Machine Learning Techniques in Predicting Bottom Hole Temperature and Remote Sensing for Assessment of Geothermal Potential in the Kingdom of Saudi Arabia

Round 1

Reviewer 1 Report

Consider better visualizations for Figures 7 as it is diffuclt to see what is being presented.

What is well logs temperature in Figure 11?

How was the surface temperature mapping results in Figure 12 obtained?

Describe the architecture of the ML algorithms.

Linear Regression is not an ML algorithm

Figrue 17. Why are there outliers at 0-5 distributions?

Consider moving some of the plots to a supplementary material. The paper is too long. 

Degrees Celsius should be written well throughput the manuscript.

How are the ML models validated?

What is the source of the data?

Paper title is too long. Shorten greatly and remove unecessary information.

Define all abbreviations in the abstract and conclusions

Provide novelty/originality statement at second to last paragraph of the introduction section.

None

Author Response

Dear Reviewer 1

Thank you very much for you valuable comments. we did follow the comments and update te manuscript.

 

regards

Essam Aboud

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Brief summary:

The study built models that connect input vectors/scalars (including temperature and magnetic field) spatial, such as geothermal energy sites around the volcanic region of Harrat Rahat. The study presented a conforming label or objective vector using training data and supervised machine-learning techniques. Solving technical problems like the absence of borehole temperature records or inaccuracies in the locations of these wells was also accomplished using machine learning.

The collected data is from gamma-ray and temperature logs for calculating heat production and geothermal gradient. This data was trained and linked with current geothermal zones, then spatial predictions.

The benefit would be accurately predicting and finding other sites from the trained data. That will help with predictions of sites for exploration and drilling operations.

The machine-learning techniques utilised in the study were outlined. An XG Boost is stated in the findings to produce the best results from two other algorithms, i.e., Linear Regression and Random Forest.

Broad comments:

Article:

Introduction

The authors explained the need for geothermal energy because it would address the modern thermal needs of societies. It informed the reader of the many large volcanic fields within the targeted area. It did not educate about machine learning. The authors need to slightly educate the reader about machine learning, especially on the approaches that they will take in their study.

Methodology

Python tools, Pandas, and Seaborn modules were used. Methods of analysing data, like exploratory data analysis (EDA), were used. Then machine learning tools to train the data were explained.

Results

The results were well presented. There are some specific comments I have made in the reviewed manuscript. They were about the presentations of the tables and the readability of some comments/lettering in the figures used.

Conclusion

The conclusion section of the article was well presented. I think it should include statements relating to machine learning. But I conclude as a reviewer that: Thermal infrared imaging for geothermal prospecting and monitoring thermal activity in the HR. LST and NDVI were captured. Therefore, regions with temperatures greater than 43 ^oC were the best geotherm spots. The predicated data found the HR region the best candidate with its high volcanic activity.

Overall, it is a good study, showing the capabilities of AI tools for the sustainability agenda.

 

Please refer to some comments in the manuscripts.

Specific comments:

The study has many acronyms, and the author does not describe all of them yet uses them in the discussion. I suggest that care should be exercised around this matter. It can be a list of acronyms provided or well-articulated along the manuscript.

The figure should be appropriately sized. They look oversized, too blown up.

The figures from other sources are stated or cited need to have permissions.

The manuscript is well structured well. There are fewer grammar issues.

The table’s design is inconsistent, and it needs to be changed to the prescripts of the journal.

 

Some questions:

 None

Comments for author File: Comments.pdf

Author Response

Dear Reviewer 2

Thank you very much for you valuable comments. we did follow the comments and update te manuscript.

 

regards

 

Essam Aboud

 

Author Response File: Author Response.docx

Reviewer 3 Report

Dear Authors,

Thank you for the opportunity to review this very interesting article. I have a few suggestions that in my opinion will affect its value for readers.

1. Lines 123-131 - I propose to reformulate this paragraph and, starting from the purpose of the article, characterize the further structure of the article (without discussing the results).

2. Figures and tables presented in paper are worth discussing in more detail - e.g. figures 24-27.

2. In Discussion, it is worth discussing the theoretical and practical implications of the obtained results in more detail.

3. The article lacked a discussion of limitations.

4. It is worth adding a section on future research.

 

5. The text is incorrectly formatted in several places - e.g. lines 313-360.

Author Response

Dear Reviewer 3

Thank you very much for you valuable comments. we did follow the comments and update te manuscript.

 

regards

 

Essam Aboud

 

 

Author Response File: Author Response.docx

Reviewer 4 Report

The submitted manuscript (sustainability-2514824) is lengthy and tedious due to the lack of an engaging comprehensive discussion that draws a clear new understanding. How this formulation can be built in machine learning is also not clear. Most figures need revising for better visualization. Although providing some comparisons, the accuracy requirement is not appropriately solved. Moreover, the conclusion "CO2 emissions in the HR volcanic field, highlighting the hotspots associated with active volcanic activities and their connection to geothermal sources....." needs more evidence. This manuscript needs major revision.

Author Response

Dear Reviewer 4

Thank you very much for you valuable comments. we did follow the comments and update te manuscript.

 

regards

 

Essam Aboud

 

 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Paper can be accepted after all abbreviations (RS and KSA) are removed from title.

Good

Author Response

Dear reviewer

kindly, have a look to the attached file.

regards

Essam

 

Author Response File: Author Response.docx

Reviewer 4 Report

-

Author Response

Dear reviewer,

kindly, have a look to the attached file.

regards

 

Essam Aboud

Author Response File: Author Response.docx