Let's take a head count. The number of people who believe that "something should be done" to prevent catastrophic global warming is approaching a critical mass. That, I think, is a good thing. But there's also an increasingly large number of people who have decided that nuclear energy is the way to get there. After all, we're going to need lots of non-carbon power. And nuclear energy offers non-carbon power. So let's go! Sure, the dirty hippies will kick and scream, but who listens to them, anyway?

Well, sometimes I wonder if the dirty hippies have a point, so I've been trying to read up on the matter. Here's a salvo from the "anti" side: David Fleming's short paper, "Why Nuclear Power Cannot Be a Major Energy Source." He relies heavily on work done by Jan Willem Storm van Leeuwen and Philip Smith. Granted, those two hardly have the final say on this matter (and Fleming is well aware of the dangers of leaning too hard on a single source), but their research seems to be respectable and worth taking seriously.

Fleming estimates that there are enough large uranium deposits on this earth to keep the nuclear industry going at its current rate for about 75 years. Obviously, if you ramp up the number of nuclear plants, that uranium's going to last for even less time. If you built, say, enough plants to provide a sixth of the world's electricity, you'd have about 24-48 year's worth of uranium. And, obviously, you'd create a lot of nuclear waste along the way that, if nothing else, would require a lot of additional energy to clean up.

Now hold up. What about getting uranium from the water? What about building fast-breeder reactors? What about all the magical solutions that will give us nuclear power forever? Fleming goes through these various alternatives, and finds them wanting. It would likely take much more energy to suck the uranium out of the water than you'd get from the uranium itself. So that's not worth it. Fast-breeder reactors are not yet commercially viable, and may never be. (In any case, they create serious proliferation headaches.) And so on. Read the paper for details.

Okay, fine. So nuclear power can't be a major source of carbon-free energy for all eternity. But, you ask, couldn't it at least work as a 'stopgap' measure, allowing us to reduce carbon emissions in the short term while our brightest scientists figure out how to power the world off solar energy in the meantime? Theoretically, yes. But there are tradeoffs. Opportunity costs. Nuclear energy is, after all, hugely expensive. The government has to subsidize the building of the plants, insure them against accidents, set up a police state to guard them against terrorists, dispose of the waste. It's a fat bill.

So if taxpayers are going to spend billions and billions of dollars in order to reduce the world's carbon emissions, it's not clear that nuclear power is the best place to park that money, especially if it's just a short-term solution. That money might be better spent on research into renewables, or on energy conservation. (One British government study argued as much.) Not to mention the fact that nuclear power depends on centralized power grids, whereas a national energy strategy based on renewables would likely rely on local minigrids. The two approaches may very well be at cross-purposes.

Now, none of this is a decisive argument against nuclear power altogether. Some of these assumptions, I take it, are heavily disputed. But I think there are very valid questions about the whole enterprise and it's not helpful to dismiss nuclear critics as mindless tree-huggers.

Update: Hmmm... not long after writing this, I noticed this post on The Oil Drum by Martin Sevior, a Professor of Physics at the University of Melbourne. He argues that "provable uranium resources amount to approximately 85 years supply at the current level of consumption with current technology." That's more or less consistent with the above. But then he also says that there are some 500 years worth of reserves that could be mined profitably. That would obviously change the picture. So is he right? Sevior's post has a lot of good information, although he doesn't address the opportunity-cost point.

Second update: If anyone really wants to get deep in the weeds, here's an old debate between Sevior and Storm van Leeuwen on, among other things, how much usable uranium's lying around.