|Annu. Rev. Astron. Astrophys. 1998. 36:
Copyright © 1998 by . All rights reserved
The mean absolute magnitudes of the SNe Ia that have been calibrated by Cepheids in their parent galaxies are MB MV -19.4 or -19.5. The same mean absolute magnitudes are calculated for explosion models that give a good account of SN Ia light curves and spectra, i.e. Chandrasekhar-mass carbon-ignitor models that eject about 0.6 M of 56Ni. Using MB = - 19.45 ± 0.2 in Equation 1, based on the 26 non-red Calán-Tololo Hubble-flow events, would give H0 = 56 ± 5, but various workers using the Cepheid-based SN Ia absolute magnitudes and physical models in various ways have obtained values of H0 ranging from about 54 to 67. Those who want a consensus value from SNe Ia with conservative errors could use, for now, H0 = 60 ± 10.
This value depends heavily on three things: the Cepheid distances of the HST SN Ia Consortium, the light-curve calculations of Höflich & Khokhlov (1996), and the observed light curves of the Calán-Tololo Hubble-flow SNe Ia. At present, the main issue about the Cepheid distances appears to be the possibility of a significant metallicity dependence. Some technical issues that bear on the accuracy of the complex light-curve calculations remain to be resolved, but the main issue here is the value of the ejected nickel mass, which in turn depends on whether the total ejected mass is Chandrasekhar, super-Chandrasekhar, or sub-Chandrasekhar. The main issue about the Hubble-flow SNe Ia probably has to do with selection bias. For example, many of the Calán-Tololo events peaked not far above the detection limit. If the Calán-Tololo events are selected from the bright end of the non-red SN Ia absolute-magnitude distribution, then the absolute-magnitude dispersions from the Calán-Tololo survey are probably too low, the mean absolute magnitude of non-red SNe Ia in Equation 1 is too bright (for a fixed H0), and most of our current estimates of H0 from SNe Ia probably are too high.
This particular reviewer, chock full of opinions and suspicions about where the errors and biases in the present analyses are likely to be, goes out on a limb (where a branch belongs) and suggests that when all has been said and done, H0 from SNe Ia will be in the 50s.
The names of colleagues from whom I have learned about SNe Ia range from Abi to Zalman and would a fill a page, but for numerous discussions at the University of Oklahoma, I must thank Eddie Baron, Adam Fisher, and Peter Nugent. Part of this review was written in Trieste during a stimulating visit to the International School of Advanced Studies (SISSA), for which I am indebted to my host, Dennis Sciama. I want to conclude by expressing my appreciation of Allan Sandage and Gustav Tammann for their insights and their perseverance.