This Cornell prof has a technique for growing rice
that he says
can double (or more) yields while using alot less water and less seed. Pretty cool, if it works. I am a bit skeptical: humans have been growing rice for an awfully long time, and if improvements of this order were available, you would think that someone would have figured it out in the last 4000 years. But the principles behind the technique are sound, so it probably works at least reasonably well.