Changing seasons: why do the radioactive isotopes slow down in summer?

There is a very interesting (and possibly very significant) story [1] doing the rounds of various [2] science [3] sites [4] at the moment (New Scientist, [4], got to it last year before the paper was published). The story relates to a simple finding from Purdue University: that radioactive decay rates may not be truly constant; they seem to vary (almost imperceptibly) according to the season, solar flare activity and with the rotation of the Sun's core (apparently one rotation every 33 days).  The article can be found on arXiv[5] (or in your trusty copy of the proceedings of "the Fifth Meeting on CPT and Lorentz Symmetry") but I will try to analyse it myself, here.

 The basic premise is this: radioactive atoms decay randomly, a single atom of Uranium could decay the second you look at it or in 5,000 years time and there is no way of knowing which it will be. Given a large enough group of Uranium atoms and they will undergo a predictable average number of decays within a given  amount of time (for example 500±5 decays every second) so while you can't say that there will be 50 decays you know that there will almost certainly be between 45 and 55.  To predict the expected number of decays per second (as well as the range into which it falls) a constant called the 'decay rate' [6] is used. Previously it was thought that is was a pretty solid constant (a fact that atomic clocks are built around) and that while the exact number may change it did so according to some well understood statistics.

It was research into the randomness of radioactive decays (for random number generators or dice as they are known) that brought the anomaly to light: when measuring the decay rate of Manganese-54 [7] dips in the count rate (actual number of decays) were noticed that strongly correlated with peaks of solar flare activity seen in December 2006 (Fig 1 [8]). According to the statistical analysis [8] the probability of such a coincidence between decay rates and a solar flare of that strength (S2 which occur ~25 times in each 11 year solar cycle) seen on 12th Dec is roughly ~$10^{-13}$ so the probability of two such events seems very small indeed.

Fig 1. "Normalized December 2006 54Mn decay data along with GOES-11 x- ray data on a logarithmic scale. For 54Mn, each point represents the number of counts in the subsequent four hour period normalized to the average decay rate
(see text), and has a fractional    $\sqrt{N}$    statistical uncertainty of ~2x10$^4$. For the GOES-11 x-ray data, each point is the solar flux in W/m2 summed over the same real-time intervals. The 12 December peak in the x-ray flux occurred at ~21:37 EST." [8]

As well as manganese similar evidence has been seen over month long periods as the sun's core rotates. The analysis [5] shows uses a Joint Power Statistic [9] (JPS) to measure the correlation between the decay rate and the inner core activity of the Sun. Again this analysis claims a remarkable accuracy (the likelihood of such a result is claimed to be 1 in $10^{12}$) but the technique used (JPS) is not one I am familiar with so I can't (yet) comment on it; once I have read the article [9] I will try to.

The final piece of evidence[10] for solar effects on decay rates comes annually: by looking at the count rates of various other radioactive isotopes (Silicon-32 [11] and Radium-226 [12]) over the course of a year these show strong but small affect (1 part in 500).

One hypothesis is that this is caused by changes in the neutrino flux. This seems strange as none of the isotopes undergo neutrino induced decay (as far as I can tell). It may be that the Weak nuclear force field that the neutrinos interact via is needed to trigger decay (although as I have not seen the relevant data on whether there is an excess or dearth of neutrinos it's hard to guess). Either way it will be giving a lot of theorists something to puzzle over and may give us a useful way of inspecting the internal working of the Sun, if it is correct it suggests that the core of the sun rotates more slowwly than the rest of it and may offer other methods of probing regions that are not easily inspected.

Assuming that there is a real effect on display here there are two things that we can do: firstly study in depth any data that emerges from the recent[13] solar activity which may show some disturbances depending on the make-up of the flare, the 4th August was a Coronal Mass Ejection (CME) rather than a radiation storm. Secondly see what emerges from other solar radiation storms which will, no doubt, be well documented.

UPDATE [27-08-1020]: a pretty good refutation of this is on the discovery magazine's "80 Beats" here


[1] Stanford University News, 23 August 2010, http://news.stanford.edu/news/2010/august/sun-082310.html

[2] slashdot, 24 August 2010, http://science.slashdot.org/story/10/08/24/0155229/The-Strange-Case-of-Solar-Flares-and-Radioactive-Decay-Rates

[3] io9, 23 August 2010, http://io9.com/5619954/the-sun-is-changing-the-rate-of-radioactive-decay-and-breaking-the-rules-of-chemistry

[4] New Scientist, June 2009, http://www.newscientist.com/article/mg20227141.400-solar-ghosts-may-haunt-earths-radioactive-atoms.html

[5] arXiv, 20 July 2010, arXiv:1007.3318v1 [hep-ph], http://arxiv.org/abs/1007.3318v1

[6] wikipedia, 24 August 2010, http://en.wikipedia.org/wiki/Decay_rate#Radioactive_decay_rates

[7] wikipedia, 5 May 2010, http://en.wikipedia.org/wiki/Manganese-54#Manganese-54

[8] arXiv, 22 August 2008, arXiv:0808.3156v1 [astro-ph], http://arxiv.org/abs/0808.3156v1

[9] arXiv, 2 Feburary 2005, arXiv:astro-ph/0502050v1, http://arxiv.org/abs/astro-ph/0502050v1

[10] arXiv, 25 August 2008, arXiv:0808.3283v1, http://arxiv.org/abs/0808.3283v1

[11] wikipedia, 15 March 2010, http://en.wikipedia.org/wiki/Isotopes_of_silicon

[12] wikipedia, 6 August 2010, http://en.wikipedia.org/wiki/Radium-226

[13] NASA,  4 August 2010, http://www.nasa.gov/topics/solarsystem/sunearthsystem/main/News080210-cme.html