Are nukes safe?

With the threat of nuclear "meltdown" adding to the troubles in Japan, and even more so adding to the worries of the US, I've been thinking about nuclear power. Darwin Catholic has a lot of good links and information on his site today. The question that people seem to be debating (and have been for a long time) is should we use nuclear power in light of its safety record.

I found a document (Comparative Assessment of Natural gas Risks) online about the safety of various means of energy production. The aim of the paper is to demonstrate the safety of natural gas vs. other power sources, so we should keep that in mind. However, it does present a rather complete analysis of data across a wide range of power sources. Here is a summary of some of the raw data, from 1969 through 2000. The data includes "severe" accidents that resulted in one or more of the following:

1) at least five fatalities or
2) at least ten injured or
3) at least 200 evacuees or
4) extensive ban on consumption of food or
5) releases of hydrocarbons exceeding 10’000 t or
6) enforced clean-up of land and water over an area of at least 25 km2 or
7) economic loss of at least five million USD(2000).

Whenever any one of the above criteria is satisfied, the accident is considered to be severe. Here are the numbers presented.

Technology	Accidents	Fatalities	Notes
Coal	1130	20,510	(a)
Oil	455	21,424
Natural Gas	159	2,273
LPG	124	4,436
Hydro	11	29,938	(b)
Nuclear	1	31	(c)

(a) Including China
(b) Banqiao and Shimantan dam failures together caused 26,000 fatalities
(c) Latent fatalities are treated separately.

By these criteria, nuclear is extremely safe. However, as noted the table does not include latent fatalities. Here's what they have to say about latent fatalities:

The presence of nuclear in these tables is primarily due to the Chernobyl accident, with a contribution from the Three Mile Island (TMI) accident to the economic losses and evacuation. Estimates of latent fatalities and latent cancers are only available for the nuclear chain for which they are of particular relevance. Delayed fatalities are likely to have occurred for the other chains with no records available; their significance per accident should, however, be incomparably smaller in comparison with the Chernobyl accident.

Later in the paper nuclear latent fatalities are given as 33,000. No figures for latent fatalities from other sources are given. This presents a dilemma. One can't compare the technologies' relative safety without all the data. Not including latent fatalities, nuclear is clearly the safest energy source there is. Including latent fatalities for only nuclear energy make nuclear look bad, but we don't know what the numbers would be for other technologies. For instance, how many people die of latent effects of chemical exposure in an oil accident?

Adding to the uncertainty is the assumption that these technologies do not result in deaths when they are working properly. For instance, should we include black lung deaths to be latent deaths from coal generation, even though they aren't associated with an accident?

In the case of nuclear, there are additional issues. The number of latent fatalities is itself a hotly disputed subject. Claims for deaths from Chernobyl range from 4,000 to 500,000. According to the IAEA's (International Atomic Energy Agency) publication "Ten Years After Chernobyl":

Compared with other nuclear events: The Chernobyl explosion put 400 times more radioactive material into the Earth's atmosphere than the atomic bomb dropped on Hiroshima; atomic weapons tests conducted in the 1950s and 1960s all together are estimated to have put some 100 to 1,000 times more radioactive material into the atmosphere than the Chernobyl accident.

They also note that the design flaws of the Chernobyl reactor have been fixed in other reactors.

The most serious deficiencies in other operating RBMK reactors are being addressed through safety upgrades. Between 1987 and 1991, a first stage of upgrading was performed on all RBMK units to eliminate the design deficiencies which contributed to the Chernobyl accident, to improve shutdown mechanisms and heighten general safety awareness among staff. There are plans for further safety improvements.

So, should Chernobyl be counted? Nuclear reactors aren't designed and built that way anymore, and the ones that were built that way have been improved, so the failure that occurred there shouldn't in theory be possible anywhere else. On the other hand, it can be argued that all technologies improve over time and if we exclude that event we should exclude events from all technologies that were not up to the latest safety standards. That would leave us with virtually no data to work with, and that would be skewed by the current state of the art in that field.

So I worked the numbers both ways - with and without nuclear latent deaths, and using the estimates in the original article of 33,000 latent deaths from Chernobyl. Understand that the truth lies somewhere between the two sets of figures.

Lastly, there is disparity in the number of deaths because some technologies are more prevalent than others. The following table shows fatalities per GWeyr (fatalities per Gigawatt of electricity years). This is IMHO a more meaningful number than just fatality numbers, because it takes into account the fact that more electricity is generated by coal and oil than hydroelectric and nuclear.

	Fatalities per GWeyr
Technology	Immediate	Latent
Coal	1.762	--
Oil	1.032	--
Natural Gas	0.196	--
LPG	16.853	--
Hydro	10.288	--
Nuclear	0.048	48 (6.4-800)

For nuclear, 48 fatalities per GWeyr is the figure used in the document. Using the 4,000-500,000 numbers results in the range shown (6.4 to 800 fatalities/GWeyr).

So nuclear is the safest or the most dangerous power source, or something in between. If we consider all historical data, nuclear energy appears to be far less safe than other methods of electricity production (assuming latent deaths from other technologies are not significant). However, if we consider the Chernobyl accident an anomaly and consider only modern, properly designed reactors, nuclear becomes the safest method of electricity production by far. It all depends on one's assumptions. So, I'd have to say that the results are inconclusive. There are arguments to be made on either side.

What did surprise me in researching this was how dangerous hydroelectric power is. To be fair, 90% of those fatalities were due to two disasters in China, and if those are considered anomalies then the safety of hydro would be in line with oil. I didn't consider this data to be anomalous because it was not due to a flaw in the design, but a natural disaster, just as the current nuclear accident in Japan is due to a natural disaster. One can argue that a bad design, once fixed, is eliminated from future consideration, but a natural disaster such as a flood or earthquake is not controllable by humans.

Posted in: energy,technology