Over at Texas A&M there's Dr. Jim Ansley, looking for ways to make sure Texas stays in the energy business. He's working on a machine that'll harvest and chip up mesquite, for production into ethanol from its cellulose fibers.
Mesquite is an awesome feedstock for this. The "food versus fuel" cranks who want us to believe we can't grow fuel because it would take farmland away, will have to face the fact that fuel feedstocks grow on marginal land and fix their own nitrogen (meaning that the "crop" requires no human inputs in order to be an economically viable resource). An acre of mesquite can yield 15 tons of chipped wood every 10 years. Back of the envelope calculations suggests this works out to at least 1500 gallons of ethanol and a net greenhouse gas (GHG) decrease of about 85% campared with reformulated gasoline.
Going over what I know about rapeseed, jojoba, willow, and so on, fuel crops that grow on marginal lands seem to yield between 50 and 200 gallons of liquid fuel per acre per year. Perhaps there's a limit, some sort of diminishing return, for plants making tissues out of sunlight and carbon from the air. It makes sense, of course, that plants can make only so much "stuff," but I guess what I'm noticing is that unless we decide to make some sort of crazy engineering project out of it, we can probably reasonably expect 200 gallons of fuel per acre per year, out of any relatively unforced system.
Hmmm... 200 gallons per acre per year...
If I drive 12000 miles in a year on 200 gallons, I'd need a 60 mpg vehicle to do so. If 300 million people copy me, then we're using the output of 47,000 square miles.
Huh, lookit that: we could supply 100% of our domestic passenger transportation needs from native plants grown as fuel crops on marginal soils, in an area about the size of Pennsylvania.