Strange, bright flashes observed on the moon for centuries and often dismissed as the lunar equivalent of UFOs may in fact be emissions of volcanic gas. A researcher says he has reviewed the evidence for so-called lunar transients and found them to occur only in areas of the moon that belch radon gas, suggesting that the flashes could be the result of dust stirred up by such emissions—possibly volcanic in origin.

"A lot of people think that this is just craziness—this is up there with UFOs," says astrophysicist Arlin Crotts of Columbia University. "But no, this is real science. And it's something people should have done 30 years ago." Other experts, although intrigued, are not yet convinced of transients or Crotts's proposed explanation.

Moon watchers since at least 1540 have reported seeing bright spots or other pinpoint distortions on the moon's surface that faded anywhere from a minute to a few hours later. Interest in these transients exploded in the late 1950s and 1960s among amateur astronomers, who cranked out many a spurious-looking report of lunar lights, Crotts says; even the Apollo astronauts claimed to see a few. "People have been wondering about this for hundreds of years, to the point where they've given up on it," he says.

To determine which, if any, sightings were legit, Crotts statistically analyzed the hundreds of documented transients—"a hair ball of a data set," he says—and found 450 sightings, most pre-1960s, that were similar in description despite occurring in different centuries or on different continents. "However you split them up, historically or by the geography of observers…, everybody sees the same [kind of] thing," he says. "It makes it easier to believe they're real."

The crater Aristarchus was the site of the most transients, followed by Plato, Grimaldi, Kepler, Copernicus and Tycho, Crotts reports in a paper submitted to the journal Icarus. These areas, which account for about 10 percent of the lunar surface, also encompass four sites where lunar missions have pinpointed emissions of radioactive radon 222 gas, a by-product of uranium 238, and the more diffuse radon by-product, polonium 210. Crotts says the odds of such overlapping are less than 1,000 to one.

One possible explanation for transients, he says, is volcanic gas escaping from the moon's interior to the surface, where it might build up under the fine lunar soil until it erupts, puffing up a visible cloud of dust. Researchers believe that Aristarchus was the most volcanically active part of the moon, leaving behind solidified lava called basalt riddled with gas pockets.

Crotts' analysis goes further than prior work in proving that transients are real, but the gas explanation is "mostly 'correlations equals causation,'" says geologist Paul Spudis of the Johns Hopkins University Applied Physics Laboratory. Transients are difficult to study, he adds, because observations are hard to confirm or replicate. Hoping to address that problem, Crotts is currently setting up a robotic imaging system in Chile to begin a systematic search for transients.