15 March 2011

Superkamiokande: Neutrinos Not Very Weird

The latest data from the Superkamiokande experiment, which measures neutrino properties and looks for proton decay are out and support a pretty simple model. The slides on the latest conference presentation by Jeffrey Wilkes and a summary of his presentation at neutel11 are available.

Basically, the experimental data from Superkamiokande supports a model where there are three massive neutrinos that oscillate back and forward from one kind to the other at well defined probabilities, with the probabilities and masses being the same for neutrinos and anti-neutrions, while strongly disfavoring many beyond the Standard Model conjectures.

My understanding of these summaries of the presentation is that sterile neutrinos and decay models outside the three neutrino oscillation model are ruled out at a seven sigma level (in physics, five sigma is usually considered definitive, while less than three sigma is consider likely to be a fluke), the existence of a tau neutrino is supported at a 3.8 sigma level, and once again, there is no evidence of proton decay, with tau over beta greater than 1.21 x 10^34 years.

The data come close to ruling out some of the predictions of decay rates of processes in a SUSY SO(10) model and as more data are accumulated, this inconsistency (or experimental support) will become more clear.

The averaged data from Superkamiokande by a couple of different methods, and the MINOS confirm each other in some cases to within two sigma. One of the Superkaiokande methods is within one sigma of the MINOS best fit, and the two Superkaimiokande methods are within one sigma of each other.

The data support the hypothesis that neutrinos and anti-neutrinos have the same masses (as opposed to the "mirror neutrino" hypothesis). The best fit atmospheric data both have a best fit for sine squared two theta of 1.0 for both neutrinos and anti-neutrinos, and the best fit for delta mass squared for neutrinos is 2.1 x 10^-3 eV^2 while it is 2.0 x 10^-3 eV^2 for anti-neutrinos.

The best fit of the study finds that the mass difference squared between electron neutrinos and muon neutrinos is 7.7 E-5 and that the mass difference squared between muon neutrinos and tau neutrinos is 2.1 E-3.

The best fits for the oscillation matrix parameters are also provided: Sine squared two theta estimates are provided for each of three oscillations: 23=0.50; 12=0.30 and 13=0.04. The sum of the square root of the three quantities ought to equal the square root of two (which would imply that the martix has probabilities that sum to 100% in all cases), which they do to within the margin of error of the measurements. The best fit for the CP violating interference phase is zero, although there is no significan constraint the the interference phase at the 90% confidence level.

UPDATE (a few minutes later): The MINOS experiment supports somewhat more weird results. Their best fit delta_m^2 measurement is 2.32x10^-3 eV^2, which is a bit more than the Superkamiokande, but unremarkable given the amount of experimental and systemic error estimated to be present in each result.

But, in their data "there is presently an interesting tension between neutrinos and antineutrinos, at the 2.3 sigma difference, in the oscillation parameters." A difference, if one does exist, implies either CPT violation (and with it Lorentz invariance violation), or that there are two different types of particles out there: neutrinos and mirror neutrinos. Generally, under three sigma results that seem to butt up against well established theory and aren't confirmed by multiple experiments with different methodologies are expected to be experimental error, but this kind of result still bears further examination.

MINOS claims to have "the World’s most precise delta_m^2 measurement" but the error bars in the Superkamiokande research in the experiemental data constrain charts look larger in MINOS to me, merely eyeballing it.

It also bears mentioning that all of the experiments so far in the presentation, at least, are ongoing, and promise significantly more precise measurements in the medium term (i.e. over the next few years to a decade).

UPDATE TWO (3-15-2011): "Xenon100 will not present new results at NEUTEL, we have to bit our nails for a few more weeks." Recall that Xenon 100 is doing dark matter direct detection studies. My Monday blog post on this had noted the rumor that this might be the case.

Some context for all of this data is available at Wikipedia.

No comments: