Review Reports - Ecological Risk Assessment of Heavy Metals Pollution in the Loskop Dam of the Olifants River System, South Africa

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Below, I am sending some suggestions regarding the article "Ecological Risk Assessment of Heavy Metals Pollution in the Loskop Dam of the Olifants River System, South Africa".

Line 35 = What could be the ecological consequences?;

Line 164 = What could be the reason for the detection of these specific metals at certain times of the year?;

Line 178 = It seems to me that calculating metal concentrations in this way does not make much sense;

Line 229, 235 = In Figure 4 and 5, the sample labels need to be corrected. Some of them overlap, making them impossible to read.

The manuscript lacks a subsection comparing the obtained research results with others already available worldwide. These results should be classified and described in relation to those from other studies.

One of my biggest concerns about this manuscript is that it is merely an environmental report. It lacks arguments and conclusions that would impact the scientific quality of the research. While the article is well-written as an environmental report, it adds nothing to the science. It should be improved to include aspects from other studies around the world, etc.

Author Response

Reviewer 1

Line 35 = What could be the ecological consequences?

Thank you for pointing out this. We agree with this comment. Therefore, we have added a statement to address the comment, “Some of the ecological consequences have been added to the introduction”. Page 2, line 47.

Line 164 = What could be the reason for the detection of these specific metals at certain times of the year?; (lines 298-301)

We agree, though a reason has already been given, this has been explained in the text; increased surface runoff from the catchment in summer when the area receives most of the rainfall and binding of metals to sediments during winter (page 14, lines 363-366).

Line 178 = It seems to me that calculating metal concentrations in this way does not make much sense;

The statement is deleted.

Line 229, 235 = In Figure 4 and 5, the sample labels need to be corrected. Some of them overlap, making them impossible to read.

We agree that some of the elements were overlapping. The figures have been corrected. See pages 10 and 11.

Some local and international studies have been included for comparison (pages 358-359 and 363-366).

This has been addressed throughout the text.

Reviewer 2 Report

Comments and Suggestions for Authors

Manuscript Title: Ecological Risk Assessment of Heavy Metals Pollution in the Loskop Dam of the Olifants River System, South Africa
Manuscript Number: sustainability-4287333

Overall Evaluation

This manuscript addresses an environmentally relevant issue by assessing heavy metal contamination and associated ecological risks in a major freshwater reservoir in South Africa. The work contributes to understanding sediment-water interactions and provides data that are useful for water quality management and policy frameworks aligned with sustainability goals. The integration of contamination indices such as EF, Igeo, CF, and ERI adds value to the ecological interpretation. However, despite the relevance of the topic, the manuscript shows several conceptual, methodological, and interpretative limitations that weaken its scientific rigor and broader applicability. The study design lacks sufficient depth, the statistical treatment is basic for the complexity of the dataset, and the discussion does not adequately link findings to broader ecological processes or global literature. There are also inconsistencies in reporting, insufficient justification of methods, and limited novelty in comparison with existing studies.

Abstract

The abstract presents results without clearly framing the research gap and does not adequately justify why this study is needed in the context of existing literature. The statement that ecological risk is low appears overly generalized without acknowledging site-specific or metal-specific variability, which may mislead readers. The description of methods is overly condensed and lacks clarity regarding sampling design and analytical robustness. It would be beneficial to include quantitative highlights (e.g., exact RI values or exceedance levels) to strengthen the scientific communication. The link to Sustainable Development Goal 6 is somewhat superficial and should be better integrated with the findings rather than presented as a generic statement.

Introduction

The introduction provides general background information but lacks a clear articulation of the specific research gap and novelty of the study. Several statements are broad and descriptive without critical synthesis of recent literature, particularly regarding sediment–water interactions and risk assessment frameworks. The justification for selecting Loskop Dam as a case study is not sufficiently developed in terms of its uniqueness or representativeness. The objectives are stated but not framed as testable hypotheses or specific research questions, which reduces the scientific clarity. The literature cited is somewhat outdated or generalized, and there is insufficient engagement with recent advances in ecological risk modeling and multivariate source apportionment.

It is recommended to include relevant literature that links disturbance, ecosystem processes, and pollution dynamics, for example:
Chaturvedi et al., 2025, Land Degradation & Development, https://onlinelibrary.wiley.com/doi/10.1002/ldr.70334

Materials and Methods

The study area description is basic and lacks ecological and hydrological context that would help interpret contamination patterns. There is no discussion of land-use gradients or upstream pollution sources in sufficient detail. The sampling design is limited, with only three sites and two seasons, which may not capture spatial and temporal variability adequately. The rationale for sample size and replication is not provided, raising concerns about statistical robustness.

The methodology for sample collection and preservation is described, but there is insufficient detail on quality assurance/quality control procedures beyond a brief mention of recovery percentages. The use of ICP-MS is appropriate, but calibration standards, detection limits, and uncertainty estimates are not clearly reported. The choice of Fe as a normalization element in EF calculations is not justified in relation to local geochemistry.

The statistical analysis is overly simplistic. The use of t-tests and ANOVA without testing assumptions (normality, homoscedasticity) is problematic. PCA is applied, but there is no discussion of data standardization, eigenvalues, or loading thresholds. More advanced multivariate techniques (e.g., cluster analysis or source apportionment models) would strengthen the analysis.

Results

The results are generally descriptive and lack depth in interpretation. Several sections repeat numerical values without synthesizing patterns or ecological implications. The presentation of physicochemical parameters does not clearly distinguish between statistically significant and non-significant findings beyond brief mentions of p-values.

In the sediment analysis, the ordering of metal concentrations is reported, but the ecological relevance of these gradients is not explored. The spatial distribution results are described without linking them to hydrodynamics or sediment transport processes. Correlation analysis is presented with high coefficients, but potential issues such as multicollinearity or spurious correlations are not addressed.

The PCA results indicate high explained variance, but there is insufficient interpretation of component loadings and no validation of the model. Figures are referenced but not critically discussed, and the seasonal clustering is not linked to environmental drivers in a mechanistic way.

Discussion

The discussion largely reiterates results rather than providing deeper ecological interpretation. The linkage between heavy metal distribution and anthropogenic activities is suggested but not supported by strong evidence or comparative studies. The role of sediment as both sink and source is mentioned but not explored in terms of biogeochemical cycling.

The interpretation of EF, Igeo, and CF indices lacks critical evaluation, particularly regarding their limitations and assumptions. The conclusion that ecological risk is low is presented without sufficient caution, especially considering that some metals exceed guideline values. The discussion does not adequately compare findings with similar studies in other regions, limiting its broader relevance.

Conclusion

The conclusion summarizes the findings but overgeneralizes the ecological safety of the system. The statement that there is no short-term risk should be qualified, considering the observed exceedances of certain metals. The implications for management and policy are not sufficiently developed, and recommendations remain generic. There is a need to highlight limitations of the study and suggest directions for future research, particularly long-term monitoring and inclusion of biological indicators.

Author Response

Abstract

Comments: The abstract presents results without clearly framing the research gap and does not adequately justify why this study is needed in the context of existing literature. The statement that ecological risk is low appears overly generalized without acknowledging site-specific or metal-specific variability, which may mislead readers. The description of methods is overly condensed and lacks clarity regarding sampling design and analytical robustness. It would be beneficial to include quantitative highlights (e.g., exact RI values or exceedance levels) to strengthen the scientific communication. The link to Sustainable Development Goal 6 is somewhat superficial and should be better integrated with the findings rather than presented as a generic statement.

Response: Agree. Thank you for the comments. We have revised the abstract section to address the comments raised.

Introduction

Comment; It is recommended to include relevant literature that links disturbance, ecosystem processes, and pollution dynamics, for example:
Response: Agree. The introduction section has been revised to highlight the research gap. More information is added to the introduction to address the comments raised.

Materials and Methods

Comments: The study area description is basic and lacks ecological and hydrological context that would help interpret contamination patterns. There is no discussion of land-use gradients or upstream pollution sources in sufficient detail.

The sampling design is limited, with only three sites and two seasons, which may not capture spatial and temporal variability adequately.

The rationale for sample size and replication is not provided, raising concerns about statistical robustness.

The methodology for sample collection and preservation is described, but there is insufficient detail on quality assurance/quality control procedures beyond a brief mention of recovery percentages.

The use of ICP-MS is appropriate, but calibration standards, detection limits, and uncertainty estimates are not clearly reported.

The choice of Fe as a normalization element in EF calculations is not justified in relation to local geochemistry.

The statistical analysis is overly simplistic. The use of t-tests and ANOVA without testing assumptions (normality, homoscedasticity) is problematic.

PCA is applied, but there is no discussion of data standardization, eigenvalues, or loading thresholds. More advanced multivariate techniques (e.g., cluster analysis or source apportionment models) would strengthen the analysis.

Response: Agree. The methodology section is revised with detailed description of the procedures used to collect data. Where possible the limitations are provided (page 446-456).

Results

Comments: The results are generally descriptive and lack depth in interpretation. Several sections repeat numerical values without synthesizing patterns or ecological implications. The presentation of physicochemical parameters does not clearly distinguish between statistically significant and non-significant findings beyond brief mentions of p-values.

Response: Agree. Even though the study mainly focused on heavy metal pollution, additional information is added to the physicochemical part of the study (page 14, 351-353; 358-359). Sediment contamination part of the study has been revised, including correlation and PCA results.

Discussion

Comments: The discussion largely reiterates results rather than providing deeper ecological interpretation. The linkage between heavy metal distribution and anthropogenic activities is suggested but not supported by strong evidence or comparative studies. The role of sediment as both sink and source is mentioned but not explored in terms of biogeochemical cycling.

Response: Agree. We have addressed the problem by including more work done in the study area (page 14, lines 358-359; 363-366; 378-384; page 15, 424-425).

The role of sediment as both sink and source is explained in terms of biogeochemical cycling (page 14, lines 373-375).

The limitations of the use of EF, Igeo, and CF indices have been stated (page 16, 446-448).

The conclusion that ecological risk is low is presented with a caution, especially considering that some metals exceed guideline values. It is stated that it is for the short-term (page 16, 469-472).

Conclusion

Comments: The conclusion summarizes the findings but overgeneralizes the ecological safety of the system. The statement that there is no short-term risk should be qualified, considering the observed exceedances of certain metals. The implications for management and policy are not sufficiently developed, and recommendations remain generic. There is a need to highlight limitations of the study and suggest directions for future research, particularly long-term monitoring and inclusion of biological indicators.

Responses: Agree. The limitations of the study are included (page 16, lines 446-456).

Reviewer 3 Report

Comments and Suggestions for Authors

1.The author employs global average shale values (Turekian & Wedepohl, 1961) as background values for calculating the enrichment factor (EF), contamination factor (CF), and geoaccumulation index (Igeo), rather than using local or regional geological background values for South Africa. This may introduce significant bias in identifying heavy metal sources, particularly if the region inherently exhibits elevated natural background concentrations. Use background values derived from undisturbed sediments or deep soils in the study area; if such data are unavailable, explicitly discuss the limitations of using global averages and interpret pollution levels cautiously in the conclusions.

2.The sediment samples were sieved at 0.2 mm (200 μm), whereas most international studies on heavy metals in sediments adopt a finer fraction of <63 μm. Coarser particles typically contain lower heavy metal concentrations, potentially leading to underestimation of pollution levels. Adopt a sieve size of <63 μm in future studies, or explicitly justify the use of 0.2 mm in the methodology and discuss its potential impact on the results. It is recommended to refer to recent research papers in the introduction. Multi-factor coupled numerical simulation and sensitivity analysis of hysteresis water inundation induced by the activation of small faults in the bottom plate under the influence of mining. Modelling and simulation on anisotropic non-Darcy seepage characteristics of gas in naturally fractured coal.

3.Principal component analysis (PCA) and correlation analysis generally require a larger sample size (recommended at 5–10 times the number of variables). With only 6 sediment samples (3 sites × 2 seasons) analyzed for 9 heavy metals, the sample size is severely insufficient, risking overfitting or spurious correlations. Clearly state the exploratory nature of this study and interpret PCA and correlation results with caution in the conclusions; recommend increasing sampling frequency and the number of sites in future work to enhance statistical reliability.

4.The distribution of heavy metals in sediments is strongly influenced by parameters such as organic matter content, iron-manganese oxides, and grain size composition. The absence of measurements for these factors prevents determining whether the high correlations between Fe, Mn, and certain metals stem from natural geochemical processes. Supplement data on total organic carbon (TOC), grain size distribution, and Fe/Mn oxide speciation, or at least discuss this omission as a study limitation.

5.The conclusion states that the sediments "do not pose any short-term risk to the environment," yet no bio-toxicity tests or bioaccumulation analyses were conducted; inferring ecological risk solely from the ecological risk index (ERI) is limited. Revise the conclusion to: "Based on sediment chemical analysis and the ecological risk index, current heavy metal pollution levels do not indicate significant short-term ecological risk," and recommend conducting studies on bioindicators (e.g., benthic invertebrates) to validate this assessment.

Author Response

Comments 1.The author employs global average shale values (Turekian & Wedepohl, 1961) as background values for calculating the enrichment factor (EF), contamination factor (CF), and geoaccumulation index (Igeo), rather than using local or regional geological background values for South Africa. This may introduce significant bias in identifying heavy metal sources, particularly if the region inherently exhibits elevated natural background concentrations. Use background values derived from undisturbed sediments or deep soils in the study area; if such data are unavailable, explicitly discuss the limitations of using global averages and interpret pollution levels cautiously in the conclusions.

Response: We agree with the comment. We have explained why the global average shale value was used for this study to calculate the enrichment factor (EF), contamination factor (CF), and geoaccumulation index (Igeo). The background sediment values are not available in South Africa. This has been included in the limitation section of the study (Page 14, lines 446-448).

Comments 2. The sediment samples were sieved at 0.2 mm (200 μm), whereas most international studies on heavy metals in sediments adopt a finer fraction of <63 μm. Coarser particles typically contain lower heavy metal concentrations, potentially leading to underestimation of pollution levels. Adopt a sieve size of <63 μm in future studies, or explicitly justify the use of 0.2 mm in the methodology and discuss its potential impact on the results. It is recommended to refer to recent research papers in the introduction. Multi-factor coupled numerical simulation and sensitivity analysis of hysteresis water inundation induced by the activation of small faults in the bottom plate under the influence of mining. Modelling and simulation on anisotropic non-Darcy seepage characteristics of gas in naturally fractured coal.

Response: We agree with the comment. The sediment samples were sieved at 0.2 mm (200 μm). The suggestion is taken for future studies.

Comments 3.Principal component analysis (PCA) and correlation analysis generally require a larger sample size (recommended at 5–10 times the number of variables). With only 6 sediment samples (3 sites × 2 seasons) analyzed for 9 heavy metals, the sample size is severely insufficient, risking overfitting or spurious correlations. Clearly state the exploratory nature of this study and interpret PCA and correlation results with caution in the conclusions; recommend increasing sampling frequency and the number of sites in future work to enhance statistical reliability.

Responses: We agree with the comment that the sample size was small. The suggestion is taken for future studies. However, it has been stated as one of the limitations of the study (Page 16, lines 449-450).

Comments 4. The distribution of heavy metals in sediments is strongly influenced by parameters such as organic matter content, iron-manganese oxides, and grain size composition. The absence of measurements for these factors prevents determining whether the high correlations between Fe, Mn, and certain metals stem from natural geochemical processes. Supplement data on total organic carbon (TOC), grain size distribution, and Fe/Mn oxide speciation, or at least discuss this omission as a study limitation.

Responses: We agree. The suggestion is taken and measurements of organic matter content, iron-manganese oxides, and grain size composition will be considered for future studies. Omission of such measurements are mentioned as limitations (Page 16, lines 452-456).

Comments 5. The conclusion states that the sediments "do not pose any short-term risk to the environment," yet no bio-toxicity tests or bioaccumulation analyses were conducted; inferring ecological risk solely from the ecological risk index (ERI) is limited. Revise the conclusion to: "Based on sediment chemical analysis and the ecological risk index, current heavy metal pollution levels do not indicate significant short-term ecological risk," and recommend conducting studies on bioindicators (e.g., benthic invertebrates) to validate this assessment.

Responses: We have revised the conclusion and included a statement, “Based on sediment chemical analysis and the ecological risk index, current heavy metal pollution levels do not indicate significant short-term ecological risk," and recommend conducting studies on bioindicators (e.g., benthic invertebrates) to validate this assessment” (Page 16, 469-472).