Christine,
I'll take that one. Too bad SI doesn't let you specify multiple recipients. Just kidding!
You said "very, very safe" in first, and then "no pollution." These are separate and distinct issues.
When the pros in the biz talk about safety, we mean the probabilities of having accidents of different magnitudes (from a leak/vent such as happened at TMI, going all the way up to the thing blowing up in a big explosion a la Chernobyl).
For any kind of accident, we try to figure out how likely it is. That is, what the probability is of having that kind of accident in a year's time. That is the *probability* side of the analysis.
Once the accident is defined as X and Y coming out of the reactor over some period of time with a certain heat content, etc., then we make estimates of the impacts on the surrounding region: its populace, agriculture, and economy. These impacts are called the *consequences* of the accident.
When we combine all of that information together, probabilities and consequences, we get an idea of the *risk* posed by the reactor, by considering all the different accidents we imagined possible, from small to large.
Generally, we find that the big accidents are less likely to occur than the small accidents. However, if an analysis is done where this is not the case, and a big accident has a relatively high probability, they usually put their heads together and figure out what they can do (by changing the equipment or procedures) to make that big accident less likely, and thereby reduce the risk.
Nuclear "pollution" from a reactor that is running normally or in routine maintenance does not leak anything that is detectable unless you have very expensive and sensitive measuring equipment, and do detailed surveys, and then perform statistical analysis of the data.
The reason for the measurement difficulty is that radiation is everywhere, from natural causes. And that the small amounts of radioactive material released from running reactors are very hard to distinguish from background radiation.
This would not be so if there were an accident of serious proportions, as are analyzed with that MACCS2 program.
For *real* releases from a major accident, hot spots of radiation could be detected with a handheld geiger counter costing about $500. For areas within a 10-mile radius of commercial power reactors (the Emergency Planning Zone), it's common to train and equip the police and fire dept. with these so that in the event of an accident, police and fire personnel will be able to determine if they are in a high-radiation area. Residents of the region could purchase these for their own use, and learn how to use them if that would help allay their fears. If I lived near a commercial reactor, I'd probably get one.
But that gets to your question of how safe they are. In fact, because of my distrust of the human factor in operating U.S. nuclear power plants, I probably would not want to live near one. There really is just no way to know if someone will do something really stupid while operating the reactor.
Other countries, notably Canada and Germany, take a very different approach from the U.S. in avoiding human factor problems.
If you read that link I posted some time back about the CANDU reactors used in Canada with the heavy water, there was mention of the CANDU making extensive use of automated control systems. This is just a fancy way of saying that the equipment is designed to protect itself from human stupidity, and that under certain conditions the reactor control system will "lock out" the controls so that the human reactor operator is prevented from causing a disaster.
German reactors are the same. If something goes wrong, and the reactor shuts itself down, the first 15 or 30 minutes of the shutdown sequence are done completely automatically, and the operators can't override anything until that most critical period has elapsed.
Finally, the light water reactors (LWR) used in the U.S. are based largely on the design of the reactors developed to power nuclear submarines. They operate under high pressures and temperatures.
If something goes wrong and the reactor shuts down, the thing is so hot that cooling water needs to be kept pumping through the reactor for days or it will melt. Newer reactor designs operate at lower temperatures, and are designed so that once a shutdown is triggered, the operators can just "walk away" and there is no need to keep pumps running or anything like that.
Let me know if anything here is unclear, or if I should try to explain anything else.
David |
