An email from a friend triggered a line of thought that fits with many comments here about how best to fit renewable power into an energy system that doesn't match its characteristics very well. In a response to last Friday's posting (which was featured in the Wall Street Journal's Energy Roundup blog) she mentioned attending a conference and hearing a proposal to use nuclear power to provide fresh water. This idea has been around for decades, and it has been implemented in a few places. As fresh water becomes increasingly scarce in many parts of the world, nuclear desalination could become attractive. But electricity is fungible, and it seems like a shame to waste hundreds or thousands of Megawatts of reliable, baseload power on an application that doesn't require either quality, at least not to the same degree as commercial and residential customers do. This seems like a perfect application for wind or solar power.
One of the perennial arguments against renewables is their intermittency or cyclicality. Adding energy storage to smooth this out--even low-cost pumped storage--affects project economics that still generally require subsidies to match the returns from conventional energy. Several of my readers have pointed out that widely dispersed wind farms are less variable in aggregate than individually, and large grids can accommodate reasonable quantities of intermittent power without requiring storage. Fair enough, but that will only take you so far, as countries with a high concentration of wind power in a small area, such as Denmark, have learned. This is one reason you see people talking about using wind to generate hydrogen, which can be accumulated non-ratably and dispensed as needed.
Desalination seems like another, even more useful way to avoid having to match the demanding standards of the power grid. We've been storing water for millennia, and there's nothing that says that the production from a desalination plant must feed directly into the water main, without spending some time in a tank or reservoir that would naturally buffer the unpredictable generation from a facility run by wind turbines. In fact, a developer could add power storage to the mix as well, by pumping the potable water uphill and using it to generate power at times of peak demand. And unlike reactors, which come in pretty large size increments, wind and solar are essentially infinitely scalable. This isn't an argument against nuclear power, which looks like an increasingly important option in a carbon-constrained world. It's just my sense that there are better uses for the steady, high-quality power output from a reactor.
Wind and solar power are now growing rapidly, from a very small base. But mightn't they grow even faster, if, instead of forcing their output to match our established usage patterns, clever developers channeled some of their efforts into applications for which the drawbacks of current renewable energy technology don't matter?