But is the sky the limit? Several scientists now say it's actually possible to have so many turbines that they start to lose power. They steal each other's wind.
Sailboat captains experience a similar phenomenon; they call it "dirty air." If you're sailing directly downwind of another sailboat, you'll slow down. That's because the lead boat creates a turbulent vortex of air behind it as the wind spills off its sails. As sailors know, "dirty air" means less power.
The blades on wind turbines make dirty air too, so engineers space the machines far apart. But wind developers want to bring the total number of turbines in the U.S. to more than 100,000, up from about 45,000 today. Would that spread the turbulent air far and wide?
Harvard physicist David Keith says that's possible. "[With] very large wind farms," he says, "we can now see long footprints that extend, in some cases, tens of kilometers downstream, where you have slower-moving wind."
Keith is one of several scientists who have designed computer simulations to see what might happen at huge wind farms. "If we're going to scale wind power up to supply a significant fraction of the global energy demand — say 10 percent of global energy demand as we get towards midcentury — then these effects begin to matter," he says. "Exactly how much they matter we still don't know."
The answer has become a kind of puzzle for atmospheric scientists. Just what, they ask, is the "saturation point" for wind power? That is, when is the wind so dirty that there's no point in building one more turbine?
So far, that point is hypothetical. To get to saturation, you'd need huge wind farms —bigger than any that exist now — with thousands of square miles packed with turbines. For the 45,000 big turbines now spread around in clusters throughout the U.S., that's not a problem.
Atmospheric scientist Mark Jacobson at Stanford University has done calculations that suggest you'd have to get up into the many millions of turbines before you'd lose a serious amount of wind. And people are thinking about how to get around the problem.
"We found that by spreading out the wind farms themselves," Jacobson says, "you reduce this impact of having low energy when you just have one wind farm with lots of turbines."
Elizabeth Salerno at the industry's American Wind Energy Association says developers are making sure they're not going to dilute the wind. Doing that would lose money.
"Our developers spend a lot of time with experts and atmospheric scientists to ensure that they are putting each individual wind turbine and the entire project in a location and a layout that's going to optimize their result," Salerno says.
But as you build more and more wind farms, spreading them out could present complications, too. You can't put them just anywhere — you need to have transmission lines reasonably close by, for example. Also, as you get more wind farms, more people are likely to complain about the view. And lots of places simply aren't windy enough to be useful.
Even with those limits, there's plenty of wind to go around, Salerno says. "We have enough wind resource in the U.S. on shore [and] on land to do 10 times over our power production today."
That's 10 times all the power produced now in the U.S from every means, including coal, nuclear and hydropower.
Salerno says no one expects wind power to come anywhere close to that. For one thing, the nation's electricity grid runs more reliably if utilities can draw on different kinds of energy that can back each other up. When the wind isn't blowing, the grid can draw on natural gas or nuclear power. Or if natural gas prices spike, people can use more wind.
At the moment wind provides about 3.5 percent of all electricity in the U.S. The wind association says a reasonable goal is to raise that to 20 percent of the nation's electricity needs by 2030. That would mean building maybe 75,000 more wind turbines, Salerno says, and not building them all in one place.