Tuesday, May 23, 2006

Planning Too Far Ahead

"There are engineering questions about the massive storage repository proposed for the Nevada desert. Certainty about its ability to keep groundwater supplies safe falls off after about 10,000 years—while the facility needs to function as planned for several hundred thousands of years."

The quote is from an interesting article on global warming and ways of dealing with it in the latest issue of Harvard Magazine. The context is a discussion of problems in storing nuclear waste as one limit to increases in the role of nuclear power.

To me, at least, the idea of worrying about effects more than ten thousand years out is so absurd as to be marginally sane. Nobody alive knows whether our species will still exist in ten thousand years, if it exists if most humans will still live on earth, or if we still live on earth what sort of society, economy and technology we will have. If things do continue more or less along current lines—not, in my view, very likely—ten thousand years of economic growth would give us a society for which a little radioactivity in Nevada groundwater would be a trivial problem. If we assume a 1% annual rate of growth in per capita real income, it takes only about 2300 years to bring the income of the average individual up to the current income of the world.

Worrying about problems ten thousand years out is particularly odd given that nuclear power is being discussed as a way of limiting global warming. Elsewhere in the article, in the context of a time horizon of 100 to 500 years, another source suggests the possibility of sea level rises of over 200 feet. I am reluctant to trust extrapolations that far out as well—but compared to 10,000 years, a hundred years is practically as close as next Thursday. And drowning areas containing a considerable fraction of the population of the globe would be a slightly more serious problem than contamination of the Nevada water table.

14 comments:

Anonymous said...

My favorite part about the super long term planning they are doing is that they plan to put glyphs and pictures, carved in stone, so that if civilization falls and there's no record of the repository that people will have warning about the dangers.

Think about what happens when some explorer finds hieroglyphs warning of some nebulous danger ahead. The archeologist or whoever is warned and backs off, right? They would never forge ahead to see what treasures would be guarded by these warnings.

Anonymous said...

I agree with you about the long term planning. However, the immediate danger of nuclear power plants makes me feel uneasy. I do not actively participate in protests or rallies or whatever, but I can't say nuclear power has my full support. Usually I am one of the last ones who use the ugly t-word, but aren't nuclear power plants a great target for terrorists? And isn't the potential of destruction of a nuclear power plant a teeny bit too immense to calm ourselves with a low probability of a maximum credible accident?

I'd like to hear what you think about it, especially because you're both an economist and a physicist.

David Friedman said...

Anonymous asks about the risk of terrorist attacks on reactors. I'm certainly not an expert, but my impression is that it isn't all that great--probably less than the risk of some other targets. Getting a reactor to explode by crashing something into it doesn't seem to be practical--at most you can produce the equivalent of a dirty bomb, raising the local background radiation level for a while--and even that would be hard.

A more serious concern is that if there are a lot of reactors all over the world, it becomes easier for lots of countries to manufacture nuclear weapons.

Anonymous said...

Only you could make a discussion of nuclear waste witty. Great!

I kinda see it as part of the never-will-be-fully-resolved moral dilemma of how much do we live for ourselves now and what weight to give to the possible negative impact of our choices on others, and other species, now and on into the future.

For example, would I personally burn down the Amazon rainforest to grow soybeans, graze cattle, harvest wood etc. to feed and house a starving planet if it would personally enrich me to the tune of a billion dollars? No, I wouldn't. I sometimes think that some would burn it down for a nickel. Money is nice, but isn't everything. It's not even most everything. If it's a choice, I'd rather the humans starve and thereby spare the plants and wildlife that have existed in balance for millions of years. Anyway, I don't see the soybeans and hamburgers making it to the starving masses around the world. So just who is benefiting from all the destruction? Burger King? Who are we to muck it all up in a span of a couple hundred years or so? Once it no longer exists, will we, as just one remaining non-extinct species, be collectively better off? What do we want: a world with more egotistical billionaires or a sustainable and livable world?

I agree with you that looking too far ahead into the future is dubious at best and can rapidly turn into science-fiction. I just wish that we humans would look back in time - as far back as necessary - to apprehend what is worth preserving for all-time's sake.

Anonymous said...

Perhaps we are at a special point in history, but all other periods of prosperity, achievement, and growth have been followed by periods of stagnation, decay, and ignorance. Given that, I'd give us good odds of not being omnipotent meta beings in 10,000 years. Any descendants of ours would likely appreciate it if we didn't make the planet too unlivable. It is perhaps insane to worry about the future of a humanity we will never see, but I feel that is a wonderful, desirable, and distinctly human trait.

That being said, I strongly support nuclear power. The problems of waste storage are not insurmountable. The most interesting idea that I've seen for storing traditional waste is to stash at the edge of a subduction zone, where we can watch over it as it slides into the earth. I haven't looked into this in detail to see how practical it is.

What I have looked into is integral fast reactors (a type a breeder reactor). And they seem to me an extraordinarily good choice for providing power. The traditional nuclear reactor has three main drawbacks
1. The waste is very long lived.
2. The reactor is not as safe as we might like.
3. There is not enough fuel for really long term use.

The first problem (as well as the problem of what to do with our current generation of waste, since it can be used as fuel) is mostly solved by fast reactors. Fast reactors still result in nuclear waste, but that waste has a half life on the order of 100 years rather than 10,000 years. I think that we can probably build things to last for a couple hundred years without too much trouble. Moreover, if we mess up, the damage will only last a few hundred years.

The second problem has largely been solved already. Nuclear power plants are now built with passive rather than active shutdown devices, which greatly reduces their chance of meltdown. (It's still instructive to note that Chernobyl failed when the automatic safety systems were shut down in order to perform a test. The total death toll for Chernobyl as of 2004 was estimated at 56 ( http://www.world-nuclear.org/info/chernobyl/inf07.htm ), which probably gives nuclear power a lower fatality per kilowatt-hour that coal (remember mining). ) That being said, part of the volatility of a nuclear power plant comes from the fact that its coolant it super-heated pressurized water. If you crack a traditional reactor all the coolant immediately blows out. The current generation of fast reactor uses liquid sodium, which isn't any better, but liquid lead has been proposed as coolant. Lead would have the distinct advantage of not boiling off when you need it the most.

The third problem is taken care of by the fact that a fast reactor really is a breeder, in that it breeds more radioactive fuel. Rather than stocking the reactor with U235, the isotope U238 is used. In a fast reactor, the U238 absorbs neutrons and becomes an isotope of plutonium that can be used as the primary neutron source. The great thing about U238 is that there is over 100 times as much of it as U235. The estimates I've seen suggest there's enough U235 to supply all the world's power needs for something on the order of 100 years. We've got enough U238 to last at least a couple thousand.

All that being said, there are a few other objections that people usually raise. One of them is proliferation, which seems to me a non-issue for nuclear plants built in countries with vast stockpiles of nuclear weapons. The reason that proliferation is an issue at all is that the present generation of fast reactors requires bred fuel to be reprocessed before it is used. The centrifuges and acid that achieve this (remember the aluminum tube debacle in Iraq) can also be used to purify the material to weapons grade fuel. A possible solution to this problem is provided by integral fast reactors which, until President Clinton canceled research on them, were showing a lot of promise. The word "integral" here means that the reactor is designed to be self contained. It is not necessary to ship fuel out for reprocessing. Instead, the bred fuel is basically electroplated onto a new rod, a process which is inadequate for producing weapons grade fuel.

This has the additional advantage of not requiring a nasty chemical slurry which must then be disposed of.

Sorry for the length of my post, but I'm mildly obsessed by this.

The following are useful links for people who want to know more.

http://academic.brooklyn.cuny.edu/physics/sobel/Nucphys/breed.html
http://en.wikipedia.org/wiki/Integral_Fast_Reactor
http://www.nationalcenter.org/NPA378.html

David Friedman said...

"For example, would I personally burn down the Amazon rainforest to grow soybeans, graze cattle, harvest wood etc. to feed and house a starving planet if it would personally enrich me to the tune of a billion dollars? ... Money is nice, but isn't everything."

I don't see why you have to bring the billion dollars into the question at all--why isn't the starving planet a sufficient motive? You seem to be evading that issue later with your claim that the soybeans don't make it to the starving masses.

What if they do make it--as, so far, they have? (Calorie consumption per capita in poor countries has been trending up for decades). On what basis do you prefer the welfare of trees to the welfare of human beings? Would you feel differently if it was you, your friends, and your children who were threatened with starvation?

Anonymous said...

trumpit: It is more of a question of, do we pump billions of tons of pollutants into the air, or put a few tons of it deep undergroud? So its a choice of people tens of thousands of years from now potentially having to deal with a bit of polluted groundwater, or people a lot sooner than that having to deal with the much more serious problem caused by (a) global warming, and (b) various pollutants like sulphides, nitrous oxides, dust, etc etc.

David Friedman said...

Bryan writes:

"Perhaps we are at a special point in history, but all other periods of prosperity, achievement, and growth have been followed by periods of stagnation, decay, and ignorance."

I'm not sure what examples you have in mind, but if you forget the term "dark ages," which I gather has been abandoned by historians as misleading, and look at population estimates instead, the claim is a dubious one. European population hit bottom about 600 A.D., with the Roman Empire barely cold in its grave, and from there rose faster and faster until the 14th century--passing the earlier Roman high about 800 A.D.

There was a distinct stutter in the 14th century, and a couple of centuries when, due in part to the black death, things didn't look so good. But after that the general trend has been up--for about the past four hundred years.

There's no guarantee that will continue, of course--but neither is there any guarantee that it won't.

Populaation isn't the same thing as prosperity--but in pre-modern societies population expansion is pretty good evidence that things are going at least relatively well.

montestruc said...

The whole issue about very long term storage is a crock from what I understand. Their is a finite long term risk associated with the uranium or thorium ore in the ground due to it's potential radioactive energy. When you mine the ore, then use it to make power, much of that energy is used up, while the decay products may be more radioactive in the short term (a few hundred years) the risk is less over the very long haul. So if you put the waste in casks that are sealed well enough for about 500 years, and put the casks back in the mines you dug the ore out of and seal them with concrete, the long term risk is much less than leaving the ore in the ground, and the short term risk (0-500 years) is all you have to worry about and that is economicaly possible.

Malcolm Kirkpatrick said...

"Montestruc" beat me to his point. I'd like to add that all over the world, rivers are cutting through ore bodies and carrying radioactive elements to sea. Volcanos emit radioactive elements in their gasses. Ore bodies leak radon. I wonder if reactor waste has to be sealed away at all, rather than diluted and poured into the ocean or mixed with dirt and dumped into the mine from which it was taken. I'd like to see the arguments on either side of that question quantified.

Anonymous said...

AIUI, the insane thing is trying to bury plutonium and uranium because we don't reprocess. That's the stuff which needs dekamillennial stability. If you burn all the fuel, you just have to bury the high level wastes, which decay below ore radioactivity levels in 500 years.

Anonymous said...

"...A more serious concern is that if there are a lot of reactors all over the world, it becomes easier for lots of countries to manufacture nuclear weapons..."

I have more of a concern for the byproducts from the former Soviet Union reactors.

Considering the tremendous amount of money required to produce fissionable materials, I don't see where it would be economically advantagious for a smaller country to misuse it's resources in this manner; especially if the country is not under any threat from other countries.

montestruc said...

Malcolm Kirkpatrick said...
"Montestruc beat me to his point. I'd like to add that all over the world, rivers are cutting through ore bodies and carrying radioactive elements to sea. ,,,, I wonder if reactor waste has to be sealed away at all, rather than diluted and poured into the ocean "

The short-term radioactivity, and danger from reactor waste is large more than for most chemically hazardous waste -- and since it is so compact, it seems clear that it is much cheaper to store it in good containers so it does not contaminate the local environment where dumped, than to dilute it. At least when considering the cost of the damage to the environment.

Yes in several hundred years that would be plausible, but the issue needs settlement in the short term.

Anonymous said...

Another point not being addressed is the likely improvement of waste cleanup technology. I would wager that within 10,000 years, humans will come up with a few new tricks for dealing with all kinds of hazardous materials.