Farmland Trace Metal Contamination and Management Model—Model Development and a Case Study in Central Taiwan
Round 1
Reviewer 1 Report
Title:
Farmland Heavy Metal Contamination and 2 Management Model – Model Development and 3 a Case Study in Central Taiwan
General comments :
The manuscript is clearly written and organized. This is particularly true at the paragraph level; some of the bigger-picture organization is a little harder to follow.
My main concern is about the use of C(t,i) which is a hypothetical concentration that includes both the soil-associated metals and the porewater metals. This is not a real concentration (it is a weighted average of two) and does not connect with the later processes – e.g., leach and infiltration, plant uptake, etc.
With the number of parameters that are used, and many that are similarly names (e.g., C(t,i) vs C(T,i)), it is hard to keep track. Can the authors please include the following:
- Decide on a consistent and revealing symbol naming scheme
- Include a short paragraph that describes the labeling scheme – this will hopefully give the reader a better overall sense for the parameters
- Include a table with all parameters, definitions, and units; also list how parameters values were determined (e.g., sources). This is generally included in Table 1 (great!) but helpful if this is refered to earlier.
There is also some confusion on labeling of other aspects – e.g., “standard” as used in latter tables and figures.
Line-by-line:
Abstract: it would be stronger if more of the abstract were devoted to results. As it currently it written, only a ~quarter of the length is devoted to discussion of findings from the study.
L45+: I am curious where most of the “billions” has gone if only 60 million has been for “expenditures for soil remediation on farmlands?”
L112+: It seems like there are a couple important assumption for the concentration-based equation: 1) that the sediment mass is not changing, 2) the desorption into the water does not significantly affect the concentration. By expressing this as concentration, it also implies a homogeneity (e.g., vertical within the system boundary). While the concentration-based equation may help with connecting to the main questions, the underlying assumption and simplifications need to be explained.
Equ 8: Three points on this equation.
- The appearance for the symbol for density (rho) is different in equation 8 than in equation 3 which can cause confusion.
- Equation 8 makes it clear that Ct,i represents both soil and water contributions in a “hypothetical combined concentration” (i.e., doesn’t match the real world) but this is not clear from the text.
- The use of “n” to represent the isotherm non-linearity is a well-established model known as the Freundlich, but should be described in a way that is consistent with the well-established model. With this in mind, Kf should be used in place of Kd (where Kd is often reserved for linear isotherms). “n” is often called the Freundlich exponent.
Eq 10: This is using the combined concentration (C(t,i) – which includes both the soil-associated and porewater aspects), but only the dissolved constituents can leach into the underlying water.
Fig 3 – looks like the top panel has green dots for the calibration sampling sites, but this is not clear based on the figure labeling (i.e., only red dot in key).
L256+: How were the samples extracted and analyzed? What were limits of quantitation? What was the data processing approach if a measured concentration was < LOD or < LOQ?
Model limits – thanks for including this section. Very important.
Table 2 – Can you add the number of measurements that when into the results shown here? Also, minor editorial aspect, but I suggest listing the MAP criteria in the opposite order
Fig 5 – please include the parameter symbol so that it is clear where it fits into the system of equation. “Sediment ratio” is not shown in Table 1, so likely some inconsistencies in labeling. What time period or scenario was considered for this calibration?
Table 3 – seems that you are missing a footnote to define control and monitoring standards. I am unfamiliar with these terms, so additional information would be much appreciated.
Fig 6 – missing x axis values.
L339 – better to give some guidance on what is meant by “long period of time.” E.g., months? Years? Decades?
L345: I’m not sure what “With” indicates for section 3.4 title, but suggest excluding this word.
Fig 8 – I can’t see the blue series (sediment – irrigation – plants) for most of the graphs. This may be because they are under another series, but need to make this visible in some way
L375: what are the units or what is the time-period for the removal rates? E.g., percent per year? Of is this for the entire simulation?
Author Response
General comments :
The manuscript is clearly written and organized. This is particularly true at the paragraph level; some of the bigger-picture organization is a little harder to follow.
My main concern is about the use of C(t,i) which is a hypothetical concentration that includes both the soil-associated metals and the porewater metals. This is not a real concentration (it is a weighted average of two) and does not connect with the later processes – e.g., leach and infiltration, plant uptake, etc.
Reply:
Thank you for your advice. We have reconfirmed all the model parameters, their definitions and sources, and compiled them in Table 1 to facilitate reading and verification.
With the number of parameters that are used, and many that are similarly names (e.g., C(t,i) vs C(T,i)), it is hard to keep track. Can the authors please include the following:
- Decide on a consistent and revealing symbol naming scheme
- Include a short paragraph that describes the labeling scheme – this will hopefully give the reader a better overall sense for the parameters
- Include a table with all parameters, definitions, and units; also list how parameters values were determined (e.g., sources). This is generally included in Table 1 (great!) but helpful if this is refered to earlier.
There is also some confusion on labeling of other aspects – e.g., “standard” as used in latter tables and figures.
Reply:
We have reconfirmed all the model parameters, their definitions and sources, and compiled them in Table 1 to facilitate reading and verification.
The definition of relevant standards has also been described in Table 3 with footnotes.
Line-by-line:
Abstract: it would be stronger if more of the abstract were devoted to results. As it currently it written, only a ~quarter of the length is devoted to discussion of findings from the study.
Reply:
The abstract has been rewritten to properly illustrate the relevant research results. See abstract.
L45+: I am curious where most of the “billions” has gone if only 60 million has been for “expenditures for soil remediation on farmlands?”
Reply:
US$60 million refers to the cost of remediation of contaminated farmland, and does not include other monitoring, investigation, and crop destruction costs.
L112+: It seems like there are a couple important assumption for the concentration-based equation: 1) that the sediment mass is not changing, 2) the desorption into the water does not significantly affect the concentration. By expressing this as concentration, it also implies a homogeneity (e.g., vertical within the system boundary). While the concentration-based equation may help with connecting to the main questions, the underlying assumption and simplifications need to be explained.
Reply:
Yes, for the simplification of the model and ease of use, we assume that the concentration of the sediment remains unchanged, and the desorption into the water does not significantly affect the concentration.
It is assumed that the mass and concentration of the sediment will not change during the process of transporting to the paddy field, and the boundary conditions of the plough layer will not be affected. See line 169-171.
Equ 8: Three points on this equation.
- The appearance for the symbol for density (rho) is different in equation 8 than in equation 3 which can cause confusion.
- Equation 8 makes it clear that Ct,i represents both soil and water contributions in a “hypothetical combined concentration” (i.e., doesn’t match the real world) but this is not clear from the text.
- The use of “n” to represent the isotherm non-linearity is a well-established model known as the Freundlich, but should be described in a way that is consistent with the well-established model. With this in mind, Kf should be used in place of Kd (where Kd is often reserved for linear isotherms). “n” is often called the Freundlich exponent.
Reply:
The notation of density has been revised to be consistent in equation 8 and in equation 3.
The model assumes that under homogeneous conditions, the total concentration of trace metals in the plough layer includes solid phase and solution phase. Detailed in line 161-163.
Thanks for your suggestion, kd has been corrected to kf. Detailed in line 156-162.
Eq 10: This is using the combined concentration (C(t,i) – which includes both the soil-associated and porewater aspects), but only the dissolved constituents can leach into the underlying water.
Reply:
We have added relevant descriptions: In order to simplify the model, atmospheric deposition is represented by dust fall (The particle size is about 10-100μm). When dust fall enters the topsoil, it is regarded as completely mixed, and the properties is the same as that of the topsoil, and does not consider the proportion of the dissolved constituents in the falling dust. See line 173-179.
Fig 3 – looks like the top panel has green dots for the calibration sampling sites, but this is not clear based on the figure labeling (i.e., only red dot in key).
Reply:
Sorry for our negligence. The colors of the sampling sites have been revised to be consistent. Please refer to Fig 3.
L256+: How were the samples extracted and analyzed? What were limits of quantitation? What was the data processing approach if a measured concentration was < LOD or < LOQ?
Reply:
Analysis method: Firstly, 1,300 sets of soil samples were screened by portable XRF (X-ray fluorescence) in this study irrigation area with grid method, and then the XRF screening results were selected to exceed the regulatory standard, and the microwave-assisted aqua regia digestion method was used for the full analysis of trace metals. Among them, there are 119 sites in area (a) and 174 sites in area (b). The detection limits of detector (710 ICP-OE) for Cr, Cu, Ni, and Zn are 1.55, 1.38, 1.24, and 1.52 mg/kg, respectively. Sample sampling and analysis were completed by the EPA project [21]. The main contaminants in the agricultural lands are Cr, Cu, Ni, and Zn with mean concentration between 296–1056 mg/kg (Figure 4). The mean concentrations from the 119 sampling sites for Cr, Cu, Ni, and Zn in study area (a) were 410, 464, 348, and 1024 mg/kg, respectively and from the 174 sampling sites in study area (b) were 390, 296, 430, and 1056 mg/kg, respectively. Since 1991, the remediation of farmlands contaminated with TMs in Changhua County has been performed continuously through many remediation projects funded by central and local governments. The above two groups of paddy field soil contamination data belong to two different projects, and the remediation work has begun in 2019. Detailed in 2.2.11. Case study.
Model limits – thanks for including this section. Very important.
Reply:
Thanks.
Table 2 – Can you add the number of measurements that when into the results shown here? Also, minor editorial aspect, but I suggest listing the MAP criteria in the opposite order
Reply:
The number of measurements was added in the Table 2, and listing the MAP criteria have been rearranged. Please refer to Table 2.
Fig 5 – please include the parameter symbol so that it is clear where it fits into the system of equation. “Sediment ratio” is not shown in Table 1, so likely some inconsistencies in labeling. What time period or scenario was considered for this calibration?
Reply:
The parameter symbol Sediment ratio b is marked in Fig 5. The description in Table 1 has also been corrected. Please refer to Table 1 and Fig 5.
The target year of the simulation was set at 40 years, mainly because Taiwan's economy began to take off rapidly in the 1970s and 1980s. There are a large number of metal surface treatment plants in this study area, and TM contaminants are discharged into irrigation channels along with the effluents. Detailed in section 2.2.11. Case study..
Table 3 – seems that you are missing a footnote to define control and monitoring standards. I am unfamiliar with these terms, so additional information would be much appreciated.
Reply:
Sorry for our negligence, Notes have been supplemented explained. Detailed in Table 3.
Soil pollution control standards: refers to the soil pollution control limits set to prevent the deterioration of soil pollution.
Soil pollution monitoring standard: refers to the concentration of pollutants that are required to be monitored for soil pollution based on soil pollution prevention purposes.
Fig 6 – missing x axis values.
Reply:
Sorry for our negligence, the X-axis numbers have been added, please refer to Figure 6.
L339 – better to give some guidance on what is meant by “long period of time.” E.g., months? Years? Decades?
Reply:
Thank you for your advice. The results of the simulation may exceed the regulatory standards in about 15-20 years, so it is more appropriate to use decades. Detailed in line 392.
L345: I’m not sure what “With” indicates for section 3.4 title, but suggest excluding this word.
Reply:
We have excluded ‘with’. Please refer to section 3.4 Irrigation channel sediment.
Fig 8 – I can’t see the blue series (sediment – irrigation – plants) for most of the graphs. This may be because they are under another series, but need to make this visible in some way
Reply:
Sorry for our negligence. Because it almost overlaps with the results of another scenario simulation, it is not easy to distinguish on the graph. In response to the above problems, we have redrawn to improve the resolution. Please refer to Fig 8.
L375: what are the units or what is the time-period for the removal rates? E.g., percent per year? Of is this for the entire simulation?
Reply:
The unit of removal rate is percent per year. It has been added in the paragraph, detail in line 430-431.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Please, see the attached report.
Comments for author File: 
Comments.pdf
Author Response
The goal of the paper is convincing, and the work-flow process used as base of the mathematic model is well explained. Understanding cumulative effects of the quality of water and sediments in the irrigation channels of the paddy soil are interesting topics, albeit very complex being the processes involved of different nature and ruled by different bio-geo-chemical factors. Indeed, the paper needs major details that can help international readers in understanding the efficiency of the approach and especially allow its replicability. The Case study representation is often poor in details and sometime vague in findings that does not allows to catch the potentiality of the management model proposed. Some input data provided (collected by previous studies), and the way of collecting them need to be more visible in the manuscript (by using also the Supporting Information).
Reply:
Thank you for your advice. In order to understand and confirm the real cause of pollution in the paddy field, this study uses mathematical methods to verify whether the results of past investigations are correct. The part of the case study has been rewritten, and the source and use of the data (parameter) have been explained in detail. Detailed in 2.2.11. Case study and Table 1.
In addition, in some paragraphs are reported consideration/findings are given as a fact, without supporting with data: Row 271 is reported “in the present model are included irrigation channel water quality, sediment HM concentrations, paddy soil HM concentrations, and other data” – which are the other data? What is the irrigation channel water quality? The definition of the water quality is given by different parameters that can be of different fields (chemical, ecotoxicological etc.). How is calculated the Runoff water quality? CWii? Being the elements considered 4, I‘m wondering which values are considered in the model: is the concentration of metals in water calculated as a median of the four trace elements (Cu, Zn, Ni, Cr?) (is a sum, a mean, a geometric mean value? Or as a ratio to reference values (in terms of Cx/Cr when Cx the element concentration in the given with respect to the national reference admissible)?
Reply:
The part of the case study has been rewritten, and the source and use of the data (parameter) have been explained in detail. Detailed in 2.2.11. Case study and 2.2.12. Model parameters.
The mean value (including the upstream, middle and downstream channel) of irrigation water quality (CWI,i) and sediment metal concentration (CS,i) were taken from the more complete survey results in 2006 [27] to represent the mean value of the past decades. Considering that the paddy field irrigation rate is fast during irrigation, in order to simplify the calculation process, this study regards the water quality of irrigation water and runoff water as the same.
The same consideration for CSi (in sediments). What is considered sediment HM concentrations. Row 306-307: In the explanation of the Model Simulation, the authors cited the Figure as and stated “In the results of this study, as shown in Figure 6, it was found that the HMs in the paddy soil have mainly come from the sediment in the irrigation channel. Looking at Figure 6 graphs and captions I was not able to catch the information included in the text. Are the graphs in Figure 6 referred to the irrigation channel? The caption states: “Model simulating the changes in HM concentrations with time”: How can be possible to understand from this figure that the paddy soils are impacted by sediments in the irrigation channel?
Reply:
We rewrite line 358 as’ In the results of this study, as shown in Figure 6, 7, and 8. ‘ to facilitate understanding, and at the same time add in line 410-412 with ‘In Figure 8, when comparing the situation with and without sediment, the difference in the final concentration of TM can be found, and the amount of change in concentration is the degree of influence of the sediment.’
Figure 6 is the simulation result of Equation 26, which means those factors such as irrigation, atmospheric deposition, fertilization, runoff, plant uptake, and sediment have been included in the simulation result. In Figure 6, the description of the Y axis has been corrected to concentration (mg/kg), CT, i, it represents the relationship between the total TM concentration in the soil and the simulation time.
In addition, did the management model defined calculate the synergistic effects in the process of leaching or uptake of metals? Both processes depend on soil microorganisms/ microbial activity and on some soil properties especially soil pH and redox potential during the analyzed period. I did not see in the Model physico-chemical data input. In addition, bioavailability of the metals in soils depends not only on organic matter in the soils but also in some depositional factors, mineralogic contents (content of oxy-hydroxy of Al, Fe, Mn clay minerals? Are these factors considered in the Model?
Reply:
Since the plough layer is regarded as a completely homogeneous mixed state, the total heavy metal concentration of the soil in the paddy field includes both the solid phase and the solution phase. In the process of percolation, the influence of the concentration gradient of heavy metals is ignored. At the same time, external factors such as pH, chemistry, physics and biology are not considered. In order to improve the usability and convenience of the model, most of the impact factors have been simplified. Detailed in 2.2.14. Model limits.
Some other uncertainty on the concentration of metals provided by fertilizers. It depends by the crops considered and by crop’s nutritional needs. Not all crop rotations require the same amount of fertilizer. Is the value considered specifically for paddy culture fertilized needs? What kind of values are considered for the simulation?
Reply:
The study area is a paddy field, the main crop is rice, and the fertilizer applied is mainly chemical fertilizer in the past decades. In the simulation process, the cumulative amount of heavy metals is calculated based on the annual average fertilizer usage and converted into the model calculation based on the heavy metal content in the fertilizer. Detailed in Case study and Table 1 The content of heavy metals in fertilizers refers to the related research results of Reference 22.
For atmospheric depositional concentrations, which dust particulates are considered: particles (PM2.5), coarse particles (PM2.5–10), and super-sized particles (PM10+). Are considered seasonal and geographical variations of atmospheric depositional concentrations being the simulation period very large (40 year)?
Reply:
In this study, atmospheric deposition refers to coal dust, factory buildings, road dust, air dust, rain inclusions and other relatively large particles (approximately 10-100 μm) that are released when fuel is burned, and which naturally fall to the ground by gravity and cause public disgust. The measurement method is to collect the amount of falling dust within a square kilometer area with a dust bucket every month. The annual average value taken from the study area from 2006 to 2015 was used as the input condition of the model, and according to its heavy metal content, it was converted into a cumulative milligram input mode per square meter per year. Detailed in line 174-179, Case study and Table 1.
Some clarification about the Evapotranspiration calculation. It is the value considered in the Model referred to the Effective ET (that is referred to a specific crop? Please, provide major details. Other details needed:
Reply:
The related data of paddy field water balance in this study area is taken from Reference 28 and 29. The water budget of each simulated component during 2004–2009, including Irrigation water, Effective rainfall, Evapotranspiration, Vertical percolation. Detailed in Case study and Table 1.
What does it mean the sentence in the row 69 –"the correlation between the HM content in the topsoil and that in the corresponding aqueduct sediments and irrigation water has been determined to be 86% and 57%, respectively”. Is it referring to a statistical correlation parameter Pearson Coefficient R2=? Or other?
Reply:
The related data was taken from Reference 13. The method of “paired t-tests” was used to compare the current results with the data gathered by TARI (Taiwan Agricultural Research Institute, Council of Agriculture). For the farmland soils with the contents of heavy metals exceeding the control standards, the correlations of the heavy metal contents in the topsoil with those in the corresponding sediment and water were determined to be 86% and 57%. Detailed in line 66-68.
Row 249: After collecting … it was concluded that the farmland’s HM contamination was closely related to the irrigation system. Is this finding referred to the paper cited beforehand? (21 – EPA Taiwan?). Please, clarify I can only suggest authors to re-organize better the paper, providing major explanation in the validation model sessions: it would be useful to provide those information that can enable to understand the functionality and what kind of data are specifically included for each scenarios(what is the meaning of absent values in the table of input data model).
Reply:
Thank you for your advice. According to the study results of Reference 13 and 21. Poor management of the irrigation system leads to an indiscriminate mix of irrigation water and discharged wastewater from the manufacturing industries, which is the main cause of farmland contamination. In order to understand and confirm the real cause of pollution, this study uses mathematical methods to verify whether the results of past investigations are correct. The description of case study, including parameter acquisition and data analysis, has been reorganized in this article. Detailed in Case study and Table 1.
Reviewer 3 Report
Interesting work, however, there is no consideration for the factor related to soil / solution pH as in an alkaline environment most of the metals are not absorbed by the plants. Moreover, we no longer use the phrase „heavy metals” but „toxic / trace metals, leaving it for the type of music.
Author Response
Interesting work, however, there is no consideration for the factor related to soil / solution pH as in an alkaline environment most of the metals are not absorbed by the plants. Moreover, we no longer use the phrase „heavy metals” but „toxic / trace metals, leaving it for the type of music.
Reply:
Thank you for your advice. In order to simplify the model and improve the convenience of its application, we ignored the impact of soil / solution pH changes on the efficiency of soil absorption of heavy metals. This is also a limitation of the model, and we have added this recommendation to the section of the model limitation. Detail in line 326-333, section 2.2.14. Model limits. In addition, the related term ‘heavy metals’ has been revised to ‘trace metals’.
Round 2
Reviewer 2 Report
The paper was substantially enhanced and can be accepted in the present revised form.
