Elsevier Editorial System(tm) for Energy Policy Manuscript Draft. Keywords: Soybean biodiesel; Sustainability Assessment; Tallow biodiesel.

Transcription

1 Elsevier Editorial System(tm) for Energy Policy Manuscript Draft Manuscript Number: JEPO-D--005R1 Title: Environmental sustainability of biodiesel in Brazil Article Type: Full Length Article Keywords: Soybean biodiesel; Sustainability Assessment; Tallow biodiesel. Corresponding Author: Prof. Suani T Coelho, Dr Corresponding Author's Institution: University of São Paulo First Author: Erica G Castanheira, MSc Order of Authors: Erica G Castanheira, MSc; Renata Grisoli, MSc; Fausto Freire, Dr; Vanessa Pecora; Suani T Coelho, Dr Abstract: Biodiesel production in Brazil has grown % since 07. It is an emergent bioenergy for which it is important to guarantee environmental sustainability. The objective of this article is to characterize the biodiesel production chain in Brazil, to identify potential environmental impacts and to analyze key drivers and barriers for biodiesel environmental sustainability. This article explores these aspects and focuses on the increasing demand for the main feedstocks for biodiesel production in Brazil: soybean oil and beef tallow. The impacts of land use and land use change on greenhouse gas emissions, biodiversity and water, as well as the energy balance were found to be critical for the environmental sustainability assessment and development of biodiesel chains. Increasing agriculture yields, diversifying feedstocks and adopting ethyl transesterification can contribute to minimize environmental impacts. It was also found that environmental impacts and conflicts between food and bioenergy can be mitigated by appropriate policies aiming at an integrated optimization of food and bioenergy production and through agro-economic-ecological zoning, allowing adequate use of land for each purpose. Despite the limitation and weakness of some sustainability tools and initiatives, certification can play an important role for the sustainability of the emerging biodiesel production in Brazil. Response to Reviewers: Reviewer #1 1.1.This is an interesting paper with useful information on soybean and biodiesel production (over 0 references!). Biodiesel in Brazil is mainly produced from soybeans (77-2% of the total production). The total area used for soybean production is approximately million hectares. Environmental impacts, land use change and biodiversity are significant problems caused by soybean production which is moving the agricultural frontier into the Amazonia but its not biofuel production that is causing it. Biodiesel production uses a small fraction the soybean produced (less than 10%). Therefore to place biodiesel in the proper context is very important and the paper does not do that clearly. For example: Section.1 addresses effects on food production and security for biofuels in general and not biodiesel specially. Section.2 again deals with the general problem of soybean expansion and not biodiesel. This is a major weakness of the impacts paper. The biofuel production should be clearly distinguished from soybean production for other uses. Response: We agree that it is important to place biodiesel in the proper context and acknowledge the text should be improved. In this context, subsection 2.1 was updated and we decided to remove

2 subsection.1 from the updated manuscript. Regarding subsection.2, we acknowledge that the general problem of soybean expansion is not only caused by biodiesel production, which we believe is now better described in the updated subsection 2.1. However, it should be noted that since 05 the increase in soybean oil national consumption was very much related with biodiesel increasing production (see new Figure c). Action: Subsection.1 was removed from the updated manuscript. Subsection 2.1 and Figure were updated with more information about the quantity of soybean and soybean oil used in Brazil for biodiesel production and food purposes. The new text and figure read: Fig. presents production, import, export and national consumption (NC) of soybean grain (Fig. a), meal (b) and oil (c) in Brazil since. The NC quantities used for biodiesel production and food purposes are also shown in fig. c. Soybean grain production increased about 7% since 01, due to both an increase in exportation and national demand. In 0-0, exported soybean represented in average 27% of the Brazilian production while in 10, it represented over %. Brazil is selfsufficient in soybean grains, meal and oil, supplying the domestic market and exporting the surplus to international markets (MAPA, ). Domestic consumption of soybean meal and oil has been growing significantly especially in the last five years. Over % of the soybean grains consumed in the domestic market are processed to co-produce soybean oil and meal (% of meal and 2% of the oil were exported in 10). Only -7% was consumed as food or feed. Oil exports have been reducing since 05 due to the growing biodiesel production (commercialized in Brazil since 05). In 10, biodiesel production required about % of the total domestic consumption of soybean oil. 1.2.On dealing specifically with GHG and energy balance in biodiesel production a Table giving the numbers from different authors would help the reader to understand the issues. Response: We thank for the suggestion. Action: A table giving the numbers of GHG and energy balance for tallow biodiesel (Table 1) was included in the subsection.2 of the revised manuscript. The GHG and energy balance results for soybean biodiesel are explained in the text. 1..Since transesterification accounts for 7% of the energy needed in the production of biodiesel a stronger discussion of the ethanol "versus" methanol routes would be very helpful. This is attempted in section. but few numbers are given comparing them. Response: Thank you for this suggestion. However, it should be noted that the 7% mentioned for transesterification are for tallow biodiesel and do not include beef production. Action: Additional information and a stronger discussion regarding ethanol "versus" methanol routes was included in the subsection.2: "Despite the difference of using a renewable (ethanol) and fossil (methanol) source, is important to compare the environmental impacts of two routes. A study performed by CENBIO () compared the production of biodiesel using methyl and ethyl route considering life-cycle assessment approach. The difference between the results of two routes, for the various environmental impact categories (e.g. climate change, acidification), is less than 10%. The results showed that there were no significant differences in the environmental impacts of these two technological routes, despite the fossil origin of the methanol imported in Brazil (Coelho et al., )." 1.. Finally there is practically nothing on the economics of biodiesel production which is heavily subsidized in Brazil. Such subsidies were introduced in the initial stages of the program which started as a social program using locally available agricultural products for small farmers in the North East

3 region of Brazil. This program failed but the subsidies remained when soybean (which is produced mainly in the South East) became the main feedstock. Presently there seems to be little social and economic justification for subsidizing such program. Response: We thank this comment but the main objective of the paper is to discuss the environmental sustainability of biodiesel; however, we agree that the objective should be clarified and the importance of the economic aspects should be emphasized. Action: In order to clarify the objective, the following sentence was included in the end of section 1: "This paper aims to address only environmental impacts of biodiesel from soybean and beef tallow in Brazil; however, it must be taken into account that economic aspects of biodiesel are the main reason for choosing these two raw materials for biodiesel in Brazil. Other feedstocks such as palm and castor oil present high opportunity costs (NAE, 05) since they are used in the food industry and others (more interesting in economic terms). Soybean oil is the main raw material because it is the by-product of the production of soybean meal for animal feed (to be exported together with the grains). In a similar way, beef tallow is the by-product of meat production to be exported. However, it must be observed that, even in the case of soybean and tallow biodiesel, most biodiesel is commercialized in auctions by Petrobras (the Brazilian Oil Company), which pays prices much higher that the final price of biodiesel blended with diesel oil in the country (diesel oil prices are controlled by Federal Government as a tool against inflation rates)." Reviewer #.1. The paper discusses those environmental impacts originating within the biodiesel production chain, using soybean and beef tallow as example (Brazil's most relevant biodiesel feedstock in terms of blending). It is a review - not an original article. As usual for reviews, it is very detailed and possesses quite a bit of interesting information, especially numbers. I think from the reader's perspective it would be preferable to include more tables in order to synthesize the information better. Response: We thank this suggestion, but we attempted to be sparing in the use of tables (as recommended in the guide for authors). Action: We decide to include one table (Table 1 in the subsection.2) and also the appendix includes a table (Table A1)..2. First, and most importantly, the paper discusses "environmental sustainability" of both beef tallow and soybean oil, but does not offer a definition of what constitutes "environmental sustainability". This is particularly important because both feedstock have a long history on deforestation (direct or indirect), problems with GMO, pesticide-use or negative impacts due to monoculture. Speaking of environmental sustainability in this context raises a positive connotation that is very difficult to defend where a clear benchmark is missing, e.g. what qualitatively/quantitatively constitutes an "environmentally sustainable biofuel". In fact, both a theoretical framework is missing and an argument ("are beef tallow and soybean sustainable" are missing, and this leaves me, as a reader, wondering on the original contribution of the paper. Response: We agree and acknowledge that it is important to define "environmental sustainability and give a theoretical framework about the concept. This was missing in the submitted version, but is now included in section. The original contribution of the paper is to characterize the biodiesel production chain in Brazil, identifying potential environmental impacts and analyzing key drivers and barriers for the biodiesel environmental sustainability development; however, discussing if soybean or beef tallow biodiesel constitute an "environmentally sustainable biofuel" is complex and difficult to define based on a single argument. Action: The following paragraphs were included in the beginning of section :

4 Environmental sustainability is a difficult concept to explain, but it can be defined as the maintenance of natural capital (Goodland, ). The two fundamental environmental services (the source and the sink functions) must be maintained unimpaired during the period over which sustainability is required. In this context, the environmental sustainability of biodiesel is critically related with life cycle impacts associated with air (GHG emissions and others), water and soil, energy balance, biodiversity (FAO, b). Another important issue involves developing strategies to ensure that as the production of biofuels increases, adequate supplies of other needed agricultural and forest-based goods are produced (FAO, b). However, to quantify the environmental sustainability of biodiesel is complex. An increasing number of countries have established initiatives to define sustainability criteria for biofuels. For instance, the European Union Directive on the Promotion of Renewable Energy Sources (RED) defined that for biofuels be counted as renewable energy, a minimum GHG saving of % is required by (comparing with fossil fuel). The directive also stipulates no-go zones for feedstock production (e.g. areas where land is deemed to be of high biodiversity value or high carbon stock, wetlands, peatland) (FAO, b). Also, there is an important initiative from Global Bioenergy Partnership (FAO/GBEP, ) defining sustainability indicators for bioenergy (FAO/GBEP, ), which are now starting to be evaluated for different countries. ".. In many cases the authors also refer to "potentials" (e.g. ethyl transesterification, which is not at economical scale despite the issue having been discussed for several years) or considerable uncertainties in key values, but are seldom able to provide hard evidence on environmental sustainability (e.g. the case of the water footprint, which might have important consequences for medium-term sustainability). Response: We aknowledge that Ethyl transesterification is not produced at large economical scale, but can be feasible in economic terms as shown by the production of company FERTIBOM. In the case of water footprint, the idea was to show how this issue has been considered in Brazil and what level of information we can indicated in a discussion related to this issue. Action: The following sentence was included: "Most industries in use methylic transesterification for biodiesel production; however, there are two companies used (anhydrous) bioethanol for biodiesel production (ANP, ). For instance, FERTIBOM (in Sao Paulo State) produces biodiesel from different raw materials (vegetable oils and also residual oils) and consider it is economically feasible since they participate in the biodiesel auctions (CENBIO, )." In the case of water footprint (WF), we include the following text to expand the idea of Gerbens-Leenes et al. (0), insert the currently situation of local studies and include a recent result for a Brazilian region: "Gerbens-Leenes et al. (0) calculated a WF for soybean biodiesel of liters per MJ (7 liters per liter of biodiesel), according to the authors in terms of the bioenergy sources considered in this study, biodiesel has the bigger WF, followed by ethanol and bioelectricity. WF studies of biodiesel in Brazil are still rare, few of them are indeed representative and this issue still needs to be deepened in the country. A preliminary study for showed that soybean biodiesel water footprint varies from 0 to 0 liters per MJ (Seabra et al., ). Another reference in southern Brazil for production of biodiesel indicates that the major contribution to degradative water use is the agricultural phase (.%) while for consumptive use the largest contribution is the oil extraction process (7.%) (Muller, 12).".. This lack in depth, secondly, also makes the authors recommendations less valuable. The authors write that "environmental impacts and conflicts between food could be mitigated by appropriate policies aiming at an integrated optimization of food and bioenergy production and through agroeconomic-ecological zoning, allowing adequate use of land for each purpose". While they correctly state that initiatives like the Amazon Moratorium are on the way to mitigate some of the impacts, these do by far not cover every environmental impact covered in the paper. Rather they focus on

5 reducing/halting indirect or direct deforestation, which strangely enough is very little discussed in the paper (the authors only reference possible indirect deforestation in the Amazon once, but do not provide information on the (existing) evidence). The role of policies in resolving conflicts in the paper is treated only superficially in the paper; it is clear that the policy framework can have beneficial consequences for biodiesel sustainability, but judging from history this process seems to be rather ineffective. Together this makes the reader feel that the paper is rather unbalanced or uncritical with regards to environmental risks of large-scale biodiesel expansion in Brazil. Response: We thank this remark and we agree that the role of policies should be further discussed. A critical point a view regarding environmental risks of biodiesel expansion in Brazil is also missing. Action: The last paragraph of subsection. was revised and a new paragraph was introduced: "According to CEO (12), despite the national and international sustainability tools and initiatives, there has been some opposition from social movements and environmental organizations both in Europe and in biofuel producing countries, which has weakened these tools and initiatives legitimacy. In fact, there are also important economic interests (market barriers), but it must be noticed that these tools are active in the country. The soybean certification scheme makes sense in the context of the certification of the several products that are produced from soybean; however, the certification schemes only exist for biodiesel and not for oil or meal.this issue is very important also because there should be coherence betweem the methods to attribute impacts bettweens the various ybean based co-products. The agro-ecological zoning is an important policy tool and has taken into account environmental, economic and social aspects as an answer to challenges in sustainable expansion of bioenergy crops. An interesting initiative was launched by the Federal Government to guarantee the sustainable expansion of sugarcane and oil palm in Brazil (two agro-ecological zonings): for sugar cane in 0 (EMBRAPA, 0) and for oil palm in 10 (EMBRAPA, 10). In this process, maps were produced showing soils, topography, climate and rainfall. This regulation enables a guidance to credit policies and use for public banks as a condition for production financing. The zoning identified areas where sugar cane crop expansion could take place. It forbids sugar cane cultivation in 2.5% of the national territory. It has identified million hectares (EMBRAPA, 0) that comply with environmental and productivity requirements, mainly from the intensification of cattle ranching, which is currently very inefficient (less than 1 head ha-1) (Strapasson et al., 12). Such initiatives could be also adopted for soybean expansion, contributing to the sustainability of soybean biodiesel. However here there is the issue of the intensification of cattle ranch as an important goal to be achieved both for these zonings and for the sustainability of cattle ranch in Brazil (and indirectly to the sustainability of biodiesel from beef tallow).".5. Third and finally, sustainability from the reviewer's perspective it is difficult to discuss sustainability excluding social impacts and economics. In fact, the authors include a discussion of food security, which is rather a social issue. So, one might also ask, what are the broader social consequences of soybean and tallow production? There are many reports available on this issue, including from Brazilian organizations/universities, and rather than being quiet on the topic, the authors should take the time to include - at least in the discussion and conclusions a broader perspective. Also economics is an important issue, as already mentioned for the case of ethyl transesterification which might be more sustainable, but doesn't seem to be economic. Response: We thank for the comments and we agree that is difficult to discuss sustainability excluding social and economic aspects. However, as presented in the introduction, the main objective of this article is to discuss the environmental sustainability of soybean and beef-tallow biodiesel. Nevertheless, some economic aspects were addressed throughout the article. Action: To clarify the economic sustainability of ethyl transesterification, the follow sentence was included in subsection.2 of the revised manuscript:

6 "Most industries in use methylic transesterification for biodiesel production; however, there are two companies used (anhydrous) bioethanol for biodiesel production (ANP, ). For instance, FERTIBOM (in Sao Paulo State) produces biodiesel from different raw materials (vegetable oils and also residual oils) and consider it is economically feasible since they participate in the biodiesel auctions (CENBIO, ).".. I would suggest a careful revision of the English which is sometimes hard to read, as well as a careful review of the (in-text) citations. For example, Schaffel and La Rovere (10) and not Schaffer and La Rovere (10). Response/Action: We agree with the comment. A careful revision of the English and citations were made..7. The research highlights should be reviewed as they do not provide interesting information. Response: We thank this suggestion and acknowledge that the highlights do not provide interesting information. Action: The revised manuscript includes the following revised highlights: - Biodiesel production (based on soybean and beef tallow) increased sharply in Brazil - Land use change, biodiversity, water impacts, GHG and energy balance are critical. - Diversifying feedstock and adopting ethyl transesterification can minimize impacts. - Environmental zoning and certification can play an important role for biodiesel sustainability.".. Biodiesel production in Brazil has grown % since 07 - please use absolute figures Action: This sentence was revised as follow: "Biodiesel production in Brazil has grown from 7 min 07 to 2.7 M min 12.".. The impacts of land use and land use change on greenhouse gas emissions, biodiversity and water, as well as the energy balance were found to be critical for the environmental sustainability assessment - the "environmental sustainability assessment" or "environmental sustainability?" Response/Action: We thank this remark. The expression environmental sustainability assessment was rectified to "environmental sustainability" in the revised manuscript..10. Biodiesel footnote 1 - put earlier, or delete. Not really necessary. Response/action: We thank this suggestion and the footnote 1 was deleted... p. 2 for biodiesel production and biodiesel production - please rewrite sentence. Response/action: We apologize for the mistake and the sentence was rewritten..12. p : The aim of PNPB was to promote the sustainable production and use of biodiesel - this is not entirely true; sustainability isn't included as one of the Program's three pillars (which are social inclusion, guarantee competitive prices and regional diversification). Response: Thanks for the comment. In fact, the objective of the National Program for Biodiesel Production was to implement, in a sustainable way, in technical and economic aspects, the production and use of biodiesel.

7 Action: This sentence was revised as follow: "The objective of the PNPB was to implement, in a sustainable way, in technical and economic aspects, the production and use of biodiesel and to allow initially the blend of 2% biodiesel (in volume) with diesel (B2) on a voluntary basis. ".. p. 7 Appendix A isn't included - anyways the numbers should be presented in text. Response/action: We apologize for the mistake. Appendix A was included in the revised manuscript... p. please cut footnote. Response/action: We agree and thank this suggestion. Footnote was deleted... p that grassland plantations are occupying increasing areas (more than % in years) - not clear what is meant with more than % in years. Response: We agree that the percentage of increase in grassland plantation area between 170 and 0 should be clarified. Action: The sentence was rewritten and a new sentence was included: "In fact, these data demonstrate that grassland plantations are occupying increasing areas (grassland plantation area increase more than % between 170 and 0), mainly in the North and Central- West regions. This increase in grassland plantation area since 170 may be an indication of the substitution of natural grasslands, since natural grassland areas have been reduced.".1. p Fig. and 5 show that soybean expansion since 1 occurred mainly on grassland I don't see this from either figure. Response: We agree that the sentence shall be rewritten to clarify which land use was converted in soybean area. Action: Sentence was revised as follow: "As can be seen in Fig., annual cropland area significantly increased between 1 and 0, which according to Fig. 5, was driven by the rise in soybean area. There was also an increase of grassland plantation and reducing natural grassland, besides the maintenance of forest plantation and native forest. Therefore, it is possible to assume that soybean expansion occurred mainly for grassland (natural or plantation).".17. p. - Malça and Freire () in a review of biodiesel LC studies identified a strong correlation between the key modeling issues addressed by the surveyed LC models and biodiesel GHG intensity. - Please provide information on how strong this correlation is (numbers). Response: We thank this suggestion; however, Malça and Freire () did not provide information regarding statistical correlation values. Nonetheless, further details were included to improve this issue. Action: We included two new sentences at the end of the 2nd paragraph of sub-section.2. The new text (revised manuscript) reads: Malça and Freire () in a review of biodiesel LC studies identified a strong correlation between the key modeling issues addressed by the surveyed LC models and biodiesel GHG intensity. This review showed that LC studies of biodiesel that do not account for LUC and N2O emissions from soil (or adopt low values), presented GHG intensities below 7 g CO2eq MJ-1. Instead, the studies that accounted for

8 higher N2O emissions from soil and LUC soil carbon emissions presented intensities above g CO2eq MJ-1... p. - Regarding beef tallow biodiesel, few studies have dealt with GHG and energy balance - the list seems quite extensive, based on the authors' citations in the text. Response: We thank for this remark. Action: The sentence was rewritten as follow: "Regarding beef tallow biodiesel, various studies have dealt with GHG and energy balance and the results also can vary significantly. Table 1 shows some results of energy output and GHG emissions for tallow biodiesel. The large range observed is mainly related with the system boundaries definition and allocation methods. Rendering (tallow production) generates the largest GHG emissions for tallow biodiesel (Prabhu et al., 0; Barber et al., 07; Niederl and Narodoslawsky, 0). Since tallow production is considered to have an inelastic supply (Brander et al., 0), GHG emissions from beef production (including LUC) are usually excluded in the studies (Brander et al., 0; Niederl and Narodoslawsky, 0; Prabhu et al., 0). Related to energy balance, according to Barber et al (07), different energy ratios are due in part to different allocation rates and have combined with different rendering energy values. Furthermore, Bruyninckx (10) results show that the most critical stages in terms of energy requirements are the transesterification process (7%), farming (%) and slaughtering and rendering (1%).".1. p. -.1 Effects on food production and security - I do not understand how food production and security can be a key driver or barrier for environmental sustainability of biodiesel in Brazil. Response: We agree. Also taking into account comments from reviewer 1), we decided to remove the text about food production and security (section.1). Action: Section.1 was removed from the updated manuscript... p. "Considering the potential benefit from biofuel production for income generation or for local energy self-sufficiency it is known that large-scale and small-scale biofuel production can co-exist and even work together in synergy to maximize positive outcomes mainly for rural development (Cotulaet al., 0; FAO, 0; Gallagher, 0)" - this is a very strong statement, and not backed by any evidence for Brazil other than citations. In fact, in Brazil there is considerable evidence that this doesn't work (cassava family farmers for the Proálcool Program, or now the case with castor production). Response: We agree. The subsection.1 was removed from the updated manuscript... p The Social Fuel Seal is a good example, but is not without difficulties which should be discussed in the paper. Response: We thank for the suggestion. Action: We included a mention/discussion to the Social Seal (positive and negative aspects as follows: "This program has a positive aspect of giving support to small producers. On another hand, entrepreneurs claim against the high costs of the program. Additionally, since in the auctions the prices paid by Petrobras (around USD L-1, 12 USD exchange rate, ANP, 12) are higher than the final price of the blend diesel-biodiesel in the pump stations, Petrobras covers the difference."

9 .22. p. 2 - "Despite the national and international sustainability tools and initiatives, there has been strong opposition from social movements and environmental organizations both in Europe and in biofuel producing countries, which has weakened these tools and initiatives legitimacy (CEO, 12a)" - I wouldn't say all ONGs/stakeholders/researchers have this view; and also that economic interests (market barriers) often may be as or more important. The initiatives presented (SFS, Amazon Moratorium) are still interesting tools, and quite active today. Response: We agree. Action: This sentence was rewritten as follow: "According to CEO (12), despite the national and international sustainability tools and initiatives, there has been some opposition from social movements and environmental organizations both in Europe and in biofuel producing countries, which has weakened these tools and initiatives legitimacy. In fact, there are also important economic interests (market barriers), but it must be noticed that these tools are active in the country. The soybean certification scheme makes sense in the context of the certification of the several products that are produced from soybean; however, the certification schemes only exist for biodiesel and not for oil or meal. This issue is very important also because there should be coherence between the methods to attribute impacts between the various soybean based coproducts."

10 Response to Reviewers Manuscript submitted for possible publication in Energy Policy Ref.: Ms. No. JEPO-D--005 "Environmental sustainability of biodiesel in Brazil" Response to Reviewers: The Reviewers offered general comments and suggestions on several topics. We thank these comments and suggestions. Following the reviewers remarks, the manuscript was revised. The annotated comments detailing the changes that were made or justification for no change (point by pointed) as raised in the reviewers comments can be found in the table below. The reviewers comments are in the left column and our response/action in the right column. Reviewer 1 Comments from reviewers Responses/actions from authors "This is an interesting paper with useful information on soybean and biodiesel production (over 0 references!). Biodiesel in Brazil is mainly produced from soybeans (77-2% of the total production). The total area used for soybean production is approximately million hectares. Environmental impacts, land use change and biodiversity are significant problems caused by soybean production which is moving the agricultural frontier into the Amazonia but its not biofuel production that is causing it. Biodiesel production uses a small fraction the soybean produced (less than 10%). Therefore to place biodiesel in the proper context is very important and the paper does not do that clearly. For example: Section.1 addresses effects on food production and security for biofuels in general and not biodiesel specially. Section.2 again deals with the general problem of soybean expansion and not biodiesel. This is a major weakness of the impacts paper. The biofuel production should be clearly distinguished from soybean production for other uses." "On dealing specifically with GHG and energy balance in biodiesel production a Table giving the numbers from different authors would help the reader to understand the issues." "Since transesterification accounts for 7% of the energy needed in the production of biodiesel a stronger discussion of the ethanol "versus" methanol routes would be very helpful. This is attempted in section. but few numbers are given comparing them." "Finally there is practically nothing on the economics of biodiesel production which is heavily subsidized in Brazil. Such subsidies were introduced in the initial stages of the program which started as a social program using locally available agricultural products for small farmers in the North East region of Brazil. This program failed but the subsidies remained when soybean (which is produced mainly in the South East) became the main feedstock. Presently there seems to be little social and economic justification for subsidizing such program." Response: We agree that it is important to place biodiesel in the proper context and acknowledge the text should be improved. In this context, subsection 2.1 was updated and we decided to remove subsection.1 from the updated manuscript. Regarding subsection.2, we acknowledge that the general problem of soybean expansion is not only caused by biodiesel production, which we believe is now better described in the updated subsection 2.1. However, it should be noted that since 05 the increase in soybean oil national consumption was very much related with biodiesel increasing production (see new Figure c). Action: Subsection.1 was removed from the updated manuscript. Subsection 2.1 and Figure were updated with more information about the quantity of soybean and soybean oil used in Brazil for biodiesel production and food purposes. The new text and figure read: Fig. presents production, import, export and national consumption (NC) of soybean grain (Fig. a), meal (b) and oil (c) in Brazil since. The NC quantities used for biodiesel production and food purposes are also shown in fig. c. Soybean grain production increased about 7% since 01, due to both an increase in exportation and national demand. In 0-0, exported soybean represented in average 27% of the Brazilian production while in 10, it represented over %. Brazil is self-sufficient in soybean grains, meal and oil, supplying the domestic market and exporting the surplus to international markets (MAPA, ). Domestic consumption of soybean meal and oil has been growing significantly especially in the last five years. Over % of the soybean grains consumed in the domestic market are processed to co-produce soybean oil and meal (% of meal and 2% of the oil were exported in 10). Only -7% was consumed as food or feed. Oil exports have been reducing since 05 due to the growing biodiesel production (commercialized in Brazil since 05). In 10, biodiesel production required about % of the total domestic consumption of soybean oil. Response: We thank for the suggestion. Action: A table giving the numbers of GHG and energy balance for tallow biodiesel (Table 1) was included in the subsection.2 of the revised manuscript. The GHG and energy balance results for soybean biodiesel are explained in the text. Response: Thank you for this suggestion. However, it should be noted that the 7% mentioned for transesterification are for tallow biodiesel and do not include beef production. Action: Additional information and a stronger discussion regarding ethanol "versus" methanol routes was included in the subsection.2: "Despite the difference of using a renewable (ethanol) and fossil (methanol) source, is important to compare the environmental impacts of two routes. A study performed by CENBIO () compared the production of biodiesel using methyl and ethyl route considering life-cycle assessment approach. The difference between the results of two routes, for the various environmental impact categories (e.g. climate change, acidification), is less than 10%. The results showed that there were no significant differences in the environmental impacts of these two technological routes, despite the fossil origin of the methanol imported in Brazil (Coelho et al., ). " Response: We thank this comment but the main objective of the paper is to discuss the environmental sustainability of biodiesel; however, we agree that the objective should be clarified and the importance of the economic aspects should be emphasized. Action: In order to clarify the objective, the following sentence was included in the end of section 1: "This paper aims to address only environmental impacts of biodiesel from soybean and beef tallow in Brazil; however, it must be taken into account that economic aspects of biodiesel are the main reason for choosing these two raw materials for biodiesel in Brazil. Other feedstocks such as palm and castor oil present high opportunity costs (NAE, 05) since they are used in the food industry and others (more interesting in economic terms). Soybean oil is the main raw material because it is the by-product of the production of soybean meal for animal feed (to be exported together with the grains). In a similar way, beef tallow is the by-product of meat production to be exported. However, it must be observed that, even in the case of soybean and tallow biodiesel, most biodiesel is commercialized in auctions by Petrobras (the Brazilian Oil Company), which pays prices much higher that the final price of biodiesel blended with diesel oil in the country (diesel oil prices are controlled by Federal Government as a tool against inflation rates)."

11 Reviewer "The paper discusses those environmental impacts originating within the biodiesel production chain, using soybean and beef tallow as example (Brazil's most relevant biodiesel feedstock in terms of blending). It is a review - not an original article. As usual for reviews, it is very detailed and possesses quite a bit of interesting information, especially numbers. I think from the reader's perspective it would be preferable to include more tables in order to synthesize the information better." Response: We thank this suggestion, but we attempted to be sparing in the use of tables (as recommended in the guide for authors). Action: We decide to include one table (Table 1 in the subsection.2) and also the appendix includes a table (Table A1). "First, and most importantly, the paper discusses "environmental sustainability" of both beef tallow and soybean oil, but does not offer a definition of what constitutes "environmental sustainability". This is particularly important because both feedstock have a long history on deforestation (direct or indirect), problems with GMO, pesticide-use or negative impacts due to monoculture. Speaking of environmental sustainability in this context raises a positive connotation that is very difficult to defend where a clear benchmark is missing, e.g. what qualitatively/quantitatively constitutes an "environmentally sustainable biofuel". In fact, both a theoretical framework is missing and an argument ("are beef tallow and soybean sustainable" are missing, and this leaves me, as a reader, wondering on the original contribution of the paper." Response: We agree and acknowledge that it is important to define "environmental sustainability and give a theoretical framework about the concept. This was missing in the submitted version, but is now included in section. The original contribution of the paper is to characterize the biodiesel production chain in Brazil, identifying potential environmental impacts and analyzing key drivers and barriers for the biodiesel environmental sustainability development; however, discussing if soybean or beef tallow biodiesel constitute an "environmentally sustainable biofuel" is complex and difficult to define based on a single argument. Action: The following paragraphs were included in the beginning of section : Environmental sustainability is a difficult concept to explain, but it can be defined as the maintenance of natural capital (Goodland, ). The two fundamental environmental services (the source and the sink functions) must be maintained unimpaired during the period over which sustainability is required. In this context, the environmental sustainability of biodiesel is critically related with life cycle impacts associated with air (GHG emissions and others), water and soil, energy balance, biodiversity (FAO, b). Another important issue involves developing strategies to ensure that as the production of biofuels increases, adequate supplies of other needed agricultural and forest-based goods are produced (FAO, b). However, to quantify the environmental sustainability of biodiesel is complex. An increasing number of countries have established initiatives to define sustainability criteria for biofuels. For instance, the European Union Directive on the Promotion of Renewable Energy Sources (RED) defined that for biofuels be counted as renewable energy, a minimum GHG saving of % is required by (comparing with fossil fuel). The directive also stipulates no-go zones for feedstock production (e.g. areas where land is deemed to be of high biodiversity value or high carbon stock, wetlands, peatland) (FAO, b). Also, there is an important initiative from Global Bioenergy Partnership (FAO/GBEP, ) defining sustainability indicators for bioenergy (FAO/GBEP, ), which are now starting to be evaluated for different countries. " Response: We aknowledge that Ethyl transesterification is not produced at large economical scale, but can be feasible in economic terms as shown by the production of company FERTIBOM. In the case of water footprint, the idea was to show how this issue has been considered in Brazil and what level of information we can indicated in a discussion related to this issue. " In many cases the authors also refer to "potentials" (e.g. ethyl transesterification, which is not at economical scale despite the issue having been discussed for several years) or considerable uncertainties in key values, but are seldom able to provide hard evidence on environmental sustainability (e.g. the case of the water footprint, which might have important consequences for medium-term sustainability)." Action: The following sentence was included: "Most industries in use methylic transesterification for biodiesel production; however, there are two companies used (anhydrous) bioethanol for biodiesel production (ANP, ). For instance, FERTIBOM (in Sao Paulo State) produces biodiesel from different raw materials (vegetable oils and also residual oils) and consider it is economically feasible since they participate in the biodiesel auctions (CENBIO, )." In the case of water footprint (WF), we include the following text to expand the idea of Gerbens-Leenes et al. (0), insert the currently situation of local studies and include a recent result for a Brazilian region: "Gerbens-Leenes et al. (0) calculated a WF for soybean biodiesel of liters per MJ (7 liters per liter of biodiesel), according to the authors in terms of the bioenergy sources considered in this study, biodiesel has the bigger WF, followed by ethanol and bioelectricity. WF studies of biodiesel in Brazil are still rare, few of them are indeed representative and this issue still needs to be deepened in the country. A preliminary study for showed that soybean biodiesel water footprint varies from 0 to 0 liters per MJ (Seabra et al., ). Another reference in southern Brazil for production of biodiesel indicates that the major contribution to degradative water use is the agricultural phase (.%) while for consumptive use the largest contribution is the oil extraction process (7.%) (Muller, 12)."

12 "This lack in depth, secondly, also makes the authors recommendations less valuable. The authors write that "environmental impacts and conflicts between food could be mitigated by appropriate policies aiming at an integrated optimization of food and bioenergy production and through agro-economicecological zoning, allowing adequate use of land for each purpose". While they correctly state that initiatives like the Amazon Moratorium are on the way to mitigate some of the impacts, these do by far not cover every environmental impact covered in the paper. Rather they focus on reducing/halting indirect or direct deforestation, which strangely enough is very little discussed in the paper (the authors only reference possible indirect deforestation in the Amazon once, but do not provide information on the (existing) evidence). The role of policies in resolving conflicts in the paper is treated only superficially in the paper; it is clear that the policy framework can have beneficial consequences for biodiesel sustainability, but judging from history this process seems to be rather ineffective. Together this makes the reader feel that the paper is rather unbalanced or uncritical with regards to environmental risks of large-scale biodiesel expansion in Brazil." "Third and finally, sustainability from the reviewer's perspective it is difficult to discuss sustainability excluding social impacts and economics. In fact, the authors include a discussion of food security, which is rather a social issue. So, one might also ask, what are the broader social consequences of soybean and tallow production? There are many reports available on this issue, including from Brazilian organizations/universities, and rather than being quiet on the topic, the authors should take the time to include - at least in the discussion and conclusions a broader perspective. Also economics is an important issue, as already mentioned for the case of ethyl transesterification which might be more sustainable, but doesn't seem to be economic." "I would suggest a careful revision of the English which is sometimes hard to read, as well as a careful review of the (in-text) citations. For example, Schaffel and La Rovere (10) and not Schaffer and La Rovere (10)." "The research highlights should be reviewed as they do not provide interesting information." "Biodiesel production in Brazil has grown % since 07 - please use absolute figures" "The impacts of land use and land use change on greenhouse gas emissions, biodiversity and water, as well as the energy balance were found to be critical for the environmental sustainability assessment - the "environmental sustainability assessment" or "environmental sustainability?" 'Biodiesel footnote 1 - put earlier, or delete. Not really necessary." "p. 2 for biodiesel production and biodiesel production - please rewrite sentence" "p : The aim of PNPB was to promote the sustainable production and use of biodiesel - this is not entirely true; sustainability isn't included as one of the Program's three pillars (which are social inclusion, guarantee competitive prices and regional diversification)." Response: We thank this remark and we agree that the role of policies should be further discussed. A critical point a view regarding environmental risks of biodiesel expansion in Brazil is also missing. Action: The last paragraph of subsection. was revised and a new paragraph was introduced: "According to CEO (12), despite the national and international sustainability tools and initiatives, there has been some opposition from social movements and environmental organizations both in Europe and in biofuel producing countries, which has weakened these tools and initiatives legitimacy. In fact, there are also important economic interests (market barriers), but it must be noticed that these tools are active in the country. The soybean certification scheme makes sense in the context of the certification of the several products that are produced from soybean; however, the certification schemes only exist for biodiesel and not for oil or meal.this issue is very important also because there should be coherence betweem the methods to attribute impacts bettweens the various ybean based co-products. The agro-ecological zoning is an important policy tool and has taken into account environmental, economic and social aspects as an answer to challenges in sustainable expansion of bioenergy crops. An interesting initiative was launched by the Federal Government to guarantee the sustainable expansion of sugarcane and oil palm in Brazil (two agro-ecological zonings): for sugar cane in 0 (EMBRAPA, 0) and for oil palm in 10 (EMBRAPA, 10). In this process, maps were produced showing soils, topography, climate and rainfall. This regulation enables a guidance to credit policies and use for public banks as a condition for production financing. The zoning identified areas where sugar cane crop expansion could take place. It forbids sugar cane cultivation in 2.5% of the national territory. It has identified million hectares (EMBRAPA, 0) that comply with environmental and productivity requirements, mainly from the intensification of cattle ranching, which is currently very inefficient (less than 1 head ha-1) (Strapasson et al., 12). Such initiatives could be also adopted for soybean expansion, contributing to the sustainability of soybean biodiesel. However here there is the issue of the intensification of cattle ranch as an important goal to be achieved both for these zonings and for the sustainability of cattle ranch in Brazil (and indirectly to the sustainability of biodiesel from beef tallow)." Response: We thank for the comments and we agree that is difficult to discuss sustainability excluding social and economic aspects. However, as presented in the introduction, the main objective of this article is to discuss the environmental sustainability of soybean and beef-tallow biodiesel. Nevertheless, some economic aspects were addressed throughout the article. Action: To clarify the economic sustainability of ethyl transesterification, the follow sentence was included in subsection.2 of the revised manuscript: "Most industries in use methylic transesterification for biodiesel production; however, there are two companies used (anhydrous) bioethanol for biodiesel production (ANP, ). For instance, FERTIBOM (in Sao Paulo State) produces biodiesel from different raw materials (vegetable oils and also residual oils) and consider it is economically feasible since they participate in the biodiesel auctions (CENBIO, )." Response/Action: We agree with the comment. A careful revision of the English and citations were made. Response: We thank this suggestion and acknowledge that the highlights do not provide interesting information. Action: The revised manuscript includes the following revised highlights: - Biodiesel production (based on soybean and beef tallow) increased sharply in Brazil - Land use change, biodiversity, water impacts, GHG and energy balance are critical. - Diversifying feedstock and adopting ethyl transesterification can minimize impacts. - Environmental zoning and certification can play an important role for biodiesel sustainability. Action: This sentence was revised as follow: "Biodiesel production in Brazil has grown from 7 m in 07 to 2.7 M m in 12." Response/Action: We thank this remark. The expression environmental sustainability assessment was rectified to "environmental sustainability" in the revised manuscript. Response/action: We thank this suggestion and the footnote 1 was deleted. Response/action: We apologize for the mistake and the sentence was rewritten. Response: Thanks for the comment. In fact, the objective of the National Program for Biodiesel Production was to implement, in a sustainable way, in technical and economic aspects, the production and use of biodiesel. Action: This sentence was revised as follow: "The objective of the PNPB was to implement, in a sustainable way, in technical and economic aspects, the production and use of biodiesel and to allow initially the blend of 2% biodiesel (in volume) with diesel (B2) on a voluntary basis. "

13 "p. 7 Appendix A isn't included - anyways the numbers should be presented in text" "p. please cut footnote" Response/action: We apologize for the mistake. Appendix A was included in the revised manuscript. Response/action: We agree and thank this suggestion. Footnote was deleted. Response: We agree that the percentage of increase in grassland plantation area between 170 and 0 should be clarified. "p that grassland plantations are occupying increasing areas (more than % in years) - not clear what is meant with more than % in years" Action: The sentence was rewritten and a new sentence was included : "In fact, these data demonstrate that grassland plantations are occupying increasing areas (grassland plantation area increase more than % between 170 and 0), mainly in the North and Central-West regions. This increase in grassland plantation area since 170 may be an indication of the substitution of natural grasslands, since natural grassland areas have been reduced. " Response: We agree that the sentence shall be rewritten to clarify which land use was converted in soybean area. "p Fig. and 5 show that soybean expansion since 1 occurred mainly on grassland - I don't see this from either figure" "p. - Malça and Freire () in a review of biodiesel LC studies identified a strong correlation between the key modeling issues addressed by the surveyed LC models and biodiesel GHG intensity. - Please provide information on how strong this correlation is (numbers)." "p. - Regarding beef tallow biodiesel, few studies have dealt with GHG and energy balance - the list seems quite extensive, based on the authors' citations in the text." "p. -.1 Effects on food production and security - I do not understand how food production and security can be a key driver or barrier for environmental sustainability of biodiesel in Brazil." "p. "Considering the potential benefit from biofuel production for income generation or for local energy self-sufficiency it is known that large-scale and smallscale biofuel production can co-exist and even work together in synergy to maximize positive outcomes mainly for rural development (Cotulaet al., 0; FAO, 0; Gallagher, 0)" - this is a very strong statement, and not backed by any evidence for Brazil other than citations. In fact, in Brazil there is considerable evidence that this doesn't work (cassava family farmers for the Proálcool Program, or now the case with castor production)." Action: Sentence was revised as follow: "As can be seen in Fig., annual cropland area significantly increased between 1 and 0, which according to Fig. 5, was driven by the rise in soybean area. There was also an increase of grassland plantation and reducing natural grassland, besides the maintenance of forest plantation and native forest. Therefore, it is possible to assume that soybean expansion occurred mainly for grassland (natural or plantation)." Response: We thank this suggestion; however, Malça and Freire () did not provide information regarding statistical correlation values. Nonetheless, further details were included to improve this issue. Action: We included two new sentences at the end of the 2nd paragraph of sub-section.2. The new text (revised manuscript) reads: Malça and Freire () in a review of biodiesel LC studies identified a strong correlation between the key modeling issues addressed by the surveyed LC models and biodiesel GHG intensity. This review showed that LC studies of biodiesel that do not account for LUC and N2O emissions from soil (or adopt low values), presented GHG intensities below 7 g CO2eq MJ-1. Instead, the studies that accounted for higher N2O emissions from soil and LUC soil carbon emissions presented intensities above g CO2eq MJ- 1. Response: We thank for this remark. Action: The sentence was rewritten as follow: "Regarding beef tallow biodiesel, various studies have dealt with GHG and energy balance and the results also can vary significantly. Table 1 shows some results of energy output and GHG emissions for tallow biodiesel. The large range observed is mainly related with the system boundaries definition and allocation methods. Rendering (tallow production) generates the largest GHG emissions for tallow biodiesel (Prabhu et al., 0; Barber et al., 07; Niederl and Narodoslawsky, 0). Since tallow production is considered to have an inelastic supply (Brander et al., 0), GHG emissions from beef production (including LUC) are usually excluded in the studies (Brander et al., 0; Niederl and Narodoslawsky, 0; Prabhu et al., 0). Related to energy balance, according to Barber et al (07), different energy ratios are due in part to different allocation rates and have combined with different rendering energy values. Furthermore, Bruyninckx (10) results show that the most critical stages in terms of energy requirements are the transesterification process (7%), farming (%) and slaughtering and rendering (1%)." Response: We agree. Also taking into account comments from reviewer 1), we decided to remove the text about food production and security (section.1). Action: Section.1 was removed from the updated manuscript. Response: We agree. The subsection.1 was removed from the updated manuscript. Response: We thank for the suggestion. "p The Social Fuel Seal is a good example, but is not without difficulties which should be discussed in the paper." Action: We included a mention/discussion to the Social Seal (positive and negative aspects as follows: "This program has a positive aspect of giving support to small producers. On another hand, entrepreneurs claim against the high costs of the program. Additionally, since in the auctions the prices paid by Petrobras (around USD L-1, 12 USD exchange rate, ANP, 12) are higher than the final price of the blend diesel-biodiesel in the pump stations, Petrobras covers the difference."

14 "p. 2 - "Despite the national and international sustainability tools and initiatives, there has been strong opposition from social movements and environmental organizations both in Europe and in biofuel producing countries, which has weakened these tools and initiatives legitimacy (CEO, 12a)" - I wouldn't say all ONGs/stakeholders/researchers have this view; and also that economic interests (market barriers) often may be as or more important. The initiatives presented (SFS, Amazon Moratorium) are still interesting tools, and quite active today." Response: We agree. Action: This sentence was rewritten as follow: "According to CEO (12), despite the national and international sustainability tools and initiatives, there has been some opposition from social movements and environmental organizations both in Europe and in biofuel producing countries, which has weakened these tools and initiatives legitimacy. In fact, there are also important economic interests (market barriers), but it must be noticed that these tools are active in the country. The soybean certification scheme makes sense in the context of the certification of the several products that are produced from soybean; however, the certification schemes only exist for biodiesel and not for oil or meal. This issue is very important also because there should be coherence between the methods to attribute impacts between the various soybean based co-products.

15 *Highlights (for review) Highlights: Biodiesel production (based on soybean and beef tallow) increased sharply in Brazil Land use change, biodiversity, water impacts, GHG and energy balance are critical. Diversifying feedstock and adopting ethyl transesterification can minimize impacts. Zoning and certification can play an important role for biodiesel sustainability.

16 *Manuscript Click here to view linked References Environmental sustainability of biodiesel in Brazil 1 2 Érica Geraldes Castanheira 1, Renata Grisoli 2, Fausto Freire 1, Vanessa Pecora 2, Suani Coelho 2* 5 1 ADAI-LAETA, Department of Mechanical Engineering, University of Coimbra, Pólo II Campus, 7 Rua Luís Reis Santos, 00-7 Coimbra, Portugal 10 2 CENBIO, Brazilian Reference Center on Biomass, Institute of Energy and Environment, 12 University of São Paulo, Av. Prof Luciano Gualberto, 1, São Paulo, Brazil Corresponding author. suani@iee.usp.br Abstract 22 Biodiesel production in Brazil has grown from 7 m in 07 to 2.7 Mm in 12. It is an emergent 2 2 bioenergy for which it is important to guarantee environmental sustainability. The objective of this 2 27 article is to characterize the biodiesel production chain in Brazil, to identify potential environmental 2 impacts and to analyze key drivers and barriers for biodiesel environmental sustainability. This article 0 explores these aspects and focuses on the increasing demand for the main feed stocks for biodiesel production in Brazil: soybean oil and beef tallow. The impacts of land use and land use change on greenhouse gas emissions, biodiversity and water, as well as the energy balance were found to be 7 critical for the environmental sustainability assessment and development of biodiesel chains. Increasing 0 1 agriculture yields, diversifying feed stocks and adopting ethyl transesterification can contribute to minimize environmental impacts. It was also found that environmental impacts can be mitigated by 5 appropriate policies aiming at an integrated optimization of food and bioenergy production and through 7 agro-economic-ecological zoning, allowing adequate use of land for each purpose. Despite the limitation and weakness of some sustainability tools and initiatives, certification and zoning can play an 51 important role for the sustainability of the emerging biodiesel production in Brazil. 5 5 Keywords: Soybean biodiesel; Sustainability Assessment; Tallow biodiesel

17 1 Introduction 1 2 Current liquid biofuel production processes rely on first-generation conversion pathways and 5 comprise two distinct products: bioethanol and biodiesel. Policies worldwide have stimulated 7 biofuel demand by setting targets and blending quotas, and have aided its development by 10 establishing support mechanisms (such as subsidies and tax exemptions in the starting point of the 12 programs) (Bringezu et al., 0). In this context, over the past five years, liquid biofuels production increased at an average annual rate of 17% for bioethanol and 27% for biodiesel, 1 17 reaching over million m (. and.1 million m of biodiesel and bioethanol, respectively) 1 in (REN, 12). Biodiesel accounted for approximately 5% of the world biofuel production 22 in 00, but this share has been rising and, in, biodiesel represented about % of the total 2 2 biofuel production Fig. 1 shows the growth in biodiesel production since 00. The columns in the figure represent the 2 contribution of different world regions for biodiesel production of the five most important 0 countries. Europe was the dominant region with increasing production since 05. North America was a distant second producer led by the United States of America (USA) until 0 when USA production fell by over 10 thousand barrels per day (mainly due to the economic downturn, 7 incentives changes for biodiesel and foreign trade policies), while growth continued in Central & 0 1 South America and Asia & Oceania. The five most important countries (% of world production in 10) were Germany, Brazil, France, Argentina and USA (EIA, 12)

18 Biodiesel production 1 (10 barrels day -1 ) 2 Asia & Oceania 00 Africa 5 0 Middle East 700 Eurasia Europe 0 10 Central & South America 0 North America 12 United States of America 1 Brazil 00 Germany 17 France 1 Argentina Source: U.S. Energy Information Administration, International Energy Statistics (EIA, 12) 2 Figure 1. Global growth in biodiesel production (00-10) In Brazil, the federal government created in 0 the National Biodiesel Production and Use Program (PNPB). The objective of the PNPB was to implement, in a sustainable way, in technical and economic aspects, the production and use of biodiesel and to allow initially the blend of 2% 7 biodiesel (in volume) with diesel (B2) on a voluntary basis. Federal law,07 (Brasil, 05) was approved in 05 to establish a mandatory target of 2% and 5% of biodiesel addition to the diesel 0 1 oil in 0 and, respectively. In July 0, the National Council of Energy Policy (CNPE) adopted % of biodiesel as the compulsory blend, which was raised to 5% (B5) in January 10, 5 7 anticipating in three years the goal set in 05 (MME, ). The fast growing interest and production of biodiesel worldwide has led to increasing concern about 51 the environmental, economic and social impacts, especially regarding land competition, air and 5 water emissions, biodiversity and the fuel versus food debate, also in Brazil (e.g. Padula et al., 5 12; GEA, 12; Janssen and Rutz, ; Diaz-Chavez, ; Lange, ; Lynd et al., ; 5 5 Schaffel and La Rovere, 10; Santos and Rathmann, 0). To counterbalance these potential

19 negative effects, it is crucial to establish an overview of current and future trends of biodiesel in 1 2 Brazil, including characterization of production chain, main impacts and several policies, standards 5 and certification schemes in place to help biodiesel sustainability assessment and development. 7 This paper provides an overview of the key challenging factors towards environmental 10 sustainability of biodiesel in Brazil, based on an overview of biodiesel production chain and 12 environmental impacts. It is organized in five sections, including this introduction. Section 2 presents a characterization of the biodiesel chain in Brazil, focused on the main feed stocks 1 17 (soybean oil and beef tallow). Section analyzes the main environmental impacts of biodiesel. The 1 key drivers and barriers for the environmental sustainability of biodiesel in Brazil are discussed in 22 section. Section 5 draws the concluding remarks. 2 2 This paper aims to address only environmental impacts of biodiesel from soybean and beef tallow 2 27 in Brazil; however, it must be taken into account that economic aspects of biodiesel are the main 2 reason for choosing these two raw materials for biodiesel in Brazil. Other feedstocks such as palm 0 and castor oil present high opportunity costs (NAE, 05) since they are used in the food industry and others (more interesting in economic terms). Soybean oil is the main raw material because it is the by-product of the production of soybean meal for animal feed (to be exported together with the 7 grains). In a similar way, beef tallow is the by-product of meat production to be exported. However, 0 1 it must be observed that, even in the case of soybean and tallow biodiesel, most biodiesel is commercialized in auctions by Petrobras (the Brazilian Oil Company), which pays prices much 5 higher that the final price of biodiesel blended with diesel oil in the country (diesel oil prices are 7 controlled by Federal Government as a tool against inflation rates)

20 2 Biodiesel production in Brazil 1 2 Biodiesel production in Brazil has grown from 7 m in 05 to approximately 2.7 million m in 5 (ANP, 12). The effective production in represented only % of the actual total 7 nominal capacity of biodiesel production (.0 million m³) (MME, 12). The regions with a higher 10 nominal capacity (produced over 7% of the biodiesel in Brazil) are the Central-West (states of 12 Mato Grosso, Mato Grosso do Sul, Goiás and Distrito Federal) and the South (states of Rio Grande do Sul, Paraná and Santa Catarina) Cerri et al. (10) presented an estimation of the total biodiesel production in the 10- period 1 and the requirement for biodiesel production to supply domestic demand by, based on data 22 from the National Energy Plan 0 (NEP) (MME, 07) and the National Plan on Climate 2 2 Change (NPCC) (Governo Federal, 0). Regarding biodiesel production, the results show a total 2 27 of. million m (NEP scenario) and.2 million m (NPCC scenario). Concerning the 2 requirement for biodiesel (B5) production to supply domestic demand by, the results vary 0 between. million m (NEP scenario) and. million m (NPCC scenario). These values show that the Brazilian requirement in can be achieved by the total nominal capacity of biodiesel production in Brazil in (.0 million m³) (ANP, 12). 7 Soybean oil is currently the main feedstock of biodiesel production in Brazil. According to ANP 0 1 (), in 0 and 10 this raw material was responsible for about 77-2% of biodiesel production. Other raw materials are beef tallow (-17%) and cottonseed oil (2%). Cerri et al. 5 (10) estimated that biodiesel will be produced from five raw materials in : soybean (5%), 7 beef tallow (27%), palm (10%), castor bean (%) and sunflower (1%). This calculation assumed a 51 decrease in the use of soybean as feedstock in the long term, an increase in the participation of other crops, the production of each oilseed in the Brazilian territory and the amount of animals 5 5 slaughtered by in Brazil. The two main biodiesel feed stocks in Brazil are presented in the 5 next subsections

21 17/77 177/7 17/7 17/0 0/1 1/2 2/ / /5 5/ /7 7/ / /0 10/1 /2 12/ 1/ 1/5 / 1/7 17/ / 1/00 00/01 01/02 02/0 0/0 0/05 05/0 0/07 07/0 0/0 0/10 10/* /12# Soybean oil 5 Soybean oil is a vegetable oil extracted from the seeds of the soybean (Glycine max), which can be 7 produced almost all over the Brazilian territory. Fig. 2 presents the historic evolution of soybean 10 cultivated area in the various Brazilian regions and the soybean yield in Brazil. Since 01/02, the 12 soybean cultivated area grew about %. Central-Western and Southern regions have the highest cultivated area, representing together % of the total cultivated area in the last three seasons 1 17 (0/0 to 10/) (CONAB, 12). The states of Mato Grosso (27%) and Goiás (%) in 1 Central-West and Paraná (%) and Rio Grande do Sul (%) in South are the major producers of 22 soybean in Brazil (CONAB, 12) 1. Although the cultivated area in Brazil has grown in recent 2 2 years, an increase in the yield associated with technological advances, management and efficiency 2 27 aspects is one of the reasons for the increase in soybean production. 2 0 Cultivated area (10 ha) Yield (kg ha -1 ) South Southeast Central-West 10 Northeast North Soybean yield (Brazil) 0 0 *Preliminar data # Estimated data 51 Source: CONAB, 12 5 Figure 2. Evolution of soybean cultivated area and soybean yields in Brazil North and Northeast are the regions where the cultivated area is lower but the area increased about 5% and 0% in the last 10 years

22 Fig. presents production, import, export and national consumption (NC) of soybean grain (Fig. 1 2a), meal (b) and oil (c) in Brazil since. The NC quantities used for biodiesel production 5 and food purposes are also shown in Fig. c. Soybean grain production increased about 7% since 701, due to both an increase in exportation and national demand. In 0-0, exported soybean 10 represented in average 27% of the Brazilian production while in 10, it represented over %. 12 Brazil is self-sufficient in soybean grains, meal and oil, supplying the domestic market and exporting the surplus to international markets (MAPA, ) Figure. Production, import, export and national consumption (NC) of soybean grain (a), meal (b) 5 and oil (c) in Brazil. The NC quantities used for biodiesel production and food purposes are also 5 0 shown in c

23 1 2 Domestic consumption of soybean meal 2 and oil has been growing significantly especially in the 5 last five years. Over % of the soybean grains consumed in the domestic market are processed to 7 co-produce soybean oil and meal (% of meal and 2% of the oil were exported in 10). Only - 107% was consumed as food or feed. Oil exports have been reducing since 05 due to the growing 12 biodiesel production (commercialized in Brazil since 05). In 10, biodiesel production required about % of the total domestic consumption of soybean oil Beef tallow 22 Beef tallow is a co-product from the beef meat industry (slaughterhouse, rendering and retail). 2 2 Appendix A shows cattle slaughter and beef tallow production and consumption in Brazil from to. In, a production of more than 0 thousand (metric) tonnes of beef tallow was 2 calculated assuming that each slaughtered cattle provides an average of kg of usable tallow 0 (Levy, ; Peres, 10). However, it is important to mention the high uncertainty in the number of cattle slaughter (statistical data only includes the slaughterhouses under sanitary inspection and vary depending on the source considered as reference) and in the quantity of usable tallow that can 7 be produced from each animal. 0 1 Beef tallow consumption in Brazil almost doubled in the last fourteen years (since 17), but little information is available about the various uses of tallow. One of the reasons is that tallow has been 5 considered a low-value co-product of the cattle beef industry and, historically, the main consumer 7 of tallow is the soap industry. The different uses of tallow in Brazil in 0 were (Andrade Filho, 5107): 70% soap production, % used as fuel in boilers, 10% for animal feed and 5% others. In 07, a study made by Aboissa Vegetable Oils presented other values (Gomes et al., 0): 1% 5 5 cleaning and hygiene purposes, % chemical purposes, 12% for biodiesel production, 10% for Soybean meal is approximately 0% the soybean grain in mass and is a high-quality, in-demand protein ingredient for animal feed and human food protein

24 animal feed and % used as fuel in boilers. Since 07 beef tallow consumption for biodiesel 1 2 production increased almost six times and in 10 over than thousand tonnes of beef tallow 5 was used to produce biodiesel. Biodiesel production was the main destination for beef tallow in 7 Brazil in 10 and represented about 72% of the total beef tallow consumed. 10 The main driving forces behind the use of tallow as feedstock for biodiesel in Brazil are the low 12 price of raw material and the fact that Brazil has the second largest cattle herd in the world (Levy, ). Biodiesel from beef tallow presents advantages in some properties (cetane number and 1 17 stability), compared with biodiesel produced from soybean oil (Moraes et al., 0) but some 1 important limitations, namely viscosity, which does not allow 100% production from beef tallow, as 22 concluded from field visits for this study Environmental impacts 0 Biofuel production has attracted the attention of stakeholders because of concerns related with of greenhouse gas (GHG) emissions (particularly from land conversion), food production, water security and biodiversity (Ravindranath et al., ). In this context, a country-specific approach to 7 life-cycle assessment (LCA) is vital when evaluating the environmental impacts of bioenergy 0 1 systems. Local conditions, such as agricultural practices, land use changes (LUC) and transport infrastructures, will have a major impact on the environmental performance of the system being 5 modeled (Panichelli et al., 0). Only a few LCA studies were performed for soybean and tallow 7 biodiesel produced in Brazil, focusing on energy and GHG balances (e.g. Cavalett and Ortega, 510, 10; Mourad and Walter, ), and more recently on impacts resulting from water (consumption and pollution), land use and LUC (Prudêncio da Silva et al., 10; Batlle-Bayer et 5 5 al., 10; Castanheira and Freire,, 12; Grisoli et al., 12). This section discusses the main 5 environmental aspects

25 1 2.1 Land use and land use change 5 Biofuel feedstock currently occupies about 1% of arable land (Berndes et al., 10) and the increase 7 in land use for biofuel production initiated a widespread debate among policy makers and 10 researchers (Witcover et al., ; Ponsioen and Blonk, 12; Lange, ; Walter et al., ; 12 Nassar et al., 10; Yang et al., 0). Land is a limiting factor for biomass production in general and three land use impact pathways can be identified (Milà i Canals et al., 07): biodiversity, 1 17 ecological soil quality and biotic production potential. Higher biofuel production can lead to 1 increased GHG emissions and loss of biodiversity if it is from new land being converted to grow 22 biomass ( direct land use change) or existing agriculture being displaced ( indirect land use 2 2 change), as discussed in Gibbs et al. (0), Fargione et al. (0), Goldemberg et al. (0). Cerri 2 27 et al. (10) assessed that Brazil will need to expand about.-. Mha of area planted with 2 oilseeds to meet the domestic demand for biodiesel until. 0 The land use in Brazil since 170 is shown in Fig.. Since 0 there was not a significant change in the total area of land use; however, there were important changes between the different types of land use, namely a decrease on natural grassland and forest plantation areas and an increase area for 7 the other uses (cropland, native forest and grassland plantations). Natural grasslands dominated the 0 1 type of land use until 5, since when grassland plantations has become the major area in Brazil. In fact, these data demonstrate that grassland plantations are occupying increasing areas (grassland 5 plantation area increase more than % between 170 and 0), mainly in the North and Central- 7 West regions. This increase in grassland plantation area since 170 may be an indication of the 51 substitution of natural grasslands, since natural grassland areas have been reduced. However, natural grasslands still has a prominent place in the context of the Brazilian cattle industry, mainly 5 5 in the Northeast and in Southern regions in Brazil

26 Land use (10 ha) Forest plantations Figure. Evolution of land use and cattle population in Brazil (170-0). Fig. also shows the historic evolution (170-0) of cattle population in Brazil and for each region. In 0 the total cattle heads in Brazil was more than 5 million (double compared to 175) and % were located in the Central-Western region and % in Northern region of Brazil (where the grassland plantations more increased). Since 00, there was a strong increase of cattle heads in Northern region (7%) as a result of a combined effect of an overall total increase in cattle (% more) and a shift in location. The total area of grassland was reduced during the 1 0 period, while the cattle population increased. This was possible due to cattle production intensification resulting from the conversion of natural grassland into grassland plantations, improved management and maintenance of grassland plantations. In addition, although cattle production in Brazil is essentially based on the use of pastures, more intensive systems through supplementary feeding on pasture or by the use of feedlots have become important in the Central- Western and Southeastern regions (Cezar et al., 05) Native forest Perennial crops Annual crops Grassland plantations Natural grassland Cattle population Catte population (10 heads) Brazil North Northeast Southeast South Central-West Source: IBGE, 07 and ; CONAB, 12 0

27 Cropland area almost doubled since 170, growing 5% of perennial cropland and % annual 1 2 cropland. The main reason for this growth is the increased area of soybean and, in a less extension, 5 by sugarcane. Area occupied by the most important perennial and annual crops in Brazil is shown in 7 Fig. 5. In 10 cropland was more diversified and maize (2%), soybean (22%), beans (10%), rice 10 (10%) and sugarcane (10%) occupied most of the area. However, in 10, most of the cropland was 12 occupied by soybean (%), maize (%) and sugarcane (%) Cropland Soybean (10 ha) Maize 1 Sugarcane Beans Rice 22 Wheat 2 Coffee 2 Cassava Others Source: IBGE, 0 1 Figure 5. Evolution of perennial and annual crops in Brazil (10-10). 5 As can be seen in Fig., annual cropland area significantly increased between 1 and 0, 7 which according to Fig. 5, was driven by the rise in soybean area. There was also an increase of grassland plantation and reducing natural grassland, besides the maintenance of forest plantation 51 and native forest. Therefore, it is possible to assume that soybean expansion occurred mainly for 5 grassland (natural or plantation)

28 .2 GHG and energy balance 1 2 The life-cycle GHG emissions of biodiesel arise directly from land use change and from the use of 5 fertilizers and fuels; and indirectly from the manufacture of feedstock inputs (e.g. fertilizers, 7 chemicals), electricity generation, transportation and transformation of raw fossil fuels, etc. 10 A wide range of GHG emissions was reported in LCA studies for biodiesel (e.g. Castanheira and 12 Freire, ; Malça and Freire, ). This wide variability of GHG emissions can be due to technological and location issues such as the biodiesel conversion route, agriculture mechanization 1 17 level, crop and farm management, and changes in land use for livestock and soybean production or 1 associated with methodological assumptions in the life-cycle (LC) calculations. According to 22 several authors, LC modeling choices can have a significant effect on the GHG calculations. Malça 2 2 and Freire () in a review of biodiesel LC studies identified a strong correlation between the key 2 27 modeling issues addressed by the surveyed LC models and biodiesel GHG intensity. This review 2 showed that LC studies of biodiesel that do not account for LUC and N 2 O emissions from soil (or 0 adopt low values), presented GHG intensities below 7 g CO 2 eq MJ -1. Instead, the studies that accounted for higher N 2 O emissions from soil and LUC soil carbon emissions presented intensities above g CO 2 eq MJ A positive net energy balance (Kallivroussis et al., 02, Pradhan et al., 0, Fore et al., ) is 0 1 also one necessary criterion for a biodiesel to be a sustainable alternative to fossil diesel. To verify if biodiesel has a positive energy balance, a LC approach must be employed, allowing 5 quantification of the renewability of biofuel delivered to consumers (Malça and Freire, 0). 7 However, within the energy analysis and LCA literature, there is lack of consensus concerning the 51 definition and calculation of energy efficiency indicators to characterize the LC energy requirements of biofuels (Malça and Freire,, 0) CO 2 fluxes from carbon stock changes 1 This means a positive energy return compared with the energy required to produce the biofuel 2 5

29 The energy balance of biofuels varies from one producer to another due, for example, to different 1 2 yields, agricultural practices, industrial technologies, distances and transport used. Also, there are 5 different methodological approaches for calculating energy balances of biofuels which make direct 7 comparisons of results extremely difficult (Mourad and Walter, ). A large body of work that 10 relies on LCA to investigate the biofuel production provides different, sometimes contradictory, 12 results for net energy values (Menichetti and Otto, 0, Bureau et al., 10). According to Cavalett and Ortega (10), 0.27 kg of crude oil equivalent is required as inputs to produce one liter 1 17 of soybean biodiesel in Brazil, which means an energy return of 2. J of biodiesel per Joule of 1 fossil fuel invested. A renewability factor of. for soybean biodiesel in Brazil was found by 22 Mourad and Walter () Recent studies performed for soybean biodiesel showed that the land conversion from forest, 2 27 savannah or grassland (improved) to soybean plantation in Brazil leads to the most significant LC 2 CO 2 emissions. Castanheira and Freire (12) and Grisoli et al. (12) calculated a wide range of 0 GHG emissions for soybean biodiesel (between 12 and 0 g CO 2 eq MJ -1 ) mainly due to alternative LUC scenarios. Emissions due to LUC represent 0 to 0% of the total GHG emissions. When LUC is not considered, soybean cultivation is the LC phase that most contributes for the GHG 7 balance. In the past, some LC studies reported a correlation between biodiesel energy inputs and 0 1 GHG intensity; however, recent LC studies for biodiesel demonstrated that taking into account soil emissions in LC assessments, namely carbon emissions due to LUC and N 2 O emissions due to land 5 use, negates the correlation between biodiesel energy inputs and GHG intensity (Malça and Freire, 7, 10; Soimakallio et al., 0; Reijnders and Huijbregts, 0). 51 Regarding beef tallow biodiesel, various studies have dealt with GHG and energy balance and the results also can vary significantly. Table 1 shows some results of energy output and GHG emissions 5 5 for tallow biodiesel. The large range observed is mainly related with the system boundaries They also showed that the soybean biodiesel production is strongly dependent on the use of non-renewable resources in the industrial processing stages, agricultural production and transport (Mourad and Walter, ; Cavalett and Ortega, 10)

30 definition and allocation methods. Rendering (tallow production) generates the largest GHG 1 2 emissions for tallow biodiesel (Prabhu et al., 0; Barber et al., 07; Niederl and Narodoslawsky, ). Since tallow production is considered to have an inelastic supply (Brander et al., 0), 7 GHG emissions from beef production (including LUC) are usually excluded in the studies (Brander 10 et al., 0; Niederl and Narodoslawsky, 0; Prabhu et al., 0). Related to energy balance, 12 according to Barber et al (07), different energy ratios are due in part to different allocation rates and have combined with different rendering energy values. Furthermore, Bruyninckx (10) results 1 17 show that the most critical stages in terms of energy requirements are the transesterification process 1 (7%), farming (%) and slaughtering and rendering (1%) Table 1. Energy output and GHG emission for tallow biodiesel Reference Energy output (per MJ input) GHG emission (g CO 2 eq MJ -1 ) Bruyninckx, 10; Brander et al., to Prabhu et al., Barber et al., Beer et al., Biodiversity 7 One global hot spot of biodiversity is the Brazilian Cerrado (Brazilian savannah), which represents 0 1 about % of the tropical savannahs world-wide (Myers et al., 00). It is the largest neotropical savannah formation in America (Eiten, 172; Furley, 1) and is the second largest biome in 5 Brazil extending over 0 million ha (Batlle-Bayer et al., 10). In the last years, there has been a 7 significant increase in agricultural and cattle production in Brazilian Cerrado. It was estimated that 51 more than half of Cerrado has been transformed into pasture, cash-crop agriculture and other uses in a time-period of only years (Cederberg et al., 0). The development of agricultural 5 5 activities (expansion and intensification) in the region has been rapidly reducing the biodiversity of 5 the ecosystems There are no results for energy requeriments in this study. 2 5

31 Cerrado is the main biome of Central-Western region in Brazil, the most important beef-producing 1 2 region in the country (% of the national beef production in 10). Almost all of this production 5 proceeds from extensive breeding systems, characterized for low animal productivity and low 7 financial payback. These unfavorable indexes reflect the inadequate management of the land-plantanimal system practiced in a large part of the cattle-breeding estates, what consequently leads to degradation of the pastures (Júnior and Vilela, 02). Cultivated pastures cover around % of the Cerrado area (Klink and Machado, 05) and most of these cultivated pastures experience some 1 17 degree of degradation (da Silva et al., 0). Pastures degradation is the most important obstacle to 1 establishing a sustainable cattle breeding in agronomic, economic and environmental terms in 22 Cerrado. Among the factors that explain the degradation of the pasture areas in the region, the low 2 2 soil fertility can be highlighted (Júnior and Vilela, 02) Due to irrigation and soil amelioration techniques, Cerrado also became an important agricultural 2 region for soybean, maize and rice production in addition to its use for cattle breeding. Currently, 0 the region is increasingly threatened by single-crop monoculture plantations mainly for soybean cultivation (Janssen and Rutz, ). In 10, 5% of soybean was produced in Central-west region of Brazil (more 0% than in 0). The expansion of soybean production replaced pasture lands 7 and small farms of varied crops. Flaskerud (0) estimated that the overall expansion of Brazilian 0 1 cropland (also for soybean) will include 51% on former pastureland, % in the Cerrado area and 7% in the Amazon rainforest. 5 Land use change in Cerrado may cause indirect deforestation of the Amazon forest. Shift from 7 small-scale farming and cattle pasture to large-scale soybean monocultures forces farmers and cattle 51 breeders to search for alternative land, which is often in the Amazon area. Valuable areas of the Cerrado need to be protected by future biofuel sustainability schemes (Lucon, 0; Janssen and 5 5 Rutz, )

32 . Water footprint 1 2 The water footprint (WF) of a given product is the volume of freshwater used to produce the 5 product, measured in terms of water volume consumed (evaporated) or polluted over the various 7 steps of the production chain (Hoekstra, 12). The WF of biofuels is highly dependent on 10 feedstock type, geographic region (local climatic, hydrological and soil conditions) and crop (or 12 livestock) management practices (Stone et al., 10; Berndes, 0). The production of energy crops for biofuel production can have substantial impacts on water 1 17 demand, especially if irrigation is used (Jumbe et al., 07; Coelho et al., 12; Gerbens-Leenes et 1 al., 0; Emmenegger et al., ). As conventional production methods, bioenergy feed stocks 22 production can have water quality impacts from fertilizer and pesticide use (Lovett et al., 0; 2 2 Goldemberg et al., 0). Furthermore, converting pastures or woodlands into cropland (both for 2 27 food or bioenergy production) may exacerbate problems such as soil erosion, sedimentation and 2 excess nutrient (nitrogen and phosphorous) runoff into surface waters, and infiltration into 0 groundwater from increased use of fertilizer (FAO, 0). According to Gerbens-Leenes et al. (0), the water footprint of biofuels ranges between 5 (ethanol from sugar beet) and (biodiesel from jatropha curcas) liters per MJ. Biodiesel water 7 footprints are nearly two to four times higher than the water footprint for bioethanol crops, because 0 1 oilseed crops are less water efficient (Gerbens-Leenes et al., 0; Singh and Kumar, ). Gerbens-Leenes et al. (0) calculated a WF for soybean biodiesel of liters per MJ (7 5 liters per liter of biodiesel), according to the authors in terms of the bioenergy sources considered in 7 this study, biodiesel has the bigger WF, followed by ethanol and bioelectricity. WF studies of 51 biodiesel in Brazil are still rare, few of them are indeed representative and this issue still needs to be deepened in the country. A preliminary study for showed that soybean biodiesel water footprint 5 5 varies from 0 to 0 liters per MJ (Seabra et al., ). Another reference in southern Brazil for 5 production of biodiesel indicates that the major contribution to degradative water use is the

33 agricultural phase (.%) while for consumptive use the largest contribution is the oil extraction 1 2 process (7.%) (Muller, 12). No WF studies were found for tallow biodiesel but Gerbens- 5 Leenes et al. () presented the WF of beef production in Brazil (10 to 22 L per kg of 7 beef) Environmental sustainability of biodiesel: key drivers and barriers in Brazil 1 17 Environmental sustainability is a difficult concept to explain, but it can be defined as the 1 maintenance of natural capital (Goodland, ). The two fundamental environmental services 22 (the source and the sink functions) must be maintained unimpaired during the period over which 2 2 sustainability is required. In this context, the environmental sustainability of biodiesel is critically 2 27 related with life cycle impacts associated with air (GHG emissions and others), water and soil, 2 energy balance, biodiversity (FAO, b). Another important issue involves developing strategies 0 to ensure that as the production of biofuels increases, adequate supplies of other needed agricultural and forest-based goods are produced (FAO, b). However, to quantify the environmental sustainability of biodiesel is complex. 7 An increasing number of countries have established initiatives to define sustainability criteria for 0 1 biofuels. For instance, the European Union Directive on the Promotion of Renewable Energy Sources (RED) defined that for biofuels be counted as renewable energy, a minimum GHG saving 5 of % is required by (comparing with fossil fuel). The directive also stipulates no-go zones 7 for feedstock production (e.g. areas where land is deemed to be of high biodiversity value or 51 high carbon stock, wetlands, peatland) (FAO, b). Also, there is an important initiative from Global Bioenergy Partnership (FAO/GBEP, ) defining sustainability indicators for bioenergy 5 5 (FAO/GBEP, ), which are now starting to be evaluated for different countries

34 .1 Land expansion versus agriculture intensification 1 2 The Brazilian annual growth rate of soybean production is estimated in 2.% by 1, close to the 5 global rate estimated (2.%) for the next ten years (MAPA, ). This may be achieved in two 7 ways (Elobeid et al., 10): land expansion or land intensification. Some authors consider that 10 expansion of land area comes with a number of environmental challenges highlighted by the recent 12 debate on direct and indirect LUC brought about by biofuel expansion (Searchinger et al., 0; Fargione et al., 0). On the other hand, there are several studies (Goldemberg et al., 0; 1 17 Goldemberg, 0; Goldemberg and Guardabassi, 0) showing positive results for Brazil and 1 also presenting the benefits for developing countries when sustainable bioenergy production occurs, 22 such as job generation in rural areas and local investments allowing significant development in such 2 2 countries In the case of soybean in Brazil, the increased production comes from additional land (land 2 expansion), rather than from higher yields (intensification). As shown in Fig. 2, soybean cultivated 0 area in /12 was. times greater than the 17/77 area, while the soybean yield in /12 was only 1.5 times greater than in 17/77. UNEP (10) analyzed that the limited potential for the expansion of cultivated lands and the need to increase agricultural production over the next decades 7 leave no alternative than land intensification. In this context, it is essential to gain a better 0 1 understanding of the yield trends and the future yield potential of biofuel feedstocks to help determine the impact of biofuel expansion on agricultural markets (Elobeid et al., 10). A more 5 detailed discussion on this subject is presented in GEA (12), in particular considering the 7 specificities between the different types of biofuels (mainly between bioethanol and biodiesel, 51 which is the focus of this analysis). Brazil has achieved a soybean yield of -000 kg/ha in the last three years (Fig. 2). An 5 5 enhancement of % in /12 since 17/77 can be related to the improvement of agricultural 5 practices. Although future productivity is critical, as it will shape emissions from conversion of

35 native landscapes to food and biodiesel, investment in agricultural research is rarely mentioned as a 1 2 mitigation strategy (Burney et al., 10; Somerville et al., 10). Experience shows that production 5 can indeed be intensified (meaning more production per unit area) whilst reducing inputs and 7 lowering the environmental impacts of agriculture. Intensification and environmental sustainability 10 are not necessarily incompatible (Somerville et al., 10). Burney et al. (10) estimated the net 12 effect of historical agricultural intensification on GHG emissions between and 05 and they found that while emissions from factors such as fertilizer production and application have 1 17 increased, the net effect of higher yields has avoided emissions of up to gigatons of carbon 1 (GtC) since. Their analysis indicates that yield improvements should therefore be prominent 22 among efforts to reduce future GHG emissions. 2 2 Also, genetic improvements can allow the growth of soybean production without excessive land-use 2 27 expansion. The genetically modified (GM) crops mostly have the characteristic of being herbicideresistant or insect-resistant. Since Brazil legalized GM soybean in 05, GM soybean has been 2 0 growing and it represented 5% of the total planted area in /12 (. million ha). In the long term, some authors expect that GMO applications will increase even more in Latin America (Janssen and Rutz, ). The use of GMO for food and biofuel production in Brazil is 7 controversial but there is more acceptance of GMO in Latin America than in Europe, as discussed 0 1 in Elobeid et al. (0) and Oplinger et al. (), as well as van Meijl and von Tongeren (0). Regarding management practices (no-tillage, reduced-tillage or conventional tillage) for soybean 5 cultivation, only few Brazilian producers used tillage systems (Castro et al., 10). In no-tillage 7 systems, the rotation of crops with species that increase plant residues on soil surface is 51 fundamental to avoid erosion and to improve nutrient cycling through nutrient mobilization from deeper soil layers (Crusciol et al., 05). However, it is likely that other factors such as 5 5 environmental stewardship, personal experience, equipment availability, switching costs, farm size,

36 time and labor availability, fuel prices, and government incentive programs will be a far greater 1 2 influence on the decision to no-till than the effect on crop yield (DeFelice et al., 0). 5 Regarding beef production in Brazil (Fig. ), it can be observed that the cattle population has been 7 rising, even with the reduction of grassland. The intensification of livestock activity occurred due to 10 the evolution of the number of cattle heads per unit of area (density): cattle density increased from 1. heads ha -1 in 175 to 1. heads ha -1 in 0 (IBGE, 07), which is still a very low density when compared to the average for developed countries. Goldemberg (0) also indicated possible 1 17 increasing grazing intensities using less land areas in livestock: in Sao Paulo State, cattle density 1 heightened in the last decade, thereby increasing area for food/bioenergy crops. Soares et al. (0) 22 showed that the overall balance on GHG emissions is positive despite the increase in intensive 2 2 animal husbandry and the corresponding replacement of cattle areas by sugarcane crops. This 2 27 intensification in Brazil can also contribute to the availability of large areas for soybean (and other 2 crops) expansion. 0.2 Biodiesel production: methanol vs bioethanol Regarding biodiesel production, it is important to identify opportunities for using bioethanol in the 7 transesterification process. In 10, more than 7% of the biodiesel produced in Brazil used 0 1 methanol to produce fatty acid methyl ester (FAME) (a total consumption of about 02 thousand m ). Most industries in use methylic transesterification for biodiesel production; however, there are 5 two companies used (anhydrous) bioethanol for biodiesel production (ANP, ). For instance, 7 FERTIBOM 7 (in Sao Paulo State) produces biodiesel from different raw materials (vegetable oils 51 and also residual oils) and consider it is economically feasible since they participate in the biodiesel auctions (CENBIO, ). 5 5 The technical and economic success of replacing methanol (fossil origin) by bioethanol to produce 5 biodiesel (resulting in fatty acid ethyl ester (FAEE) instead of FAME) is expected to be a

37 challenging factor for the environmental sustainability of biodiesel, particularly for Brazil. 1 2 Chemically, the methyl and ethyl routes are very similar; however, in practice there are differences 5 between these 2 routes, namely reaction time, catalyst amount and reaction temperature. According 7 to Hamelinck et al. (07), difficulties in the separation phase are a major barrier in ethyl ester 10 production for any feedstock used. 12 Beyond the technical aspects, the main limiting factors for the implementation of ethyl transesterification are the price and availability of bioethanol. Brazil has the cheapest bioethanol 1 17 price in the world but is strongly depend on the geographical location and on fluctuations over the 1 time. On the other hand, methanol prices are relatively constant along time, but witnessed a sharp 22 increase recently. Concerning bioethanol availability, Brazil is the second biggest producer in the 2 2 world with million m (REN, 12) and according to Goldemberg and Guardabassi (0), 2 27 there is still a significant potential for increase the bioethanol first-generation technology 2 production. A preliminary estimative shows that 1.2 million m of bioethanol would be required for 0 the total Brazilian biodiesel production (% of the actual production). Despite the difference of using a renewable (ethanol) and fossil (methanol) source, is important to compare the environmental impacts of two routes. A study performed by CENBIO () 7 compared the production of biodiesel using methyl and ethyl route considering life-cycle 0 1 assessment approach. The difference between the results of two routes, for the various environmental impact categories (e.g. climate change, acidification), is less than 10%. The results 5 showed that there were no significant differences in the environmental impacts of these two 7 technological routes, despite the fossil origin of the methanol imported in Brazil (Coelho et al., 51)

38 . Sustainability tools and initiatives: strength and weakness 1 2 Sustainability assessment of biodiesel is critical at a methodological and practical level because of 5 the critical issues to be assessed: i) agricultural practices (use of fertilizers and pest control 7 techniques, intensive versus extensive farming); ii) competition with food, feed and fiber 10 production for use of water and tillable land; iii) regional market structure; iv) logistics and 12 distribution networks (including biomass transport); v) conservation of biodiversity; vi) cost; and vii) uncertainties resulting from climate change. In this context, it is crucial that the environmental 1 17 impacts are evaluated and quantified in order to provide a rational basis for assessing the long-term 1 viability and acceptability of individual biodiesel supply chain options. To deal with these issues, 22 several initiatives have also been started by governments, industry players and civil society to 2 2 develop criteria for sustainable production of biofuels After the establishment of the Brazilian program, several sustainability initiatives have been 2 established, initially based on strategic, logistic and social issues, mainly focusing on North and 0 Northeast regions for poverty mitigation. After several difficulties to develop local crops such as castor in semi-arid Northeast region, other social sustainability tools where introduced, as discussed here. Taking into account the difficulties to address the production from small farmers due to lack 7 of capacity building and economic factors (Obermaier et al., 10) and considering the shift for 0 1 soy-based biodiesel (large scale farmers in Central-West), the Program has established a Social Fuel Seal - a regulatory instrument designed to promote social inclusion throughout the new fuel 5 production and value chain. The Social Fuel Seal establishes the conditions for industrial producers 7 of biodiesel to obtain tax benefits and credit. In order to receive the seal, an industrial producer 51 must purchase feedstock from family farmers and smallholders enter into a legally binding agreement with them to establish specific income levels and guarantee technical assistance and 5 5 training (MME, ). This program has a positive aspect of giving support to small producers. On 5 another hand, entrepreneurs claim against the high costs of the program. Additionally, since in the

39 auctions the prices paid by Petrobras (around USD L -1, 12 USD exchange rate, ANP, 1 2) are higher than the final price of the blend diesel-biodiesel in the pump stations, Petrobras 5 covers the difference. 7 The Amazon Moratorium was established with the objective to map protected areas, to set up an 10 adequate monitoring system and to work out arrangements for land use in the Amazon area. The 12 Amazon Moratorium is regarded as a valuable initiative to reduce the negative impact of enhanced soy cultivation (for food, feed and biofuels) with respect to the Amazon rainforest (Lucon, 0; 1 17 Janssen and Rutz, ). The Task Force Sustainable Soy is a proponent of the current Amazon 1 Moratorium applied in order to achieve responsible land use in the Amazon biome. The Task Force 22 Sustainable Soy is the platform of a group of Dutch companies involved in soy production and 2 2 marketing. Participants of the Task Force come from the oils and fats, processing, animal feed, meat 2 27 and dairy sectors. The Task Force recognizes the need for more attention to the ecological and 2 social consequences of the expansion of South American soy farming (Janssen and Rutz, ). 0 According to CEO (12), despite the national and international sustainability tools and initiatives, there has been some opposition from social movements and environmental organizations both in Europe and in biofuel producing countries, which has weakened these tools and initiatives 7 legitimacy. In fact, there are also important economic interests (market barriers), but it must be 0 1 noticed that these tools are active in the country. The soybean certification scheme makes sense in the context of the certification of the several products that are produced from soybean; however, the 5 certification schemes only exist for biodiesel and not for oil or meal. This issue is very important 7 also because there should be coherence between the methods to attribute impacts between the 51 various soybean based co-products. The agro-ecological zoning is an important policy tool and has taken into account environmental, 5 5 economic and social aspects as an answer to challenges in sustainable expansion of bioenergy 5 crops. An interesting initiative was launched by the Federal Government to guarantee the

40 sustainable expansion of sugarcane and oil palm in Brazil (two agro-ecological zonings): for sugar 1 2 cane in 0 (EMBRAPA, 0) and for oil palm in 10 (EMBRAPA, 10). In this process, 5 maps were produced showing soils, topography, climate and rainfall. This regulation enables a 7 guidance to credit policies and use for public banks as a condition for production financing. The 10 zoning identified areas where sugar cane crop expansion could take place. It forbids sugar cane 12 cultivation in 2.5% of the national territory. It has identified million hectares (EMBRAPA, 0) that comply with environmental and productivity requirements, mainly from the 1 17 intensification of cattle ranching, which is currently very inefficient (less than 1 head ha -1 ) 1 (Strapasson et al., 12). Such initiatives could be also adopted for soybean expansion, contributing 22 to the sustainability of soybean biodiesel. However here there is the issue of the intensification of 2 2 cattle ranch as an important goal to be achieved both for these zonings and for the sustainability of 2 27 cattle ranch in Brazil (and indirectly to the sustainability of biodiesel from beef tallow) Concluding remarks Biodiesel production in Brazil is an emergent bioenergy for which it is important to guarantee the 7 environmental sustainability. Aiming at contributing for this discussion, this paper characterized the 0 1 biodiesel production chain in Brazil, identified potential environmental impacts and analyzed key drivers and barriers for biodiesel environmental sustainability, focusing on the main biodiesel feed 5 stocks in Brazil: soybean oil and beef tallow. The main environmental impacts of soybean and beef 7 tallow biodiesel chains in Brazil were identified and discussed. Impacts related with land use and 51 land use change, namely GHG emissions and biodiversity, as well as energy intensity and water impacts were found to be critical for biodiesel environmental sustainability. Increasing agriculture 5 5 yields, diversifying feed stocks and adopting ethyl-transesterification process can contribute to 5 minimize the impacts. The impact of bioenergy production on food prices, GHG balance, quantity

41 and quality of water can be mitigated by appropriate policies aiming at an integrated optimization of 1 2 food and bioenergy production. Conflicts between food and biodiesel can be avoided through agroeconomic-ecological zoning for soybean and for beef, similarly to the ones already made for 5 7 sugarcane and oil palm in Brazil (EMBRAPA, 0; 10), allowing adequate use of land for each 10 purpose. Despite the limitation and weakness of some sustainability tools and initiatives, policies, 12 standards and certification schemes also can play an important role on the sustainability assessment and development of the emerging biodiesel production in Brazil. Although Brazilian social and 1 17 environmental legislation applicable to biofuels covers all the aspects pointed out in the certification 1 schemes, it is a fact that in many cases the legislation is not correct accomplished, indicating that a 22 better enforcement is required Acknowledgements 0 The authors are grateful to CNPq (BIOACV Project 7/10-7) for the financial support. The research presented in this article was also supported by the Portuguese Science and Technology Foundation (FCT) projects: MIT/SET/00/0 (Capturing Uncertainty in Biofuels for 7 Transportation. Resolving Environmental Performance and Enabling Improved Use) and 0 1 PTDC/SEN-TRA/71/10 (Extended well-to-wheels assessment of biodiesel for heavy transport vehicles). The authors are grateful to FCT/CAPES for the approved project in 5 : Avaliação da sustentabilidade ambiental de bioenergia através da Avaliação de Ciclo de 7 Vida. Érica Castanheira gratefully acknowledges financial support from FCT, through grant 51 SFRH/BD/0/

42 References 1 2 ABIOVE,. Brasil Complexo Soja: Evolução das Cotações FOB-Brasil. Associação Brasileira das Indústrias de Óleos Vegetais ABIOVE. Available at: (accessed November 10, ). 5 Andrade Filho, M., 07. Aspectos técnicos e econômicos da produção de biodiesel: o caso do sebo bovino como 7 matéria-prima. Dissertação de Mestrado em Energia, Universidade Salvador, Bahia, 1 p. 10 ANP,. Anuário Estatístico Brasileiro do Petróleo, Gás Natural e Biocombustíveis-. Ministério de Minas e 12 Energia, ANP-Agência Nacional do Petróleo, Gás Natural e Biocombustíveis. ISSN-5. ANP, 12. Produção Nacional de Biodiesel Puro - B100 (metros cúbicos) por produtor e estado Ministério 1 de Minas e Energia, ANP-Agência Nacional do Petróleo, Gás Natural e Biocombustíveis, Superintendência de 17 Planejamento e Pesquisa. Available at: (accessed August 10, 12). 1 Barber, A., Campbell, A., Hennessy, W., 07. Primary energy and net greenhouse gas emissions from biodiesel made 22 from New Zealand Tallow. CRL energy report 0-b, report to Energy Efficiency and Conservation Authority 2 (EECA). CRL Energy Ltd, Lower Hutt. 2 2 Batlle-Bayer, L., Batjes, N. H., Bindraban, P. S., 10. Changes in organic carbon stocks upon land use conversion in 27 the Brazilian Cerrado: A review. Agriculture, Ecosystems and Environment 7, Beer, T., Grant, T., Morgan, G., Lapszewicz, J., Anyon, P., Edwards, J., Nelson, P., Watson, H., and Williams, D., 01. Comparison of Transport Fuels. Stage 2 Study of Life-cycle Emissions Analysis of Alternative Fuels for Heavy Vehicles. Australian Greenhouse Office, Final Report EV5A/2/FC. Berndes, G., 0. Future biomass energy supply: The consumptive water use perspective. International Journal of 7 Water Resources Development, 2, Berndes, G., Bird, N.D., Cowie, A., 10. Bioenergy, Land Use Change and Climate Change Mitigation. IEA 0 1 Bioenergy. Available at (acessed December 2, 12). Brander, M., Hutchison, C., Sherrington, C., Ballinger, A., Beswick, C., Baddeley, A., Black, M., Woods, J., Murphy, 5 R., 0. Methodology and Evidence Base on the Indirect Greenhouse Gas Effects of Using Wastes, Residues, and Byproducts for Biofuels and Bioenergy. Report to the Renewable Fuels Agency and the Department for Energy and 7 Climate Change. Report PR A, 0 November Brasil, 05. Lei federal n.07, de de janeiro de 05. Dispõe sobre a introdução do biodiesel na matriz energética brasileira; altera as Leis nos.7, de de agosto de 17,.7, de 2 de outubro de 1 e 10., de 0 de 5 dezembro de 02; e dá outras providências. Diário Oficial da União, Brasília, Seção 1 - /1/05 5 Bringezu, S., Schuetz, H., O Brien, M., Kauppi, L., Howarth, R., McNeely, J., 0. Towards Sustainable Production 5 and Use of Resources: Assessing Biofuels. United Nations Environment Programme, Paris, France, 102 pp

43 Bruyninckx, I., 10. The use of tallow-sourced biodiesel in New Zealand. Master of Engineering Studies. Department 1 of Mechanical Engineering, The University of Auckland, New Zealand, November Bureau, J., Disdier, A., Gauroy, C., Tréguer, D., 10. A Quantitative Assessment of the Determinants of the Net 5 Energy Value of Biofuels. Energy Policy,, Burney, J. A., Davis, S. J., Lobell, D. B., 10. Greenhouse gas mitigation by agricultural intensification. Proceedings of the National Academy of Sciences, 107, Castanheira, É.G., Freire, F., 12. Life-cycle GHG assessment of soybean biodiesel. Proceedings of the International Symposium on Sustainable Systems and Technology. 12 IEEE International Symposium on Sustainable Systems and Technology (ISSST), 5 pages Castanheira, É.G., Freire, F.,. Greenhouse gas assessment of soybean: implications of land use change and 1 different cultivation systems. Journal of Cleaner Production, vol. 5, pp Castro, A. M. G.; Lima, S. M. V.; Silva, J. F. V (Ed.), 10. Complexo Agroindustrial de Biodiesel no Brasil: 2 Competitividade das Cadeias Produtivas de Matérias- Primas. Brasília: Embrapa Agroenergia, p. 2 2 Cavalett, O., Ortega, E., 0. Emergy, nutrients balance and economic assessment of soybean production and 27 industrialization in Brazil. Journal of Cleaner Production 17, Cavalett, O., Ortega, E., 10. Integrated environmental assessment of biodiesel production from soybean in Brazil. Journal of Cleaner Production, -70. Cederberg C., Meyer D., Flysjö A., 0. Life cycle inventory of greenhouse gas emissions and use of land and energy in Brazilian beef production. SIK Report No 72. June 0. 7 CENBIO,. Comparação da ACV de Biodiesel produzido a partir de Óleo de Soja e Gordura Bovina via rota 0 Metílica e Etílica. CENBIO Centro Nacional de Referência em Biomassa. Final Report Project BIOACV nº 17/10-7, financial support the CNPq. São Paulo,. CEO, 12. Roundtable on Responsible Soy - the certification smokescreen. Corporate Europe Observatory (CEO), 5 May 22, 12. Available at: 7 Cerri, C.C., Bernoux, M., Maia, S. M. F., Cerri, C. E. P., Junior, C. C., Feigl, B. J., Frazão, L. A., Mello, F. F.C., Galdos, M. V., Moreira, C.S., Carvalho, J. L. N., 10. Greenhouse gas mitigation options in Brazil for land-use 51 change, livestock and agriculture. Sci. Agric. (Piracicaba, Braz.), 7, Cezar, I. M., Queiroz, H. P., Thiago, L. R., Cassales, F., Costa, F. P., 05. Sistema Produção de Gado de Corte no 5 Brasil: Uma Descrição com Ênfase no Regime Alimentar e no Abate (Brazilian Beef Production Systems: A 5 Description with Emphasis on Animal Nutrition and Slaughtering). Campo Grande, MS: Documentos 1, Embrapa 5 Gado de Corte, 0 p

44 Coelho, S. T. ; Agbenyega, O. ; Agostini, A. ; Erb, K. ; Haberl, H. ; Hoogwijk, M. ; Lal, R. ; Lucon, O. ; Masera, O. ; 1 Moreira, J. R.. Land and Water. Linkages to Bioenergy. In: Global Energy Assessment (Davis, G., Goldemberg, J., 2 orgs). International Institute for Applied Systems Analysis and Cambridge University Press. Viena Coelho, S. T.; Silva, G. A.; Freire, F.; Grisoli, R.; Castanheira, E.; Garcilasso, V. P.; Nogueira, A. A comparative Life 7 Cycle Assessment of biodiesel from soybean oil and beef tallow in Brazil. Energy for Sustainability. Sustainable Cities: Designing for People and the Planet. Coimbra,. 10 CONAB, 12. Levantamento de safras. CONAB - Companhia Nacional de Abastecimento (National Company for 12 Food Supply), Brasília. Available at: (accessed July, 12). Crusciol, C.A.C., Cottica, R.L., Lima, E.V., Andreotti, M., Moro, E., Marcon, E., 05. Persistência de palhada e 1 17 liberação de nutrientes do nabo forrageiro no plantio direto. Pesq. Agropec. Bras., 0, -. 1 Da Silva, J. E., Resck, D. V. S., Corazza, E. J., Vivaldi, L., 0. Carbon storage in clayey Oxisol cultivated pastures in the Cerrado region, Brazil. Agriculture, Ecosystem and Environment, 10, DeFelice, M. S., Carter, P. R., Mitchell, S. B., 0. Influence of tillage on corn and soybean yield in the United States 2 and Canada. Online. Crop Management. Available at: Diaz-Chavez, R.A.,. Assessing biofuels: Aiming for sustainable development or complying with the market? 2 Energy Policy,,. 0 EIA, 12. International Energy Statistics. Prepared by the U.S. Energy Information Administration (EIA). Available at: (accessed December, ). Eiten, G., 172. The Cerrado vegetation of Brazil. Bot. Rev.,, Elobeid, A. E., Tokgoz, S., Yu, T.-H., 0. Assessment of Biofuel Resource Elasticity in APEC Economies. APEC 0 Energy Working Group, APEC# -RE-01., December 0. 1 Elobeid, A. E., Tokgoz, S., Yu, T.-H., 10. Mitigating Land Use Changes From Biofuel Expansion: An Assessment of Biofuel Feedstock Yield Potential in APEC Economies. Agricultural and Applied Economics Association Annual Meeting, July -27, 10, Denver, Colorado. 7 EMBRAPA - Brazilian Agricultural Research Corporation. Agro-ecological Sugarcane Zoning, 0. Available at: (accessed on 2 October 12). 51 EMBRAPA - Brazilian Agricultural Research Corporation. Agro-ecological Palm oil Oil Zoning, 10. Available at: (accessed on 2 October 12). 5 5 Emmenegger, M. F., Pfister, S., Koehler, A., Giovanetti, L. de, Arena, A. P., Zah, R.,. Taking into account water 5 use impacts in the LCA of biofuels: an Argentinean case study. Int J Life Cycle Assess, 1,

45 FAO, 0. The State of Food and Agriculture Biofuels: prospects, risks and opportunities. Food and Agriculture 1 Organization (FAO), Rome, Italy. 2 FAO, a. FAOSTAT - Food Balance Sheets. Available at: 5 FAO, b. Biofuels and the sustainability challenge: A global assessment of sustainability issues, trends and policies 7 for biofuels and related feedstocks. Trade and Markets Division - Food and Agriculture Organization of the United Nations, Rome,. Available at: FAO/GBEP,. Global Bioenergy Partnership - Sustainability Indicators for Bioenergy First Edition. December. ISBN Available at: 1 Fargione, J., Hill, J., Tilman, D., Polasky, S., Hawthorne, P., 0. Land clearing and the biofuel carbon debt. Science 17 :12 1 Flaskerud, G., 0. Brazil s Soybean production and impact. North Dakota State University of Agriculture and 22 Applied Science, NDSU Extension Services, July Fore, S. R., Porter, P., Lazarus, W.,. Net energy balance of small-scale on-farm biodiesel production from canola 2 and soybean. Biomass and Bioenergy,, Furley, P.A., 1. The nature and diversity of neotropical savanna vegetation with particular reference to the Brazilian 2 0 cerrados. Global Ecol. Biogeogr.,, 22-. GEA, 12. Global Energy Assessment - Toward a Sustainable Future. Cambridge University Press, Cambridge, UK and New York, USA and the International Institute for Applied Systems Analysis, Laxenburg, Austria. 7 Gerbens-Leenes, P. W., Mekonnen, M. M., Hoekstra, A. Y.,. A comparative study on the water footprint of poultry, pork and beef in different countries and production systems. Value of Water Res. Rep. Ser. No.. UNESCO- 0 IHE, Delft, the Netherlands. 1 Gerbens-Leenes, W., Hoekstra, A.Y., van der Meer, T. H., 0. The water footprint of Bioenergy. Proceedings of the National Academy of Sciences, 10, Gibbs, H. K., Johnston, M., Foley, J. A., Holloway, T., Monfreda, C., Ramankutty, N., Zaks, D., 0. Carbon payback 7 times for crop-based biofuel expansion in the tropics: the effects of changing yield and technology. Environ. Res. Lett.,, Goldemberg, J., 0. The Brazilian Experience with Biofuels. Innovations Journal,, Goldemberg, J., Guardabassi, P., 0. Are biofuels a feasible option? Energy Policy, 7, Goldemberg, J.; Coelho, S.T.; Guardabassi, P., 0. The sustainability of ethanol production from sugarcane. Energy 5 5 Policy,,

46 Gomes, M., Biondi, A., Brianezi, T., Glass, V., 0. O Brasil dos agrocombustíveis, gordura animal, dendê, algodão, 1 pinhão-manso, girassol e canola: impacto das lavouras sobre a terra, o meio ambiente e a sociedade. São Paulo: 2 Repórter Brasil - Centro de Monitoramento de Agrocombustíveis (CMA), p. 5 Goodland, R. The concept of environmental sustainability. Annual review of ecology and systematics. Vol. 2,. 7 Available at: Governo Federal, 0. Plano Nacional sobre Mudança do Clima (PNMC). Governo Federal-Comitê Interministerial 10 sobre Mudança do Clima, Brasília, DF, Brazil. 2p. Available at: 12 Grisoli, R. Nogueira, A. Castanheira, E. G. Freire, F. Silva, G. A. Coelho, S., 12. Emissões de gases de efeito de estufa no ciclo de vida do biodiesel de soja produzido no Brasil. III Congresso Brasileiro em Gestão do Ciclo de Vida 1 17 de Produtos e Serviços Novos desafios para um planeta sustentável, pp. 2-2, 0 a 0 de setembro de 12, Maringá PR Brasil. ISBN: Hamelinck, C., Schober, S., Mittelbach, M., Verolet, J., Dehue, B., 07. Fatty acid ethyl esters. Final report for Lot a 22 of the Bioscopes project, June Hoekstra, A. Y., 12. The hidden water resource use behind meat and dairy. Animal Frontiers, 2, IBGE, 07. Banco de Dados Séries Estatísticas & Séries Históricas. Rio de Janeiro: IBGE, IBGE,. Banco de Dados Séries Estatísticas & Séries Históricas. Rio de Janeiro: IBGE,. Janssen, R., Rutz, D. D.,. Sustainability of biofuels in Latin America: Risks and opportunities. Energy Policy,, Jumbe, C., Msiska, F., Mhango, L., 07. Report on National Policies on Biofuels Sector Development in Sub-Saharan Africa. Final Working Draft from Food, Agriculture and Natural Resources Policy Analysis Network (FANRPAN) to 0 WIP Renewable Energies, Germany, for Compete Competence Platform on Energy Crop and Agroforestry Systems for 1 Arid and Semi-arid Ecosystems Africa. Júnior, G. B.M., Vilela, L., 02. Pastagens no Cerrado: Baixa Produtividade pelo Uso Limitado de Fertilizantes. 5 EMBRAPA - Empresa Brasileira de Pesquisa Agropecuária - Centro de Pesquisa Agropecuária dos Cerrados, 7 Ministério da Agricultura, Pecuária e Abastecimento. Planaltina, DF, Dezembro, 02. ISSN Kallivroussis L., Natsis, A., Papadakis, G., 02. RD-Rural Development: The Energy Balance of Sunflower 51 Production for Biodiesel in Greece. Biosystems Engineering, 1, 7-. Klink, C.A., Machado, R.B., 05. Conservation of the Brazilian Cerrado. Conservation Biology, 1, Lange, M.,. The GHG balance of biofuels taking into account land use change. Energy Policy,,

47 Levy, G.,. A inserção do sebo bovino na indústria brasileira do biodiesel: análise sob a ótica da Economia dos 1 Custos de Transação e da Teoria dos Custos de Mensuração. Dissertação apresentada para obtenção do título de Mestre 2 em Ciências. Universidade de São Paulo - Escola Superior de Agricultura Luiz de Queiroz, Piracicaba. 5 Lovett, A.A., Sunnenberg, G.M., Richter, G.M., Dailey, A.G., Riche, A.B., Karp, A., 0. Land use implications of 7 increased biomass production identified by GIS-based suitability and yield mapping for Miscanthus in England. BioEnergy Research, 2, Lucon, O., 0. Overview of sustainability assessment tools for biomass production in Latin America. Final Report 12 prepared for the BioTop project supported by the European Commission in the 7th Framework Programme, April 0. Lynd, L.R., Aziz, R.A., de Brito Cruz, C.H., Chimphango, A.F.A., Cortez, L.A.B., Faaij, A., Greene, N., Keller, M., 1 17 Osseweijer, P., Richard, T.L., Sheehan, J., Chugh, A., van der Wielen, L., Woods, J., van Zyl, W.H.,. A global conversation about energy from biomass: the continental conventions of the global sustainable bioenergy project. 1 Interface Focus 1, Malça, J., Freire, F.,0. Renewability and life-cycle energy efficiency of bioethanol and bio-ethyl tertiary butyl ether 2 2 (bioetbe): assessing the implications of allocation. Energy The International Journal,, Malça, J., Freire, F., 0. Energy and Environmental Benefits of Rapeseed Oil Replacing Diesel. International Journal 27 of Green Energy., Malça, J., Freire, F., 10. Uncertainty Analysis in Biofuel Systems: An Application to the Life Cycle of Rapeseed Oil. Journal of Industrial Ecology,, 2-. Malça, J., Freire, F.,. Life-cycle studies of biodiesel in Europe: A review addressing the variability of results and modeling issues. Renewable and Sustainable Energy Reviews,, 1. 7 MAPA,. Soja Mercado interno. Ministério da Agricultura, Pecuária e Abastecimento MAPA. Available: 0 (accessed July 10, 12). 1 Menichetti E., Otto, M., 0. Energy balance and greenhouse gas emissions of biofuels from a life-cycle perspective. In: R.W. Howarth and S. Bringezu (eds) Biofuels: Environmental Consequences and Interactions with Changing Land 5 Use. Proceedings of the Scientific Committee on Problems of the Environment (SCOPE). International Biofuels Project 7 Rapid Assessment, 22- September 0, Gummersbach Germany. Cornell University, Ithaca NY, USA. Milà i Canals, L., Bauer, C., Depestele, J., Dubreuil, A., Knuchel, R. F., Gaillard, G., Michelsen, O., Müller-Wenk, R., 51 Rydgren, B., 07. Key Elements in a Framework for Land Use Impact Assessment within LCA. Int J LCA, 12, MME, 07. Plano Nacional de Energia 0. Ministério de Minas e Energia, Brasília, DF, Brazil. 12v. 5 MME, 12. Boletim Mensal dos Combustíveis Renováveis. Ministério de Minas e Energia, Secretaria de Petróleo, Gás 5 Natural e Combustíveis Renováveis, Departamento de Combustíveis Renováveis. Edição n.º, Janeiro/

48 MME,. Portal do Programa Nacional de Produção e Uso do Biodiesel. Available at: 1 (accessed March 22, ). 2 Moraes, M.S.A., Krause, L.C., da Cunha, M.E., Faccini, C.S., de Menezes, E.W., Veses, R.C., Rodrigues, M.R.A., 5 Caramão, E.B., 0. Tallow biodiesel: properties evaluation and consumption tests in diesel engine. Energy Fuels, 22, 71-. Mourad, A. L., Walter, A.,. The energy balance of soybean biodiesel in Brazil: a case article. Biofuels, 10 Bioproducts and Biorefinery, 5, Muller, G.T., 12. Emprego da pegada hídrica e da análise de ciclo de vida para a avaliação do uso da água na cadeia produtiva do biodiesel de soja. Dissertação de mestrado. Programa de Pós-graduação em Recursos Hídricos e 1 17 Saneamento ambiental da Universidade Federal do Rio Grande do Sul. 1 Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B., Kent, J., 00. Biodiversity hotspots for conservation priorities. Nature, 0, NAE, 05. Cadernos NAE: Processos estratégicos de longo prazo Biocombustíveis. NAE: Núcleo de Assuntos 2 Estratégicos da Presidência da República, nº. 2 (jan. 05). Brasília, 05. ISSN Nassar, A.M., Antoniazzi, L. B., Moreira, M. R., Chiodi, L., Harfuch, L., 10. An allocation methodology to assess 2 GHG emissions associated with land use change Final Report. ICONE (Institute for International Trade Negotiations). 0 Niederl, A., Narodoslawsky. M., 0. Ecological Evaluation of Processes Based on By-Products or Waste from Agriculture: Life Cycle Assessment of Biodiesel from Tallow and Used Vegetable Oil. In: Bozell, J., et al., Feedstocks for the future; ACS Symposium Series ; American Chemical Society: Washington, DC, 0, p 2. 7 Obermaier, M.; Herrera, S.; Rovere, E.L., 10. Análise de problemas estruturais da inclusão da agricultura familiar na cadeia produtiva de biodiesel. IV Congresso Brasileiro de Mamona e I Simpósio de Oleaginosas Energéticas, João 0 Pessoa, PB. 1 Oplinger, E.S. M. J., Martinka, J. M., Gaska, K. A., Schmitz, C. R.,. Wisconsin Soybean Variety Test Results. A5, Department of Agronomy and Plant Pathology, University of Wisconsin-Madison. Available at Padula, A. D., Santos, M. S., Ferreira, L., Borenstein, D., 12. The emergence of the biodiesel industry in Brazil: Current figures and future prospects. Energy Policy,, Panichelli, L., Dauriat, A., Gnansounou, E., 0. Life cycle assessment of soybean-based biodiesel in Argentina for 5 export. Int J Life Cycle Assess,, -. 5 Peres, S., 10. Aproveitamento de Resíduos de Animais para Produção de Biodiesel. II Seminário Biodiesel Fonte 5 de Energia das Oleaginosas em Pernambuco. Recife, - Maio de

49 Ponsioen, T.C., Blonk, T.J., 12. Calculating land use change in carbon footprints of agricultural products as an 1 impact of current land use. Journal of Cleaner Production, Prabhu, A., Pham, C., Glabe, A., 0. Detailed California-Modified GREET Pathway for Renewable Diesel Produced 5 in California from Tallow (U. S. Sourced). Alternative Fuels Section and Fuels Section of the California Air Resources 7 Board. Stationary Source Division Release Date: July, 0 Version: 1.0. Pradhan, A., Shrestha, D. S., Van Gerpen, J., Duffield, J., 0. The energy balance of soybean oil biodiesel 10 production: a review of past studies. American Society of Agricultural and Biological Engineers 51, Prudêncio da Silva, V., van der Werf, H. M. G., Spies, A., Soares, S. R., 10. Variability in environmental impacts of Brazilian soybean according to crop production and transport scenarios. Journal of Environmental Management 1, Ravindranath, N.H., Sita Lakshmi, C., Manuvie, R., Balachandra, P.,. Biofuel production and implications for land use, food production and environment in India. Energy Policy,, Reijnders, L., Huijbregts, M., 0. Biogenic greenhouse gas emissions linked to the life cycles of biodiesel derived 2 from European rapeseed and Brazilian soybeans. JCleanProd, 1(), REN, 12. Renewables 12 - Global Status Report. Renewable Energy Policy Network for the st century. 2 Available at: (accessed July 10, 12). 0 Santos, O. I. B., Rathmann, R., 0. Identification and analysis of local and regional impacts from the introduction of biodiesel production in the state of Piauí. Energy Policy, 7, 0 0. Schaffel, S. B., La Rovere, E. L., 10. The quest for eco-social efficiency in biofuels production in Brazil. Journal of 7 Cleaner Production,, Seabra, J., Hernandes, T., Galdos, M.,. Water Footprint and Water Consumption for the Main Crops and Biofuels 1 Produced in Brazil. American Geophysical Union, Fall Meeting, abstract #HM-05. Searchinger, T., Heimlich, R., Houghton, R. A., Dong, F., Elobeid, A., Fabiosa, J., Tokgoz, S., Hayes, D., Yu, T.-H.,. Use of U.S. Croplands for Biofuels Increases Greenhouse Gases Through Emissions from Land-Use Change. 7 Science,, Singh, S., Kumar, A.,. Development of water requirement factors for biomass conversion pathway. Bioresource 51 Technology, 102, 1-. Soares, L.H.B., Alves, B.J.R., Urquiaga, S., Boddey, R.M., 0. Mitigação das Emissões de Gases de Efeito Estufa 5 pelo Uso de Etanol de Cana-de-açúcar Produzido no Brasil. Circular Técnica 27. Embrapa, Seropédia, Rio de Janeiro. 5 ISSN

50 Soimakallio S., Makinen T., Ekholma T., Pahkala K., Mikkola H., Paappanen T., 0. Greenhouse gas balances of 1 transportation biofuels, electricity and heat generation in Finland: dealing with the uncertainties. Energy Policy, 7, Somerville, C.,Youngs, H., Taylor C., Davis, S. C., Long, S. P., 10. Feedstocks for Lignocellulosic Biofuels. Science 7 (),, Stone, K.C., Hunt, P.G., Cantrell, K.B., Ro, K.S., 10. The potential impacts of biomass feedstock production on 10 water resource availability. Bioresource Technology, 101, Strapasson AB, Ramalho-Filho A, Ferreira D, Vieira JNS, Job LCMA (12) Agro-ecological zoning and biofuels: the Brazilian experience and the potential application in Africa. In: Johnson FX, Seebaluck V (eds) Bioenergy for 1 17 sustainable development and international competitiveness - the role of sugar cane in Africa. Routledge, London, UNEP, 10. Sustainable agriculture and the sustainable use of agricultural biodiversity: Concepts, Trends and challenges. Fourteenth meeting Nairobi, 10- May 10. Available at: van Meijl, H., von Tongeren, F., 0. International Diffusion of Gains from Biotechnology and the European Union s 2 Common Agricultural Policy. Agricultural Economics,, Walter, A., Dolzan, P., Quilodrán, O., de Oliveira, J. G., da Silva, C., Piacente, F., Segerstedt, A.,. Sustainability 2 assessment of bio-ethanol production in Brazil considering land use change, GHG emissions and socio-economic 0 aspects. Energy Policy,, 0-1. Witcover, J., Yeh, S., Sperling, D.,. Policy options to address global land use change from biofuels. Energy Policy, 5, 7. 7 Yang, H., Zhou, Y., Liu J., 0. Land and water requirements of biofuel and implications for food supply and the environment in China. Energy Policy, 7,

51 Appendix A. Cattle slaughter and beef tallow production, consumption (and uses), import and export in Brazil Table A1. Cattle slaughter and beef tallow production, consumption (and uses), import and export in Brazil N NE CW SE S Total N NE CW SE S Total % % % 7% % % 2% 1% 0% 1% 1% 0% % 5% (t) % 2% 0% 0% 0% 5% % 1% % % 2% % 0% 1% (t) Total Cleaning&hygiene 270 Soap production 22 Chemical 7 Biodiesel 0 27 Animal Feed 72 Boilers (as fuel) Others * Data from IBGE (). Cattle slaughter* (nº heads) Beef tallow production # (t) Beef tallow import Beef tallow export National beef tallow use + (t) # Estimation based on cattle slaughter and assuming a yield of kg tallow/cattle slaughtered (Levy, ; Peres, 10). + Estimation based on the beef tallow production plus imports and minus exports (FAO, a). It was considered that no tallow was imported and exported in Brazil in 10 and due to the lack of data and taking into account that these percentages are very low (less than 5%) in the last years. Estimation based on the percentage of different uses in 0 (Andrade Filho, 07), 07 (Gomes et al., 0) and 0-10 (ANP, ).

52 Table1 Table 1. Energy output and GHG emission for tallow biodiesel. Reference Energy output (per MJ input) GHG emission (g CO 2 eq MJ -1 ) Bruyninckx, 10; Brander et al., to 7.2 Prabhu et al., Barber et al., Beer et al., There are no results for energy requeriments in this study.

53 TableA.1 Table A1. Cattle slaughter and beef tallow production, consumption (and uses), import and export in Brazil N NE CW SE S Total N NE CW SE S Total % % % 7% % % 2% 1% 0% 1% 1% 0% % 5% (t) % 2% 0% 0% 0% 5% % 1% % % 2% % 0% 1% (t) Total Cleaning&hygiene 270 Soap production 22 Chemical 7 Biodiesel 0 27 Animal Feed 72 Boilers (as fuel) Others * Data from IBGE (). Cattle slaughter* (nº heads) Beef tallow production # (t) Beef tallow import Beef tallow export National beef tallow use + (t) # Estimation based on cattle slaughter and assuming a yield of kg tallow/cattle slaughtered (Levy, ; Peres, 10). + Estimation based on the beef tallow production plus imports and minus exports (FAO, a). It was considered that no tallow was imported and exported in Brazil in 10 and due to the lack of data and taking into account that these percentages are very low (less than 5%) in the last years. Estimation based on the percentage of different uses in 0 (Andrade Filho, 07), 07 (Gomes et al., 0) and 0-10 (ANP, ).

54 Figure1

55 Figure2