LBL study of supernovas has surprising results

January 17, 1992

By Judith Goldhaber

Supernovas, those cataclysmic explosions that signal the death of a star, are far more common in galaxies like our own Milky Way than had previously been believed, according to the results of a search conducted by a team at LBL.

In an experiment involving an automated telescope and a sensitive electronic detection system, members of LBL's Automated Supernova Search team have found 20 supernovas, most of them in the last three years. Analysis of the data suggests that supernovas occur at least once every 30 years in galaxies like our own, and maybe even more often. In the past, supernovas were thought to occur in Milky Way-type galaxies no more than once in 100 to 300 years.

The finding was announced last week by team member Saul Perlmutter at a meeting of the American Astronomical Society in Atlanta, and is published in the Jan. 1 issue of Astrophysical Journal Letters.

The discovery could encourage experiments designed to detect neutrinos and gravity waves emitted during supernova explosions. Perlmutter reported that the greatest increase over previous estimates was for the type of supernova known as Type Ic, which is thought to involve the collapse of the core of a dying star. More than half of the supernovas in the nearby universe are of this type.

The LBL Automated Supernova Search experiment uses an unmanned robotic telescope coupled to a sensitive charge-coupled device (CCD) camera to keep a close watch on 1,800 "nearby" galaxies (those within about 100 million light years of earth). Because the system is completely automated, it can look at each of these galaxies daily or every few days, at the rate of about one galaxy per minute. Each image is immediately compared to a reference image taken earlier, and the computer looks for a new bright spot. The system makes it possible to identify new supernovas within days of their birth.

In five years of searching, the research team found 20 supernovas, 18 of them in the type of galaxy known as "late spiral," which includes the Milky Way. Late spiral galaxies account for only about 45 percent of nearby galaxies, but they were found to contain almost all the supernovas. To obtain the rates, the scientists selected the three years for which image analysis was completely automated.

The new higher rates are particularly encouraging, says team co- leader Richard Muller, since it increases the odds for detecting a supernova in our own galaxy in the near future. [Supernova 1987A, discovered recently in the skies of the southern hemisphere -- the first supernova visible to the unaided eye in more than 400 years -- was not in the Milky Way proper, but in a nearly "satellite galaxy," known as the Large Magellanic Cloud.]

Muller explained that the new supernova frequency rate actually tallies well with estimates based on historical records left by civilizations all over the world. These records reveal a total of seven supernova sightings in the last one thousand years -- all of which must have occurred within our own galaxy, since they were seen with the naked eye. Adjusted for the fact that most of the Milky Way galaxy is invisible to observers on Earth, this "historical rate" would predict an average of about one supernova every 30 years. However, in recent years astronomers have tended to discount that figure, since it conflicts with the much lower rate (100-300 years) based on photographic studies of external galaxies. Some recent studies, involving both photographic and visual observations, have suggested a higher rate, though not as high as that revealed by the LBL study.

Muller speculates that the disparity between the previous rates and the new LBL finding is due to the fact that many supernovas are so dim, and fade from the skies so rapidly, that only a highly sensitive and systematic search of the LBL type could be expected to find them. This first phase of the LBL automated search was conducted with UC Berkeley's Leuschner telescope, located about six miles from the Lab. Many of the supernovas observed were close to the limit of sensitivity of the experiment, which suggests, Perlmutter says, that many more supernovas might be discovered with a better system farther from city lights. To this end, the team has directed their efforts to the design and construction of a new, more sensitive automated telescope system. When completed in about two years, the new system will be installed at a major observatory.

Other members of the search team include Heidi Marvin Newberg, Carl Pennypacker, Timothy Sasseen, and Craig Smith, all of LBL and the UC Berkeley Space Sciences Laboratory.