Elusive patterns seen in solar neutrino data.

Elusive patterns seen in solar neutrino data

For 20 years, a detector deep in the Homestake gold mine near Lead, S.D., has recorded the arrival of neutrinos apparently produced by the thermonuclear reactions at the suns's core. Three new analyses of the accumulated data now reveal that month-to-month and year-to-year variations in the number of neutrinos detected show intriguing correlations with changes in solar activity.

Because neutrinos are the only known particles that reach Earth directly from the sun's core, they provide a unique window on the sun's inner workings. By establishing correlations between neutrino detection and solar phenomena, researchers hope to obtain a more complete picture of solar activity, enabling them to test theoretical explanations for why the number of solar neutrinos detected on Earth falls short of the sun's predicted rate of neutrino production.

Reporting in the Nov. 29 NATIRE, Todor Stanev of the University of Delaware in Newark and his colleagues describe a "strong" correlation between the neutrino detection rate and the monthly sunspot number. Their statistical analysis indicates that the detection rate decreases as the number of sunspots increases (SN: 4/21/90 p.245).

In the same issue, Lawrence M. Krauss of Yale University reports an apparent link between the number of neutrinos detected and small shifts in the frequencies of certain types of solar oscillations. Previously, measurements made from 1977 into 1988 of the frequencies at which the sun's surface rises and falls had revealed that these frequencies increase slightly as the solar cycle shifts from a minimum to a maximum (SN: 7/7/90, p.7).

In a statistical analysis to appear in ASTROPHYSICAL JOURNAL, John N. Bahcall of the Institute for Advanced Study in Princeton, N.J., and William H. Press of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Mass., also found significant patterns, similar to Staney's, in the Homestake data. But their analysis reveals that only the last two-thirds of the data set show a clear time variation that may be connected with the solar cycle.

In addition, because the detection rate appears to follow a pattern that changes more smoothly and regularly than the actual sunspot count and tends to lag behind surface activity. Bachall and Press postulate that the neutrino detection rate may be directly related not to support number but to an unrecognized effect coupled to the solar cycle.

The significance of all these proposed correlations remains unclear. "There's a consistent view that there's something statistically very peculiar going on," Press says. "But people have different ideas about what it all means."

And statistical analyses are fraught with uncertainties. "Searching for correlations between any a priori unrelated phenomena can both subtle and dangerous," Krauss warns in the introduction to his paper.

One key problem is the paucity of data. "We're stuck with the fact that all the [neutrino detection] experiments have low counting rates," says Kenneth Lande of the University of Pennsylvania in Philadelphia.

"You can get a number for the average rate over many years, but if you want to get time variations or correlations with sunspots or anything else, you need a much higher counting rate. Until we build better detectors with higher counting rates, we're stuck with large uncertainties in the statistics."

Moreover, recent results from the Kamiokande detector in Japan and ambiguous preliminary data from the Soviet-American galium experiment in the Soviet Union provide little help in solving the solar-neutrino puzzle (SN: 9/1/90, p.141).

"Right now, it's a puzzle with no single good answer because at least two unlikely things have to happen to make it work out," Press says. "In other words, any explanation that I can think of requires two coincidence, and that seems one too many for me."