Distant Supernovae and the Accelerating Universe

Abstract

The observation of SN 1997ff at redshift 1.7 has been claimed to refute alternative models such as grey dust or evolution for the faintness of distant supernovae, leaving only an accelerating Universe as a viable model. However, a very simple one parameter evolution model, with the peak luminosity varying as an exponential function of cosmic time, converts the flux vs. distance law of the critical density matter-dominated model into that of the concordance Omega_matter = 0.3 flat vacuum-dominated model with an error no larger than 0.03 mag over the range 0-2 in redshift. A grey dust model that matches this accuracy can easily be contrived but it still fails by overproducing the far-IR background or distorting the CMB. Models that involve oscillation between photons and axions could emulate an exponential function of cosmic time without violating these background constraints. Clearly a better and well-tested understanding of the Type Ia supernova explosion mechanism and the origin of the correlation between the decay rate and luminosity is needed before any effort to reduce statistical errors in the supernova Hubble diagram to very small levels.