Report: Brazilian Ethanol is Sustainable

For those who are expecting a Brazilian debunking, I am going to have to disappoint you. My previous debunking was not addressed at the issue of whether Brazilian ethanol is sustainable, but rather whether their example can be exported to the U.S. Whenever the topic of Brazilian sugarcane ethanol has come up, my response is generally that from what I have read, it appears to be a pretty good deal. Furthermore, I have never seen evidence to dispute the high EROEI claims of sugarcane ethanol. However, I will usually note that there are few comprehensive reports that have examined the process in detail, and I would feel more comfortable about the positive assessments if someone did such a study.

My wish has been granted. IEA Bioenergy has recently publicized a report entitled “Sustainability of Brazilian bio-ethanol”. The report was commissioned by The Netherlands Agency for Sustainable Development and Innovation, and is in my opinion the most important endorsement of Brazilian ethanol to date. The work was conducted by the Copernicus Institute at Utrecht University in the Netherlands and at the University of Campinas in Brazil. The 136-page report is publicly available here (1.2 meg PDF).

The purpose of the study is explained in the introduction to the report:

The Dutch society recognizes the need for sustainable production and use of biomass. This has been expressed by environmental groups and the Parliament. The Dutch government decided to seek solutions by developing sustainability criteria and certification of biomass by a commission sustainable production of biomass. Between January 2006 and July 2006 these criteria have been developed. Parallel to these developments, in February 2006 this project was commissioned by SenterNovem on behalf of the Dutch Ministry for Housing, Spatial Planning and the Environment. As Brazil is one of the most likely countries to export bio-ethanol from sugar cane to the Netherlands, the sustainability of Brazilian bio-ethanol is the main topic of this report.

The main objective of this report was a comparison of Dutch sustainability criteria and the current Brazilian practice, and quantification of the consequences for ethanol production in terms of production method and production costs if these sustainability criteria are applied. To this end, the Dutch sustainability demands for bio-ethanol were investigated, including stakeholder consultation in the Netherlands, and an extensive assessment of the current ecological, economic and social impacts of ethanol production based on sugar cane in Brazil was carried out.

This is precisely the kind of study that has been needed to verify that claims of Brazilian ethanol sustainability are on sound scientific footing. According to the report, there are some areas of concern, but none that should prevent Brazilian ethanol from meeting the sustainability criteria:

While the current study contains many different types of uncertainties, no prohibitive reasons where identified why ethanol from São Paulo principally could not meet the Dutch sustainability standards set for 2007. In many impact categories, Brazilian ethanol from sugar cane scores average to (very) positive, see also Table I for a summary. For a number of other criteria, problems are identified, but it also appears that these may differ strongly between the individual plants. Furthermore, for most of these issues, measures can be identified to improve performance (when needed).

For the future and the whole of Brazil, too many uncertainties remain to determine whether also additional criteria from 2011 onwards can be met. First of all, it is as yet unclear how additional land use for sugar cane may cause indirect / induced land-use, and how large the actual impacts will be on land use, biodiversity etc. Second, it is also uncertain whether and how the Dutch sustainability criteria will deal with these indirect impacts, as these criteria are not yet clearly defined.

It is important to recognize that sustainability criteria lead to higher production costs – depending on the strictness of the sustainability criteria, we estimate additional ethanol costs of up to 56%, though in case mechanical green harvesting is applied, additional ethanol costs are estimated at 24%.While the latter may not necessarily be prohibitive given current oil prices, it is clear that some financial support is most likely needed to stimulate sustainable ethanol production.

The report examined a number of sustainability criteria. However, in this essay I will mention only two: EROEI and soil erosion.

What’s the EROEI?

One the issue of sugarcane ethanol EROEI, which has been debated here a great deal, the study mentioned two different literature reports. The first was by Oliveira et al. in 2005, and it concluded that the EROEI was between 3.1 and 3.9. The second report was by Macedo et al. in 2004 and it concluded that the EROEI was between 8.3 and 10.2. (Note that the “bad” EROEI was still over double the EROEI of corn ethanol.) Due to the huge disparity between the two papers, the authors took a look at the underlying numbers, and concluded that the discrepancy involved the amount of diesel used in the agricultural operations process. They ultimately tracked down another paper that agreed with the Macedo study, so they reasoned that the diesel consumption numbers used by Oliviera were erroneous. They therefore concluded that an EROEI between 8.3 and 10.2 was legitimate.

Not surprisingly, the greenhouse gas (GHG) emission reduction for sugarcane ethanol was estimated to be >80%. EROEI and GHG emissions are very closely related, such that a renewable energy source possessing a high EROEI should demonstrate a high level of GHG emission reduction.

Soil Erosion

One area that did not fare as well as sugarcane ethanol advocates have often advertised is on the issue of soil erosion. I have been told a number of times that there is no erosion from sugarcane production, or that production is managed such that the topsoil actually increases over time.

While the report noted that the erosion is lower than for crops such as corn, it did note:

Soil erosion in sugar cane is generally limited compared to conventional agricultural crops such as corn and soybeans, although the exact difference is dependant on local conditions. However, soil losses for sugar cane may vary dramatically from 0.1 t/ha/y to 109 t/ha/y, depending on many factors, such as the declivity, the annual rain fall, the management and harvesting system, etc.

The report did state that data on erosion from sugarcane production was limited, and that there were some studies that suggested little to no erosion. The report also indicated that the erosion issue should be studied more closely, and that a soil erosion management plan is required. They state that new compliance criteria are to be developed for 2011. In concluding the section on soil erosion, they state:

Soil erosion during sugar cane production can be a site-specific problem. Soil erosion rates under sugar cane production are limited compared to conventional cropland, but are likely higher compared to pastures. Data on soil erosion rates under various land use types are however uncertain. Soil erosion can be prevented in various ways, although it cannot be avoided completely. Consequently, only in case very strict soil erosion rates are applied (which goes beyond the approach applied in existing certification systems and guidelines) soil erosion could be an important bottleneck for certification. As far as soil erosion can be prevented, the costs are likely in the order of magnitude of a few percent of the production costs of ethanol. We conclude that soil erosion can be regarded in general as having a medium impact factor on soil erosion.

My guess is that the last sentence in that section was supposed to read “We conclude that soil erosion can be regarded in general as having a medium impact factor on the sustainability criteria.”

Implications for Tropical Countries

Based on the findings in the report, it suggests that many tropical countries have the potential for sustainable fuel production. This should be particularly true of any country that can grow excess sugarcane according to Brazil’s methods. The major caveat is that the soil erosion issue must be appropriately managed. Not only would this help certain countries achieve some level of energy security, but excess fuel produced for export would open up new opportunities for farmers, factory workers, etc. and generate income for the country.

Implications for the U.S.

The reason the Netherlands commissioned this study is that they want to transition to fuels that are produced in a sustainable manner. If Brazil or other tropical countries can produce enough fuel for export, it will benefit the U.S. just like it will benefit the Netherlands. However, the U.S. does have an import tariff in place that penalizes Brazilian ethanol in order to protect (unsustainable) homegrown corn ethanol.

One way the Brazilian example does not benefit the U.S. is in providing a template for success. As I have argued previously, Brazil’s particular situation is not applicable in the U.S. As I wrote in an article for World Energy Source, the U.S. uses 7 times the energy per capita that Brazil does. Our supply/demand imbalance gap is 16.9 barrels per person per year. Theirs was 0.2 barrels per person last year, prior to the opening of a new Petrobras platform earlier in the year (immediately after which they declared energy independence). Furthermore, we rely on a crop (corn) in the U.S. that is much less energy efficient, and has ten times the soil erosion of sugarcane production. Finally, we are not in a tropical climate, and therefore have a much shorter growing season than does Brazil.

Brazilian ethanol expert Milton Maciel, has echoed these arguments:

Sugar cane ethanol from Brazil is NOT a realistic target or a comparable model for USA ethanol from corn. It is very easy to replace all gasoline when you would only need 8 billion gallons per year and you have a generous plant that thrives rain-fed under tropical conditions, occupying less than 1% of a country’s arable land, to produce alcohol to replace 50% of all that gasoline. However, this cannot be extrapolated for USA’s conditions, neither for corn, not even for sugar cane in Southern states. So, realistically, let’s understand that sugar cane ethanol in Brazil is mangoes and corn ethanol in USA is apples.


The Brazilian example does suggest some avenues ripe for exploration in the U.S. There are certain crops that are far less erosive than others. According to Table K.1 in the report, soil erosion for sugarcane and corn was 1.24 and 12.0 (t/ha/y) respectively. Note that erosion from corn is 10 times the erosion from sugarcane. Another listed crop, potatoes, had about half the erosion level of corn. However, the level of erosion for potatoes is still not sustainable, so we need to look to other crops if we are to maintain the integrity of our topsoil.

Imagine a couple of scenarios. First, imagine a variety of sugarcane that is bred/engineered to withstand more temperate climates. Imagine the Midwest covered in sugarcane instead of corn, and we would have taken a big step toward sustainability. Alternatively, imagine a plant that currently thrives in the Midwest, does not contribute to soil erosion, and produces sugar (easily converted to ethanol). I could envision something like Miscanthus, with an engineered gene(s) that allows it to produce sugar. If you have ever seen the ancestor of corn – teosinte – and compared it to modern corn, this idea does not seem out of the question. Again, if we could pull something like that off in the U.S., it could offset some of the decline in conventional oil production without exacting the high environmental price of alternatives like coal-to-liquids.

Selective Breeding Turned Teosinte into Modern Corn


The sustainability of Brazilian ethanol appears for the most part to be as advertised. The indications are that the EROEI is at least 8.3, which would actually make it better than for gasoline. (Incidentally, that brings to mind the question of why Brazil would want to export any ethanol; I would use my ethanol internally and would export my oil instead). The sustainability criteria that were used in this study are an example of what we should be doing for all of our fuel sources, and we should encourage those that meet these standards.

20 thoughts on “Report: Brazilian Ethanol is Sustainable”

  1. Robert, thanks for posting this information. Our debate last week kept hanging on corn-based ethanol even though I wanted to talk about ethanol production in general. It is obvious from this study that sugar based ethanol the route to take instead of grain based. There are plenty of plant species that can be grown in the US that have high sugar value. This includes sugar beets, sweet sorghum (which is availble in my state of Texas), Jerusalem artichokes, fodder beets, and sweet potato.

    This is an old study , but gives some insight on sugar production options. The study focused on Southeast US. I would expect that 15 years later the yeilds would be improved in both crop productions and ethanol extraction.

    There are practices that can be put into place to minimize soil erosion and limit the amount of fertilizer that is required for these species. I hope we will see more research going into this segment of the ethanol industry.

    Now where is Gary with his catchy connect the output to the input jingle?


  2. Good-bye rainforests, unfortunately.

    That is one of the sustainability criteria, and current production is not near the rainforests. Your concern is legitimate, though. If the rest of the world is not as concerned about sustainability as are the Dutch, production may expand toward the rainforests.



  3. Eric said, “Now where is Gary with his catchy connect the output to the input jingle?”

    Right here Eric,

    My problem is not with ethanol as a fuel, but with the sustainability of corn-based ethanol, and why corn ethanol has gained so much momentum in the U.S. (Soemething President Bush echoed today in his address at St Louis when he said we need to find an alternative to corn as a feedstock for ethanol.)

    If the Dutch have made a compelling case that cane-based ethanol is sustainable that’s good enough for me. (For the record, my grandparents came from the Netherlands.)

    But I do suspect that if they were to study corn ethanol using the same methods, they wouldn’t find corn ethanol as made in the U.S. to be sustainable. As we discussed in the other thread, growing corn to convert it into ethanol is a gross example of making a liquid fuel by consuming “non-renewables.” (Those non-renewables include not only fossil fuels, but the fertile soil the American Heartland was endowed with and perhaps our most precious resource — water.) The hard truth is that corn ethanol in the U.S. is not really about renewable fuels, but is instead a direct result of national farm policy.

    Based on the Dutch study, it seems there is a very good chance that cane-based ethanol could pass the “connect the output to the input test,” something our corn ethanol industry could not do.

    My only concern about the Brazilian cane ethanol model is the dependence of their agriculture on dirt cheap, manual labor. Making cane ethanol is more efficient than corn ethanol, but part of that reason is the ready availability of inexpensive, manual peasant labor — something we don’t especially want in the U.S. (I’m sure you can understand why our corn farmers are reluctant to wade into their corn fields swinging machetes, and instead prefer driving diesel fuel-hungry tractors and corn pickers.)

    BBM said, “Good-bye rainforests, unfortunately.”

    That is a concern BBM, and would have to be part of any sustainability study. Fortunately, the Brazilian cane fields are not in the Amazon rain forest, although some are in their Atlantic coastal rain forests. However, a rain forest is a rain forest.

    Best Regards,

    Gary Dikkers

  4. I expect that these developments (which are good news certainly) will put enormous pressure on Brazilian markets to expand production markedly so that ethanol can be exported to the US and Europe in effect becoming a “Saudi Arabia” for EtOH.

    This would be a welcome development in many ways (reducing mid east geopolitical leverage, etc), but I’m sure that it will eventually place further pressure on ecologically sensitive areas.

  5. bbm,
    the production of sugarcane is far away from the rainforest.
    The amazon basin soil is not good for sugarcane production. Also there is a lot of rain in that region (about 4.000mm of rainfall/year compared to about 1.500mm/year in Sao Paulo) so the sugarcane cannot mature, reducing drasticaly the sugar content.

    And Gary, your concerns about labor conditions are going to diminish as there is a federal law that all sugarcane must be cut by machines at the most in 2020. Some states as Sao Paulo have state laws that prohibited the burning of sugarcane near cities, powerlines, roads,… therefore reducing the area that mills can use manual labor.


  6. Robert,

    I came to this story from a link at EB,

    Right underneath your story was one about some guys in Australia who have developed a non-GM yeast strain that can digest xylose. I don’t really understand biochemistry all that well yet (though I will next semester), but how might their discovery affect A) corn and sugar cane EtOH yields B) energy available from bagasse (would more coal/NG need to be used in Brazil?) C) the industry in general.

    Thanks for your excellent writing

    David Huck

  7. but part of that reason is the ready availability of inexpensive, manual peasant labor

    Then they should use the food required to feed the peasants as part of the EROEI equation. It would be interesting to see the effects of the balance.

  8. how might their discovery affect A) corn and sugar cane EtOH yields B) energy available from bagasse (would more coal/NG need to be used in Brazil?) C) the industry in general.


    The sugars in corn and sugar cane are already 100% fermentible with normal yeast. However, the sugars you get from processing biomass is not. Cellulosic biomass yields a significant fraction of xylose, which those yeasts you mention can attack. So, they could significantly boost cellulosic ethanol yields.



  9. I understand that rainforest territory is currently not used for sugarcane production, but if EtOH production is scaled up by a factor of 10 or more (to supply foreign markets) I think that it is likely to displace some food production that may then relocate to rainforest territory.

    If not, then great.

  10. ok, only 1% of the countries arable land is used for sugar cane and that supplies 50% of brazils oil. Tad Patzek says we use 14x as much oil (you say 7x as much ENERGY per capita). Since there are 2x as many americans we can assume it will take 2×14 = 28 times as much land to supply 50% of the US’s oil. Which means if 30% of the brazils rainforest gets cut down we can substitute 50% of the US’s oil with brazil ethanol?

  11. heh I understand that my “back of the envolope” calculations are probably horribly flawed. I guess I need to find how much is considered arable land and what % of that is rainforest. But am I missing anything else?

  12. Thanks for an excellent post, Robert. This kind of sustainability analysis is exactly the kind of analysis that needs to be applied to all kinds of biofuels, so that we don’t inadvertendlty trade the evil we know (oil), for the evil we don’t (massive soil erosion, nitrification of our waterways, etc. etc.).

    You wrote, “Incidentally, that brings to mind the question of why Brazil would want to export any ethanol; I would use my ethanol internally and would export my oil instead

    Brazil would want to export ethanol, rather than oil, if it is able to sell the ethanol at a sufficient price permium relative to oil. I would imagine that those, like the Dutch, who are focused on purchasing a sustainable fuel would be willing to pay more for that fuel than for oil, justifying Brazil’s export of ethanol instead of oil.

    Brazil is going to have to do some interesting cost-benefit calculations to maximize the profit they can generate from their domestic fuel resources – both sustainable/renewable and otherwise.


    Jesse Jenkins

  13. Expanding Brazilian sugarcane production in the 1990’s displaced crop food production and grazing on the same land. That food production moved elsewhere, including into former rain forest. So yes, while technically not growing sugarcane in the rain forest, increasing demand for sugarcane IS putting pressure on these areas. This has been documented.
    Ready, arable, cultivatable land is finite.

  14. Yet another piece of evidence for non-sustainable:
    “Brazil’s ethanol push could eat away at Amazon –
    U.N. official voices his concerns ahead of energy visit by Bush”

    Basically, what I’ve said previously is now being documented. The drive for sugarcane production in Brazil is forcing other agricultural commodities, such as soybean production discussed in this article, further into what was once rainforest.

    Ready, arable, cultivatable land is finite.

  15. Apparently your readers do look for information, instead of just hopping on the ‘hit-ethanol’ bandwagon. But what is published and marketed worldwide is not necessarily the whole story.

    Amazon forest 1 – Sugar cane (anywhere in the world) requires a dry period for maturation. Rainfall in the Amazon region is much too high all year round. Not the place to plant if you want sugar content! As President Lula joked recently: if cane was profitable in the Amazon, the Portuguese would have planted it there many centuries ago!

    Amazon forest 2 – As with drugs, the economics of deforestation is demand-driven by the rich. The main driver for forest cutting is lumber. And this lumber trade exists and is triving because of increased demand from the US and Europe for hard woods. Here we are again penalizing the supply side and patting consumers on the head.

    Amazon forest 3 – The Amazon forest is not the legal Amazon region, which encompasses since the 70s (for fiscal and local politics reasons) many states, such as Maranhão and Piaui, which are as Amazonic as a beach in Bahia. Many arguments speak of cane in these states as having replaced forest – pure ignorance.

    Amazon forest 4 – Growth of cane area has not occured at the forest`s expense. Just look at the map. New areas today are mostly in middle Brazil (Minas, Goiás and Mato Grosso) and even in São Paulo.

    Amazon forest 5 – For the last 30 years, Brazilian volume of cane per hectare, as well as sugar content per ton of cane have been increasing at astounding rates.

    Amazon forest 6 – The main macroethical debate is how to justify or penalize what the rich countries did to their forests in the last centuries versus what the poor and not-so-poor should do today and in the future. This was one of the background points for Kyoto, and has been systematically eliminated from most commentaries.

    As you see, there`s always more than meets the eye.

    Heitor Moura, from Rio de Janeiro

  16. Part of the problem people are concerned with is that new cane plantations (for example in the state of Mato Grosso do Sul, where I now live) will drive out sustainable extensive cattle ranching to newly burned down forest regions. On the other hand, in the US, huge areas of farmland are planted with corn. Why not move ethanol production to Brazil and move extensive cattle ranching to the US. Economically and environment-wise it should make sense…

Comments are closed.