THE ABOVE CHART IS THE MATHEMATICAL FACT OF THE EPA's DATA. One has to wonder why months after the detection no one else in authority or expertise has so clearly and unequivocally stated what the mathematics of the EPA's data actually state? On the contrary, the EPA has it listed as a non-detect; the BRAWN team from the University of California Berkeley Unclear Engineering Department has interpreted the data to mean, "In statistical terms, this is "consistent with zero," i.e., a non-detection"
The facts of the mathematics belie the BRAWN team's "interpretation". I have postulated in the past that the University of California Berkeley Unclear Engineering Department's BRAWM team is all BRAWM and NO BRAINS; the following analysis hopefully will provide the BRAWN team with some insight and allow them to publicly correct their "interpretation" with the actual facts above.
The EPA data lists a detection of 0.0000013 pico Curies per cubic meter (pCi/M3) of Plutonium 238. They also identify a "Combined Standard Uncertainty" (CSU) of 0.000003 pCi/M3. The CSU is just a fancy way of saying that the EPA is guessing at how uncertain they are of their result. Even though it is just a guess, it is treated as if it were a mathematically calculated Standard Deviation (Sigma). The range of possible results given the EPA's number is represented by the blue bell curve shaped line in the chart above. The most likely answer is found under the highest peak of that blue bell curve. The extreme ends of the curve are that answer + and - three times the CSU (guessed Sigma). The less space there is under the curve, the less likely that particular answer is to be the correct one.
Looking at the Chart above, it is clear to see that 67% of the probable answers are to the right of the zero value, and 33% of the probable answers are to the left of the zero value. It is those values that are the actual probability of Plutonium detection. Of course one has to ask why does the EPA even lists detections with negative values, how is that possible? A negative value is only possible because the EPA massages the numbers before they release them; They subtract off a value which represents how wrong they guess their instruments are in terms of the actual detection quantity.
Moreover, notice that the EPA took the sample over 5 days, March 11-15. A more competent interpretation of the data provided by the EPA would increase the strength of the detection to 0.0000065 pCi/M3, and raise the detection level confidence to 98%. The significance of the measurement time frame is that Fukushima melted down on March 11; the travel time to Riverside would indicate that all of the detected Plutonium 238 would have been captured by the air filter no earlier than March 15. In short, that means the detection should not have been averaged over 5 days, but rather 1 day; hence, the detection is actually 5 times greater than the EPA indicated. A quick analysis of other CSU guesses from other detections shows they do not fluctuate much with the air sample size or number of days measured; hence, no change in the CSU.
The result is shown in the next chart, a 98% certainty of Plutonium 238 detection. All the exact same information holds true for the EPA's additional detection of Plutonium 239 during the same time period.
However, the analysis does not stop there. For the exact same March 11-15 time period, the EPA also has detections for Uranium-234, Lead-212, and Uranium-238; all of which are part of the nuclear chain for Plutonium 238. The fact they were all detected together makes them a veritable fingerprint identification of Plutonium 238 coming from a fuel rod. The detection of Plutonium 239 was also found with a similar fingerprint of Uranium 235. By March 15, there would be no more publicly listed Plutonium or Uranium testing from the EPA, but they continued Gamma Spectrometry until March 18. Those results indicated a finding of Thallium-208, Bismuth-212, Lead-212; once again, further confirming the fingerprint of Plutonium 238 from a fuel rod.
But that is not the last nail in the coffin; the ratio of detected Plutonium 238 to detected Plutonium 239 was one to one. Pu 238 has a much shorter half life than Pu 239. The Pu 238/ Pu 239 ratio one would expect to find, if this detection were from old above ground nuclear weapons testing, is 0.026. The fact that the ratio is much greater is a direct indication that the Pu 238 was a result of Fukushima. (even TEPCO admits this for Japan's fallout).
The question still remains, what is the health effect for the people in Riverside, California? I can't answer that directly. But, I did do my own risk assessment assuming I lived in Riverside, and assuming that one Plutonium atom decaying in my lungs would lead to lung cancer (an assumption of unknown risk). It is important that the resultant calculation be understood within the context of its assumptions and ground-rules, follow the link if you want to see the results; then ask why its up to one downwind website in Missouri to provide you with information that should be locally forth coming as a result of your own State and Federal tax dollars.
UPDATE 6/3/11
I have added a link to the National Institute of Standards and Technology that gives some good insight into the math and proper application of Combined Standard Uncertainty.
http://physics.nist.gov/Pubs/guidelines/sec5.html
No comments:
Post a Comment