Leaching Runoff Fraction for Nitrate and Herbicides on Sugarcane Fields: Implications for Grey Water Footprint

Round 1

Reviewer 1 Report

Dear authors,

Please find below comments on your article titled “Leaching runoff fraction for nitrate and herbicides on sugar-cane fields: implications for grey water footprint”, where you discuss results from your field experiments and compare and contrast them with theoretical values obtained from commonly used methods in the literature.

In general, I found this to be an interesting and well-written paper. I think that this should be of interest to the journal readers. However, I believe that some improvements should be made to clarify some aspects of the work. Please find below some comments on these suggested changes.

-          In line 71, you mention “Tier 1”. I think that the three tiers need to be briefly described here, as most readers are unlikely to know what these are without needing to read the cited reference.

-          In the paragraph beginning on line 121, you introduce the three treatments for trial 1. To me, it is very unclear what these treatments are on from this paragraph. (Only two are mentioned initially, before the control treatment is mentioned later on.) Could you please re-write this to be clearer. Perhaps, something like this: “The treatments comprised 120 kg N ha^-1 of ammonium nitrate applied via broadcast over residue mulch (N_120bro), 120 kg N ha^-1 of ammonium nitrate incorporated in the soil at 0.1 m on both sides of the sugarcane row (N_120inc), and the control treatment without nitrogen fertilization (N_zero), according to the methodology proposed by Castro et al. [29].”

-          In lines 129-133, it is unclear whether the additional chemicals and the herbicides were also applied to the control area, or just to the two other treatment areas.

-          The statistical analysis section is very brief and should be expanded slightly. For example, it is stated that the homogeneity of variance was checked, but it is not clear what the variables are that are being considered. The statement about the Tukey test being used for comparing significance seems incorrect. I think that you mean that the Tukey test was used to assess whether means for different group were significantly different from each other. I suggest being a bit more explicit on what exactly the groups being compared are, e.g. what are the variables being compared.

-          In Table 2, you include some letters, a, b, A and B. It is not clear from the text what these letters mean. Could you please explain this clearly in the table caption?

-          Given that standard errors are included in Table 2, I think it would help to link to the equation(s) used in the calculations of these values.

-          I think that adding “replicate” as a column would make Table 3 clearer.

-          It looks like the words “preview” and “previews” are being incorrectly used (e.g. lines 127 and 139). I think that these should be “previous”. If I am incorrect, this requires a short explanation of what is meant here.

Dear Editor,

Thank you for sending the paper titled “Leaching runoff fraction for nitrate and herbicides on sugar-cane fields: implications for grey water footprint” for review. In this paper, the authors discuss results from their field experiments and compare and contrast them with theoretical values obtained from commonly used methods in the literature.

In general, I found this to be an interesting and well-written paper. I think that this should be of interest to the journal readers. However, I believe that some improvements should be made to clarify some aspects of the work. I have suggested some changes in my comments to the authors.

Author Response

Dear Reviewer, on behalf of my colleagues, the manuscript “Leaching runoff fraction for nitrate and herbicides on sugarcane fields: implications for grey water footprint” has been revised following the suggestions of the Reviewers, and it is now being resubmitted. We greatly appreciate your kind and careful revision, and we would like to give answers and clarifications to the questions and criticisms raised by you in the above general comments. All modifications were made intending to meet the Reviewer’s desires as much as possible.

Thank you, Fabio Scarpare

-        In line 71, you mention “Tier 1”. I think that the three tiers need to be briefly described here, as most readers are unlikely to know what these are without needing to read the cited reference.

Dear Reviewer that is correct. There are two other tiers (Tiers 2 and 3). Although outside the scope of this study, a brief description of these tiers (2 and 3) is now in the introduction (third paragraph). Moreover, in the discussion section (third paragraph), we mentioned Tier 2 and 3 and how limited this approach can be used in large-scale GWF assessment.

-          In the paragraph beginning on line 121, you introduce the three treatments for trial 1. To me, it is very unclear what these treatments are on from this paragraph. (Only two are mentioned initially, before the control treatment is mentioned later on.) Could you please re-write this to be clearer. Perhaps, something like this: “The treatments comprised 120 kg N ha^-1 of ammonium nitrate applied via broadcast over residue mulch (N_120bro), 120 kg N ha^-1 of ammonium nitrate incorporated in the soil at 0.1 m on both sides of the sugarcane row (N_120inc), and the control treatment without nitrogen fertilization (N_zero), according to the methodology proposed by Castro et al. [29].”

Thank you for your suggestion. Done.

-          In lines 129-133, it is unclear whether the additional chemicals and the herbicides were also applied to the control area, or just to the two other treatment areas.

The herbicides were applied in all treatments, including the control. We added this information to the text (line 139).

-          The statistical analysis section is very brief and should be expanded slightly. For example, it is stated that the homogeneity of variance was checked, but it is not clear what the variables are that are being considered. The statement about the Tukey test being used for comparing significance seems incorrect. I think that you mean that the Tukey test was used to assess whether means for different group were significantly different from each other. I suggest being a bit more explicit on what exactly the groups being compared are, e.g. what are the variables being compared.

We expanded the statistical analysis, including the missing information - lines 283-289.

-          In Table 2, you include some letters, a, b, A and B. It is not clear from the text what these letters mean. Could you please explain this clearly in the table caption?

Indeed, the lower- and upper-case letters needed to be more explicit in the table caption. Thank you for this observation. Now it is corrected.

-          Given that standard errors are included in Table 2, I think it would help to link to the equation(s) used in the calculations of these values.

Dear Review, including these equations in Table 2 does not add any relevant information to the reader since they cannot be extrapolated to another context than the specific conditions of this experiment. For this reason, we prefer not to present these equations in Table 2.

-          I think that adding “replicate” as a column would make Table 3 clearer.

Dear Reviewer, in Table 3, each line is already composed of three replicates (for SO trial – stated in line 127) and four replicates (for JA trial – stated in line 148). To draw attention to this information, “replicates” was included in the first line of 2.2 Section “Field and Lab measurements.”

-        It looks like the words “preview” and “previews” are being incorrectly used (e.g. lines 127 and 139). I think that these should be “previous”. If I am incorrect, this requires a short explanation of what is meant here.

Dear Reviewer, you are right. Thank you for your observation. There were three instances of this error. They are all corrected now! (Lines – 132, 138, and 146).

Reviewer 2 Report

Great work with results in this study provides significant insights into the impact of sugarcane cultivation on water resources and identifies opportunities for more sustainable agricultural practices. Providing a more accurate assessment of the grey water footprint (GWF) for sugarcane production by measuring the leaching runoff fractions of nitrate and herbicides in the Brazilian sugarcane environment; These findings demonstrate that management practices, such as split and incorporated nitrogen application, can significantly reduce the nitrate load on the environment. This suggests that agricultural practices can be modified to reduce the impact of sugarcane cultivation on water resources.

The manuscript is organized about materials and methods; results are clearly presented however I have the following comments.

1)     In the section of 2.3. Runoff and leaching assessment (from Line 185): Could you explain in more detail how the leaching and runoff assessments were conducted, including the Franke et al. Tier 1 approach and PestLCI approach? And 2.4. (Line 270) authors consider to provide more to clarify how Grey Water Footprint assessment performed, and what assumptions were made in the calculations?

2)     Could the authors have some explanation why is different between average α_Frank and α_PestLCI. How do the findings of this study compare to similar studies conducted in other sugarcane-growing regions? Are there any notable differences or similarities beside the factors of tropical environments (surface cover, geographic)

3)     How might the management practices examined in this study be implemented on a larger scale in the Brazilian sugarcane industry?

4)     What other pollutants or chemicals commonly used in sugarcane cultivation could be assessed using the methods employed in this study, and what are the potential implications for GWF estimations?

5)     could the authors give some comments how the results of this study might be relevant to other agricultural sectors, beyond sugarcane cultivation, that use similar chemical inputs and face similar water resource management challenges?

6)     Do you have any comments of the properties and application methods of different pesticides affect their leaching runoff fractions, and how can this information be used to refine the empirical models for estimating these fractions? (vary across different soil types, climate conditions, and management practices)

Author Response

Dear Reviewer, on behalf of my colleagues, the manuscript “Leaching runoff fraction for nitrate and herbicides on sugarcane fields: implications for grey water footprint” has been revised following the suggestions of the Reviewers, and it is now being resubmitted. We greatly appreciate your kind and careful revision, and we would like to give answers and clarifications to the questions and criticisms raised by you in the above general comments. All modifications were made intending to meet the Reviewer’s desires as much as possible.

Thank you, Fabio Scarpare

The manuscript is organized about materials and methods; results are clearly presented however I have the following comments.

1)     In the section of 2.3. Runoff and leaching assessment (from Line 185): Could you explain in more detail how the leaching and runoff assessments were conducted, including the Franke et al. Tier 1 approach and PestLCI approach? And 2.4. (Line 270) authors consider to provide more to clarify how Grey Water Footprint assessment performed, and what assumptions were made in the calculations?

Dear Reviewer, we do not understand your comment. Section 2.3 is split into subsections explaining how the leaching and runoff assessments were conducted for field trial data (section 2.3.1) for Franke et al. Tier 1 (section 2.3.2) and for PestLCI (section 2.3.3). Regarding section 2.4 (GWF), the main assumption refers to using distinct alphas, which is explicitly in the paragraph.

2)     Could the authors have some explanation why is different between average α_Frank and α_PestLCI. How do the findings of this study compare to similar studies conducted in other sugarcane-growing regions? Are there any notable differences or similarities beside the factors of tropical environments (surface cover, geographic)

The first paragraph of the discussion explains why there is a difference between average α_Frank and α_PestLCI.

“The PestLCI underestimated the losses of diuron and hexazinone in the two harvests evaluated. This model was originally developed to meet European production scenarios [26,54]; therefore, its applicability to other geographic areas can be compromised [55]. Some essential practices in tropical environments, such as vegetation cover, were not incorporated into the model and can explain this result [56]. The maintenance of straw on the soil surface plays a vital role in the dynamics of pesticides. The amounts of plant material maintained in the experiments (~8 kg ha^-1), which are not considered in the PestLCI, can substantially interfere with the losses of the herbicides on the surface and in-depth. In addition to the geographic factor, the difference of 0.2 m in depth assessed for herbicide loss between PestLCI (1 m) and our agricultural experiment (0.9 m) may have contributed to the lower leaching runoff fraction obtained by the model since herbicides tend to degrade in the soil profile.”

3)     How might the management practices examined in this study be implemented on a larger scale in the Brazilian sugarcane industry?

In Brazil's central-southern region, which corresponds to ~90% of the sugarcane cultivation area, the green cane harvest without a burn is highly adopted, as stated in the Introduction (first paragraph). For incorporated nitrogen fertilization application, while machinery has been developed for this purpose, adopting this technique is more restricted to sandy soils with moderate to high slopes. However, we were not able to estimate this proportion. Unfortunately, we could not find any statistics about that!

4)     What other pollutants or chemicals commonly used in sugarcane cultivation could be assessed using the methods employed in this study, and what are the potential implications for GWF estimations?

The seventh paragraph of the discussion already addresses this point.

“Besides fertilizers, various other agrochemical types usually applied in sugarcane areas, including natural substances with no toxicological characteristics and risks (e.g., vegetable extracts and pheromones used to control pests) to chemical molecules such as herbicides, insecticides, ripeners, and adhesive spreading agents.”  Next, we show some GWF results with a higher order of magnitude numbers for herbicides GWF addressed in the sugarcane production system in Brazil.

5)     could the authors give some comments how the results of this study might be relevant to other agricultural sectors, beyond sugarcane cultivation, that use similar chemical inputs and face similar water resource management challenges?

Although other crops explored in the same region as soybean, corn, and beans also rely on chemical inputs (fertilizer, herbicides, and other chemical molecules), some aspects of sugarcane cultivation/management are unique, which makes a comparison to other crop cultivation system difficult (if not inappropriate). 

For example, the sugarcane cultivation system without burn (called green cane management) adopted in Center-South Brazil includes the deposition of large amounts of plant litter on the soil after each harvest, ranging from 10 to 20 Mg of dry matter per ha. No other crops in Brazil have such mulch that influences the production process. In the introduction's first paragraph, we state sugarcane culture's main characteristics.

“However, sugarcane was the second-largest National pesticide market in 2018, with ~12% of all sales in the country [3]. Moreover, due to a large amount of biomass produced, substantial amounts of nitrogen (N) fertilizer in commercial sugarcane fields are used compared to many other crops (4-6). Furthermore, ~1000 weed species are estimated to inhabit the sugarcane agroecosystems, competing for nutrients, water, and light [7]. Therefore, various pesticides are applied simultaneously throughout the season [8-9].”

 

6)     Do you have any comments of the properties and application methods of different pesticides affect their leaching runoff fractions, and how can this information be used to refine the empirical models for estimating these fractions? (vary across different soil types, climate conditions, and management practices)

Pesticide properties that affect their leaching and runoff fractions include solubility, adsorption, and degradation rates. Highly soluble pesticides are more likely to leach into groundwater and surface water. At the same time, those with strong adsorption to soil particles are more likely to remain in the soil and not migrate to water sources. Pesticides with slower degradation rates can persist in the environment for extended periods, increasing the potential for leaching and runoff. These properties can be measured through laboratory experiments and used to parameterize models that predict the behavior of pesticides in the environment.

Empirical models for estimating pesticide leaching and runoff fractions can be refined by incorporating data on soil properties, climatic conditions, and management practices. Different soils have varying capacities to retain pesticides, so models that take soil type into account can improve predictions of pesticide movement. Climatic conditions such as rainfall and temperature can affect pesticide behavior by influencing soil moisture and degradation rates, so models incorporating climate data can improve predictions of pesticide transport. Management practices such as soil preparation, irrigation, and application rate can also influence the movement of pesticides, so models that consider these factors can improve predictions of the fate of pesticides in the environment.

Reviewer 3 Report

Comments and Suggestions for Authors

Review of Manuscript ID= sustainability-2344390

The paper is about Leaching runoff fraction for nitrate and herbicides on sugar-2 cane fields: implications for grey water footprint. The paper is well written but they can improve it if they can get time.

Based on the general comment, I do have some minor concerns that should be clarified.

 1.       Tier 1 approach is already outdated which is why they are using tier 2 and other improved versions of the model or tool; it is not clear to me why are you comparing your result with Tier 1.  What are these more specific datasets they used for Tier 2 and you can’t get them? As far as you are doing and measuring things at ground level, what is the difficulty to compare your result with at least Tier 2?

2.      Please increase the visibility of figure 3. Zoom it please, split A and B to have two figures rather than one so that the details in A and B can be easily identified.

3.      The two models such as SWAP and WOFOST are independent but in your expression, it seems one is embedded in another, please make it clear

4.      Why do the authors prefer to use such models like SWAP, WOFOST or PEST instead of others? There are more similar models to use like DSSAT, Aqua crop, and others, and why the author's SWAP/WOFOST

5.      How accurate your model result is?

6.      In you your conclusion or method, please write a detailed explanation as to how your hypotheses were confirmed.

There is not that much problem with the quality of the English language, it is well-written.  

Author Response

Dear Reviewer, on behalf of my colleagues, the manuscript “Leaching runoff fraction for nitrate and herbicides on sugarcane fields: implications for grey water footprint” has been revised following the suggestions of the Reviewers, and it is now being resubmitted. We greatly appreciate your kind and careful revision, and we would like to give answers and clarifications to the questions and criticisms raised by you in the above general comments. All modifications were made intending to meet the Reviewer’s desires as much as possible.

Thank you, Fabio Scarpare

Based on the general comment, I do have some minor concerns that should be clarified.

1. Tier 1 approach is already outdated which is why they are using tier 2 and other improved versions of the model or tool; it is not clear to me why are you comparing your result with Tier 1.  What are these more specific datasets they used for Tier 2 and you can’t get them? As far as you are doing and measuring things at ground level, what is the difficulty to compare your result with at least Tier 2?

Dear Reviewer, the comparison of Tier 1 against Tier 2 or Tier 3 is out of the scope of this study. However, this study addresses Tier 1 because it is the main assumption used in GWF studies since most of its applications are made to provide insights at large-scale areas (e.g., watershed, counties, and even continental spatial scale). Indeed, this aspect needed to be clarified in the text. That is why some changes have been made in the introduction (third paragraph) to give the reader an idea of why we are addressing especially Tier 1 and no other Tiers.

 

2. Please increase the visibility of figure 3. Zoom it please, split A and B to have two figures rather than one so that the details in A and B can be easily identified.

We increased the figure visibility but did not split figures A and B because we wanted to be consistent with Figure 5.

3. The two models such as SWAP and WOFOST are independent but in your expression, it seems one is embedded in another, please make it clear

That is correct; SWAP and WOFOST are two independent models. However, there is a SWAP/WOFOST embedded version (used in this study). SWAP contains three crop growth routines: a simple module, a detailed module for all kinds of crops (WOFOST, WOrld FOod STudies), and a detailed module for grass (re)growth. However, to clarify this aspect, we include “embedded.”

 

“In SWAP, an embedded generic mechanistic crop growth model WOrd FOod STudies - WOFOST [37-38] was used based on the calibration by Scarpare [39] for sugarcane above-ground biomass production under rainfed conditions in São Paulo State.” Line 228.

 

4. Why do the authors prefer to use such models like SWAP, WOFOST or PEST instead of others? There are more similar models to use like DSSAT, Aqua crop, and others, and why the author's SWAP/WOFOST

All process-based simulation crop models have their strengths and weaknesses. We used the SWAP/WOFOST model for water balance estimation interacting with plant growth because we already had this model calibrated for rainfed sugarcane growth in this region (Scarpare et al., 2011). 

The Parameter Estimation algorithm PEST was used to estimate the saturated hydraulic conductivity inversely (using the soil moisture and other hydraulic parameters like teta saturation, teta residual, alfa, and n from van Genuchten soil water retention model). We used this approach because we no longer had the soil samples for Ksat laboratory analysis. 

 

5. How accurate your model result is?

The leaching and Runoff components errors were around 5%.

6. In you your conclusion or method, please write a detailed explanation as to how your hypotheses were confirmed.

We changed the text in the first paragraph in the conclusion section to address this point.

Reviewer 4 Report

Dear authors.

The work is very interesting but some issues should be addressed.

In key words you should add also nitrates and GWF.

Row 89: Also mention here the PestLCI model which you use as you have mentioned it in row 188.

Row 94: Rewritten the sentence, you cannot use "some chemicals"

Rows 114-116: Put reference for data mentioned in this sentence.

Also, put reference for the data mentioned in Table 1.

Please also present in Figure 2 some cross-section overview.

Table 2 should be mentioned before  in text, not after the table.

Section 3.2. Rows 323-325:  Please mention that methods are explained in section 2.3...

Section 5-Conclusions: Rephrase the first sentence... i.e. "Our experiments showed that..."

Please put in Conclusions also what do you suggest? Building a new model  or reconstruction of some existing models to be used in this case...?

Some minor spelling and grammatical errors can be found in the manuscript, so I suggest that you read and check the manuscript.

Author Response

Dear Reviewer, on behalf of my colleagues, the manuscript “Leaching runoff fraction for nitrate and herbicides on sugarcane fields: implications for grey water footprint” has been revised following the suggestions of the Reviewers, and it is now being resubmitted. We greatly appreciate your kind and careful revision, and we would like to give answers and clarifications to the questions and criticisms raised by you in the above general comments. All modifications were made intending to meet the Reviewer’s desires as much as possible.

Thank you, Fabio Scarpare

In key words you should add also nitrates and GWF.

Usually, Journals do not encourage repeating words from the title to the keywords. That is why “nitrate,” “grey water footprint,” and “herbicide” words were not included in. However, the Sustainability Journal allows the inclusion of up to 10 pertinent keywords specific to the article yet reasonably common within the subject discipline. For this reason, ending up we included them.

Row 89: Also mention here the PestLCI model which you use as you have mentioned it in row 188.

Done

Row 94: Rewritten the sentence, you cannot use "some chemicals"

We rewrite this sentence, excluding the “some chemicals” expression.

Rows 114-116: Put reference for data mentioned in this sentence.

Old and well-established studies from the early last century, e.g., Köppen climate classification (1936), Thornthwaite climate classification (1929), and reference crop evapotranspiration from Penman-Monteith (1948), are widely recognized by the scientific community and, therefore, are usually not cited. Especially because Scientific Journals usually ask to avoid ancient citations in the text. For this reason, we prefer not to cite this study.

Also, put reference for the data mentioned in Table 1.

Dear Reviewer, we apologize, but we did not understand your statement. What reference should we include in Table 1 if all information comes from our trials?

Please also present in Figure 2 some cross-section overview.

We included a photo with a cross-section overview.

Table 2 should be mentioned before in text, not after the table.

Done

Section 3.2. Rows 323-325:  Please mention that methods are explained in section 2.3...

Done

Section 5-Conclusions: Rephrase the first sentence... i.e. "Our experiments showed that..."

Done

Please put in Conclusions also what do you suggest? Building a new model  or reconstruction of some existing models to be used in this case...?

This statement is already in the conclusions in the first paragraph.

“For nitrate GWF estimations, under the absence of more detailed information from the study area, we recommend adopting 5% instead 10% under Brazilian production conditions. The same type of herbicide recommendation cannot be made from the presented results since only two molecules were targets in our evaluation. However, our results suggest that the herbicide global leaching runoff average is also out of order of magnitude.”