Cosmological Supernovae and the Measurement of Omega

While most people spent the past winter holidays taking a relaxing break from work, members of the distant supernova search devoted long hours to hunting for signatures of the violent death of stars that exploded billions of years ago. This data run was part of an ongoing search that has already yielded the most distant supernova ever observed and which may eventually tell us about the matter content of the universe.

Figure 1 The 4-meter Mayall Telescope at Kitt Peak National Observatory (photo by Mark Hanna, courtesy National Optical Astronomy Observatories).

The supernova search is part of an effort to measure the mass density of the universe. There is strong observational evidence and theoretical arguments that one class of spectroscopically distinguishable supernovae, the Type Ia, is a good standard candle. With a peak magnitude of MB = -18.86 -5log(50/H0), its importance lies in the fact that it can be studied to measure distances on cosmological scales. By studying the curvature in the Hubble diagram (a plot of redshift vs brightness), we aim to measure the deceleration of the universal expansion, the parameter q0. Using the standard Friedmann universe with zero cosmological constant in which Omega is two times q0, this measurement allows us to find the present mass density.

Figure 2 The solid curve shows the calculated lightcurve for 1992bi for the best fit q0=0.1 based on the template lightcurve for nearby supernovae. The dotted curves are for q0=0.1 (upper curve) and for q0=0 (lower curve). The inner error bar, sc, shows the combined uncertainty from photometry measurements; the outer error bar includes the intrinsic dispersion of Type Ia supernovae.

Our earlier discovery of supernova 1992bi at z=0.458 demonstrated the feasibility of our search technique and gives us a first trial measurement of Omega = 0.2+/-1.25. Figure 2 shows measured and expected lightcurves in different scenarios. It will take tens of supernovae in order for us to home in on a precise value for Omega; our current telescope run is intended to show that we can study these supernovae in the required batches of multiple scheduled discoveries.

Our supernova search strategy is as follows. Several nights are used to take reference images of fields with known high redshift galaxy clusters. Then several weeks later, the same fields are re-observed and transported over the computer network to Berkeley for cleaning and analysis. These new images are compared to the references almost entirely with computers, which match and subtract the images and then scan the subtraction for point-source candidate supernovae. The entire process is completed within a day or so and scheduling is arranged so that we have telescope time to immediately follow up our candidates both photometrically and spectroscopically.

Figure 3 The upper plot is an image of a galaxy taken in 1992. The lower plot is the same galaxy taken at the INT on Dec 22. The new source to the right of the galaxy is one of our supernova candidates. The Berkeley distant supernova search team includes Saul Perlmutter, Carl Pennypacker, Gerson Goldhaber, Ariel Goobar, Bruce Grossan, Reynald Pain, Silvia Gabi, Alex Kim, Matthew Kim, and Ivan Small (CfPA, Lawrence Berkeley Laboratory, and Space Sciences Laboratory). Our observations were made in December and January at the 2.5 meter Isaac Newton Telescope in the Canary Islands with Richard McMahon (IofA) and Mike Irwin (GRO) and in February at the Kitt Peak National Observatory's 4 meter telescope with Marc Postman (STScI) and Tod Lauer (NOAO). In addition, many observers at many observatories helped in getting follow-up.

In total, we collected around sixty R and I band fields, each with hundreds of galaxies at redshifts from 0.2 to 0.6. Although data is still streaming in and has not been completely analyzed, this run appears to be a great success. We have a number of good candidates for which we are busily constructing lightcurves and determining supernova type. Figure 3 shows the discovery and reference images of one of these events.

Looking towards the future, we have just completed a memorandum of understanding with the Royal Greenwich Observatory. This agreement calls for the building of a 4-chip CCD camera. With thinned CCD chips, this camera will have a much higher quantum efficiency than the previous system, and so will increase our detection rate by more than the four times gained by the larger field of view. We will also be guaranteed telescope time for both searching and follow-up. With our proven supernova detection technique, future observations will add many more events to our already existing pool of cosmological supernovae. Within a few years, we will be measuring Omega, and begin to learn about the total content of dark matter in the universe.