A mix of perennial grasses and herbs might offer the best chance for the U.S. to produce a sustainable biofuel, according to the results of a new study. But making that dream a reality could harm local environments and would require developing new technology to harvest, process and convert such plant material into biofuels such as ethanol.

Biofuels have become controversial for their impact on food production. The ethanol used in the U.S. is currently brewed from the starch in corn kernels, which has brought ethanol producers (and government ethanol mandates) into conflict with other uses for corn, such as food or animal feed. Already, corn ethanol in the U.S. has contributed to a hike in food costs of 15 percent, according to the Congressional Budget Office, and the U.N. Food and Agriculture Organization blames corn diverted to biofuels for a global increase in food prices.

To see if nonfood plants could be a source of a biofuel the way corn is, researchers followed six alternative crops and farming systems in so-called marginal lands over 20 years, including poplar trees and alfalfa. Such marginal lands face challenges such as soil fertility and susceptibility to erosion.

The new analysis found that conventional crops such as corn had the highest yield of biomass that can be turned into biofuel on marginal lands, although their ability to reduce CO2 is harmed by tilling, fertilizing and other CO2-producing activities necessary to turn them into fuel. (Such factors have caused considerable scientific disagreement over whether ethanol from corn delivers any useful greenhouse gas reductions, although the researchers find that even corn provides some climate benefits as long as oil production and combustion is included in the comparison.)

In contrast, the grasses and other flowers and plants that grow naturally when such lands are left fallow—species such as goldenrod, frost aster, and couch grass, among others—can deliver roughly the same amount of biofuel energy per hectare per year if fertilized, yet also reducing CO2 by more than twice as much as corn. "When biofuel is produced from such vegetation, the overall climatic impact is very positive," says lead researcher Ilya Gelfand of Michigan State University. The research was published in Nature on January 17. (Scientific American is part of Nature Publishing Group.)

By taking those field results and feeding them into a computer model that calculated how much such marginal land was available within 80 kilometers of a "potential biorefinery," Gelfand and colleagues found that 21 billion liters of cellulosic ethanol could be produced in this way per year from roughly 11 million hectares of currently fallow land in 10 Midwestern states.

Such a glut of cellulosic biofuel, if realized, would reduce greenhouse gas emissions—compared with oil that otherwise would have been burned—by 44 teragrams (44 billion kilograms) per year. That is "the same as the CO2 emissions from 10 million medium-sized cars, each with an annual run of 20,000 kilometers," wrote climate researchers Klaus Butterbach-Bahl and Ralf Kiese of the Karlsruhe Institute of Technology in Germany in a commentary on the research, also published in Nature. It would also satisfy 25 percent of the 80-billion-liter target for cellulosic ethanol production in 2022 set by the U.S. government in 2007.

"It is good that the authors are attempting to focus on land that is not already used for food production," says agricultural expert Timothy Searchinger of Princeton University, who was not part of the study. But the research suggests that even if researchers maximized the capacity to grow biofuels on all marginal lands, "the amount of cellulosic ethanol it could produce is only enough to provide 1.5 percent of U.S. transportation fuel by 2020." And the expected yields are about 20 percent less than predicted by a U.S. Department of Energy analysis of biofuel potential in 2011.

Such cellulosic ethanol from native plants would also require technological breakthroughs to efficiently convert plant leaves, stems and other inedible parts into fuel. Whereas a few such cellulosic biorefineries are being built or exist at the prototype scale, none have the capacity to cope with such a mix of perennial grasses and herbs. In fact, one of the key needs for such sustainable biofuels to move forward remains "a profitable biorefining process," Gelfand notes, along with a better understanding of ecological impacts. As it stands, such cellulosic biorefineries get their materials either from the residue of conventional crops, such as corn stover, or from harvesting trees.

Turning marginal lands into biofuel farms could also have a negative impact on the local environment. Marginal areas are often currently set aside for conservation, both as a means to provide a habitat for wildlife as well as a way to protect it from agricultural runoff into waterways.

And, even if such cellulosic ethanol became a reality, it might still come into conflict with food production. "If fuel became sufficiently valuable, it might well displace food crops, as is now the case with corn grain ethanol," Gelfand says. But "expanding fuel production to marginal lands not suitable for food production is a way to relieve the pressure on productive cropland to produce fuel."