Members of the Subaru high-redshift supernova search group reported on June
21st (IAU Circular No. 7649, the discovery of seven
new supernovae. All are very faint (at least 5 million times fainter than the
faintest stars visible to the unaided eye) and very far away (at least a
thousand times further than the nearest large galaxy, M31 in the constellation
of Andromeda). The team’s success is largely due to Subaru Telescope’s large
8.2 meter primary mirror and the wide field camera, Suprime-Cam.

Supernova explosions are intrinsically very bright, shining with the light
of over a billion Suns. This makes them visible out to very great distances.
Equally important, the peak brightness of a supernova explosion is quite
predictable, making them useful as “standard candles” for gauging the distance
directly to the host galaxy in which they’ve occurred. This makes them a
powerful tool for studying the structure and evolution of our Universe. Recent
research based on supernovae similar to the ones just announced has caused
astronomers to reconsider their understanding of how the Universe has been
expanding since the Big Bang.

Finding enough distant supernovae to carry out this kind of research isn’t
easy. Supernovae shine brightly for only about a month and are exceedingly
rare objects. Our own Milky Way galaxy, for example, produces only one or
two supernova explosions every hundred years. The research team used the
tremendous light-gathering power of Suprime-Cam on Subaru Telescope to first
obtain reference images of the nighttime sky containing many thousands of
faint, distant galaxies. They then obtained a second set of identical 1-hour
exposures one month later to see if any of their galaxies now contained a
supernova (Figure 1). They found 23 promising candidates. To confirm that
the objects really are supernovae and to assign them redshifts, the team used
FOCAS on Subaru Telescope a few days later to obtain spectra for the first
eight candidates (Figure 2). Seven were confirmed as supernovae. These
supernovae ranged in brightness from 22.8 to 24.3 magnitudes at the time of
their discovery in late May and have redshifts ranging from 0.2 to 1.0. An
object with a redshift of z=1.0 indicates that we are seeing it at a time
when the Universe was only about 40% of its present age of ~13.5 billion
years and at a distance of about 8 billion light years from the Earth. The
number of supernovae so far discovered with a redshift greater than 0.9
totals just twelve, with the Subaru team providing three on their very first

Supernovae come in a number of different types, each class having its own
characteristic light-curve and spectral appearance. The type best suited to
cosmological studies is the type Ia supernova. This type is intrinsically
the brightest and also the most homogeneous in terms of the peak brightness
they achieve. Six of the seven discovered supernovae are the type Ia.

When intermediate mass stars (up to a few times the mass of the Sun) reach
the end of their lives, they lose their outer layers to leave behind just
their bare inner-most cores. These stellar cores are very hot and extremely
dense, but not sufficiently so to ignite nuclear reactions within the
nuclear “ashes” of carbon and oxygen accumulated over the stars’ billion-
year lifetime. If one of these dead stars happens to reside in a binary
star system, it’s possible that as the companion star reaches the end of
its life, material lost from its surface will be captured by the dead star.
If the dead star is already close to the Chandrasekar mass limit (~1.4 solar
masses), the newly accumulated material will tip the balance beyond the
point where the stellar core can support itself against the pull of its own
gravity and cause the dead star to begin contracting. The star’s density
rises rapidly, re-igniting nuclear fires that burn so violently that the
dead star is totally consumed in a tremendous explosion … a type-Ia
supernova explosion!

This research is part of the Supernova Cosmology Project. Members of the
Subaru high-redshift supernova search group are: M. Doi, H. Furusawa,
F. Nakata, M. Ouchi, N. Yasuda, S. Miyazaki, N. Kashikawa, Y. Komiyama,
Y. Ohyama, M. Yagi, K. Aoki, I. Hook, S. Perlmutter, and G. Aldering.


[Figure 1:]
Reference and Discovery images of SN2001cv.The background image shows
approximately 3% (5 arcmin x 7 arcmin) of the area covered by a single
exposure using Suprime-Cam. The close-up boxes measure about 10 arcsec on
a side.

[Figure 2:]
FOCAS spectral data for SN2001cv. The blue line shows the observed spectrum
of the supernova slightly contaminated with light from the host galaxy. The
thicker black line shows a cataloged spectrum taken from Kirshner et al.,
1993, ApJ vol. 415, p. 589 of a nearby (much brighter) supernova for
comparison. Although the FOCAS data is quite noisy because the supernova
is so faint, it’s still good enough to confirm that the new object is
indeed a supernova (of type Ia) and that it has a redshift of z=1.039 based
on emission lines measured in the host galaxy.