Assessment of Soil Contamination Using GIS and Multi-Variate Analysis: A Case Study in El-Minia Governorate, Egypt

Round 1

Reviewer 1 Report

Dear Authors

For a better understanding in-depth, it is a need for time to work on this topic. Furthermore, the achievement of potential benefits by using current technology is also dependent on the extensive research work for more exploration. Although the experiment is well organized, I suggest a major revision due to the following deficiencies.

Abstract

1. Introduce the need for study in 1-2 lines.

2. Please give a clear-cut problem statement that is tackled in the current study.

3. Provide design of experiment in a single line.

4. Results are fine.

5. The conclusion is ok.

6. Add the targeted beneficiary audience who will get benefits from this research.

7. Also, give clear-cut recommendations while describing the best treatment. I suggest writing the application rate as well with the best treatment abbreviation for more and better understanding.

Author Response

For a better understanding in-depth, it is a need for time to work on this topic. Furthermore, the achievement of potential benefits by using current technology is also dependent on the extensive research work for more exploration. Although the experiment is well organized, I suggest a major revision due to the following deficiencies.

Abstract

1. Introduce the need for study in 1-2 lines.

We have clarified  the need for the study , Line 15-16 showed the need for the study

2. Please give a clear-cut problem statement that is tackled in the current study.

Line 13-15 illustrated the main problem that the study focus in abstract and also line 99-104 in introduction

3. Provide design of experiment in a single line.

We have added in line 16-17 in abstract  and also in section 2.3

4. Results are fine.
5. The conclusion is ok.
6. Add the targeted beneficiary audience who will get benefits from this research.

We clarified in Line 405-408 showed the beneficiaries of the current results

7. Also, give clear-cut recommendations while describing the best treatment. I suggest writing the application rate as well with the best treatment abbreviation for more and better understanding.

Line 408 to 4011 show the recommended strategies for mitigation of the contamination of soils

Reviewer 2 Report

This is an interesting manuscript providing assessment of Soil Contamination Using GIS and Multi-Variate Analysis data in from an area of the world with particular interest. I have however some points of concern regarding the methodology followed:

1. You should be able to summarize the problems in existing literature before mentioning your study aims. Abstract and conclusion parts should be enhanced. Mention point and non-point sources.

2. How probable sources can be identified? Give in detail. Authors should include rainfall precipitation during the sampling year because rainfall also play an important role for the leaching of metals. Why the authors not discussed seasonal variation? What about the precision of the analytical process? Below diction limit details?

3. Geochemical normalization for CF and CD estimation should better use metal from 'relatively uncontaminated site' from the study area. The use of mean continental shelf contamination data is not appropriate for this parameter.

4. I do not agree with the soil contamination analysis. My view is that grain size analysis (texture) is required prior to chemical analysis, aiming to remove the selective tendency of heavy metals to adsorb at finer soil.

5. In terms of sampling strategy, I would like to understand the reasoning behind stations location. Are these stations related to point and non-point pollution sources? A table with the location of each sampling station and possible impact on the water quality contamination will be helpful for the readers not familiar with the study area.
   
   
6. In the result and discussion, not to be completed. Link the results with with recent literatures. Authors mostly mentioned about results cambered with the standards. What kind of natural and anthropogenic sources? Show their areas and metal concentrations in the areas and link the concentration profiles/indexes and the specific source areas as discussed in the manuscript.

7. Conclusion add some management tactics based on this assessment.

Author Response

This is an interesting manuscript providing assessment of Soil Contamination Using GIS and Multi-Variate Analysis data in from an area of the world with particular interest. I have however some points of concern regarding the methodology followed:

1. You should be able to summarize the problems in existing literature before mentioning your study aims. Abstract and conclusion parts should be enhanced. Mention point and non-point sources.

We have summarized the problems  in lines 99-104

“The limitation of fresh irrigation water is one of the most serious problems of agricultural production in Egypt, which makes farmers irrigate with water from unsafe sources, or in a more precise sense, polluted, such as agricultural drains and renewable groundwater. Also, the excessive use of chemical fertilizers to increase the yield is one of the sources of pollution in the Egyptian lands.”

The pollution sources in the current study are non-point sources

2. How probable sources can be identified? Give in detail. Authors should include rainfall precipitation during the sampling year because rainfall also play an important role for the leaching of metals. Why the authors not discussed seasonal variation? What about the precision of the analytical process? Below diction limit details?

- Authors should include rainfall precipitation during the sampling year because rainfall also play an important role for the leaching of metals.

The pollution sources of the study area  can be identified based on the field observation.

 We have clarified in section 2.1 Lines 121-125 the effect of rainfull

“The study area is described as arid climate with hot summer, warm winter, high evaporation, and low rainfall intensity. The meteorological station of El Minia Governorate (EMA, 2020) records for 35 years (1985– 2020) as follows; the main monthly temperature ranges between 12.9℃ in January and 30.2℃ in Aug. Rainfall is rare throughout this region; the total mean precipitation is 28.0 ml/ year.”

- Why the authors not discussed seasonal variation?

Due to the rare rainwater that receiving seasonally, the concentration of heavy metals does not affect by rain water

- What about the precision of the analytical process? Below diction limit details?

Actually we used ICP-MS which has high accuracy analysis.in addition we have not gotten values under the detection limit

iCAP™ RQ ICP-MS (thermofisher.com)

3. Geochemical normalization for CF and CD estimation should better use metal from 'relatively uncontaminated site' from the study area. The use of mean continental shelf contamination data is not appropriate for this parameter.

Previous studies showed that background values used in calculating contamination factors were varied. Also, due to the excessive use of chemical fertilizers in Egypt since the 60s on the one hand and the lack of a reference background for heavy elements under the conditions of the Egyptian lands, on the other hand, it was better to use mean continental shelf contamination data.

4. I do not agree with the soil contamination analysis. My view is that grain size analysis (texture) is required prior to chemical analysis, aiming to remove the selective tendency of heavy metals to adsorb at finer soil.

Particle size distribution has been done prior to chemical analysis. The obtained data showed that the study area has only clay texture class with no variation.

Lines 127-128, and section 2.3

5. In terms of sampling strategy, I would like to understand the reasoning behind stations location. Are` these stations related to point and non-point pollution sources? A table with the location of each sampling station and possible impact on the water quality contamination will be helpful for the readers not familiar with the study area.

The study area was selected based on field observations, through which it was shown that farmers use, in varying degrees, the El-Moheet drain water in the irrigation process, which is an agricultural drainage channel and the waste of the sugar factory in the Abu Qirqas Center of Minya Governorate, in addition to dumping garbage. Map 1 showed the location of soil samples

We collected soil samples based on mixed methods (one sample per 400 m from both sides of the drain) on the other hand we collected some random samples far from the drain

6. In the result and discussion, are not to be completed. Link the results with with recent literature. Authors mostly mentioned about results cambered with the standards. What kind of natural and anthropogenic sources? Show their areas and metal concentrations in the areas and link the concentration profiles/indexes and the specific source areas as discussed in the manuscript.

We linked the results with recent literature regarding the subject of the study

What kind of natural and anthropogenic sources?

The previous answered  refer to anthropogenic sources and Line 99-104

I appreciate the helpful comments

Reviewer 3 Report

This manuscript aims to classify areas of heavy metal soil pollution in a zone which is illegally irrigated with polluted water and where high variability exists in soil fertilization by farmers. Although it clearly elaborates on a very relevant subject – i.e. soil pollution as risk for entire ecosystems – it is not clear to me what is the added value of this study. Further than adding knowledge to local pollution patterns, which is only relevant to local authorities, the manuscript does not place results in a broader context. The discussion section is very poor and lacks in-depth analysis and interpretation of results. I believe the manuscript could mainly bring knowledge on the relationship between soil pollution and  fertilization/irrigation measures by taking into account field-scale results of the questionnaire, but these data have not fully been incorporated in the study. It is a pity that results of AHC, PCA, DC, … are nowhere linked and discussed together. The use of kriging for establishing pollution patterns is artificial and shows bullet-eyed patterns which could be prevented by an improved sampling scheme and fine-tuning of and knowledge on kriging parameters. I don’t think the sampling scheme has been chosen randomly by software, but rather based on personal experience or field accessibility as they appear to lay on a straight line. I also do not agree that pollution is higher in the north than in the south of the study area. I believe the bullet-eyed pattern, reflecting strong spatial variability, is due to inappropriate sampling and strong spatial variability caused by variable fertilization and illegal irrigation schemes, not due to the urban area (in that case the top of the study area – closest to the urban area – should show strongest pollution which is not so). Regarding clustering, the authors do not show the locations of both clusters on a map, nor do they explain why they are different and how this is important for the general conclusion.

Line 133-136: How much soil sample was taken for 1 digestion (in grams)?

Line 58 – 60: If these elements don’t have a high environmental pollution impact, why are they incorporated in this study? This seems to be contradictory.

Line 60 – 61: I would rephrase this sentence to highlight the goal of the paper, e.g. ‘Due to this variability, spatial classification of soil contamination is inevitable for establishing appropriate remediation measures by decision makers.’

Line 68 – 69: Why is kriging chosen over other geostatistical methods?

Line 79-80: The authors elaborate a lot on PCA but not on AHC. I would expand the introduction on AHC.

Line 89: Replace ‘maintain’ by ‘remediate’. Maintain means soil pollution will keep on going.

Section 2.1: This section lacks a description of the soils under study (profile description, texture, parent material).

Line 100-101: Replace this phrase to section 2.2.

Section 2.2: It would be interesting to add the results of the questionnaire in Supplementary Material.

Figure 2: I do not see added value of this figure as all is written out in the text, too. The ‘Determine Land Use’ step is nowhere discussed. Does it influence the results?

Section 2.4: Instead of using background values from literature, the results might be more realistic and thus practically applicable by taking either the lowest value in the data set as background value, either the values of a sample collected in an unpolluted environment somewhere near the current study area.

Line 163-164: I do not see why the formula is shown here, it is also not properly introduced in the text. Regarding kriging, PCA and AHC, I think it is better to only briefly summarize in the Materials and Methods section how these techniques are used in this study, and then just refer to a proper reference for further information on the techniques themselves.

Table 2: It would be good to also show the experimental and modelled variograms in a figure, for instance in supplementary material.

Line 245 – 247: What are possible sources of urban pollution and how do they reach the study area? If by wind, then wind direction should be incorporated in the discussion.

Line 248 – 251: This hypothesis does imply that fertilization is only composed out of Co as otherwise also the other elements should show higher concentrations in the south. It does not seem very likely.

Line 253 – 257: What is the significance level? Can the p-values be shown in the graph, too? Why are only some correlations discussed and what do they learn us? A correlation above 0 is always positive, it is only a poor correlation (perhaps not significant).

Line 297 – 298: You can also group the observations with PCA by adding them to the biplot. Such a graph can be very informative.

Author Response

This manuscript aims to classify areas of heavy metal soil pollution in a zone which is illegally irrigated with polluted water and where high variability exists in soil fertilization by farmers. Although it clearly elaborates on a very relevant subject – i.e. soil pollution as risk for entire ecosystems – it is not clear to me what is the added value of this study. Further than adding knowledge to local pollution patterns, which is only relevant to local authorities, the manuscript does not place results in a broader context. The discussion section is very poor and lacks in-depth analysis and interpretation of results. I believe the manuscript could mainly bring knowledge on the relationship between soil pollution and fertilization/irrigation measures by taking into account field-scale results of the questionnaire, but these data have not fully been incorporated in the study. It is a pity that results of AHC, PCA, DC, … are nowhere linked and discussed together. The use of kriging for establishing pollution patterns is artificial and shows bullet-eyed patterns which could be prevented by an improved sampling scheme and fine-tuning of and knowledge on kriging parameters. I don’t think the sampling scheme has been chosen randomly by software, but rather based on personal experience or field accessibility as they appear to lay on a straight line. I also do not agree that pollution is higher in the north than in the south of the study area. I believe the bullet-eyed pattern, reflecting strong spatial variability, is due to inappropriate sampling and strong spatial variability caused by variable fertilization and illegal irrigation schemes, not due to the urban area (in that case the top of the study area – closest to the urban area – should show strongest pollution which is not so). Regarding clustering, the authors do not show the locations of both clusters on a map, nor do they explain why they are different and how this is important for the general conclusion.

Actually the manuscript highlighted the following :

• This manuscript presents to the reader a model of land pollution that may differ from the rest of the lands around the world, as the agricultural holding area is very small (0.1-0.25 hectares ) , so the methods of agricultural management, irrigation methods, and also fertilizer additions are differing.
• Illustrated the current spatial distribution of each heavy metals separately of the study area
• Using PCA and AHC to delineate the pattern of heavy metals
• Apply the DC on the results of PCA to show the degree of contamination to be started the following management programs .

Line 133-136: How much soil sample was taken for 1 digestion (in grams)?

We collected 500 grams for various analysis but we just use 0.25 gram for ICPMS

Line 58 – 60: If these elements don’t have a high environmental pollution impact, why are they incorporated in this study? This seems to be contradictory.

Actually we illustrated the previous results, this sentence regard special condition of water of Nile river 

Line 60 – 61: I would rephrase this sentence to highlight the goal of the paper, e.g. ‘Due to this variability, spatial classification of soil contamination is inevitable for establishing appropriate remediation measures by decision makers.’

We have considered and rephrased the sentence

Line 68 – 69: Why is kriging chosen over other geostatistical methods?

We tested several models such as IDW and Kriging , even among kriging models  we tested , and finally selected the most suitable model according to the parameters mentioned in sections 2.5 ,  3.2 and table 2

Line 79-80: The authors elaborate a lot on PCA but not on AHC. I would expand the introduction on AHC.

We have explained AHC in introduction section   line  84-89 and methods section 2.6  

Line 89: Replace ‘maintain’ by ‘remediate’. Maintain means soil pollution will keep on going.

We replaced

Section 2.1: This section lacks a description of the soils under study (profile description, texture, parent material).

We improve the description of soil and study area

Line 100-101: Replace this phrase to section 2.2.

We have moved to section 2.2

Figure 2: I do not see added value of this figure as all is written out in the text, too. The ‘Determine Land Use’ step is nowhere discussed. Does it influence the results?

Figure 2 just shows the steps of the work

Section 2.4: Instead of using background values from literature, the results might be more realistic and thus practically applicable by taking either the lowest value in the data set as background value, either the values of a sample collected in an unpolluted environment somewhere near the current study area.

Previous studies showed that background values used in calculating contamination factors were varied. Also, due to the excessive use of chemical fertilizers in Egypt since the 60s on the one hand and the lack of a reference background for heavy elements under the conditions of the Egyptian lands, on the other hand, it was better to use mean continental shelf contamination data.

Regarding sample collected in an unpolluted environment somewhere near the current study area.

Its good idea we will consider in the future work and will collect sediment samples.

Line 163-164: I do not see why the formula is shown here, it is also not properly introduced in the text. Regarding kriging, PCA and AHC, I think it is better to only briefly summarize in the Materials and Methods section how these techniques are used in this study, and then just refer to a proper reference for further information on the techniques themselves.

We have considered and removed the formula from the text

Table 2: It would be good to also show the experimental and modelled variograms in a figure, for instance in supplementary material.

The data in table 2 is more clear than the figures variogram  as the resolutions is note high

Line 245 – 247: What are possible sources of urban pollution and how do they reach the study area? If by wind, then wind direction should be incorporated in the discussion.

The source of pollution of the study area is wast water that used for irrigation, fertilizers, and pesticides

Line 248 – 251: This hypothesis does imply that fertilization is only composed of Co as otherwise also the other elements should show higher concentrations in the south. It does not seem very likely.

I agree with the reviewer that fertilizers are consist not only Co but also many other elements, however, the reasons for increasing the Co maybe refer to the excessive of fertilizers or geochemical composition, but we excluded the geochemical as a reason as the soil formation of the study area  as the same source (alluvial sources)

Line 253 – 257: What is the significance level? Can the p-values be shown in the graph, too? Why are only some correlations discussed and what do they learn us? A correlation above 0 is always positive, it is only a poor correlation (perhaps not significant).

We have considered and added the significance level , figure 5

Line 297 – 298: You can also group the observations with PCA by adding them to the biplot. Such a graph can be very informative.

Three Factors were resulted based on the PCA, those factors represented 42.22%, 16.1% and 12.9 % of the total data, respectively. Biplot represented two factors F1 and F2 in two dimensions which express the strength of the relationship between the variables in both factors.

We added the observation to the biplot

I appreciate the helpful comments

Round 2

Reviewer 1 Report

Dear Authors

I am satisfied with the changes. The paper can be accepted after fine English modifications.

Regards

Author Response

thank you 

we reviewed the language 

Reviewer 3 Report

1. The discussion section is still very poor and still lacks in-depth analysis and interpretation of results (not only showing and describing results). Results of AHC, PCA, DC, ... should be linked and discussed together. 2. It is still not clear where both clusters are located, why different results are obtained for both clusters and how this is important for the general conclusion. 3. "We collected 500 grams for various analysis but we just use 0.25 gram for ICPMS" --> mention the 0.25 gram in the M&M section where you explain ICP-MS

Author Response

Here is a RESPONSE TO THE EDITORS AND REVIEWERS

of “Assessment of Soil Contamination Using GIS and Multi-Variate Analysis: A Case Study in El-Minia Governorate, Egypt”

by Amr A. Hammam, Wagih S. Mohamed, Safa Essam-Eldeen Sayed, Dmitry E. Kucher, and Elsayed Said Mohamed

The authors would like to thank the editors and reviewers for their thoughtful review of the manuscript. They raise important issues, and their inputs are very helpful for improving the manuscript. We agree with almost all their comments, and we have revised our manuscript accordingly.

Response to comments from reviewer3

1. The discussion section is still very poor and still lacks in-depth analysis and interpretation of results (not only showing and describing results). Results of AHC, PCA, DC, ... should be linked and discussed together.

- We came through the results and discussion section and made some improvements as follows:

1. In section “3.4. Cluster analysis based on PCA”, we added some sentences and compare the results
2. Section “3.5. we added a new section regarding the Land Use of the study area.

Lines 314-318 cover the link between PCA and AHC

4. Section “3.6” covers the link between AHC and DC
5. It is still not clear where both clusters are located, why different results are obtained for both clusters, and how this is important for the general conclusion.

Figure 9 we added to the manuscript and show the Land Use of the study and the location of cluster 1 (C1) and cluster 2 (C2)

3. "We collected 500 grams for various analyses, but we just use 0.25 gram for ICPMS" --> mention the 0.25 gram in the M&M section where you explain ICP-MS

Line 157: we mentioned the used wight of soil sample in heavy metals determination with Inductively Coupled Plasma Mass Spectrometry.

We appreciate the reviewer's efforts for the useful comments

Author Response File: Author Response.docx

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

The current study entitled “Assessment effects of anthropogenic activities on soil contamination using multivariate analysis: A case study in El-Minia governorate, Egypt.” is good. For a better understanding in-depth, it is a need for time to work on this topic. Furthermore, the achievement of potential benefits by using current technology is also dependent on the extensive research work for more exploration. Although the experiment is well organized, yet I suggest a major revision due to the following deficiencies.

Major Concerns

Title

• Please modify the title it is descriptive.

Abstract

• Systematic abstract is missing. Introduce the need for study in 1-2 lines.
• Please give a clear-cut point problem source as a problem statement that is tackled in the current study.
• Give logical reason for the selection of current strategy i.e., multivariate analysis.
• Quantitative data is also important to support your conclusion. I request the authors please carefully check and rewrite the results part in the abstract. Please provide a percentage increase or decrease in the result part.
• Please provide a conclusive conclusion that is withdrawn through research in a single line.
• Please conclude with a statement that shows a knowledge gap covered, potential beneficiaries and specific recommendations as well.
• Give future prospective in a single line. At least declare one best treatment. Please do not make vague statements as in science no space for that.
• As per standard suggestions, please avoid using title words as keywords

Introduction

• Please follow the title in the introduction section, i.e., anthropogenic activities, multivariate analysis, El-Minia governorate, Egypt and anthropogenic activities, knowledge gap, hypothesis and aims.
• Also, provide a novelty statement at the end. What new things authors have done or correlated in this research compared to old ones?
• Would you please give a single line about the knowledge gap which your research has covered along with the SMART (specific, measurable, achievable, realistic and time-specific) hypothesis statement?

Material and methods

• ok.

Results

• ok.

Discussion

• Please provide definite mechanisms associated with the results. The discussion part is very weak. Please incorporate at least 3-4 paragraphs showing major mechanisms due to which authors got such results

Conclusion

• Add the targeted beneficiary audience who will get benefits from this research.
• Also, give clear-cut recommendations
• Give future prospective regarding this research.

Reviewer 2 Report

the paper titled "Assessment effects of anthropogenic activities on soil contamination using multivariate analysis: A case study in El-3 Minia governorate, Egypt" discusses the application of some statistical methods on trace elements concentration in soils. Statistical methods are given by multivariate tools (PCA and cluster analysis) and by geostatistical (spatial) analysis. However in both cases the authors do not take into account that they manage "compositional data" (Aitchison, 1982; 1986 and subsequent strong developments; see for example the book and paper of Peter Filtzmoser, Vera Pawlowsky-Glahn, Juan Jose Egozcue and others...) and that the Euclidean space is not the adequate framework in which to perform data treatment. PCA and cluster analysis are very sensible to this aspect, as well as the covariance (bivariate statistics) and autocovariance (spatial analysis) structure. This means that all the results could be biased giving misleading information and would be re-considered in the light of recent developments in CoDA (Compositional Data Analysis) framework.