Annu. Rev. Astron. Astrophys. 1992. 30: 359-89 Copyright © 1992 by Annual Reviews. All rights reserved

3.4 Thermal Emission

The recognition around 1970 that supernovae near maximum light emit a thermal spectrum from a photosphere (Branch 1990 and references therein) opened up a new way to estimate extragalactic distances that is independent of all intermediate distance calibrations. The observed flux from a supernova is compared to an absolute flux that is calculated on the basis of the radius and temperature of the photosphere. The radius is the product of the instantaneous velocity at the photosphere. inferred from blueshifts of spectral features, and the time elapsed since the explosion. The explosion time can be derived, in the spirit of Baade's (1926) method for variable stars, by considering more than one time of observation (Branch & Patchett 1973) or more directly from an extrapolation of the observed premaximum light curve (Arnett 1982a).

A thorough examination of the broadband photometry of SNe Ia (L88) reveals that although the energy distribution of SNe Ia is distinctly non-Planckian during most phases of their evolution, at about 25 days after maximum light the energy distribution from the U band (0.36 µ) to the K band (2.1 µ) resembles that of a blackbody having a temperature of 5500 ± 500 K. At that phase most SNe Ia exhibit a blueshift of the red Si II absorption feature within the range 9500 ± 500 km s^-1 (Branch et al 1988, Barbon et al 1990), which we take to be the velocity at the photosphere. With a rise time to maximum light of 19 ± 2 days, the time since the explosion is 44 days and the radius becomes 3.6 ± 0.4x10¹⁵ cm. On the assumption of blackbody emissivity at 5500 ± 500 K, the absolute blue magnitude is then M_B = - 17.9 ± 0.7. The blue light curve drops by 2.2 mag during the first 25 days after maximum light (L88), so the corresponding absolute magnitude at maximum is M_B = -20.1 ± 0.7. This very simple estimate constitutes an appeal to equilibrium and makes no allowance for the effects of the extension of the atmosphere and the nongrey, scattering-dominated nature of the opacity. Jeffery et al (1992) recently have used a more detailed form of the thermal expansion approach to estimate a peak M_B = -19.8 for SN 1990N. Until detailed model atmospheres based on accurate opacities are developed and applied to SNe Ia, however, the external error in the thermal-emission estimate will not be known.