Observing the Geometry of the Universe

The classic approach to measuring , the ratio of the universe's actual density to the critical density, is to observe the apparent magnitudes of objects believed to be "standard candles" as a function of redshift. The difficulty of this measurement lies in finding the standard candle at the high redshifts needed to be sensitive to deceleration of the Hubble flow, as well as understanding how to correct for any evolution in the characteristic luminosity of the candle over the age of the universe. Type Ia supernova offer the possibility of such an evolution-free candle, since they are self-contained thermodynamic events that broadcast detailed information about their internal physical states in the form of time-dependent spectra and lightcurves. Unfortunately, although supernovae are bright enough to be studied at high redshifts, they make a poor choice for a telescope time assignment proposal since they are rare, transient, and unpredictable. The Supernova Cosmology Project, run by Saul Perlmutter (Berkeley) and an international team of collaborators, however, has developed an approach to solve these problems and turn supernovae into predictable, practical cosmological tools (Perlmutter et al. 1995).

Just after a new moon, Perlmutter and collaborators observe dozens of high-galactic-latitude fields on telescopes such as the CTIO 4-m. With a wide-field camera, each image contains hundreds of galaxies at redshifts 0.3 to 0.7. Just before the next new moon, the same fields are reobserved and the images are compared, thus checking tens of thousands of high redshift galaxies to find the ten or so showing the new light of a supernova that was not there on the previous observation (Figure 1). The supernovae generally do not have time to reach maximum light, with only ~3 weeks (~2 weeks in the supernova rest frame) between the after- and before-new-moon comparison images. Rapid data analysis makes it possible to identify the supernovae within hours of the observation, so that the follow-up photometry and spectroscopy can begin immediately. In short, this search technique allows "batches" of pre-maximum- light supernova discoveries to be "scheduled" just before new moon, the ideal time to begin follow-up spectroscopy and photometry; the follow-up can be scheduled as well.

This strategy has now been used to discover ~30 supernovae at redshifts z = 0.35 to 0.65, in three observing campaigns with NOAO telescopes, and coordinated work at telescopes around the world. The well-corrected wide-field prime-focus camera at the CTIO 4-m provides the excellent image quality necessary to detect the supernovae. The novel queue-scheduling of the WIYN 4-m is well-suited to following the light curves for the "batches" of supernova discoveries that this technique provides--and the data especially benefits from the image quality of this new telescope.

The first results from the Supernova Cosmology Project are now beginning to be reported. Figure 2 shows the first seven supernovae discovered plotted on a magnitude-redshift diagram (Perlmutter et al. 1995, 1996). (For comparison, the relatively low-redshift supernovae studied at the CTIO telescopes by the Calan/Tololo project (Hamuy et al. 1995) are also plotted.) The first high-redshift supernovae suggest that we live in a relatively high-mass universe, with near unity. If the universe is flat, the data also imply that the cosmological constant is not an appreciable factor in the expansion of the universe--an important point, given the current discrepancy between the estimated ages of the oldest globular clusters and the higher values of the current range for the Hubble constant.

The results from these seven supernovae will soon be strengthened by the inclusion of the next ~20 supernovae, discovered by this project in two scheduled "batches" at the CTIO 4-m on 19-20 November 1995 and 16-17 March 1996. Each batch of ~10 supernovae was found using two search nights compared to two previous nights. These supernovae are currently being followed with photometry and spectroscopy at the WIYN and Keck telescopes. As soon as they have faded enough that the host galaxy can be observed, we will hear the next installment of information on these cosmological measurements.