Contrast different power sources in a disaster

I’m not trying to prove any kind of power is inherently safe or dangerous, just trying to learn the relative risks of various kinds of electrical power generation at central plants under extreme, but not impossible conditions. First, let’s list some general kinds:[ol][li]Coal fired[]Gas fired[]Nuclear[]Hydro[]Geothermal[]Solar[]Wind farms[]Wave[/ol][/li]Now assume various natural disasters.[ol][li]Hurricane/typhoon[]Earthquake[]Tsunami (only applicable for coastal units)[](small but significant) Object from outer space impact[/ol]How do these power sources compare for safety?[/li]
If I’ve left out major sources of power or disaster, please add to my list.

Just to get things started, here’s how I see any of these disasters affecting the plants:[ol][li]A mess, fuel all over, possibly fire[]A mess, fuel all over, possibly fire[]A mess, fuel all over, possibly fire, radiation hazard, meltdown, explosions (but not nuclear bomb-type)[]Dam burst, downstream decimated[]Steam & hot water everywhere, explosions[]A pile of pipes and panels[]Blades spinning off, towers fallingSplat! Decimation of facilities. We’ll have worse problems other places.[/ol]Obviously some of these are harder to clean up and some may not be rebuildable after (if geothermal shifts the magma chamber, it might not be possible to generate energy there anymore).[/li]
Any better analysis?

Are you measuring human casualties? Property damage? Overall power released?

In terms of human casualties I’d say dams breaking are the worst possibly equalled by earthquakes. This of course depends where the disasters happen. A dam break may not kill anyone if there is no one down stream for a long way. On the other hand, in China, a dam break is considered responsible for 170,000 deaths (both directly and indirectly). (cite)

Earhquakes likewise can cause massive loss of life and wreckage or none at all.

Tsunamis obviously can kill huge numbers as it did in Indonesia a few years ago. Japan also had great loss of life from the tsunami but nothing like Indonesia experienced.

Coal is more vague but according to this article it shortens the lives of 24,000 people a year in the US alone. (cite)

Hurricanes are also likewise devastating. Particularly because they can affect huge areas.

Object from outer space can be everything from no big deal to end life on the planet.

In comparison nuclear power barely registers…even if you include Chernobyl.

Can I ask why you have explosions down for nuclear but not for oil, natural gas or coal?

Yes, if they directly relate to a power plant’s reaction.

A dam break is localized. A radiation release can be widespread.

But we’re considering only power plant-related factors in this thread.

But we’re only considering power plant-related factors in this thread.

But we’re only considering disaster-related factors in this thread, not long-term, non-disaster factors.

But we’re only concerned with power plants in this thread.

But we’re only considering power plants in this thread.

Obviously you have a personal problem with nuclear energy, but let’s try to look at this fairly, OK?

Please add them to the list. I don’t think explosions would be appropriate for solar, hydro, wave or wind (although some solar uses a liquid, and that could explode).

Any plant including some solar ones with a boiler can explode.

But not photovoltaic. And how does a wind turbine explode, other than spinning off blades? How does a wave generator explode?

As I said, boilers = explosions. You can also have explosions without boilers, but some power sources without a boiler are unlikely to explode.

the first “plant-facility” problem from the quake in japan was a massive fire followed by a major explosion at an oil refinery.

i was actually just wondering about that–about how everyone will be all down on nuclear energy now but oil plant fires and explosions are common–so common they are a bit of a “meh” catastrophe compared to the nuclear plants.

There is a huge emotional content to nuclear plants, more so in Japan.

As I’ve mentioned it in another thread I’ll link to it here.

The Vajont dam disaster in Italy was as a result of hydroelectric power generation. over 2000 people killed.
Every major engineering undertaking has its risks.


You repeated over and over that you are looking at power plants in this thread.

In terms of people killed from power generation:

Coal far and away beats anything else in terms of people dead.

I suspect oil is right in there with coal.

Hydro I would call as second.

After that I think the numbers become so small they are barely worth noting in the big scheme of things.

Nuclear power plant deaths are vanishingly small in the scheme of things. 4,000 for Chernobyl and that is a standout. Nothing else comes close from nuclear power plants.

Yes, from natural disasters, as stated in the OP.

From natural disasters?

Trying to actually tackle the questions the OP posits:

Any plant with flammable fuel can have the stored fuel ignite or explode or have an out-of-control fuel-fed fire in the machinery. But the adverse consequences are pretty immediate and pretty local. If you live a mile from the plant, you’re very unlikely to be injured. And the excitement is over in minutes or a few hours tops.

Any plant with a high pressure system, e.g. a boiler, can have a burst. Again the direct effects are local and short-lived.

Any plant with large or rapidly-moving parts can have a mechanical failure, ejecting a bunch of uncontrolled kinetic energy. Again the direct effects are local and short-lived.

Hydro plants have the unique issue that in addition the the kinetic dangers above, the potential energy stored in the retained pool is vastly larger than the plant’s output. So a malfunction which breaches the dam can release vastly larger quantities of energy & have non-local effects far downstream. Depending on the pool size, the excess energy can be dumped downstream for hours, not minutes.

Nuclear plants have the high pressure & mechanical risks metioned above. Plus the obvious radiation leak risks. Which are both potentially non-local and potentially long-term. As we’re seeing in Japan, despite some fairly good containment design, it’s possible for things to get out of hand.

Finally, for any plant of any type, the human impact of the failure depends a lot on where the people are versus the plant. The same failures have very different consequences for plants surrounded by urban areas vs. plants in desolate unpopulated areas.
I think it’s be interesting for the OP to clarify whether he was interested in ranking plant types by worst-case possible outcomes or by statistically most likely outcomes or by historical outcomes to date.
I think it’s fair to say that nuclear plants have the worst worst-case of the various types, with some hydro plants (e.g. Three Gorges Dam) also having pretty horrific worst-case failure modes.

All the other types have failure modes so benign by comparison that they hardly warrant inclusion in the worst-case list.
As others have pointed out above, an actual policy risk analysis requires a lot more thinking than just a ranking of worst worst cases. Actual historical failure likelihood information, as well as statistics about quantities of plants by type and site-specific isses like surrounding population and downstream river course topography would be necessary to answer questions like “What is the safest way to generate the electricity needed for Alabama?”

Not statistical at all, more anecdotal, but here’s an example of a coal “natural disaster”: Aberfan, which killed 116 children and 28 adults.

My opinion correlates closely with the previous post by LSLGuy.


How did you gather that from what I posted?

Do a search on my posting history on nuclear energy. I actually want to see new plants built (with the latest technology as well as reactors that will consume waste from “normal” reactors).

You are trying to compare damage from power generation to damage form natural disasters.

I am still not sure what measure you want to use.

Frankly, you should do your own work on this.

It is easy enough to Goole various disasters and tally the death toll or property damage caused by them.

How so?

Coal plants don’t generally blow up but slag heaps from coal plants have caused serious damage.

No American has been killed by nuclear power. Chernobyl got 4,000. Minus Chernobyl deaths due to nuclear power barely register.

Not sure how you are adding this up.

Stop and think for a moment then.

Consider it like this. You have a hundred mile length of shoreline. At twenty mile intervals you have,
a) A coal powered plant,
b) A nuclear plant
c) A gas fired plant
d) A hydro-powered plant
e) A geothermal plant.

A thirty foot high Tsunami hits the shoreline, compare and contrast the effect and the results on each of the power plants. Repeat the exercise with other disasters, and other types of power plant.

Get it? The numbers of deaths from normal everyday use of the plants themselves are irrelvant here, no matter how many thousands have coughed themselves to death over the years.

Hydro plant is potentially the worst.

Of all the rest not one has killed anyone after such a disaster (except, perhaps, for workers who might die in a boiler explosion or such).