Re: RADSAFE digest 3143

[Holloway3@aol.com]

Norm writes:

<< According to my friends at the Tooth fairy Project here's the answer to 
one of
 your questions:
 
 > Q. How do you differentiate Sr90 from bomb tests and from reactor 
emissions?
 
 > A. You can't.  However, if you find (like we have) that there has been
 > virtually no change in Sr90 levels for children born in the early 1980s, 
late
 > 1980s, and early 1990s, it's got to be something other than old bomb test
 > fallout decaying.  The only other answer is reactor emissions. >>

Norm, 

Thanks for your efforts in obtaining this information. 
It would be interesting to see the geographic location of the teeth.  
Apparently, the Tooth Fairy Project is not yet releasing any information, 
even on their analytical techniques.  (I note that as of  the middle of 1998, 
they had collected 400 teeth).  There are many other possible answers, in 
addition to your favorite one.   I can think of a number of other 
explanations as to why the Sr-90 levels might be almost constant over a 
decade.  For example, how can you rule out a change in food sources to 
sources that might have a higher Sr-90 content, enough to counteract the 
decline due to decay?  People's diets do change over time and one thing that 
is certain is that Sr-90 is not at the same level in all foods.   The time 
period from the early 1980s to the early 1990s corresponded to a big 
improvement in the economy.  You cannot rule out that this improvement may 
have changed the average diet and with it, perhaps the average Sr-90 content. 
 There are many other possibilities, contrary to the claim that it can only 
be due to reactors.   The answer favored by the Tooth Fairy Project, that of 
reactor emissions, does not make sense if the emissions are insignificant.  
Incidentally, the time period mentioned above, where the claim is made that 
there is virtually no change in teeth, covers only a 10 year period - short 
in comparison to the 29 year half -life of Sr-90.  A large decrease would not 
be expected even if all of it is from fallout.    I am not sure what is meant 
by "virtually no change" - perhaps they are admitting a small decrease? In 
any event the preliminary results you mention are not especially remarkable.  
One other important factor is the geographic origin of the teeth.  Sr-90 from 
fallout is not evenly distributed over the country.  Considerable variations 
are to be expected.  In general, rainy areas of the country will have a lot 
more Sr-90 than dry areas and there may also be variations due to latitude, 
acidity of rainfall and soil type.  I understand that the teeth collection is 
coming from many different locations.   They are also going to have to take 
into account the large regional differences in the Sr-90 content of the soil. 
 Unless the project also studies food sources and the Sr-90 in those food 
sources, then it would seem to be of limited value. For example,  one could 
imagine, if your friends are right, that some particular food, grown a few 
miles from a reactor, might be elevated in Sr-90 and that this elevation 
could then be traced to particular teeth.  But oddly enough, the idea of 
measuring Sr-90 near a reactor, in food crops, does not seem to have been 
included in the Tooth Fairy project.  It is odd that they do not want to try 
the obvious and most straight forward approach, that of trying to trace the 
Sr-90 from its source to the teeth.  

But Sternglass and Gould are doing something that is new for them.  They are 
producing measurements of their own, although perhaps not personally doing 
the work.  The big advantage of doing one's own measurements is that you 
learn something about the limitations of your results and the uncertainty of 
the measurements (all measurements have uncertainty, whether stated or not).  
Previously, they worked mostly with death rates of various sorts produced by 
various government agencies.  I suspect that no one really knows the quality 
of that data and especially how the quality has changed over several decades. 
This is an issue that they never seem to have mentioned.  

To give an example of the pitfalls of drawing conclusions from data produced 
by others, I will cite the recent paper by Mangano in Environmental 
Epidemiology and Toxicology (2000) 2, pp. 32 - 36.  Mangano attempts to show 
an effect of the Rancho Seco Power plant by citing slight increases in the 
level of Iodine-131 present in Sacremento's pasteurized milk.  He mentions 
that while the reactor was closed for repairs, the I-131 content was 1.91 pCi 
(presumably per liter) and after restart, it rose to 2.67 and 2.54 in 1988 
and 1989 and fell to 1.82 in 1990. What he did not mention was the 
uncertainty in those measurements.  If these measurements are typical, they 
are very near the detection limits and the uncertainty (error term)  may be 
quite high, perhaps as much as 40%.  So there is a good possibility that the 
numbers he cited have overlapping error bars and therefore the slight 
differences are not of any statistical significance and his conclusions are 
faulty.  Whatever the size of the error term, he was working with numbers 
that are extremely low and most investigators learn not to make too much out 
of differences between numbers that are very near the detection limit. 

R. Holloway
holloway3@aol.com
************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html