hardly see any description of methods in this section. In addition, a detailed description of various statistical analysis methods/tools. PCA,  and cluster analysis, etc., should be provided.

 

The results and discussion section is again not written very well. In fact, there is no discussion that is a must for a scientific research article.  Study results would have been more useful if intuitive and critical discussion in light of similar previous studies was included. I am wondering what aims authors attempted to achieve in applying PCA for the current analysis. Similarly, the explanation of cluster analysis results is also very confusing and weakly structured.

 

Conclusions should be made to the point and should support the study results. Study limitations, future recommendations, and relevant policy implications are clearly missing. Lastly, the manuscript also has several typos and language issues that should be fixed

Author Response

Reviewer 2

I want to thank you very much for your valuable comments, which are all taken into consideration. In response to your comments, please review the amendments attached to the manuscript.

(x) Moderate English changes required

• The entire manuscript has undergone rigorous linguistic revision.

Beginning with abstract, which is very confusing and poorly written throughout. Further, several acronyms in the abstract have been used without defining/introducing them before.

• The abstract has been reconstructed with defining all the acronyms (lines 1-19).

In the introduction, there is no coherency and smooth flow of information. Instead, various concepts are randomly presented. A review of existing relevant literature is very inadequate as well.

• The introduction was rephrased to be more coherent, and regarding the literature reviews, we tried to include all studies that have a solid relationship to the idea under study (lines 21-56).

It is also recommended to highlight the rationale/motivation behind selecting the Montney Formation as a case study.

• A paragraph introducing Montney Formation and its importance has been added (lines 57 - 67).

 

The materials and methods section should be split into sub-sections as “sample collection,” experimental program,” to facilitate readers' understanding. I can hardly see any description of methods in this section. In addition, a detailed description of various statistical analysis methods/tools. PCA, and cluster analysis, etc., should be provided.

• The methodology section has been reconstructed as you suggested (lines 94 – 168).

The results and discussion section is again not written very well. In fact, there is no discussion that is a must for a scientific research article.  Study results would have been more useful if intuitive and critical discussion in light of similar previous studies was included. I am wondering what aims authors attempted to achieve in applying PCA for the current analysis. Similarly, the explanation of cluster analysis results is also very confusing and weakly structured.

• Results and discussion has been rearranged in a proper way (lines 169-275). Also, the aim behind using the PCA has been clarified (lines 129-134).

Conclusions should be made to the point and should support the study results. Study limitations, future recommendations, and relevant policy implications are clearly missing. Lastly, the manuscript also has several typos and language issues that should be fixed

• The conclusions have been connected to the results. Recommendations and implications with future work were added (lines 310-315). All the language issues were revised and fixed.

Reviewer 3 Report

Authors show a new application to the two-step hyerarchical clustering, widening the scope in terms of fields of application. Two-step clustering is a powerful tool to deal with large amounts of variables and objects and this paper will definetely help readers to mine their data and improve their researches.

In order to colaborate to improve the scientific soundness of this manuscript I would like to state some comments to authors:

- Some references should be reviewed, for example line 52, the format has been missed.

- Some abreviatures, being common in the field of application of the analytical chemistry techniques (for example S1,S2,S3 in pryrolisis), could be very briefly introduced because some readers may be not acquainted with them when they were looking for reading the statistical technique for their own applications.

- Although the two-step cluster is a well stablished statistical method, in my opinion, authors ought include a paragraph introducing it, provided the novelty of the tool to potential readers.

- Additionally, authors could include some details about their application as the distance method used (Euclidean or max likehood), the linkage criteria and how to select of number of clusters.

- Finally, and given the fact that the two-step clustering is designed for large sets of data, I would aim authors to include a comparison between the results obtained when standard clustering methods are applied vs. two-step. I have in mind that because in this manuscript authors are applying the technique to a relatively small dataset.

Once there questions are resolved, authors will be able to give more power to their two last conclusions.

Author Response

Reviewer 3

I want to thank you very much for your valuable comments, which are all taken into consideration. In response to your comments, please review the amendments attached to the manuscript.

Does the introduction provide sufficient background and include all relevant references?

• The introduction was rephrased to be more coherent, and regarding the literature reviews, we tried to include all studies that have a solid relationship to the idea under study (lines 21-56).

Are the methods adequately described?

• The methodology section has been reconstructed (lines 94 – 168).

Are the results clearly presented?

• Results and discussion has been rearranged in a proper way (lines 169-275).

- Some references should be reviewed; for example line 52, the format has been missed.

• The references issues have been revised and fixed.

- Some abreviatures, being common in the field of application of the analytical chemistry techniques (for example S1,S2,S3 in pryrolisis), could be very briefly introduced because some readers may be not acquainted with them when they were looking for reading the statistical technique for their own applications.

• All the acronyms were defined in the abstract and in the methodology.

Although the two-step cluster is a well established statistical method, in my opinion, authors ought to include a paragraph introducing it, providing the novelty of the tool to potential readers.

• A section regarding the TwoStep method has been added to the methodology (lines 135-168). Also, a paragraph about the novelty has been added to the introduction (lines 54-56).

Additionally, authors could include some details about their application as the distance method used (Euclidean or max likehood), the linkage criteria and how to select of number of clusters.

• We have added another clustering method (the hierarchical cluster algorithm) and compared the results of the two ways (lines 135-168 and lines 242-275).

Finally, and given the fact that the two-step clustering is designed for large sets of data, I would aim authors to include a comparison between the results obtained when standard clustering methods are applied vs two-step. I have in mind that because in this manuscript, authors are using the technique to a relatively small dataset.

• Regarding the final comment, it's right that the TwoStep algorithm is used for a large data set, and our data is considered to be non-sufficient, but the application of this algorithm is more powerful than using the hierarchical cluster analysis and more consistent with the geochemical results.

Round 2

Reviewer 2 Report

Thank you very much for providing the revised version of the manuscript. The suggested modifications have made the manuscript suitable for publication.