Using Optimized Spectral Indices and Machine Learning Algorithms to Assess Soil Copper Concentration in Mining Areas
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsReview comments for sustainability- 2894957
This manuscript presents a fresh investigation of using four machine learning models to explore the optimal spectral predictors for soil Cu. The contents are well-explained. However, there are several issues that the authors must give attention to while revising the manuscript. Before possible publication of this study, following major issues require to be addressed.
(1) Line 13: The sampling location of this mining sites should be mentioned in the abstract. Please add this information.
(2) Line 50-81:The review of the current study on remotely monitoring soil heavy metal concentration is precise but not very comprehensive. The unique characteristic of relationship between soil Cu concentration and spectral indicess are not fully discussed in the Introduction. A better outline and more updated citations are expected.
(3) Line 82-83: There seems to be a missing information in this sentence. “Machine learning algorithms …… estimating heavy metal soil.”? Maybe “estimating the concentrations of heavy metal”? Please check.
(4) Line 112: The information of study area appears not very clear. The geographical location, latitude and longitude and geomorphologic features of the study area should be supplemented. If the information about the mining location is confidential that cannot be made public, you can ignore this comment. But please explain why.
(5) Line 195: This is only a personal comment, the author can optionally modify. I think the network flow diagrams are very important information in machine learning. Can you provide one or two key diagrams to help us understand?
(6) The contribution of considered indices and their interactions, which is examined and discussed in great detail in this study, are not noted in the Conclusions section. It is therefore need to be indicated.
(7) There are a few typos or non-standard expressions in the manuscript. In addition, a space should be normally provided between the value and the unit. Some spaces in the text are included and some are not. The authors need to read through the entire manuscript to check for corrections.
(8) The resolution of the images and diagrams in the manuscript is too low, such as Figure 1, Figure 6, Figure 11 and Figure 12. Please change as the high-definition versions.
(9) The authors misses some important references about soil propertiy. There are some work about this issue are suggested to be cited, such as:
An, R.; Wang, Y.X.; Zhang, X.; Chen, C.; Liu, X.; Cai, S. Quantitative characterization of drying-induced cracks and permeabilityof granite residual soil using micron-sized x-ray computed tomography. Sci. Total Environ. 2023, 876, 163213. https://doi.org/ 10.1016/j.scitotenv.2023.163213
Comments on the Quality of English Language
Minor revise is necessary.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors- The introduction should clearly outline the problem statement, research objectives, and significance of the study in addressing soil copper concentration in mining areas.
- Provide more details on the hyperspectral reflectance data collection process from abandoned copper mining areas, such as the equipment used, sampling methods, and spatial coverage, as this will enhance the transparency and reproducibility of the study.
- Further explanation of the machine learning models for predicting soil copper concentration would help readers understand the methodology better. Please clarify how these models were trained, validated, and tested, which can increase the credibility of the results.
- Elaborate on the importance of using spectral indices combined with machine learning algorithms for accurately estimating soil heavy metals. Also, discuss how these indices were selected and their role in enhancing prediction accuracy.
- Please provide a detailed interpretation of the results obtained, especially regarding the importance of the scores of the optimised spectral indices and their contribution to the RF model. This will enable readers to grasp the significance of the findings and the implications for soil protection strategies.
- In the introduction part, authors should buttress more on soil heavy metal pollution; kindly use: 10.1007/s11104-023-06256-4; 10.1080/03650340.2022.2146100.
- Ensure that the conclusions drawn from the study are well-founded and supported by the results. Additionally, provide clear recommendations for sustainable strategies for soil remediation, improvement, and environmental protection based on the research outcomes.
- The similarity index is relatively high, authors should kindly revised to reduce it.
moderate language editing required
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsAfter reading the manuscript entitled “Using Hyperspectral Remote Sensing Technology to Assessing soil Copper Concentration and Promote Sustainable Soil Protection”, I found that, the studied topic is interesting and worth studying. Lots of studies were conducted and results were obtained. The manuscript is relatively well written. Therefore, this manuscript can be considered for publication with some revisions, after the following suggestions.
(1) The novelty of this work was not clearly presented. Please follow the literature review and show the knowledge gaps identified and link them to your research objectives.
(2) Line 45. As you known, As heavy metals, copper (Cu) pollution, and excessive Cu in the soil might be due to many reasons, such as gold mining (zhang JH, et al. (2019)Dynamic characteristics of heavy metal accumulation in the farmland soil over Xiaoqinling gold-mining region, Shaanxi, China. Environmental Earth Sciences. 2019,78(1),25. DOI10.1007/s12665-018-8013-2.), why you just mentioned “opencast coal mines” here? And while, I was confused that the references” [1~3] did not seem to be related to “opencast coal mines”. What makes me feel even more strange is that the study area is one abandoned Cu mining area. Why?
(3) Lines 60 ~61. “ this technology can monitor soil Cu concentration more efficiently on a large scale”. “The hyperspectral reflectance of 647 surface soil samples (0–20 cm) was collected from the abandoned Cu mining area” , did the Hyperspectral Remote Sensing Technology also have this advantage?
(4) Some of the Figures are not clear at all and must be improved to evaluate the results. specifically Figure 1, Figure 5, Figure 6, Figure 11, and Figure 12. In addition, Figures should be sorted in the order in which they appear.
(5) The comparison of this product with other materials is highly recommend to be carried out.
(6) The English of this paper should be improved for better understanding of the paper.
(7) There are also a few issues in the presentation. For example, the formats of references are not uniform, and more articles from the journal should be cited; The number of words in the abstract far exceeds the standard.
Comments on the Quality of English Language(6) The English of this paper should be improved for better understanding of the paper.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments and Suggestions for AuthorsDear authors,
The research presents a significant component of the optimization of machine learning for Cu contamination in mining areas.
The conclusions and findings of the Opt-TBIs method's application are presented in the study in a logical manner, implying that it is a useful tool in the situation of Cu pollution.
If additional pollutants show up in the soil at the same time as copper, it presents a problem. Is it possible to determine these too? alternatively does it impact the Cu model in any way? Since there are multiple chemical species—rather than just one—in the soil, I would like to comprehend these concepts.
Good luck!
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 5 Report
Comments and Suggestions for AuthorsIt is unclear exactly which soils (soil types) were contaminated and with what copper compounds. What mineral phases are carriers of the pollutant element?
It is necessary to describe the sample preparation procedure in more detail, especially in terms of decomposition for elemental analysis.
What ratios of acids were used?
How were samples prepared for analysis? To what particle size was the final grinding of the samples carried out (if any) or were the sifted aggregates finer than 2 mm decomposed?
What sample of soil and with what approximate particle size was decomposed (this is important from the point of view of assessing the representativeness of the sample).
How many parallel samples were decomposed for each point?
It is also necessary to describe the conditions of AAS and ICP-MS changes.
Since the samples were analyzed by two methods with different matrix interferences, and therefore the accuracy of the obtained values, we recommend indicating whether there was a difference between the results of measuring the copper concentration in the same samples, but obtained by different methods (usually there is a difference).
Table 2 - The value for copper should be given a confidence interval and rounded to the appropriate number of significant figures. Typically at this level the accuracy of measurements is such that it is necessary to round to whole numbers (except 4.8)
277-279, 345 - round values to whole numbers. And in other similar places in the text.
Why is the coefficient of determination used for comparison rather than the more familiar correlation coefficient r?
I recommend that all graphs also show the value of the correlation coefficient r.
By correlations. Graphs 9-12 and 16 show that concentrations above 1000 were found only in a few samples, and for the overwhelming majority they were significantly lower. How will the obtained correlations and conclusions change if we discard points above 1000?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe revision of the manuscript is detailed and comprehensive. However, there are still some minor problems. Most of the figures in the manuscript with pdf version are unusually low in definition. Some of the diagrams，such as Figure 14, have weird size proportions, too. The authors must modify these issures.
Comments on the Quality of English LanguageThe English writing needs further polishing.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsMajority of the images/pictures are blurred, kindly improve the resolutions.
Comments on the Quality of English LanguageOk
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for Authorsthe issues the reviewers mentioned should be addressed completely, unfortunately, they did not do so. There are also a few issues in the presentation. For example, the formats of references are not uniform, and more articles from the journal should be cited; The number of words in the abstract far exceeds the standard
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 5 Report
Comments and Suggestions for AuthorsDetailed answers to the questions asked allowed the reviewer to better understand the essence of the research being carried out and the grounds that served as the basis for the conclusions made.
At the same time, since the question about the predominant forms of copper compounds that form the basis of pollution in the region under consideration actually remains unanswered, the conclusions made by the authors cannot be considered correct and universal, as stated in the conclusions to the article. An answer was given to my question about the mineral forms of the compounds, which shows that this question has not been studied in this region, or the authors do not have this information; they included all possible forms in incomprehensible combinations. It is clear that different copper-containing mineral components can give different spectral patterns. In order to claim the universality of the conclusions drawn, it is necessary to conduct methodological laboratory model experiments with the introduction of various copper-containing minerals and soluble salts, both mono-compounds and mixtures, in different concentrations into clean soil, and experiments with technogenically polluted soils how to validate the resulting models
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 3
Reviewer 3 Report
Comments and Suggestions for Authorsall the issues I cared about were addressed by the authors in the present manuscript, so , in my view, the manuscript can be accepted at the present state.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 5 Report
Comments and Suggestions for AuthorsI believe that the latest additions and changes made by the authors make it possible to publish the work in its present form
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
