The Water Footprint of Biodiesel Produced from Sunflower in South Africa

Round 1

Reviewer 1 Report

The major and minor comments are provided in the following.

(1) The abstract should be rewritten. The statement "At the level of processing...." has 490 characters! It is not comprehensible and requires English proofreading.

(2) Why did the authors not include the grey water footprint in their analysis? In sunflower production, fertilisers are also used, which worsen the quality of the runoff. Thus they affect the grey water footprint.

(3) The authors claimed that " ... sunflower irrigation data were taken from the Orange -Riet irrigation scheme database and relevant literature" . What does "relevant literature" mean? References of some of this are needed in the text.

(4) page 3, line 143: What does „lgp” mean?

(5) For which years is the analysis conducted? The authors use months, without specifying years.

(6) Tabela 1: There is no caption

(7) Tables 2 and 3 mostly repeat values from the text. Please consider whether it is worth putting the same values in two places in the article ?

(8) The entire article needs to be checked for stylistic errors.

(9) The authors analysed two options: rain-fed and irrigation production system. Was naturally occurring rainfall taken into account in option 2 (with irrigation)? This question needs more discussion in the article.

Author Response

Thank you for the valuable comments and suggestions.

1. The abstract was completely refined and certain parts were rewritten for better understanding.
2. We do understand that the grey water footprint will also contribute to the total water footprint. The lack of data on fertiliser application, however, meant that were were not in a position to calculate grey water footprint. We did include this as a recommendation for future research.
3. "and relevant literature" was removed from the sentence.
4. lgp is the length of growing period. LGP was defined in the text in brackets where we referred to lgp.
5. We added the year in the discussion of the data in Section 2.1
6. The table was out of place in the text, but we moved it to the caption.
7. We see the concern of the reviewer. We still think that the table helps the reader to quickly see the values. If it is not too serious concern we would prefer to keep the tables.
8. We had another look at the article. Any specific concerns/suggestions will be greatly appreciated.
9. It was explained in the introduction section that the irrigation system provides supplementary irrigation to meet rainfall shortfall of the crop water requirement.

Reviewer 2 Report

This paper gives a detailed analysis of water footprints from sunflower produced in South Africa. As such, it is relevant for the scientific community. It also falls within the scope of the journal WATER. However, before it can be published some major changes are needed. What I miss in the paper is the context and perspective. Why is it needed to do such a detailed analysis of the water footprint (WF) of this specific crop? Just to arrive at a better number? Also a comparison of the outcome with results of other studies is missing. E.g. with the database of Mekonnen and Hoekstra available on the website of the water footprint network. They also made calculations of sunflower per province in South Africa. Another issue is that WFs heavily rely on yields, because they are expressed per unit of yield. Yield levels of sunflower are not discussed. See for example the paper: Yield and seasonal water productivity of sunflower as affected by tillage and cropping systems under dryland conditions in the Limpopo Province of South Africa. Agricultural Water Management, Volume 98, Issue 10, August 2011, Pages 1641-1648.

Another issue might be the competition of food versus fuel. Why would South Africa use its valuable and scarce water on a biodiesel? Sunflower is also an ingredient for oils. This could be addressed in a discussion section.

An issue is also the sustainability assessment that leaves out the other crops irrigated in the basin. If they were also included, the study outcomes might be completely different not finding that sunflower production is sustainable.

Additional issues are:

There is a difference between water use and water consumption. Consumption means that the water is lost, use does not necessarily mean that the water is lost. It can also be used and flow back to where it came from. I suggest to adapt this in the paper (e.g. line 33).

Line 40, 41 that introduces the grey WF is not correct. Pollutants are assimilated to accepted water quality levels.

Line 44. There are huge differences between the WF and LCA concept. Only the quantification step is basically the same.

Section 2, materials and methods

The paper would improve a lot if first a description would be given of the system, e.g. the location, characteristics of sunflower production, etc. For example in a separate section. It is difficult for a reader to first encounter the data. First give what you want to calculate, next provide your data sources.

Line 99 and 100. Here the two locations are introduced. They are representative for rainfed and irrigated agriculture for sunflower. It could very well be that also other circumstances differ leading to other yield levels and thus other WFs. This is also reflected in the growing periods that differ between the locations. How can we interpret the different WFs then?

Line 107. What are the secondary data needed for the sustainability assessment?

2.2. Water footprint estimation

I assume that in the blue WF calculations it is assumed that crop water requirements are met? This might not be the case.

Figure 1. Here the calculation steps and input are given. I would put this before the equations. Next, I would put calculations as such, now no difference can be seen between input data and calculations. I am also wondering about Potential evapotranspiration. There is no incoming arrow.

Relate the sentences in line 152-160 to the calculations in lines 121-146 and the Figure.

The title of subsection 2.2.2 is missing.

Equation 7. Units should be m³/GJ not m³/ ton.

Sustainability assessment

It is needed to explain how water availability is calculated. I assume using the 80% EFR rule?

Line 204-205. I think it is not valid to compare the total blue WF of sunflower to water availability in this way. Probably the blue WF comes on top of the WFs of the other crops in the area. It is therefore needed to make an analysis of the total agricultural blue WF and compare this to available water. Sunflower is probably only one of the crops and water consumption is far larger than only the WF of the sunflower.

3. Results

Lines 230-232 can go out.

3.1. This results section starts with describing the method and system. It can go there where it belongs.

Figure 2. This figure shows differences between irrigated and rainfed production. Differences might be caused by different agricultural management, e.g. growth stages. The explanation of the differences shown in figure 2 can be shortened, and more detail is needed on management causing the differences (maybe also climate?).

Line 258. The fact that irrigated sunflower has a longer growing season is not a result. It could explain why the WF is larger. This aspect needs some attention.

The description of what we see in Table 1 is too long. More explanation is needed instead. Last colums of the table. I think the authors mean WF green+blue. Next, they should give the abbreviations here.

I miss a discussion section. At least a comparison with results of other studies are needed.

Figure 2. I think this figure should include the agricultural blue WF and not only the one of sunflower. Based on this figure alone, one can never conclude that sunflower production in the Orange river basin is sustainable (line 373).

Author Response

Thank you for the valuable comments and suggestions.

1. A discussion was included to relate the results with that reported in the literature.
2. Yield is a very important determinant of the water footprint yes. 
3. The sustainability assessment was not done by comparing the WFblue of sunflower with water availability. It was actually the total WFblue of all uses (agriculture, domestic, industry) that was compared with water availability. The sentence was refined to remove the confusion. 
4. "Used" was replaced with "consumed".
5.  Corrected the explanation of grey water footprint. 
6. We removed the part of the sentence stating that LCA relates to WFA.
7. A description of the system was added as requested. Section 2 was also restructured to start with the study area, the system and methods, before the data requirements are addressed. 
8. We agree that rainfed and irrigated yields can be expected to differ considerably. Difference in yields can also be caused by various management practices and other factors. 
9. The secondary data needed for the sustainability assessment was explained in the text. 
10. The sub-heading was added (it is now 2.3.2 in the revised version though)
11. the unit was corrected in Equation 7
12. An explanation of the calculation of the water availability is included to show that the 80% EFR was used. 
13. The description of the sustainability assessment was improved to remove the confusion that the sustainability is calculated by comparing the WF of sunflower with water availability. It is actually the total blue water footprint (agriculture plus domestic plus industrial).
14. The description of the system was moved to where it belong.
15. Figure 2 was replaced by a table.
16. The argument that the graph showing the sustainability assessment should not only include the water footprint of sunflower is 100% correct. The WFblue in Figure 3 does include all of the blue water that was consumed in the basin (agriculture, industry, and domestic). This confusion was removed by including the information in the discussion of the methods. 

Reviewer 3 Report

The manuscript ‘The water footprint of biodiesel produced from sunflower in 2 South Africa’. However, there are points that need to be clarified for the paper to be suitable for publication.

 

 

1. Material and methods. SAPWAT model. A reference to this model is required as well as a deeper description of the model and the main advantages/complementarities with CROPWAT

Lines 108-110 ‘The SAPWAT4 and 108 CROWAT 8.0 models were used to reinforce the data obtained for irrigated and rain-fed 109 sunflower respectively.’ The role of both models applied is not clearly explained… Figure 1 should be further explored and analysed. The irrigation requirement is linked to the ‘total blue water use’ and also effective irrigation that comes from SWAP is also connected to ‘Total blue water sue’. The reader will not understand the difference between irrigation requirement, and effective irrigation without further detailed information. At a first sight it seems that the authors summed these two components.

 

2. A description of the biodiesel system should be provided. During biodiesel production there is the production of crude glycerol (by-product), which organic fraction contains fatty acid methyl esters (FAMEs), free fatty acids (FFAs), and glycerides. The fats considered in eqs. 7 and 8 are related to which fats, exactly? These equations represent the allocation between the water consumed for biodiesel production and then the by-products… but this procedure is not clear enough…

 

3. Sustainability assessment. Line 221-222 ‘Further to the blue water scarcity assessment, the water stress was also assessed using the water stress coefficient, which is denoted by Ks as proposed by the FAO [32].’ The reference is not correct… the original reference is this one:

 ‘Allen, R.G.; Pereira, L.S.; Raes, D.; Smith, M. FAO Irrigation and Drainage Paper No. 56—Crop Evapotranspiration; FAO: Rome, Italy, 1998.’ Furthermore, this reference is outdated…. There is more recent work focused on blue water scarcity. Hoekstra AY, Mekonnen MM, Chapagain AK, Mathews RE, Richter BD (2012) Global Monthly Water Scarcity: Blue Water Footprints versus Blue Water Availability. PLoS ONE 7(2): e32688. doi:10.1371/journal.pone.0032688

 

4. Results. Lines 234-235: ‘The sunflower water footprint calculations in this study were based on two different production systems for sunflower growth, i.e., irrigation and rainfall water.’ Two scenarios are analysed. But sunflower can not be produced using rainwater and be complemented with irrigation for instance increase productivity. What is the conventional situation of sunflower cultivation, it is only rainfeed?

 

4. Figure 2. Please, have in mind that this is a scientific paper. Whether the purpose is to report all the values, the authors should present them in a table. Presenting a graph and indicating in each bar the value, gives the idea this a more a report than a paper. And the graph itself could be a better graphical aspect.

 

Table 1. ‘This is a table. Tables should be placed in the main text near to the first time they are cited.’ This should be an error. This couldn’t be the legend of the table.

 

Table 3. Report the WF results for m3/ton or me/GJ… Both units in the same table has no sense. Why the amount of water (m3/L) is the same for both rain feed and irrigated sunflower cultivation? Furthermore, this table seems unnecessary. These values are repeated in Table 4. The structure of section 3 should be rearranged. First the authors should present the WF results of the entire system, Table 4, and then explain the contributions at the processing and farm level… Table 4 shows that the WF hotspot is the farm level, so a deeper analysis on this stage is required…

‘Source: Author’s calculations.’ This is not necessary.

Table 2 with biodiesel characteristics should be moved for section 2, to the description of the system…

 

Figure 2 should be graphically improved. And other water scarcity index instead of Ks should be used.

Author Response

Thank you for the valuable comments and suggestions.

1. Reference was added.
2. A discussion of the biodiesel system was inserted.
3. The reference was changed accordingly.
4. There are sunflowers that are produced 100% rainfed, then also sunflower that is produced with supplementary irrigation. This explanation was inserted in the text. 
5. Figure was replaced with table.
6. Table 1 was moved to the correct place in the text.
7. The units have been corrected.
8. The table has been split not to have m3/ton and m3/GJ in the same table.
9. The sustainability assessment was not done based on Ks values, but rather the water scarcity index as reported by Hoekstra et al. on blue water scarcity in major river basins..

Round 2

Reviewer 2 Report

The authors addressed many of the issues of the first review round. This improved the paper a lot. However, I still have concerns about especially the sustainability assessment. The authors are from South Africa and thus familiar with the situation in the country. They have access to specific information and data sources. This might enable them to perform a more accurate sustainability assessment than presented in here. This study only shows the difference between rainfed and irrigated sunflower production on a farm scale level, indicating that irrigation results in smaller total green and blue water footprints per unit of oil. However, the impact of blue water on the environment is larger than the impact of the green water footprint.

They used blue water footprint data on the Orange river basin from Hoekstra et al. 2012. That database however covers the whole world, and it can be assumed that it is not so precise and only provides a rough estimate per basin. For the sustainability assessment I would suggest to do a water balance calculation indicating the water flows in the basin and their sizes. Next, you might do an assessment of the additional water footprints of the sunflower production. In your conclusions it is not possible to state that sunflower production is sustainable. Also do not give references in your conclusions [47] but use them earlier to do your sustainability assessment.

Additional remarks:

2.1 Study location

This section is too small. Please describe this study location in terms of location in the basin, precipitation, water availability, crops grown there etc. This is relevant for the study. The description of the biodiesel cycle can then be shortened. We only need to know that is needs some water.

Check all the references.

E.g these are 3 corrected references:

Gerbens-Leenes, P.W. and Hoekstra, A.Y., 2013. Water footprint quantification of energy at a global level. In: Dallemand, J.F. and Gerbens-Leenes, P.W. (eds.) (2013) Bioenergy and water, Joint Research Centre, European Union, Ispra, Italy.

Gerbens-Leenes, W., Hoekstra, A.Y and Van der Meer, T.H., 2009. The water footprint of bioenergy. Proceedings of the National Academy of Science (PNAS), 106 (25), 10219-10223.

Take the most recent water footprint manual from 2011.

Author Response

Dear Reviewer,

Thank you so much for the valuable comments that helped us to improve the manuscript. We have addressed the remaining concerns/comments as follows:

1. A short discussion was added after the discussion of the comparison of the WF of rainfed and irrigated sunflower (Table 4) to put the results in context. 
2. The discussion of the sustainability assessment was refined. The emphasis was less on sustainability assessment, and more on a blue water scarcity assessment. It is recognised that sustainability assessment is much more complex than merely considering water use. For the purpose of this paper, the assessment of the blue water scarcity during the growing period of irrigated sunflower is considered sufficient to tell the story we want to tell with this paper. We also agree that local data may be possibly be more accurate for the purpose of the sustainability assessment. For this paper we would rather stick to what we have done, and we have included a recommendation for future research to conduct a comprehensive sustainability assessment using local data rather than data from global studies. 
3. The conclusions were also refined after we have implemented the necessary corrections. 
4. The discussion of the Study location was improved, and a map was added to show the location of the study area within South Africa. 
5. References were corrected.

Reviewer 3 Report

The paper was further improved. There are only a few typos that should be corrected, such as wwublue above equation 6; the units presentations, for instance in equation 7, the units are not within []. The same comment for the reminign equations. 

Author Response

Dear Reviewer,

Thank you for these and the previous comments. We do appreciate your contribution to improve the manuscript. We have addressed the remaining comments as follows:

1. Typos were corrected 
   1. wwu was changed to wu
2. The units of Equations 7-12 were put within [] as suggested.

Thank you for your time.

Round 3

Reviewer 2 Report

The authors addressed all suggestions for improvement.