Identifying the Source of Heavy Metal Pollution and Apportionment in Agricultural Soils Impacted by Different Smelters in China by the Positive Matrix Factorization Model and the Pb Isotope Ratio Method

Round 1

Reviewer 1 Report

The authors have addressed on the concerns of the original submission. Accept as is. Thanks.

Author Response

Response to Reviewer 1 Comments

• Point:The authors have addressed on the concerns of the original submission. Accept as is. Thanks.

Response: We are grateful to the anonymous reviewer for your time reviewing this manuscript. Thanks for your comments.

Reviewer 2 Report

Review Manuscript Ref.: sustainability-1208068

Review: The manuscript is well structured and present interesting information on heavy metal pollution. However, it is very difficult to follow the corresponding changes stated in the cover letter as no reference line(s) are described relative to the new version of the manuscript. Yet, based on their response, these are some of the aspects that still require further (scientific) explanation. English has been improved. Additional comments/questions: 

1. Regarding my previous comment: “Based on literature data, the authors stated that the source of relatively high concentration of Zn, Cd and Pb in soil is the smelting plant. However, this statement lack of scientific proof as it is unknown the concentration of such elements in the output of such process. Further explanation and justification is required”

Although PMF can simplify the high-dimensional variables by using the correlation matrix and covariance matrix and transform them into several comprehensive factors – as has been stated by the authors. Does this statistical tool consider the operational conditions of the corresponding smelter and/or environmental conditions? The authors seem to unknown the fact that the concentration of metals in the output material from the smelter depends on the operational conditions.

2. The authors have also stated that “Pb isotope ratio method is considered reliable for source analysis [10–12] due to its tiny isotope fractionation, even if the physical and chemical conditions of the system change (e.g., smelting, coal burning, and coking), their isotopic composition will not change generally”.

Can the authors explain with examples why they consider Pb isotope ratio method a reliable source analysis? Is there a way to demonstrated reproducible results with such technique?

Although the isotopic composition “will not change generally”, there may be some conditions in which it does change. What’s the effect of the concentration of Pb?

3. Previously I have also commented on Fig. 5, which describes a single trend line, but five R2-values each one determined (apparently) based on only two points.

Although the Authors have stated in their cover letter that “the number of samples is small in the statistical, the samples are all representative in the research area”, they do not provide further information on how these R2 were determined and, for the same reason it is arguable the way how the Authors concluded that smelting is the main source.

4. 7 not the best way to show the heavy metal element content ratio fingerprint in the agriculture soil around Zhuzhou Smelter and Huludao Zinc Plant distribution in the overall trend. Due to the presence of a continuous line between the different elements, it is not clear what’s between Pb-As, As-Hg, Hg-Cr etc. Please, see my comment in the previous revision.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

References 34, 35, 37 and 39 need supplementation. For printed version deliver publishing company name and authors’ names (when available). For internet resources deliver URL and access date.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

There were no substantial revisins made.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 4 Report

Dear authors, 

You have done a good job in revising the manuscript. congratulations.

Just a slight English polishing. 

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

This study assessed the contribution of smelters, agricultural and others to cause heavy metal pollution in two regions of China. I think there's a good description of two smelters and their importance to the economic activity of their regions. I also think a good description of study background is provided for the study. I think the study is well designed and results well articulated. However, there are some grammatical errors that has to corrected. 

Reviewer 2 Report

The manuscript is well structured and present interesting information on heavy metal pollution. Certainly, mining activity and the corresponding extractive process to recover metals can have a significant impact in the surrounding area. However, in this research, the lack of information related to metal concentration in the output materials coming from the referred smelters does not allow to sustain several of the conclusions stated by the authors. In addition, the authors do not include any information regarding environmental regulations, particularly related to the maximum permissible limit of heavy metals in soil. These are some of the aspects that require further (scientific) explanation. English writing must also be reviewed by the authors.

Additional comments/questions: 

Abstract:

Line 17: It should read “different smelting plants” rather than “different smelt plants”.

Line 22: It should read “were seriously polluted” rather than “were seriously pollution”.

Section 1

Line 42: What the authors means with “such as Pb, Cd, Hg, As et al. [4]”. Please re-write and clarify.

Lines 43-45: Why the pollution status of heavy metals in agricultural soil is worsen? Why the authors stated that the source of heavy metals in agriculture soil is unknown, by taking into consideration that in the abstract it is clearly described that such pollution is caused by different smelting plants? Please clarify.

Line 55: What if the source component spectrum and source contribution rate is negative?

Lines 75-79: How Cd ended up contaminating agriculture soil?

Line 94: What’s the “Spearman's correlation coefficient analysis”? Please include the corresponding reference.

Section 2:

Lines 163-164: What’s DZ/T 163 0184.12-1997? Please add the corresponding reference.

Lines 171-172: Please write the corresponding sub-index in Ci and Si.

Line 173: What’s GB 15618—2018? Include the corresponding reference.

Line 186: What’s “mass fraction of the jth element in the kth source”

Lines 211-213: The authors stated “the skewness values of all metals except As were greater than unity, showing that these metals positively skewed towards lower concentrations”. However, by observing table 2, such statement is not clear at all. What the authors mean with “skewness” and “skewed”?

Line 214: What the authors mean with “steeper”?

Lines 215-216: The authors gave some figures on the rise of concentration of some metals, but they do not state against what these figures are being compared to? This also leads to the question what’s the maximum permissible heavy metal concentration in agriculture soil and/or in water and/or in the air? Further explanation is required.

Lines 217-219: The authors stated described some percentages for Cd, Pb, Zn and Hg with which they conclude that the pollution of these four metals was “severe”. Further explanation is required to understand how it is possible to conclude such statement based on percentages? Also, why the authors did not use percentages when they refer to As, Ni, Cu, Cr, Zn, 215 Pb, Hg and Cd in lines 215-216?

Line 232: What the authors mean with “background values of soil”?

Line 233: What’s the standard?

Lines 236-238: The authors stated “spatial distributions of concentrations of soil heavy metals in this area would have a higher possibility of being influenced by extrinsic factors”. What are the extrinsic factors stated by the authors? Please, explain further.

Line 239-241: The authors stated that both research areas were polluted by Cd, Pb, Hg and Zn. However, it is not described how these meals ended up in the agriculture soil. Further explanation is required.

Lines 252-253: Where such standard are coming from? Please clarify.

Line 297: What the authors mean with “rose”?

Section 3.1.3: Based on Figs 3 and 4, there is a correlation between the concentration of metals and the location of the corresponding smelting plants, which is expected by taking into consideration the research area described in section 2.1. Can the authors give further information regarding the concentration of metals coming out from the smelting plant?

Section 3.1.4: What the author means with correlation between heavy metals? Is there any physical chemical association between the different elements? How type of compounds are they forming? Are they stable/unstable?

Section 3.2:

Line 362: It should read “Many related studies have shown that Cd, Pb and Zn ARE generally associated with medium…”  rather than “Many related studies have shown that Cd, Pb and Zn WERE generally associated with medium…”

Line 364: It should read “smelting lead and zinc ores, volatile oxides ARE generated…” rather than “smelting lead and zinc ores, volatile oxides WOULD BE generated…”

Lines 359-367: Based on literature data, the authors stated that the source of relatively high concentration of Zn, Cd and Pb in soil is the smelting plant. However, this statement lack of scientific proof as it is unknown the concentration of such elements in the output of such process. Further explanation and justification is required.

Lines 368-373: It is understood that the main source of Mg, Ca and Fe is the chemical decomposition of natural rocks. How the natural concentration of these metals compare with the concentration of the output material coming either from the mining and/or the smelting processes?

Lines 375-383: Regarding the presence of As, Al and Cr, the results allow to partially agree with the authors statement on the application of chemical fertilisers as the main source. Notice that early in the manuscript, the authors describe that such elements are part of the minerals being processed in the corresponding plants. How much is the concentration of such metals in the chemical fertilisers mentioned by the authors? Do the comply with local regulations? And again, how does the concentration of such elements compare with the concentration of the output material coming either from the mining and/or the smelting processes?

Lines 392-397: Need to be reviewed based on my previous comments.

Section 3.3

Lines 438-443: Fig. 5 describes a single trend line, but five R²-values each one determined (apparently) based on only two points. How the authors can ensure reproducibility of these results by taking into consideration that (apparently) each R²-value was determined from only two points?  

Line 444: It should read “…the smelting, Vehicle exhaust dust…” and not “…the smelting, ehicle exhaust dust…”

Lines 448-450: The authors stated that “Pb from smelting, vehicle exhaust dust and fertilizer might be the main input pathway for anthropogenic Pb to the soils, but not by soil parent materials”. What the authors mean with soil parent material?

Lines 467-473: This paragraph should be included in the conclusion sections. Also, what are the corresponding advantages and disadvantages in heavy metal source analysis by PMF and isotope ratio methods? Please describe further.

Section 3.4:

Lines 485-486: In which state Cu, Zn, Cd and Pb are released from the corresponding smelter and in which concentration?

Fig. 7: Why the authors decide to use a graph with steps and peaks rather than bars? As it is, the graph is not readable. Further explanation is required, otherwise please change the format of the graph to bars. What “content ratio to Zn” means? Please add the corresponding description in the caption of the figure.

Lines 494-497: It is not clear the correlation between the “sharp peaks” and the profile of the agriculture soil around smelting areas. Further explanation is required.

Fig. 8: The same as Fig. 7.

Conclusion must be modified based on these comments/suggestions.  

Reviewer 3 Report

The article is interesting but it requires a careful, comprehensive revision. First of all, description of methods applied in the data analysis should be more "reader friendly". Some essential information were not delivered, hindering understanding of the Authors' concepts.

For example, in description of Source Apportionment by Pb Isotope Ratio Method:

According to the text in lines 132-137, samples of 4 different materials were collected. 2 samples of factory soil, a single sample of fertilizer, not mentioned number of vehicle exhaust dust (collection method not described), 2 samples of a dust from exhaust pipe from a motor vehicle (collection method not described) and 2 samples from deep soil layers. 
Problem 1. The description of samples collection suppose very small number of samples. Actually, statistical verification of any hypothesis related to parameter in the groups of materials is discarded.
Problem 2.
For a limited number of samples, where the determination coefficient values come from? 
Problem 3.
In l.  433-436 “The prerequisite for using the isotope ratio method is the significant difference of the isotope ratios in the end-members. The result of one-way ANOVA analysis was a significant difference in Pb isotopic compositions among the smelting, fertilizer, soil parent materials and vehicle exhaust dust (F > 1, P < 0.050)”. ANOVA verifies a null hypothesis H0 whether two or more population means are equal. Samples from different populations can be distinguished by levels of a grouping variable. Rejection of H0 means that statistically significant difference occurred for at least 2 levels of the grouping variable, but it doesn’t mean significant differences between all variables. 

Minor concerns:

The aim of the study was not clearly formulated.
l. 114 what “… about 8 hm2.” means?
l. 145 correct PH to pH in all occurrences.
The Pollution Index (l. 168)and The Nemerow Integrated Pollution Index (l. 174) are not parameters resulting from statistical analysis. These are parameters describing soil pollution.
l. 155 the reference to DZ/T 0223-2001 is required.
l.163 the reference to DZ/T 0184.12-1997 is required.
l. 173 the reference to GB 15618—2018 is required.
l. 181 the reference to State Environmental Protection Administration is required.
l. 195 what information provides eq. 6 to a reader?

Reviewer 4 Report

The paper entitled "identifying the source of heavy metal pollution and apportionment in agricultural soils impacted by different smelters in China by Positive Matrix Factorization Model and Pb Isotope ratio method" was submitted for possible publication. Kindly refer to the following comments.

1. There as no discussion and comparison on the maximum contaminant level of each heavy metal pollutant to justify that the research area is really polluting the agricultural soil.
2. there was no clear explanation as to why Pi and NIPI will be used since they are both measures or index of pollution.
3. The sampling points considered  in Huludao Zinc Plant was more distributed than that of Zhuzhou City Smelter.  From Figure 1, sampling points from Zhuzhou Smelter were concentrated into 3 points while it is concentrated in 4 points in Huludao Zinc Plant. There should be a mention of what is the implication of these sampling points to the  statistical results. This might yield higher coefficient of variation for Huludao Zinc Plant. 
4. Skewness and Kurtosis was used to describe the metal concentrations in both Smelting plants.  It is better that contamination was expresses as mean +/- SE.  This will describe more the heavy metal contamination.  The coefficient of variation (CV) is relative to the other Smelting plant and not relative to a standard value.  Then what about the negative skewness of As?
5. The skewness and kurtosis plus CV must not be used as an indicator of pollution. Pollution is identified based on a standard parameter based on a pronounced policy like EPA, WHO, etc. 
6. A test on significance of skewness and kurtosis might have made sense.
7. I cannot understand what you mean by over-standard rate in line 266.
8. What is the point of comparing your values with other smelting areas in China in table 3?
9. Figure 2 is in chaos.  What are the units of measurement of the x-axis? What does the colors stands for?
10. You mentioned in section 3.1.3 that Cr and Ni ha some similar features, but your correlation matrix in Table 4 does not say so. 
11. The findings stated in lines 306 to 312 was not seen in any of the findings presented.
12. Section 3.1.4 must discuss the the significance in correlation. What dos the significance in correlation between Cd and Pb, between Cd, As and Pb and so on. Do we expect that if Cd is high, Pb and As will also be high? Something like that.
13. The R-squared values in Figure 5 is week since some are only 2-data points. There should be more data  points considered. 
14. Why is Xinxiang and Fujian considered as point of comparison in Figure 7? what is the significance of Xinxiang and Fujian in this study?
15. The contribution ratio in Figure8 is better of presented as a pie graph.
16. The authors have use several statistical treatments on a single data set which have compounded the level of significance for the overall tests. It is so surprising that no post hoc analysis was employed to explain the significance of each tests on difference or relationship.