3.3 m and from Type Ia Supernovae
The use of type Ia supernovae for measuring cosmological parameters is covered elsewhere in this volume by Filippenko (nearby supernovae and determinations of H0) and by Perlmutter (distant supernovae and m and ). Hence, these objects will not be discussed in much detail here, except to highlight their potential, and to summarize some of the main difficulties associated with them so that they can be compared relative to some of the other methods discussed in this review.
The obvious advantage of type Ia supernovae is the small dispersion in the Hubble diagram, particularly after accounting for differences in the overall shapes or slopes of the light curves (Phillips 1993; Hamuy et al. 1995: Reiss, Press & Kirshner 1997). In principle, separation of the effects of deceleration or a potential non-zero cosmological constant is straightforward, provided that (eventually) supernovae at redshifts of order unity can be measured with sufficient signal-to-noise and resolution against the background of the parent galaxies. The differences in the observed effects of m and become increasingly easier to measure at redshifts exceeding ~ 0.5. In principle, the evolution of single stars should be simpler than that of entire galaxies (that have been used for such measurements in the past).
At the present time, however, it is difficult to place any quantitative limits on the expected evolutionary effects for type Ia supernovae since the progenitors for these objects have not yet been unequivocally identified. Moreover, there may be potential differences in the chemical compositions of supernovae observed now and those observed at earlier epochs. In principle, such differences could be tested for empirically (as is being done for Cepheid variables, for example). It is also necessary to correct for obscuration due to dust (although in general, at least in the halos of galaxies, these effects are likely to be small; a minor worry might be that the properties of the dust could evolve over time). In detail, establishing accurate K-corrections for high-redshift supernovae, measuring reddenings, and correcting for potential evolutionary effects will be challenging, although, with the exception of measurements of the cosmic microwave background anisotropies (discussed in Section 9 below), type Ia supernovae may offer the best potential for measuring m and .
The most recent results based on type Ia supernovae (Perlmutter et al. 1997 are encouraging, and they demonstrate that rapid progress is likely to be made in the near future. Currently, the published sample size is limited to 7 objects; however, many more objects have now been discovered. The feasibility of discovering these high-redshift supernovae with high efficiency has unquestionably been demonstrated (e.g. Perlmutter, this volume). However, systematic errors are likely to be a significant component of the error budget in the early stages of this program.