Earthquake Zones and Nuclear Power Plants - A Clarion Call?


This map shows a heatmap of 175,000 4.5+ magnitude earthquakes since 1973 based on data from the USGS (United States Geological Survey). And worldwide locations of nuclear power stations using information from the IAEA (International Atomic Energy Agency).

An alarming number of nuclear power plants are located within active earthquake zones. Will this be a clarion call for more stringent safety standards?

According to the maptd blog:

Following the incidents at nuclear power plants in Japan after the earthquake I was wondering which power stations around the world are near active earthquake zones.

With this in mind, I've created a map combining two sets of information:

* A heatmap of every 4.5+ magnitude earthquake since to 1973 - around 174,000 in total
* The location of 248 atomic energy plants, including numbers of reactors. Represented by blue markers.

The seismic data are from the United States Geological Survey and the nuclear power station information the International Atomic Energy Agency. The map is built from two Google Fusion tables, here and here.

More like this

Is the glass half empty or half full?

I see a small percentage of all nuclear power plants located in the earthquake zones. And other than the Japan incident (earthquake plus tsunami), no problems have resulted from these few plants since 1973 as a result of earthquakes.

I would like to see the map modified showing chemical plants and storage facilities which also pose carcinogenic risk. They exist, but are not on the map.

Healthphysicit: And I have a hangnail and that is not on the map either. The conflation and earnest cross-indexing of risk within power generation options is bad enough. Must we also suffer this with unrelated things like chemical plants and storage facilities?

I do agree, though, that it could be a lot worse. And while we're at it, let's acknoweldge that those crazy antinuclear New Zealanders were right all along!

In truth, Fukushima did OK with the quake itself. the problem of the tsunami was greater. The possibility of such as tsunami was known and recognized and purposefully ignored by the regulators under pressure from the power company. And THAT, folks, is the problem with nuclear power. It is not true that it can't be done safely. But it is demonstrably true and utterly undeniable that it often WILL be done unsafely.

Greg, you are being ridiculous.

First, you use the word "safe" as if it has an absolute meaning. It does not. There is nothing "unsafe" or "safe" about nuclear power or anything unless you define those terms.

What makes the New Zealanders right about anything? You have to look at differences in population densities (especially production facilities), and available natural resources, as well as risks in order to determine a "best" energy path.

Nuclear power is done "safely". It could be "safer". So could automobiles, food safety, and chemical plants. You seem to be cherry-picking.

Healthpysicist. I agree that a word like "safe" is an oversimplification.

Nuclear power is done "safely". It could be "safer". So could automobiles, food safety, and chemical plants. You seem to be cherry-picking.

Hahahaha. Very good. Enjoy your fruit salad.

One of the problems that is involved is that the design basis of the reactors was set in the 1960s. As a result information on tsunami's that was found in the 1990s was not included. Note that TEPCO has had to do a lot of retrofitting of 1970s for seismic events (see TEPCO on Wikipedia) in fact their major plant on the Sea of Japan was shut down after 2007 for 3 years to retrofit.
However due to various things, the info that a tsunami of 14 m was possible did not get incorporated into the retrofit of the plant.
Note that the actual shaking of this event was not that extreme VII on Modified Mercali scale so well built structures should have been able to ride it out. So unless the plant is in a Tsunami prone zone there is no lesson on seismic here.

Fukuashima Daini got hit by the same quake and tsunami and is in safe cold shut down. There is probably something helpful in comparing the two as to why one plant took the beating and is still fine while the other broke down. Onagawa directly across from the epicenter only had a fire, now out, and is also in cold shut down.

adela, good question. Fukuashima Daini did have some problems. The cooling systems on three of the reactors failed but were restarted, but then because of flooding the heat sink (i.e., the Pacific Ocean) was not really available ... emergency cooling was implemented successfully, but undesirable extra pressure built up in one or two of the systems. There was a plan to release gases to reduce pressure but that seems to have not happened, or if it did, it happened without notice. One worker was killed there.

The difference seems to be in the level of damage suffered by the cooling systems.

healthphysicist: I'm glad to see I'm not the only one writing parodies of NPA's!

An alarming number of people live in earthquake prone areas. Unless you want to waste a huge amount of generated power, you put your generating facilities somewhere near the place the power is used. That's how you get nuclear power (and coal, and wind) near people and earthquake zones.

By Robert S. (not verified) on 25 Mar 2011 #permalink

The difference between the two Fukuashima plants could tell us if it is a design, procedure or just a shit happens issue.

I saw somewhere that the plants were built for an 5.7m tsunami which most likely would be based on a historical average and the back up power systems were at 10m above sea level. One more story in building height could have made all the difference.

Note the Daiichi plants were about 10 to 15 years older than the Danai plants, (1967 to 1973 for DaiIchi and 1981-1986 for Dani). This alone can make a difference. For example the Daiichi plants were designed to the 1952 Kern County Earthquake standards, while its possible lessons from the 1971 San Fernando Earthquake were incorporated in Dani. In engineering disasters make the best learnings, in Earthquake Engineering every big event teaches something new, that needs to be considered.
Re #10 in the case of Japan there are no earthquake free zones, the whole country is earthquake prone. So in that case locating where earthquakes aren't big is not much of an option.

I saw somewhere that the plants were built for an 5.7m tsunami which most likely would be based on a historical average and the back up power systems were at 10m above sea level. One more story in building height could have made all the difference.

This, in and of itself, isn't really much of a problem. Earthquakes are just one of many potential threats to a nuclear plant. As with most issues dealing with structures and the ability to handle stresses it comes down to being an engineering and design problem. Engineers can design a structure to handle pretty much any stress. Add more steel, concrete, cross-bracing -- brute force engineering design 101 stuff -- can carry you a long way. Add exotic materials and active isolation systems and engineers can design a plant that can handle anything mother nature can dish out.

This goes for safety and monitoring systems also. You can always add another redundant safety system. Stack them up and, as long they are truly independent, you can stack the odds in your favor to where the odds of the sun going nova are greater.

The real issue is cost. The trick is how do you design for forces and equipment failures while keeping the cost low enough to make the plant practical to build and keep the energy derived competitive.

By all reports so far the Fukuashima Daini plant weathered the earthquake in fine fashion. The problem was the tidal wave. Engineers were given a design parameter of something on the order of the plant handling a two meter wave and they likely built in a substantial safety margin. What they got was a three meter wave. The problem was not in the engineering design but the design goals they shot for. Goals that were selected based on estimated risks, probabilities, and a desire to keep costs low.

I'm doubtful that a simple ban on building on faults is the way to go. What about faults we don't know about. Of assumed dead geologic zones that break free?

The problem here was that the various risks were not estimated correctly and appropriate safety margins were not included. That would have prevented the problems to start with. Of course the unexpected happens. Which is why you cave redundant systems. The Fukuashima Daini plant was designed with backup diesels, batteries, and the option of getting external power. The wave took out he diesels, the earthquake and/or wave removed the external connections, and the batteries were only good for a short time. Triple redundancy defeated with a single blow. More modern nuclear plants are said to have more redundancy.

Sounds to me like what we need is better evaluation of risk, greater safety margins in design, and more layers of redundancy in case anything does go wrong. If that makes designing and building nuclear power plants prohibitively expensive ... then so be it.

In Japan there is a tradition of not saying no. Their preferred phrase is 'yes, but' followed by a laundry list of requirements and flaming hoops you have to jump through to get to do what you want. Perhaps we could learn something from this fine Japanese tradition.

thanks jeff for keeping us informed of on these matters.

@Health: Agree with the general tenor of all your comments. (And, on chemical plants, too. As any Texas or Louisiana resident knows, they're plenty more common in hurricane zones than in earthquake zones.)

@Lyle @altin @adela: Per Nova's episode on the quake/tsunami, nobody allowed for ground being lowered by quake into their calculations. I don't know about the Fukushima plants, but, in cities on the coast, in many cases, their seawalls were high enough -- if the ground hadn't subsided a meter or more.

@All ... Fossil fuels in general kill at a 4,000-1 ratio vs. nuclear, yet people don't get so worried. It's the parallel to the old flying vs. driving death rates issue.

@Health ... I wrote a newspaper column in similar vein the week after 9/11. I used different deaths, such as those from diabetes (as the Type II explosion was just starting to explode), smoking and a couple of other things.

I don't know whether this is more an issue of humans being poor at risk assessment or hiding their heads in the sand over what they dimly recognize is true on risk assessment.

I'm doubtful that a simple ban on building on faults is the way to go. What about faults we don't know about. Of assumed dead geologic zones that break free?