Nearby Supernova Factory announces 34 supernovae in one year
Best rookie year ever for a supernova search
BERKELEY, CA — The Nearby Supernova Factory (SNfactory),
an international collaboration based at Lawrence Berkeley
National Laboratory, today announced that it had discovered
34 supernovae during the first year of the prototype
system’s operation — all but two of them in the last four
months alone. The announcement was made at the 201st
meeting of the American Astronomical Society in Seattle.
“This is the best performance ever for a ‘rookie’ supernova
search,” said Greg Aldering of Berkeley Lab’s Physics
Division, principal investigator of the SNfactory. “We have
shown we can discover supernovae at the rate of nine a
month, a rate other searches have reached only after years
of trying.”
The goal of the SNfactory is to discover and carefully
study 300 to 600 nearby Type Ia supernovae, many more than
have been studied so far. Knowledge of nearby supernovae
will allow better use of observations of very distant
supernovae to measure the expansion history of the
universe; distant supernovae are the key to understanding
the mysterious dark energy that comprises some two-thirds
of the density of the universe.
A torrent of data
Aldering credits the SNfactory’s rapidly increasing
discovery rate to “data pipeline” software developed by
Michael Wood-Vasey of Berkeley Lab’s Physics Division. The
pipeline is filled with up to 50 gigabytes of data a night
(a gigabyte is a billion bytes) from wide-field cameras,
built and operated by the Near Earth Asteroid Tracking
program (NEAT), on remote telescopes at Mount Palomar in
Southern California and at Haleakala on Maui, Hawaii.
By means of a custom high-speed link with Mount Palomar,
part of the High Performance Wireless Research and
Education Network (HPWREN) spearheaded by Hans-Werner
Braun of the San Diego Supercomputer Center, and an
existing link to Maui, the data moves to the National
Energy Research Scientific Computing Center (NERSC) at
Berkeley Lab.
“NEAT sends us images of about 500 square degrees of the
sky each night,” says Wood-Vasey. “The software we’ve
developed automatically archives these at NERSC and
notifies NEAT, which is one of the services we provide
in exchange for the use of their images.”
Aldering adds that “the SNfactory owes much of its success
to NERSC’s ability to store and process the vast amounts
of data that flow in virtually every night.” NERSC’s
Parallel Distributed Systems Facility, devoted to
high-energy physics, astrophysics, and nuclear science,
gives the SNfactory team instant access to up to 2
terabytes of imaging data (a terabyte is a trillion
bytes), with another 8 terabytes accessible in
longer-term storage.
Standard candles to light the universe
The Nearby Supernova Factory grew out of the international
Supernova Cosmology Project, also headquartered at
Berkeley Lab. In 1998, the Supernova Cosmology Project
and the competing High-Z Supernova Search team announced
that the universe is expanding at an accelerating rate —
a surprising discovery that resulted from comparing the
relative brightness (magnitude) of several dozen Type Ia
supernovae at high and low redshift. Type Ia supernovae
are used as astronomical standard candles because of
their exceeding brightness and remarkably similar
optical characteristics.
The universe’s accelerating expansion points to the
existence of some kind of “dark energy,” but while
there are several theories about its nature, there
is no way to choose among them. A new generation of
experiments, using distant supernovae, is required to
eliminate those ideas that don’t work and learn more
about those that might.
First, researchers need to know much more about Type Ia
supernovae themselves. “Our motive is to establish a
much better sample of nearby Type Ia supernovae, against
which the brightness of distant supernovae can be
compared to obtain relative distances,” says Aldering.
At present, astronomers can determine the distance to a
well-studied Type Ia supernova with an accuracy of five
percent; the SNfactory expects to improve on this
accuracy.
Aldering adds “This has to be a large, homogeneous, and
well-calibrated sample, containing supernovae which can
be studied in great detail.” Measurement goals include
refining the relationships among a Type Ia’s redshift,
the width of its light curve — the time it takes to
reach maximum brightness, then slowly fade — and the
features of its spectrum over a wide range of wavelengths.
“We also want to measure the intrinsic colors of a Type
Ia at every stage, so we’ll know the effects of
intervening dust,” says Aldering, “and we want to know
what difference the ‘metallicity’ of the home galaxy
makes — that is, the presence of elements heavier than
helium. We’ll be able to do that by finding and studying
supernovae in dim, metal-poor galaxies that are
overlooked by other nearby supernova searches.”
Catch an exploding star
A decade ago the Supernova Cosmology Project developed
the technique of finding very distant supernovae “on
demand,” by systematically searching the same patches
of sky at intervals, then subtracting the two images
from each other. Any bright spots left over, once
spurious signals had been eliminated, were candidate
supernovae.
The Nearby Supernova Factory employs a similar method,
using images generated by NEAT. NEAT’s primary mission
is to discover and track asteroids and comets, especially
those that could pose a threat to Earth. While its
automated telescopes search the solar system, they
incidentally image many thousands of galaxies. The
telescopes revisit the same regions roughly every six
days during a typical 18-day observing period; when a
supernova appears in one of those galaxies, the
SNfactory can find it.
Aldering says the NEAT search strategy has distinct
advantages for finding the kind of nearby supernovae
of most interest. “It’s a blind search,” he says. “It
doesn’t target specific galaxies but looks at whatever
galaxies happen to be in an image — about 40 per image.”
One result is that the nearby supernovae found by the
SNfactory tend to be a little farther away than those
found by searches that target previously catalogued
galaxies. In the astronomical jargon, they are more
likely to be “in the Hubble flow,” Aldering explains,
“so it’s less likely their redshifts are disturbed by
the gravitational pull of neighboring galaxies,”
meaning their redshifts are good indicators of their
distance.
Wood-Vasey explains that subtracting the supernovae
from the hundreds of images generated each night first
requires software that eliminates spurious signals.
“We have to flag defects such as cosmic ray hits,
electronic noise, occasionally even ice crystals that
can form on the detector system.”
At present, while the SNfactory is still in its
prototype phase, candidates must be confirmed by human
eye. A half dozen undergraduates at the University of
California at Berkeley have been trained to inspect
the images and identify real supernovae.
“We’re working on improving the software’s subtraction
process,” says Wood-Vasey. “The software must sift
through billions of objects. It does well, but as
always the challenge is getting it to work all the time.
The better it performs, the more time our dedicated
undergraduates will have for more interesting supernova
studies.”
The next step: SNIFfing out supernovae
The SNfactory has found so many supernovae so quickly
that follow-up observations — to confirm supernova
types, determine redshifts, study full spectra, and
identify home-galaxy conditions — have not been
possible for all of them. This will change when the
French members of the collaboration complete
construction of a revolutionary new spectrograph, to
be installed next September on the University of
Hawaii’s 2.2-meter telescope on Mauna Kea.
Dubbed SNIFS, for Supernova Integral Field Spectrograph,
the fully automated instrument will “go far beyond what
we can do now,” says Aldering. SNIFS will simultaneously
obtain 225 spectra covering the target supernova, its
galaxy, and the surrounding sky, through two channels
equipped with separate CCDs optimally sensitive to blue
light and red light. At the same time, SNIFS corrects
telescope tracking and measure atmospheric light
absorption by monitoring neighboring stars. Permanently
mounted on the telescope, SNIFS will be available to
the SNfactory 20 percent of the time.
“We’ve already proved that the SNfactory can handle a
huge amount of data and identify nearby supernovae in
the Hubble flow at high, predictable rate,” says
Aldering. “We’re well on our way to providing an
essential tool in the effort to identify the nature
of the dark energy.”
Members of the Nearby Supernova Factory are, from
Berkeley Lab, Greg Aldering, Brian C. Lee, Stewart Loken,
Peter Nugent, Saul Perlmutter, Robert Quimby, James
Siegrist, Lifan Wang, and Michael Wood-Vasey; from the
Laboratoire de Physique Nucleaire et de Haute Energies
de Paris, Pierre Antilogus, Pierre Astier, Delphine Hardin,
Jean-Michael Levy, and Reynald Pain; from the Institute
de Physique Nucleaire de Lyon, Yves Copin and Gerard
Smadja; and from the Centre de Recherche Astronomique de
Lyon, Gilles Adam, Roland Bacon, Jean-Pierre Lemmonier,
and Arlette Pecontal.
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.
Additional information
More on the Nearby Supernova Factory
http://snfactory.lbl.gov/
More on Mount Palomar Observatory, operated by the
California Institute of Technology
http://www.astro.caltech.edu/observatories/palomar/
More on the Maui Space Surveillance System and the Maui
High Performance Computing Center, operated by the United
States Air Force
http://www.af.mil/news/Nov2000/n20001108_001676.shtml,
http://www.mhpcc.edu/
More on the the Jet Propulsion Laboratory’s Near Earth
Asteroid Tracking program (NEAT), funded by NASA
http://neat.jpl.nasa.gov/
More on the High Performance Wireless Research and
Education Network (HPWREN), operated by the San Diego
Supercomputer Center and the Scripps Institution of
Oceanography and sponsored by the National Science
Foundation
http://hpwren.ucsd.edu/
More on the National Energy Research Scientific Computing
Center (NERSC), operated by the U.S. Department of Energy
and headquartered at Berkeley Lab
http://www.nersc.gov/
IMAGE CAPTIONS:
[Image 1:
http://www.lbl.gov/Science-Articles/Archive/assets/images/2003/Jan-07-2003/Palomarcomp.jpg (54KB)]
The Nearby Supernova Factory uses images from wide-field
cameras built and operated by the Jet Propulsion
Laboratory’s Near Earth Asteroid Tracking program (NEAT).
One camera is mounted on the Oschin Telescope at Mount
Palomar Observatory.
[Image 2:
http://www.lbl.gov/Science-Articles/Archive/Phys-SNfactory-Aldering-img2.html (85KB)]
NEAT images are sent to the Nearby Supernova Factory
through a special link in the High Performance Wireless
Research and Education Network (HPWREN), helping
connect Mount Palomar Observatory to the National
Energy Research Scientific Computing Center (NERSC) at
Berkeley Lab.
[Image 3:
http://www.lbl.gov/Science-Articles/Archive/assets/images/2003/Jan-07-2003/SNsubtraction.jpg (12KB)]
This supernova, found in archived data, illustrates the
subtraction technique used by the Nearby Supernova
Factory to find supernovae in NEAT data. NEAT cameras
image an area of the sky for reference (left). They
image the same area again several days later (center).
The first image is automatically subtracted from the
second, and any remaining bright spots may be supernova
candidates.
[Image 4:
http://www.lbl.gov/Science-Articles/Archive/Phys-SNfactory-Aldering-img3.html (37KB)]
The SNIFS spectrograph, built by the French members of
the SNfactory team, will be permanently installed on the
University of Hawaii’s 2.2-meter telescope on Mauna Kea.
