On Mar 19, 1:42*pm, Alan wrote:
In message , "Nightjar
\"cpb\"@" wrote

He actually said the evidence is that there had been three in 3,000 years.

Then surely, statistically, there are not enough occurrences to make any
meaningful prediction of it being a 1 in 1000 year occurrence? With so
few points you wouldn't know where you were on the curve. All it can be
said is that it happened before and it is probably that it will happen
again - and within the 40/50 year predicted life-span of the facility.

Exactly correct. Insufficient data to establish any statistics. Random
events.
Maybe a curve could be established in 100,000 years.