FW: RADSAFE digest 2867

["Dukelow, James S Jr" <jim.dukelow@pnl.gov>]

I was catching up on some back RADSAFE Digests when I happened on the 11 January
2000 posting by Peter Crane, who wrote:

-----Original Message-----

	<snip>

 
Rather,
nuclear power was presented as something preternaturally safe -- so much
so that the Executive Summary of the 1975 Rasmussen Report (issued
shortly after the creation of the NRC) compared the likelihood of being
harmed by a nuclear accident to that of being injured by a falling
meteor.  Is it any surprise, therefore, that the Three Mile Island
accident came as an immense shock to the American people?  When a
serious accident occurs, and it is of a type that the Government's
nuclear experts had deemed not to be credible, how credible are the
experts at that point?  
   At this point, the apologists for nuclear power will be objecting
that nobody was hurt at TMI, and that TMI really proved how forgiving
and how safe nuclear technology is.  That's the best possible spin one
can put on it.  But to the mass of the population, the lesson of the
partial core melt at TMI is a simpler one:  you told us that kind of
accident couldn't happen, and it did.

	<snip>

   For myself, in 25 years of observing the nuclear industry and the
nuclear regulatory establishment, the most striking phenomenon is the
never-ending quest to find scapegoats.  Does anyone ever mention that
four years before Three Mile Island, the nuclear industry suffered a
staggering blow from one of its own, when Westinghouse repudiated its
fuel supply contracts, and at once, a large number of plant orders were
hurriedly canceled?  

	<snip>


JSD Comment:  Crane makes a number of assertion in his posting that seem to me
suspect, on which I have not able to comment, but the two items I selected have
some interesting misconceptions and misstatements of history.

The Rasmussen Study Executive Summary has a table asserting that, for 100
nuclear reactors, the frequency of an accident causing more that 100 fatalities
is once in 100,000 years of operation, and the frequency of an accident causing
more than 1000 fatalities is once in a million years of operation.  The table
gives exactly the same frequencies for fatalities due to meteor strikes.  I
think Crane's outrage at this assertion is based on faulty intuition rather than
data.

A paper in Nature, 6 Jan 1994, v. 367, pp. 33-39, "Impacts on the Earth by
asteroids and comets: assessing the hazard", by Clark Chapman and David
Morrison, estimates an individual's chance of dying from an asteroid/comet
impact as 1 in 20,000, with an upper bound on the frequency of 1 in 3,000 and a
lower bound of 1 in 250,000.  The major contributors to this estimated risk are
the impacts of objects greater than 600 m diameter and less than 5 km diameter,
which they estimate as occurring once every 70,000 to 500,000 years, with
estimated consequences of 1.5 billion people killed.  Contributions to the
overall risk from larger and smaller objects are significantly smaller.  They do
not fully consider the contribution to risk from objects several hundred meters
in diameter falling into the ocean and causing tsunamis that kill several tens
to hundreds of millions of people around the ocean basin impacted, although they
note that tsunamis might increase the risk tenfold.  I think if you look at more
recent studies of the likely quantitative consequences of nuclear accidents, you
will find them in the same ballpark with the risk of meteor/asteroid impact,
when it is rightly considered.

Second, one of the striking aspects of the Rasmussen Study was that it suggested
that accidents other than the large break loss of coolant accident that had been
used as the design basis for reactors build up to that time were greater
contributors to the risk of core melt.  After TMI, you could go back into the
Rasmussen Report and find the TMI accident identified and its risk reasonably
estimated.  Since both the early WASH-740 Study and the Rasmussen Report
contemplated serious reactor accidents and reported consequences (in WASH-740
and Rasmussen) and estimated frequencies (in Rasmussen), it is not correct to
say "you told us that kind of accident couldn't happen".

Finally, Crane gets the flow of causation on the Westinghouse fuel fiasco
exactly backward.  Westinghouse had sweetened a number of sales of reactors to
utilities with a promise to deliver uranium fuel for a set period of time at a
set price.  They had not put in place any contracts with uranium suppliers for
those quantities of uranium.  They had sold uranium short.  Then the Arab Oil
Embargo happened, increasing the price of oil by a factor of three or four
overnight.  Coal, gas, and uranium prices followed oil prices right up.
Although utilities had long argued that the demand for electricity was inelastic
-- people had to have a certain amount and would pay any amount for it -- that
turned out not to be the case.  Lots of utilites found themselves with nuclear
plants tentatively on order that they weren't going to need to meet reduced
demand.  So they cancelled orders.  Rumor around the industry was that
Westinghouse ended up paying around a billion dollars in cash, goods, and
services to the utilities that sued it after it abrogated the uranium contracts.

As an aside, about ten years later Westinghouse management again blew off around
a billion dollars on real estate speculation that turned sour.  A few years
after that they bought CBS, sold off their nuclear and electrical businesses,
changed their name to CBS, and became a media company, where they have an
opportunity to do some real damage to society.

Best regards.

Jim Dukelow
Pacific Northwest National Laboratory
Richland, WA
jim.dukelow@pnl.gov

These comments are mine and have not been reviewed and/or approved by my
management or by the U.S. Department of Energy.
************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html