According to the Government's official radiation safety advisers, it is only by making such assumptions that the risks of nuclear waste disposal can be properly assessed.
The National Radiological Protection Board yesterday recommended tough new limits governing the safety of waste disposal. The annual risk of fatal cancer or hereditary defects resulting from nuclear waste disposal should be less than one in a million a year, the board proposes.
This is 10 times more stringent than previous recommendations, according to Dr John Cooper, who heads the board's environmental assessments division.
However, the board has relaxed one constraint: the nuclear industry no longer has to prove that radiation doses will be within the limits for all time; instead it need calculate them 'only' for the next million years.
The future scenario for west Cumbria - where the nuclear industry proposes to bury its waste - will only apply some 300,000 years hence, when the maximum flux of radioactivity from an underground repository would hit the biosphere. Dr Cooper recommended, for the purposes of calculation, 'if you place a community of subsistence farmers in the area where the flux of radioactivity is highest and you control the risk to them, then in all probability you will be controlling the risk to everyone'.
Radioactivity will eventually leak out of any underground repository, generating risks. The board's basic criterion, Dr Cooper said, was that people alive in the future should be given a level of protection at least equivalent to that accorded to people alive now.
If the annual risks turn out to be 10 times greater (one in 100,000) than recommended, then the nuclear industry must consider other sites and other ways of handling the waste, the board says.
A spokeswoman for UK Nirex, the industry's waste disposal company, welcomed the report.' It is too early to say if Sellafield will meet the one in a million criterion. We are hoping to meet that target, but need to do more assessment work,' she said.
Patrick Green, from Friends of the Earth, said: 'The whole concept of putting forward a safety case for timescales that transcend human experience is essentially an impossible task.'