Spaceflight Now: Breaking News

Hubble to put supernova theory to the test
Posted: January 14, 2000

These images of the same patch of sky, taken a few weeks apart with the most recent image at bottom, reveal a bearby Type Ia supernova, SN1999BE, at the edge of galaxy CGCG 089-013 in the constellation Cancer. Caught approximately one week after its peak brightness, the supernova is still virtually as bright as its entire host galaxy. Photo: LBNL
In the spring of 2001, NASA's Hubble Space Telescope will catch nearby supernovae in the act of exploding at prescheduled times, the targets to be supplied "on demand" for the first time. Spectra from these nearby supernovae will be used to calibrate measurements of the accelerating expansion of the universe.

Peter Nugent and Daniel Kasen, members of the Supernova Cosmology Project headquartered at the Department of Energy's Lawrence Berkeley National Laboratory, along with Edward Baron and David Branch of the University of Oklahoma, will use the ultraviolet spectra of nearby Type Ia supernovae (roughly, those less than a billion light-years away) to determine what effects their evolutionary history may have on their use as "standard candles" to make fundamental cosmological measurements.

"Since all Type Ia supernovae have nearly the same brightness, they have been used to compare the distance and redshift of the galaxies in which they occur, and thus to measure the expansion rate of the universe," says Nugent, an astrophysicist with the National Energy Research Scientific Computing Center (NERSC) at Berkeley Lab. "The conclusion that the universe is accelerating, reached in 1998 by the Supernova Cosmology Project and our colleagues in the High-Z Supernova Search Team, was based on comparing dozens of Type Ia supernovae spanning a huge range of redshifts."

Type Ia supernovae are similar because their progenitor stars are white dwarfs that accrete mass from their partners in a binary system, exploding in a thermonuclear blast when the mass exceeds a specific limit, 1.4 times the mass of our sun.

However, very distant supernovae -- those up to eight billion light-years away -- exploded when the universe was only half its present age, and their progenitor stars may have been poor in elements heavier than hydrogen and helium. "Hydrogen and helium and a little bit of lithium were made in the big bang, but heavier elements were made later in stars and supernovae," says Nugent. Even in the earlier universe the age of individual galaxies varied greatly, but since nearby supernovae occur in an older universe, some of their progenitors may be richer in metals.

The ratio of heavier elements to hydrogen and helium is known as metallicity. Nugent and his colleagues used the IBM SP supercomputer at NERSC to develop a theory of how metallicity should affect the spectra and light curves (patterns of rising and falling brightness) of supernovae, investigating whether differences between nearby and distant supernovae might lead to systematic errors in comparing their redshifts and magnitudes.

Hubble Space Telescope. Photo: NASA
"We studied what the spectrum should look like if a supernova's progenitor star is much poorer or much richer than our sun in iron, cobalt, and nickel. We concluded that distinct differences would be evident in the ultraviolet part of the spectrum, although there should be virtually no difference in the optical spectrum," says Nugent, adding that "since the Supernova Cosmology Project and the High-Z Supernova Search Team based their results on the optical part of the spectrum, this is good news for the accelerating universe."

Meanwhile, Berkeley Lab's Greg Aldering and other members of the Supernova Cosmology Project had used the telescope of the Near Earth Asteroid Tracking System (NEAT) on Mount Haleakala on Maui, operated by NASA's Jet Propulsion Laboratory (JPL), to develop a method for finding nearby supernovae "on demand," similar to the method the Supernova Cosmology Project had earlier devised to find distant supernovae.

They scanned wide areas of the sky, then covered the same areas again some weeks later to see if any supernovae had appeared. The wide search was necessary because Type Ia supernovae occur only once or twice a century in a typical galaxy. Before the Supernova Cosmology Project's collaboration with JPL, most nearby supernovae were found by amateur astronomers or by small dedicated telescopes looking at one galaxy at a time. Using NEAT, just over one night of observing time was needed to find five nearby supernovae, three of which were Type Ia's.

"Now we know how to find nearby supernovae on schedule, and we know what we should see when we look at them," Nugent says. Based on these expectations, NASA committed observing time on the Hubble Space Telescope this spring. "Unlike much of supernova astrophysics, where theorists struggle to explain unexpected observations, this is a case where theory will drive observation."

During the spring search NEAT will scan 500 square degrees of sky -- an area equivalent to 2,000 full moons -- every night for two weeks, then make repeat images of the same areas. As the later images are made, NERSC's computing facilities will subtract the two sets from one another in real time, processing some 50 gigabytes of data each night. The bright spots left once the images have canceled one another are supernova candidates.

A host of observatories will be involved in selecting the final observational choices, including the Las Campanas Observatory in Chile operated by the Carnegie Institution of Washington, the Kitt Peak National Observatory and the Cerro-Tololo Inter-American Observatory operated by the National Science Foundation's National Optical Astronomy Observatories, and the Lick Observatory operated by the University of California Observatories.

Beginning in April, 2001, the Hubble Space Telescope will observe chosen supernovae every week for six weeks, yielding light curves and ultraviolet spectra that will allow a precise correlation of metallicity and distance. "Presently there is a seven to 10 percent uncertainty in the distance of Type Ia supernovae," says Nugent. "By eliminating systematic uncertainties, we hope to be able to improve them as distance indicators."

The observational team will include Nugent, Aldering, and Saul Perlmutter of Berkeley Lab, Mark Phillips, associate director of Las Campanas Observatory, and Adam Riess of the Space Telescope Science Institute.

In hopes of collecting many thousands more supernovae in future, the Supernova Cosmology Project at Berkeley Lab has joined with other groups to propose a satellite, the SuperNova/Acceleration Probe (SNAP). "These Hubble observations could affect how SNAP collects data and what conclusions we could draw from it," says Nugent. Spectral studies requiring dedicated Hubble time would be routine with every one of SNAP's supernova observations.

An artist's concept of the proposed SNAP satellite. Photo: LBL
Theoretical studies of how evolution may affect cosmological measurements using both Type Ia and Type II supernovae were reported by Nugent and his colleagues at the annual meeting of the American Astronomical Society in San Diego, CA, this January. Among the federal agencies supporting the theoretical and observational studies of nearby supernova metallicity have been the Department of Energy's Office of Science, NASA, and the National Science Foundation.

The Berkeley Lab is a U.S. Department of Energy national laboratory located in Berkeley, California. It conducts unclassified scientific research and is managed by the University of California.