Perspectives

ENVIRONMENTAL SCIENCE:
How Green Are Biofuels?

Global warming and escalating petroleum costs are creating an urgent need to find ecologically friendly fuels. Biofuels--such as ethanol from corn (maize) and sugarcane--have been increasingly heralded as a possible savior (1, 2). But others have argued that biofuels will consume vast swaths of farmland and native habitats, drive up food prices, and result in little reduction in greenhouse-gas emissions (3-5). An innovative study by Zah et al. (6), commissioned by the Swiss government, could help to resolve this debate by providing a detailed assessment of the environmental costs and benefits of different transport biofuels.

To date, most efforts to evaluate different biofuel crops have focused on their merits for reducing greenhouse-gas emissions or fossil fuel use. Some studies suggest that corn-derived ethanol in the United States (7) and Europe (8) consumes more energy than it produces; others suggest a modest net benefit (2). Relative to petroleum, nearly all biofuels diminish greenhouse-gas emissions, although crops such as switchgrass easily outperform corn and soy (9). Such comparisons are sensitive to assumptions about local growing conditions and crop by-products, but even more important, their focus on greenhouse gases and energy use is too narrow.

The arguments that support one biofuel crop over another can easily change when one considers their full environmental effects. A key factor affecting biofuel efficacy is whether native ecosystems are destroyed to produce the biofuels. For example, regardless of how effective sugarcane is for producing ethanol, its benefits quickly diminish if carbon-rich tropical forests are being razed to make the sugarcane fields, thereby causing vast greenhouse-gas emission increases (4). Such comparisons become even more lopsided if the full environmental benefits of tropical forests--for example, for biodiversity conservation, hydrological functioning, and soil protection--are included (10, 11).

Another environmental cost that varies among biofuels is trace-gas emissions. For example, crops that require nitrogen fertilizers, such as corn or rapeseed, can be a significant source of nitrous oxide, an important greenhouse gas that also destroys stratospheric ozone. When nitrous oxide emissions are compared among ethanol-producing crops, grasses and woody coppice become more favorable, whereas corn or canola may be worse for global warming than simply burning fossil fuels (3).

In the debate about different biofuels, one can easily be overwhelmed by the "apples and oranges" problem: Each biofuel has certain benefits and potential costs, and there is no common currency for comparing them. This is where Zah et al. have broken new ground by devising a conceptual scheme to evaluate different biofuels using just two criteria: greenhouse-gas emissions and overall environmental impact.

The authors compare gasoline, diesel, and natural gas with 26 different biofuels produced from a wide range of "crops." They assess the total environmental impact of each fuel by aggregating natural resource depletion and damage to human health and ecosystems into a single indicator, using two different methods (12). The second key criterion for each fuel is its greenhouse-gas emissions relative to gasoline.

The findings of Zah et al. are striking (13). Most (21 out of 26) biofuels reduce greenhouse- gas emissions by more than 30% relative to gasoline. But nearly half (12 out of 26) of the biofuels--including the economically most important ones, namely U.S. corn ethanol, Brazilian sugarcane ethanol and soy diesel, and Malaysian palm-oil diesel--have greater aggregate environmental costs than do fossil fuels (see the figure, top panels). Biofuels that fare best are those produced from residual products, such as biowaste or recycled cooking oil, as well as ethanol from grass or wood. The findings highlight the enormous differences in costs and benefits among different biofuels.

Emissions versus other impacts. Although most biofuels reduce greenhouse gas emissions relative to gasoline, many, including soy (top right) and corn (top left), do not fare well in terms of other environmental impacts (6). Second-generation biofuels produced from non-food plants such as switchgrass (bottom) may perform better in environmental terms. CREDIT: (SOYBEANS) SCOTT BAUER/USDA AGRICULTURAL RESEARCH SERVICE, BUGWOOD.ORG; (CORN) DEWEY LEE/UNIVERSITY OF GEORGIA, BUGWOOD.ORG; (SWITCHGRASS) RICHARD OLD/XID SERVICES, INC., BUGWOOD.ORG

Zah et al. excluded from their analysis so-called second-generation biofuels, such as those made from the breakdown of plant cellulose or lignin, because of insufficient data. Such biofuels could be produced from nonfood plants--such as prairie grasses or trees grown on marginal land (15), or algae cultivated in aquaculture (16)--reducing the use of food crops for biofuels (see the figure, bottom panel). Some second- generation biofuels appear particularly promising in terms of their benefits and costs for biofuel production (5).

Not all biofuels are beneficial when their full environmental impacts are assessed; some of the most important, such as those produced from corn, sugarcane, and soy, perform poorly in many contexts. There is a clear need to consider more than just energy and greenhouse-gas emissions when evaluating different biofuels and to pursue new biofuel crops and technologies. Governments should be far more selective about which biofuel crops they support through subsidies and tax benefits. For example, multibillion-dollar subsidies for U.S. corn production appear to be a perverse incentive from a rational cost-benefit perspective.

References and Notes

1. S. Pacala, R. Sokolow, Science 305, 968 (2004).
2. A. E. Farrell et al., Science 311, 506 (2006).
3. P. J. Crutzen, A. R. Moiser, K. A. Smith, W. Winiwarter, Atmos. Chem. Phys. Discuss. 7, 11191 (2007).
4. R. Righelato, D. V. Spracklen, Science 317, 902 (2007).
5. J. Hill, E. Nelson, D. Tilman, S. Polasky, D. Tiffany, Proc. Nat. Acad. Sci. U.S.A. 103, 11206 (2006).
6. R. Zah et al., Ökobilanz von Energieprodukten: Ökologische Bewertung von Biotreibstoffen (Empa, St. Gallen, Switzerland, 2007).
7. D. Pimentel, T. Patzek, G. Cecil, Rev. Environ. Contam. Toxicol. 189, 25 (2007).
8. S. Ulgiati, Crit. Rev. Plant Sci. 20, 71 (2001).
9. P. R. Adler, S. J. Del Grosso, W. J. Parton, Ecol. Appl. 17, 675 (2007).
10. G. Bala et al., Proc. Nat. Acad. Sci. U.S.A. 104, 6550 (2007).
11. W. F. Laurance, Biol. Conserv. 91, 109 (1999).
12. The authors use Swiss environmental impact points, which measure how much the environmental impacts exceed legal limits (see www.esu-services.ch/download/Frischknecht-2006-EcologicalScarcity-Paper.pdf) and the European Eco-indicator, which quantifies damage to human health and ecosystems (see www.pre.nl/eco-indicator99/default.htm).
13. A figure summarizing the findings of (6) is available as supporting material on Science Online.
14. W. F. Laurance, Science 318, 1721 (2007).
15. D. Tilman, J. Hill, C. Lehman, Science 314, 1598 (2006).
16. A. Melis, T. Happe, Plant Physiol. 127, 740 (2001).

10.1126/science.1153103