I often hear this sort of argument.

"From a purely carbon perspective, our research indicates that putting perennial biofuel crops on landscapes that are dominated by annual row crops will have a positive effect on soil carbon."

The finding "seems to walk you right into the food for fuel debate," DeLucia said, referring to the controversy over using agricultural land for fuel production. But because the U.S. is already devoting about 20 percent of its corn crop to ethanol production, he said, it would make sense to eventually use that land to produce a much higher yielding biofuel feedstock that has the added benefit of increasing organic carbon in the soil.

I don't buy it. It's not as if there are no better policy options. This report starts with an assumption that crop lands will be used for biofuels, and then seeks less harmful biofuel systems.

"From the time that John Deere invented the steel plow, which made it possible to break the prairie sod and begin farming this part of the world, the application of row crop agriculture to the Midwest has caused a reduction of soil carbon of about 50 percent," said Evan DeLucia, a professor of plant biology at Illinois and corresponding author on the new study.

Any debate on the environmental consequences of using plants to produce liquid fuels should also consider how each option affects soil carbon, DeLucia said.

"The biggest terrestrial pool of carbon is in the soil. The top meter of soil holds more than three times the amount of carbon stored in either vegetation or the atmosphere, so if you do little things to change the amount of carbon in the soil it has a huge impact on the atmosphere and thus global warming." . . .

Currently, ethanol is produced by fermenting the starch in corn kernels, but significantly more liquid fuel energy can be harvested from the stems and leaves of plants. The technology for producing this "cellulosic" ethanol is still quite expensive, but many believe that it will displace corn ethanol as the technology advances.

About 20 percent of the corn crop currently goes into ethanol production in the U.S., DeLucia said, "so we began with the hypothesis that it might be good for soil carbon to put a perennial biofuel crop on the landscape instead of corn."

The researchers analyzed published estimates of changes in soil organic carbon in landscapes converted from natural or agricultural land to biofuel crops. They focused on corn, sugar cane, Miscanthus, switchgrass and native prairie grasses. They also evaluated the impact of harvesting and using corn stover (the plant debris left over after corn is harvested) as a cellulosic biofuel source.

Their analysis showed that converting native land (grassland or forest) to sugarcane dramatically reduced soil carbon, creating a carbon deficit that would take decades to repay. While perennial grasses add carbon to the soil each year, DeLucia said, it could take up to a century for the sugar cane to rebuild soil carbon to former levels on native land.

Harvesting the corn residue for cellulosic ethanol production also reduced the carbon in the soil. The more plant residue was removed, the more the soil carbon declined.

Planting perennial grasses on existing agricultural lands had the most beneficial effect on soil carbon, the researchers found. Although there was an initial drop in carbon as fields were converted from corn to Miscanthus, switchgrass or native perennial grasses, the loss was fairly quickly offset by yearly gains in soil carbon as the grasses became established.

From a carbon perspective perennials are better than annuals and leaving crop trash in fields is better than removing it. It would be better still to not remove the biomass for cellulosic biofuel feedstock. This isn't a trivial issue since "The biggest terrestrial pool of carbon is in the soil. . . , so if you do little things to change the amount of carbon in the soil it has a huge impact on the atmosphere. . ."

Are there better uses for such cropland? If it was converted to permanent pasture of native perennial grasses and then grazed by ruminants that thrive on such fare a lot of food could be produced while sequestering a lot of carbon. If that food reduced demand for grain fed ruminants then less cropland would be needed for maize and it too could be converted to permanent pasture, sequestering even more carbon.

I'd like to see a study that quantified this sort of scenario. How much food could be produced in this way? How much carbon could be sequestered? How does it pencil, especially when existing subsidies are included? What policies could enable this seemingly more rational land use? What are the differences in fossil fuel usage, a meaningful question when we consider that the justifications for biofuel include energy import considerations?

Update:

There's an interesting commentary at Biopact about the same report.

Another point to note is that turning cellulose into biofuels is not the most efficient way to use a given stream of biomass. A considerable number of studies shows that on a farm-to-wheel basis, using biomass for the generation of electricity used in (plugin-) electric vehicles is much more efficient than converting that biomass into a liquid or gaseous fuel for use in internal combustion engines. A still more efficient way to use biomass is in combined heat and power (CHP) or for pure heat generation.

A recent Canadian study, for example, showed that using solid biofuels generate heat presents a five to 10-fold increase in the capacity to offset greenhouse gas emissions, compared to first-generation biofuels (earlier post). Cellulosic ethanol would be more efficient than first-generation fuels, but they remain liquid energy sources - that is, the cellulose undergoes an energy-intensive conversion process - to be used in the rather inefficient internal combustion engines.

The future of biomass for transportation may well be the production of energy crops for biochar, the waste-energy of which is used to generate carbon-negative electricity for use in electric vehicles (driving such a car would mean that you would not simply be generating "zero emissions", you would actually generate "negative emissions" and take CO2 out of the atmosphere -- see "the strange world of carbon-negative energy"). Such integrated biomass energy and char concepts, would be coupled to smart-grids and to the other renewables that are going to power the electric transportation future.

The CHP system discussed in Energy Mates that used a Stirling heat engine for electricity generation is a notable example that adds yet another efficiency hack. At this rate we may reach the point where greens will be running their HVAC systems with the windows open while the kids scream around in their electric vehicles, all madly cackling about the CO2 that they are sequestering while they conspicuously consume energy.