This forum is about wrong numbers in science, politics and the media. It respects good science and good English.
Part of the problem with nuclear power is that the dangers are vastly exaggerated by the anti-nuclear lobby. Fictitious body counts are generated by an act of numerical prestidigitation based on the bogus, and unfortunately still dominant, linear no-threshold model of radiation exposure. The same model is used to declare an elevated level of background radiation in an area either hazardous or safe, with an excessive bias towards the precautionary principle. Although this can lead to a high cost, when a great swathe of land is declared unfit for human habitation, the dangers are quite often largely in people's minds because they have been conditioned by years of anti-nuclear propaganda into a state of radiophobia.
In reply to James, the main method of avoiding the cost of nuclear accidents serious enough to lead to land being written off is to design the plants so that such accidents have a return period of at least several thousand years, and you can then probably forget about such an event occurring in a specific country in the time window for which nuclear fission reactors are likely to be around.
The way it has worked with nuclear power so far is that governments have been prepared to tolerate the small risk of a nuclear accident, and treat it as being something similar to a natural disaster like a hurricane or earthquake where they would pay most of the bill, in return for perceived benefits for that country from using nuclear power such as improved energy security and energy independence associated with having to import less fossil fuel, reduced pollution from deploying fewer coal-fired power stations, reduced CO2 emissions (if you believe in CAGW), and in the unique case of the UK, a reduced probability of governments being brought down by coal mining strikes. If you wanted a situation where private industry pays for such accidents, the first step would be remove the large amount of conservatism that is believed by many to be present in the current radiation safety limits. At the current level of conservatism, only governments can pick up the bill.
The Chernobyl and Fµkµshima incidents were obviously unexpected in the countries where they occurred, but I don't think similar incidents could occur in the UK and most other Western countries. The Greenies probably couldn't believe their luck when a nuclear accident happened in a country as outwardly sophisticated as Japan. I remember it used to be speculated that India was a strong candidate for a future nuclear accident due to them having an everyday life safety culture even worse than the old Soviet Union, with a very high road safety death toll and such practices as railway passengers being expected to cling to the outside of a train, but they seem to run a tight ship when it comes to their nuclear plants.
Ironically the Fµkµshima accident probably occurred as a result of trying to prepare for the possibility of such an accident. It looks to me like the Japanese have copied British siting policy for nuclear plants (the UK built Japan's first nuclear plant), which is to stick them in a sparsely populated area next to the sea. If you put a plant next to the sea, it reduces the land area that could be written off by a factor of two, and you've also got an unlimited supply of emergency cooling seawater if things get desperate. I suspect these selected seaside locations also have a prevailing wind direction where the wind blows out to sea. But the Japanese seem to have been highly complacent about the tsunami run-up hazard, which isn't applicable in the UK, where we would only consider storm surges, and these would have a return period of ten thousand years, (I think there is a weather related tsunami-like event called a meteotsunami which is taken into account in the consideration of UK storm surges.) In the country that supplied the Fµkµshima plant, the USA, they are much more cautious about a tsunami hazard where relevant.
One thing that may not be appreciated about nuclear accidents is that they may have benefits as well as costs. According to Mikhail Gorbachev, the last ever Soviet president, Chernobyl was the key event in triggering the collapse of the Soviet Union:
"The nuclear meltdown at Chernobyl 20 years ago this month, even more than my launch of perestroika, was perhaps the real cause of the collapse of the Soviet Union five years later. Indeed, the Chernobyl catastrophe was a historic turning point: there was the era before the disaster, and there is the very different era that has followed."
Chernobyl appears to have inflicted huge damage on the confidence of the Kremlin in Soviet science and technology. Luckily for us in the West, they appear to have adopted a Greenpeace-like view of the incident where they thought hundreds of thousands of people, if not millions, would die. Subsequently the official death toll after an assessment by the IAEA in the early 00s has turned out to be only 50, with 4000 estimated long-term deaths. If the Kremlin had known the IAEA figures back in the late 1980s, the Soviet Union and the Cold War might still be going strong.
On thorium reactors, I would advise people to be sceptical. The Greenies have a tendency to be keen on the idea of there being some sort of Green version of nuclear power waiting in the wings, as it has the useful effect of increasing the amount of dithering by our technically-challenged politicians on making decisions on the current working version of nuclear power. For decades this contribution to political dithering has been reliably provided by nuclear fusion, and the Green NGOs have always been ambiguous about whether they support nuclear fusion or not. But a few years ago, Greenpeace decided to clarify their position, and as might be expected they're against it, as discussed in this article:
After Greenpeace came out against nuclear fusion, the idea of thorium reactors, which I don't recall ever being previously mentioned in the news media, has been increasingly talked up as the new Green version of nuclear power. THe main proponent of thorium power in the UK seems to be the former Friends of the Earth activist Bryony Worthington, who was heavily involved in drawing up the UK's Climate Change Act. Worthington, who is now in the House of Lords, was previously in charge of anti-nuclear power campaigning at FoE, so her promotion of thorium reactors looks a bit suspicious.
I agree getting the risk down is the whole point, but all the major nuclear accidents seem to have involved unexpected things happening. Calculating risks of that is quite hard. Sure there is zero to minimal risk of tsunamis taking out any of the UK's power plants, that doesn't stop a plane crashing on them though, or more likely someone intentionally driving a plane into one.
So suddenly we need to design nukes to withstand a direct hit from an A380 with full tanks at the fastest you can hit the ground with one. I'm no architect but I guess that has to be basically an impossible to achieve level of safety. I guess this also has never been considered as the previous "worst case scenario" would have been the Russians firing nuclear missiles at nuclear power stations. But if it had come to that we would have been irretrievably Fµkµushima'd anyway, so there was no point guarding against it.
Flying planes into nuclear reactors...
There are two very important things to remember.
Planes have the smallest safety factor in engineering. Planes are really expensive because every pound matters. Weight is near the top of the list to minimize. Safety factors increase weight.
Designing the containment structure for nuclear reactors DO NOT have much weight consideration. Put a plane up against a wall of concrete, and the concreted needs to be pretty chinzy to lose.
Trying to hit the reactor is the bigger challenge. If there are cooling towers, those cooling towers are great targets. They aren't reactors. The reactors are typically in much smaller buildings... Much harder to hit.
The storage facilities attached to the Nuclear Reactor are perhaps a better target, but those facilities are usually pits. Hit the building and the plane travels right over the dangerous materials. The steeper the dive you try to make in a big airplane to hit such a building the harder the task...
Nuclear Facilities are terrible actual terror targets. They are awesome as virtual terror targets though.
Why don't we put the reactors underground?
That might be a stupid question coming from a nuclear engineer but I'm not one.
I agree hitting a small target with an A380 or 747 flying near-vertically at as fast as it can go without breaking up mid-air is a tough call (much much harder than flying a jumbo stable and level into the side of a building), but you really, really don't want anyone to succeed at it. For most western nukes that really is the worst case scenario, other than an internal problem leading to a meltdown. So people are going to (not entirely without reason) insist you engineer for it.
The best first line defence against planes might be missiles.
If you have missiles ready to respond the window of opportunity depends on finding some set of conditions where missiles cannot be used.
By the way, I am rather less sanguine about nuclear power than I was because it occurs to me that if we are prepared to let an unsupervised literature degreed green activist write the Climate Change Act (Bryony Worthington who was lent to the government by Friends of the Earth and now a Baroness) who, I wonder, will they get to direct our nuclear power policy and will they be looking to over protect us or make an "I told you so " situation?
PS I can't say I'm happy that the government (in the form of the MOD who are answerable to the government, in principle anyway)are going to allow our nuclear deterrent be looked after by a private sector company.
I guess our troops are too busy being sacked or sent to act as spectators at the olympics to look after any military hardware themselves....though it seems to me the MOD is very good at spending money on the wrong thinsg and then not letting to troops near any of it in case they want to use it.... such is life.
Dive bombing required special engineering. I suspect that special overrides would be involved in diving a fly by wire plane in the ground. Unless the concrete containment of the tertiary containment unit is really really badly done, it is going to beat the plane. Flying a plane which is safety factor limited against a building that is safety factor limited give much higher chance of success. The trade centers failed, not in their collapsing, but in not surviving just a little longer before collapsing. It is possible that had they continued using asbestos in the process instead of the slightly less capable material, it might have survived longer.
Building reactors underground isn't a Nuclear Engineers job, it is a Civil's worry.
Looking at the responses on the issue of planes flying into nuclear plants, it may be worth adding a few things to the thread.
As DaveE is suggesting, there have been studies carried out which have concluded that nuclear plants can withstand impact by airliners. This link gives an example of a US study carried out in 2002 which considered impact by a Boeing 767:
Note that the validation status of structural impact computer models is way above that of the computer models we normally talk about in this forum, namely climate models. The Greenies only accept computer models which give politically correct results. so they would reject any computer model which demonstrates nuclear safety, whilst at the same time accepting the far more dubious climate models as being settled science.
Brad's point that an aircraft is pretty flimsy in comparison with a nuclear plant concrete outer wall is illustrated with this Youtube video clip (narrated by Carol Vorderman), which shows an F4 Phantom disintegrating when it hits a concrete wall:
One point that could be made regarding a repeat of a 9/11 incident in the future is that the 'easier' route to gaining control of a plane, by hi-jacking the plane, is effectively now closed down. It is now much more difficult to smuggle weapons on to a plane due to increased security at many airports and the cockpit area of airliners is now effectively sealed off. Also the attitude of airline crews and passengers to being hi-jacked is substantially different following 9/11. Before 9/11 it made sense to co-operate with a hi-jacking and treat it as being a bit like an armed robbery; after 9/11 it makes a lot of sense to attempt to mob and overpower the hi-jackers if you think they intend to crash the plane or blow it up. So I think a future 9/11 incident would require a terrorist organisation to recruit an airline pilot for a suicide mission, or have some of their people train and get jobs as airline pilots to be used as 'sleeper agents'.
On JMW's suggestion that ant-aircraft weapons could be used to defend nuclear plants, I think the prefered practice is to send up military fighter jets to investigate any slightly suspicious behaviour by airliners, and the most likely targets in the UK are expected to be skyscrapers and landmark buildings in London rather than nuclear plants, which as Brad points out could be tricky to hit. Most countries accept the idea that it may be necessary to shoot an airliner down if it is seen as a significant threat, but there is one exception to this - Germany (where JamesV is resident). In Germany they did introduce a law that allowed commercial airliners to be shot down, but it was ruled as being unconstitutional as it contravened the human rights of the terrorists (there was a rumour that German greens were behind this ruling):
On underground nuclear plants, the idea does exist, but only for small scale plants of not more than a few tens of MW. Sweden built an underground nuclear plant called Ågesta about 50 years ago that was under a suburb in the capital city, Stockholm. There is a proposed modern day small scale reactor called "Hyperion", which supposedly has a hundred orders, which is intended to be buried underground. As I understand it, burying nuclear plants underground is regarded as too expensive for normal sized plants, and is probably not compatible with the fairly high water table that occurs at many sites.