Thank you.
My name is Gordon Thompson. I'm the executive director of the Institute for Resource and Security Studies in Cambridge, Massachusetts. I am a research professor at Clark University in Massachusetts.
I've been working on nuclear safety and security issues for about 30 years. Some of those projects have been in Canada. For example, I worked for the Ontario nuclear safety review in 1987 and for the Senate energy and environmental committee in 2000. I have prepared a report for Greenpeace Canada assessing the Nuclear Liability and Compensation Act. I assume Greenpeace will make that report available to you. The report contains my own views and is not in any way dictated by the client, Greenpeace.
The report covers a range of issues. I'd like to focus here on a technical issue that I believe is important for the committee. That issue is the probability and magnitude of a release of radioactive material to the environment. My thesis in this area is that the committee--and through it, the Canadian Parliament--has not been properly informed about the risk of a large release of radioactive material.
Mr. Hénault, who is before you, stated at a conference in Toronto in October that the liability limit of $650 million has in part been set because it addresses what he terms “foreseeable accidents” rather than catastrophic Chernobyl-type accidents. What Mr. Hénault is actually referring to is what is known in the industry as a design-basis accident. That is an accident that a nuclear power plant is designed to accommodate, and that is what he's referring to when he describes an accident as foreseeable.
We know that a very large release to the environment occurred from the Chernobyl nuclear power plant in 1986. That was certainly an event outside the design basis. It's also common knowledge that in 1979 there was an event at the Three Mile Island nuclear power plant in Pennsylvania, also outside the design basis. But that event did not lead to a large release of radioactive material. However--and this is not as widely known as I believe it should be--at some other nuclear power plants then operational in the United States, the same sequence of events would have led to a large release of radioactive material. And had the accident continued further down the sequence that it was following, it could have led to a large release at Three Mile Island.
The historical occurrence of these events suggest that their analogue in the future is a foreseeable event. Indeed, there is a large body of technical analysis to show that these two events are not aberrations. They are instances of design-basis accidents, and such an accident could occur at any nuclear power plant anywhere in the world.
The technical analysis that covers this area is known as probabilistic risk assessment, PRA, or sometimes probabilistic safety assessment, PSA. The Canadian nuclear industry, the Nuclear Safety Commission, and its predecessor, the Atomic Energy Control Board, are aware of the field of probabilistic risk assessment and have conducted studies in this genre.
I regret to say that the quality of these studies and their completeness does not reach the level that was set for this field of technical inquiry by the U.S. Nuclear Regulatory Commission in its study, NUREG-1150, which was published in 1990. That is a point of reference from which one can judge the quality and completeness of technical studies of severe accident potential at nuclear power plants.
Now to the implications of this field of study and the two historical events I mentioned for the Nuclear Liability and Compensation Act, specifically the liability limit of $650 million.
First, studies done by the Canadian nuclear industry show potential exposures of the public to radiation, which, when monetized at the rate used for occupational protection decisions in the nuclear industry, yield health costs that can be in excess of $50 billion--clearly an amount far in excess of $650 million.
As another example, Defence Research and Development Canada estimated the cleanup costs of a dirty bomb event at the CN Tower in Toronto. Depending on the cleanup standards used in this analysis, the cost of cleanup could reach $250 billion. The amount of radioactive material that was assumed for that study is one two-thousandth part of the inventory of that same radioactive isotope in the reactor core of an existing CANDU nuclear power plant.
The Canadian nuclear industry claims that the probability of such an event at a nuclear power plant is extremely low. I dispute that finding. The dispute is a technical issue, which clearly is not appropriate to argue before the committee at this time. However, it is important to note that we know from available evidence about the premiums set by nuclear insurers to provide insurance to nuclear power plants. These insurers assume a probability of release many orders of magnitude in excess of the probability claimed by the Canadian nuclear industry. I would therefore suggest there's a case for the committee on that fact alone to be extremely cautious in accepting the industry's claims of a very low probability.
In conclusion, I would argue that the committee, and indeed Parliament, before enacting this proposed legislation, should request that the Canadian government provide a much more thorough and complete and open analysis of the risk of a large release of radioactive material and the offside costs of such a release.
Thank you.