They're all non-controversial, and in the cases of peanuts and mustard, can be plugged directly into a crop rotation including corn. All of them have a better oil yield per acre than does hemp, and excepting soy, all of them need less water. (Also, oats can be grown in regions where soy won't since oats have a complete protein and high oil content.)
Flax only needs more water when it is harvested primarily for linen; when harvested for linseed, it needs less water than hemp to reach maturity, and linen is both stronger and has a better hand than does an equivalent hemp fabric. (Seed flax fibers can be used for cordage, but not for fabric.) A crop rotation of flax, sunflowers, peanuts and mustard would provide the soil with a complete nutrient cycle (peanuts are legumes which are nitrogen fixers; mustard is an excellent fallow year cover crop since it will grow without much attention; flax is excellent for wet springs and drier summers, while sunflowers are better for years with a dry spring) while providing more oil per acre than hemp (39 gal/acre), corn (18) cotton (35gal/acre) or soy (48gal/acre). The gallon per acre yield for flax is 51 gal/acre, mustard is 61 gal/acre, sunflowers are 102 gal/acre and peanuts are 113 gal/acre.
Flax does everything that hemp does except make the feds panic, while sunflower stalks and shells can be turned into paper and building materials (though not thread, cord or rope) and peanut shells can be ground and compressed for a carbon-balanced wood pellet equivalent. The meal left over from pressing mustard oil is an excellent organic pesticide.
Right now, they're all more expensive to grow than soy or corn because there's not a glut of automated machinery for sunflowers, flax and peanuts. It's much easier to get a cheap #2 corn harvester to attach to the Deere than it is to get a sunflower harvester. But that could change if sunflowers, flax or peanuts would have the same sort of price supports that soy and corn have. We don't need 80% of the corn we produce; it gets fed to animals, fractioned and converted into all of the products it does because we have excess corn and we have to do SOMETHING with it. If we'd put the cows back to grazing, the pigs back to foraging and eating scraps and the chickens to insect control, we could put something like 60% of the land that is currently corn and soybeans into oil crop production. That's approximately 82 million acres of corn and 74 million of soy (2005 levels.) If 60% of each were diverted to oil crops, while the remaining 20% of excess land was turned over to managed pasture, we could produce on average 7.6 billion gallons of oil. Cows, sheep and goats can digest a lot of the waste from oil crops (though not mustard), returning it via their manure to the fields from which it came.
Conversion of oil to biodiesel is on a .8 ratio (i.e. 1 gallon of oil equals .8 gallons of usable biodiesel). That means with an averaged yield (so assuming that 25% of corn and soy farmers go to one of each of the four crops mentioned and then rotate through the cycle) of 81 gallons of oil per acre, and 93 million acres converted to oil crops, we can cover our annual oil needs. (We need about 715 million gallons of oil each year at current usage rates.) We can also fractionate gasoline out of a vegetable oil, though we will have to retool our refineries to do so. The better option would be to set a cut-off year and switch to the production of diesel engines completely. Using a vegetable derived oil is actually easier on diesel engines (the sulfur is added to lubricate them, but is unnecessary with biodiesel) and is net carbon neutral. If those 93 million acres were converted to hemp, though, we wouldn't produce enough oil, and monocultures of that nature are environmentally dangerous. If everything is planted in the same crop and something develops or discovers that it likes the taste of that crop, whatever it is will breed and eat. Diversifying the crops we use for oil and food will actually lessen our dependence on fertilizer and pesticides because a more diverse farm has fewer pest issues and recovers quicker!
Here are the problems: Managed pasture is more labor intensive and requires a greater degree of quick wits and psychological flexibility than does current monoculture industrial farming, so we would have to a) make the culture truly support farmers and make farming a life where people make a white collar salary and b) recognize that while meat may be an important part of our diets, it is not the focus of our diets. But we need to do that anyway. We will also have a severe labor shortage for a few years while people get with the program and learn to be farmers again, and we will have to restore our agronomy education system. Food costs will rise for a while, we'll have to support the farmers for the first few years, ADM, Monsanto and Con-Agra will have to shift to oil processing or die, and we will have to send people and knowledge to other countries instead of direct food aid. 60% of the processed food we eat will go away and we'll have to relearn to cook. Fast food will become very expensive or have to switch to soy. There are not a lot of flax, sunflower, peanut and mustard specific fertilizers, herbicides and pesticides out there. All four are hard to patent and hybridize into obedience (corn and soy do hybridize obediently) so the seed companies will not have an interest in switching from corn and soy development to flax, mustard, sunflowers and peanuts. Mustard is an allergen.
I think it's worth it enough that I'm going back to school in 2007 to get a degree in agricultural economics.
My data is from
http://journeytoforever.org/biodiesel_yield.html and wikipedia and the USDA.