(1) Longer Half Life Fallout Detected in Saint Louis on 9/14/2011
(2) The Detected Half life is Indicative of Neptunium 239
(3) Under a Long Term Np-239 Fallout Scenario, Explosive Dispersal of the Corium Becomes Risk Mitigating
(4) Based on the Jet Stream, the Primary Fallout Areas Were Canada, Midwest, and the Yankee States.
Decay Charts
For further background data please view this video
Fukushima Fission Hit Ground Water on August 11: Detected in Saint Louis on August 20
Great blog - really defines what it means to be a citizen-scientist. Keep up the outstanding work. (It outshines my efforts about 40-fold.)
ReplyDeleteCan you explain the radon daughter ingrowth more clearly? In the video, you seem to say that the second peak should be expected at 92 hours because that corresponds with the half-life of radium. I do not think this explanation is complete or correct. It's a fairly technical problem to deal with, so perhaps you're just glossing over the details here?
EPA document 9315, Table 7.14, gives ingrowth correction factors as a function of time. If I had your raw data, I would employ these factors to remove the counts from the daughter products from the overall count rate. Is this what you've done?
I'd appreciate hearing more about this. I can't accept the conclusion without knowing more about how the second peak was removed.
Aaron Datesman
(www.tinyrevolution.com)
The in-growth is from Radon-222. The alpha affects of Rn222 and its daughter Po218 are blocked by the ziploc bag containing the sample. The half life of Rn222 is roughly 92 hours; shortly after that period is when the remaining daughters Pb214 and Bi214 would be at their maximum (as seen in the chart). I excised those data points from the analysis.
ReplyDeleteTrying to remove the noise from those daughters in the rest of long half life analysis would actually increase the noise level(that's a function of the granularity of the data and location uncertainty of the radon in the sample bag)
However the key is not the n'th degree accuracy of any particular data point but rather in the over all trending of the data in question.
Ms. X,
ReplyDeleteI think this is not correct. The number of decays of radon (and of the two radon daughters) are continually decreasing. They do not occur all at once at 92 hours.
The proper way to go about this, I think, is to make a guess about what the initial concentrations of neptunium and radon are, calculate forward what the measured decay curve should look like, and then iterate on the initial guesses until the curves correspond.
Then you can remove the contribution from radon and evaluate whether you obtain the proper half-life for neptunium. If you would be willing to share the original data with me, I would be happy to collaborate on this.
Great work and great blog. I support what you've done absolutely - I just don't think this particular piece is fully fleshed out.
You are correct. The data cleaning technique I employed would only start to hold water if the daughter products had longer half lives than the parent (not vis versa as I had used). And thanks for making me revisit my thought process, it gave me another less troubling but also likely candidate to investigate as being the long half life component.
ReplyDeleteIn regards to your data request; I suspect the sampling frequency,its disparate nature, and the sampling error is not at a level of granularity that will support the type of analysis you suggest.
The true path forward is for the powers to be to release raw data on what THEY have detected. But, I will post the raw data over the weekend so you can give it a shot, if you like.
Thanks for contributing.
The requested raw data is now available at
ReplyDeletehttp://pissinontheroses.blogspot.com/2011/09/raw-data-of-saint-louis-neptunium-239.html
Thanks for sharing the data - that's really fantastic of you. I worked on it a bit last night and will let you know if I can figure anything out. It's certainly very interesting. And worrying. Very, very worrying.
ReplyDeleteI'm a little time-constrained, so I haven't finished to my satisfaction, but the solution I'm moving toward is that a system comprised of two decay chains could explain the readings you obtained. The two chains are
ReplyDeleteSr -> Y -> Zr and
Rn -> progeny (short-term) -> Pb
For certain initial concentrations, the activity in the Sr chain actually increases over time scales of several days, while the Rn decays quickly. The superposition of those two trends produces an overall curve like the one that you measured. What the half-life of the overall decay signal turns out to be depends mostly upon the initial concentrations in the sample.
It's possible that the second half of the Rn decay chain (Pb -> Bi -> Po -> Pb) will produce the bump at around 90 hours. I haven't built that piece into my Excel model yet.
Maybe I can do that this week. It's not hard now that I got the hang of it.
Thanks for the contribution. Based on a rapid analysis, Y-90 is a viable candidate; the signature would fit my dispersal propagation model of fission driven ground water steam.
ReplyDeleteThe lack of a significant alpha increase in the raw data at 9/15/2011 17:06:50 from the unshielded test source would tend to preclude the RN decay chain. I would suspect a low probability secondary decay chain with a short half life; may be Pa-235 from Np-239.
Unfortunately none of this bodes well for us or our food crops.