In February 2008, a group of researchers led by Tim Searchinger of Princeton’s Woodrow Wilson School published a paper (PDF) in Science Express called, “Use of U.S. Croplands for Biofuels Increases Greenhouse Gases Through Emissions from Land Use Change.”

Their conclusion was startling: the government policy of supporting biofuel production, which had begun in the 1970s and accelerated with the Energy Act of 2005, was increasing, not reducing, greenhouse gas emissions. Under President Bush, support for domestic biofuels — through tax credits, protective tariffs, and escalating consumption mandates, and more — was really the federal government’s only organized effort to combat climate change. And if Searchinger and his colleagues were correct, the biofuel was actively contributing to climate change.

The researchers’ argument went like this: If government incentives inspired biofuel makers to divert a billion bushels of corn away from food and feed uses to ethanol plants, that billion bushels of corn would have to be grown somewhere else to keep up with food/deed demand — say, in what had previously been grasslands in Argentina. And by turning perennial grassland into an annually planted, highly fertilized cornfield, you’re generating greenhouse gas emissions — that could turn out to overwhelm greenhouse gas reductions from replacing petroleum-based gasoline with ethanol.

Not even cellulosic ethanol survived the analysis intact. “Biofuels from switchgrass, if grown on U.S. corn lands, increase emissions by 50%,” the authors concluded.

Understandably, the Searchinger analysis dropped like a bomb in biofuel circles. Critics seized on it as proof that the biofuel program was at best a feeble response to the looming threat of climate change; boosters thundered that the assumptions were all wrong. But indirect land-use changes inspired by the ethanol program could no longer be ignored. Another big boost to ethanol mandates came with the 2007 Energy Act, but the law also required the EPA to come up with a total lifecycle analysis of biofuels’ impact on greenhouse gas emissions.

Obama’s EPA on Tuesday delivered its much-awaited proposal for evaluating the greenhouse gas-reducing performance of biofuels — which, to the biofuel industry’s chagrin, includes indirect land use changes. Well, actually, the agency delivered two options for reckoning with land use and is seeking comment on which one will make it into the final rule, to be issued on November 30.

The first option would “assumes a 30 year time period for assessing future GHG emissions impacts and values equally all emission impacts, regardless of time of emission impact (i.e., 0% discount rate),” the EPA document states. “The second option assesses emissions impacts over a 100 year time period and discounts future emissions at 2% annually.”

In the document, the EPA includes a table showing how each kind of biofuel — from ones that exist in commercial form (corn-based ethanol, soy-based biodiesel) to ones that don’t (switchgrass and corn stover-based cellulosic ethanol) — fares in greenhouse gas terms under each scenario.

Under the 30-year scenario, corn-based ethanol looks like a truly horrible idea. Ethanol made in natural-gas-powered plants result in a 5 percent increase in GHG emission, compared to petroleum-based gasoline. For ethanol from coal-powered plants, the gain would be a jaw-dropping 34 percent.

Things look much rosier under the 100-year scenario, because over time replacing gasoline with ethanol “pays back” the one-time effect of land-use changes. Under this scenario, natural-gas-powered ethanol delivers a 16 percent cut in GHG emissions, while coal-powered ethanol productions results in a relatively modest 13 percent hike in emissions. Significantly, neither would pass the EPA’s test that biofuels deliver at least a 20 percent GHG cut to qualify under the Renewable Fuel Standard mandates.

Under both models, cellulosic forms look like big GHG winners — although they have yet to be made at commercial scale. Soy biodiesel, for its part, adds to GHG by 4 percent under the 30-year horizon, and subtracts 22 percent under the 100-year model.

The EPA will be seeking public comment on which scenario it should use for reckoning GHG performance. Biofuel producers would like to scuttle the land-use tests completely. During a press conference Tuesday (listen to it here), Renewable Fuels Association President Robert Dinneen called indirect land-use considerations far too complicated to get a handle on. He encouraged the EPA to focus on “apples to apples” comparisons between biofuels and gasoline.

I called Tim Searchinger, lead author of the land-use study, to get his perspective on the proposed rules with regard to life-cycle calculations. He made what I consider to be a devastating critique of the 100-year scenario as an analysis tool. He pointed out that scientists are calling for steep cuts in overall greenhouse gas emissions by 2050: anything less courts climate chaos.

He also criticized the 2 percent annual discount for emissions into the model. “Discounting is a concept from economics — it makes sense to discount, say, revenue streams. But it makes no sense to discount emissions. It’s just wrong.” He added: “The long horizon and discount rate ignore the true cost of emissions. We’re talking about ice caps melting and the methane being released from tundra. Do we really have a hundred years” to wait for GHG cuts?

The 30-year horizon is, at least theoretically, much more reasonable, Searchinger told me. But even here, he adds, the EPA is using “wildly optimistic” assumptions to come up with its GHG projections. For example, he said, the agency neglects to account for the potential conversion of peat land in Southeast Asia — a tremendous carbon sink — into palm plantations to offset soy diverted to biodiesel in the U.S. If even 2 percent of peatland gets flattened for that purpose, soy biodiesel’s GHG footprint will have expanded dramatically.

Of course, in the case of corn-based ethanol, all of this is academic. The proposed EPA rule would grandfather in as much as 15 billion gallons of “first-generation” biofuels, regardless of GHG performance. Last year, the industry produced 9 billion gallons. Few observers believe that the U.S. corn ethanol industry is capable of churning out more than 15 billion gallons under any circumstances.

“Corn ethanol is a done deal,” Searchinger told me. “There’s no stopping it.” The question going forward, he said, will be the land-use changes caused by other forms of ethanol, ones that do well under the EPA scenarios, like cellulosic and sugarcane.

In related news, the Obama administration rolled out a host of new supports for the ailing ethanol industry, in addition to the $5 billion-$7 billion it already gets in tax breaks and other goodies. From The New York Times:

The White House made its first major statement on ethanol on Tuesday, mustering three Cabinet members to outline a plan to shield corn ethanol producers from the credit crisis, work with them to cut their use of natural gas and coal in ethanol production, and nudge the auto industry toward production of vehicles that can use ethanol at concentrations of up to 85 percent.

In pursuing these goals, the Secretaries of Agriculture and Energy, Tom Vilsack and Steven Chu, along with the administrator of the Environmental Protection Agency, Lisa Jackson, announced during a press conference the formation of a “Biofuels Interagency Working Group,” comprised of the three agencies.

In the above-linked press conference, the Renewable Fuels Association’s Dinneen praised those moves. “The president has sent an incredibly important signal that biofuels are going to be a key component in his strategy to address energy, economic, and environmental challenges,” he said.