Earlier this month, two independent studies in the journal Science dropped a bomb into the already controversial world of biofuels. To cop the New York Times‘ lede, the studies found that:
Almost all biofuels used today cause more greenhouse gas emissions than conventional fuels if the full emissions costs of producing these “green” fuels are taken into account…
Yesterday afternoon, when I finally got around to reading the articles, my chin hit the floor. The NYT was far too gentle: they don’t just show that biofuels have worse GHG emissions than gasoline, but drastically worse emissions—and for virtually every type of biofuel, including cellulosic ethanol (except in some highly specific conditions).
For really the first time, the studies are factoring in the carbon lost from land conversion. The authors argue (persuasively, in my opinion), that it’s crooked accounting to simply do a GHG analysis of crops versus petroleum. After all, the crops used for biofuels don’t grow in a vacuum. What really happens is that new land—Indonesian rainforest, Brazilian woodlands, American grassland — is cleared and ploughed to make way for biofuel feedstock crops. Existing agricultural land, of course, is already in production for food and fiber.
The clearing is a death sentence for wildlife in some of the most biodiverse places on earth. It also releases huge amounts of carbon into the atmosphere—called the “carbon debt.” In fact, the carbon debt run up by land conversion is, in most cases, far more than is saved by subsituting biofuels for petroleum products. (Details on the studies’ are below the jump; abstracts are here and here.) It would take decades at best, centuries at worst, to repay the carbon debt. And this when we need steep emissions reductions now.
Look, I’m sure there will be further debate, and maybe even counter-studies. (The biofuels industry appears to be fighting back already.) But in a way, uncertainty could be the real problem for biofuels, as well as for the latest fad in climate policy, low-carbon fuel standards. Either biofuels are a climate catastrophe, as these studies indicate; or we have no idea what biofuels do to the climate because experts don’t agree. And that second option is the best case scenario, at least in the near term.
Find this article interesting? Please consider making a gift to support our work.
The worst case scenario, of course, is what the studies show. Corn ethanol is plain awful, needing many decades to repay its carbon debt even when planted in abandoned cropland. But corn ethanol is benign compared to biodiesel. The best biodiesel scenario from the study (soybeans planted on Brazilian grassland) would take nearly four decades to repay its carbon debt. Other biodiesel scenarios are even worse: palm oil planted on former southeast Asian tropical rainforest takes 86 years to repay its carbon debt; biodiesel from soybeans on tropical rainforest would take at least 300 years; biodiesel from palm oil on peatland rainforest would take more than 400.
To be fair, the authors point out that future generations of very specific kinds of cellulosic ethanol—municipal waste or desert algae, for example—could have positive GHG benefits. (And, presumably, the boutique biodiesel from french fry oil, which can never scale to meaningful production levels, is not quite so awful.) But the hope for a benign biofuel future is predicated on our demand for fuel not inducing futher land conversions. And that’s a very big “if.”
In any case, those future climate-friendly fuel sources are mostly hypotheticals. Real-world stuff isn’t so promising. I quote:
…if American corns fields of average yield were converted to switchgrass for [cellulosic] ethanol, replacing that corn would still trigger emissions from land use change that would take 52 years to pay back and increase emissions over 30 years by 50%.
I’m bearish on biofuels. And if you want my opinion, biofuels are really just the latest in a long line of Hail Mary’s that try to avoid ending Americans’ car addiction. There’s always a technological miracle around the corner—just wait a few more years—then we’ll have plug-in hybrids running off clean or surplus energy; then we’ll HyperCars or FreedomCars that can cross the continent on a tank; then we’ll have Segways that will revolutionize personal travel. I could go on. In the meantime, we avoid the bread and butter fixes we’ve known about for decades, and we keep sending ever greater amounts of carbon into the atmosphere.
This is really happening. Right now, maybe the single biggest threat to good cap-and-trade programs like the Western Climate Initiative is that policymakers will avoid capping transportation fuels, which are easily our largest source of emissions. The latest attempt to punt involves adopting a low-carbon fuel standard (LCFS), which is supposed to be a big boon for biofuels. (Depending on its construction, an LCFS can work like a cap on absolute emissions or a cap on the intensity of emissions.) The LCFS would quantify the lifecycle carbon emissions of every fuel stream entering the economy. To begin with, that’s an analysis of mind-bending complexity. But now, in light of the new studies in Science, it appears that an honest LCFS would either favor conventional oil or simply be technically infeasible because the accounting is so incredibly complex.
Or—and this is the biggest danger—the LCFS would use an accounting like we’ve seen in the past: one that favors biofuels, but that downplays land conversion or other factors. In that case, the carbon reductions could be a chimera.
There is a very real chance that LCFS, or something like it, will trump a general cap on transportation emissions. If that happens I am going to start calling it cap-and-pray. Pray that some new miracle solution can keep us all in our cars with no climate consequences.
For the record, I don’t pretend to be an expert in biofuels or in life-cycle analysis. But remember, these are peer-reviewed papers by respected researchers in a top science journal. (And in case you’re wondering, the authors’ affiliations don’t indicate a prediliction to anti-ag positions; they hail from places like Iowa State University, the University of Minnesota, and the USDA.)
Finally, I’ve hunted around a little—not a lot—for counters to these studies. I couldn’t find much, but please do share if you know of any! In the meantime, my take is that the findings are damning, even if they’re not the last word on the subject.
Postscript: A commenter points me to a letter from two government researchers that objects to one of the two studies (the Searchinger et al. study). To drastically summarize here, the letter makes the following principal claims: 1) US ethanol production won’t rearch the levels used in the study because of current legislation; 2) Average corn yields in the US and other countries will increase, meaning that le
ss land will be converted; 3) Brazil and China will not convert much additional land because of current legislation and practice in those countries; 4) Ethanol refining processes can become more efficient in the future; and 5) There’s a lot of biomass already available from forest growth, crop residues, and other sources.
Without spending too much more time on the keyboard for this post, I think that most of letter’s claims are debatable, at best. (But go read it for yourself, and tell me why I’m wrong!) In any case, the letter concludes by acknowledging what I take to be the central problem — uncertainty:
…indirect land use changes are much more difficult to model than direct land use changes… While scientific assessment of land use change issues is urgently needed in order to design policies that prevent unintended consequences from biofuel production, conclusions regarding the GHG emissions effects of biofuels based on speculative, limited land use change modeling may misguide biofuel policy development.
Exactly. But “limited land use change modeling” is what we’ve seen in the past. That’s what got us into biofuels. Uncertainty of this nature is a double-edged sword, one that should cut against biofuels until we’re confident they’re not a danger.