Michael Grunwald has a terrific piece in Time on the environmental devastation wrought by the recent biofuels push in Europe and North America. Yes, the ethanol boom has (ever-so-slightly) reduced our dependence on foreign oil, but it's also helped speed up the pace of deforestation around the world—deforestation that, recall, accounts for 20 percent of all manmade carbon emissions—which ultimately means that biofuels are worsening, not ameliorating, the climate crisis:

Backed by billions in investment capital, this alarming phenomenon is replicating itself around the world. Indonesia has bulldozed and burned so much wilderness to grow palm oil trees for biodiesel that its ranking among the world's top carbon emitters has surged from 21st to third according to a report by Wetlands International. Malaysia is converting forests into palm oil farms so rapidly that it's running out of uncultivated land.

But most of the damage created by biofuels will be less direct and less obvious. In Brazil, for instance, only a tiny portion of the Amazon is being torn down to grow the sugarcane that fuels most Brazilian cars. More deforestation results from a chain reaction so vast it's subtle: U.S. farmers are selling one-fifth of their corn to ethanol production, so U.S. soybean farmers are switching to corn, so Brazilian soybean farmers are expanding into cattle pastures, so Brazilian cattlemen are displaced to the Amazon. It's the remorseless economics of commodities markets. "The price of soybeans goes up," laments Sandro Menezes, a biologist with Conservation International in Brazil, "and the forest comes down."

That explains why a recent study in Science found that corn ethanol and soy biodiesel actually produce nearly twice the emissions of regular gasoline, once you take deforestation (direct and indirect) into account. Ethanol made from cane sugar is somewhat cleaner, though, as Grunwald reports, even Brazilian sugarcane fields are starting to encroach on the Cerrado, a lush savanna slightly south of the Amazon, and it's still too early to gauge what ultimate impact large-scale production might have.

Meanwhile, cellulosic ethanol—ethanol made from wood, grass, or non-edible plant parts like corn cobs—may seem more sustainable still, but even that's not a given. If, say, switchgrass ethanol ever becomes commercially viable, then sure, we could theoretically limit production to land otherwise unsuited for agriculture, but how do we know that will actually happen? As demand for cellulosic ethanol increases, some farmers may find it more profitable to start growing fuel instead, which could in turn, as Grunwald's soy example shows, lead to increased deforestation elsewhere. As we learned during the recent climate talks in Poznan, it's fantastically difficult to craft incentives that prevent people from hacking down forests to make way for a more valuable commodity.

--Bradford Plumer