| A measurement of the average energy density of the universe based on the magnitude-redshift relation of Type-Ia supernovae with 0.008 < z < 0.83 is presented. A detailed analysis of two representative supernovae, SN 1996bc and SN 1996bh, from the High-z Supernova Search Project is given as well as an examination of possible systematic errors in the experiment. We also discuss a CCD camera for the ARC 3.5-meter telescope which we have constructed for this project. Various cosmological models, parameterized by the average density of ordinary baryonic matter (ΩM) and vacuum energy (ΩΛ) have been fit to the luminosity distances of 37 supernovae from this project. In addition, the data have been combined with all of the other extant supernova distances by adding 43 published objects from the Supernova Cosmology Project. When fit with a flat cosmological model (ΩM + ΩΛ = 1), the set of distances from the High-z Supernova Search as well as the combined data set are inconsistent with a matter-dominated universe (Ω M = 1) at the 4 and 6 σ levels respectively. Relaxing the flatness constraint, and using a Gaussian prior probability of Ω M = 0.3 ± 0.1, we find the probability of Ω Λ > 0 to be 98% for the High-z data and 99.7% for the combined data. At a redshift of 0.5, the apparent peak luminosities of the Type-Ia events in our sample are on average 0.25 magnitudes fainter than they would appear in an ΩM = 0.2 universe with no vacuum energy density. We estimate the overall systematic uncertainty in the photometry to be approximately 0.1 magnitudes. We also discuss possible alternative explanations for this result including obscuration by intergalactic dust, evolution in the population of supernovae, and selection effects. Finally, we show how some of the same photometric techniques used for the high redshift supernovae can be applied to other transient sources by presenting an optical lightcurve for GRB 971214 taken with our instrument. |
