The beautiful V838 Monocerotis is a well known variable star on the outskirts of our galaxy. However, this new discovery comes as a surprise to astronomers (NASA).

Most discoveries of new celestial objects are made by ultra-modern observatories peering deep into space with ever-increasing precision. But sometimes it’s down to meticulous record keeping of decades of astronomical images before a discovery reveals itself.

Scientists using the Digital Access to a Sky Century at Harvard (DASCH) have announced the discovery of a new class of variable star after digitizing a century’s worth of photographic plates.

The Harvard College Observatory maintains more than 500,000 glass photographic astronomical plates taken between 1880 and 1980, making this the most comprehensive record of the night sky for this period. Each and every part of the sky has been surveyed between 500 to 1000 times.

This fact alone creates an amazing tool for modern astronomers to analyze the half-million plates for changes in star brightness that may have occurred over the 100 years.

The first result to come from this ongoing digitization program is the discovery of a new class of variable star that exhibits extreme changes in brightness over short time periods. This new class appears to vary in brightness by a factor of two with a periodicity of 10-100 years. So far, astronomers are baffled by the observation.

Most stars exhibit some variability in brightness — even our sun varies by about 0.1 percent in energy output throughout its 11-year cycle — but this new class varies by a factor of two: a variability of 200 percent.

Harvard-Smithsonian astronomers Sumin Tang, Jonathan Grindlay and Edward Los discovered three objects all with this extreme variability, a type of variable that has never been seen before.

The discoverers followed up the DASCH survey with spectroscopic analysis to see what these stars were made of and they appear to be made of the same stuff. Interestingly, they are older and slightly less massive than our sun.

“The behavior exhibited by these three stars can’t be readily explained by any models of stellar variability we know of, so they may indeed be a new class of variable stars,” Mike Simonsen, Vice President of the American Association of Variable Star Observers (AAVSO), told Discovery News.

Although drawing this kind of conclusion from such a small sample of stars may seem a little premature, Simonsen points out that it is unlikely to be a coincidence.

“The cause of the variability might be dust obscuration events, but it is hard to explain how you can produce and sustain obscuring dust for that long a time period.”

If something (like dust) isn’t blocking the light from these stars from view, could the variability be caused by internal changes in the stars? Perhaps they are coming to the end of their lives after depleting the hydrogen fuel being supplied to their cores?

“The variability might be caused by evolutionary changes occurring during shell burning flashes or helium core flashes, but these stars don’t seem to be that evolved,” he added.

(Helium flashes occur in intermediate mass stars when the supply of hydrogen to the internal fusion reactions of the star’s core is exhausted and helium burning begins, creating a bright flash on the surface of the star.)

Simonsen also pointed out that although the behavior of these stars seems unique now, we are most likely glimpsing a short variable period in their evolution.

So by using this retroactive technique of digitizing archival photographic plates, we have been given a unique opportunity to see a portion of stellar evolution that we could only see by analyzing decades-worth of historical observations.

“Stellar evolution takes place on cosmic time scales, not human timescales, so we rarely witness things like stars evolving to another phase in their evolution. There are few databases with 100 years, or more, of stellar data. The AAVSO is one.” — Mike Simonsen.