Annu. Rev. Astron. Astrophys. 1997. 35: 309-355 Copyright © 1997 by Annual Reviews. All rights reserved

5.2. Supernova 1987A

SN 1987A, a peculiar variant of SNe II-P, is described extensively in many other reviews (see Section 1); it is mentioned here only briefly. In general, its spectral evolution resembled that of SN 1992H (Figure 12), as can be seen, for example, in Menzies (1991). Jeffery & Branch (1990) presented an analysis that showed that to a considerable extent the evolution of the line spectrum during the first 100 days could be understood on the basis of simplifying assumptions such as resonant-scattering line source functions and LTE line optical depths. One important aspect of the SN 1987A spectrum is that narrow emission lines from the circumstellar ring were present (Wampler & Richichi 1989), and these became dominant at 3 years Wang et al 1996).

Danziger et al (1988) found that around day 530, the peaks of several emission lines (most notably [O I] 6300, 6364) shifted rapidly to bluer wavelengths. This was probably due to the formation of dust in the ejecta (Lucy et al 1991), which is consistent with the nearly simultaneous increase in the decline rate of the optical light curves and the rapid growth of an IR excess. The lines remained blueshifted even in the very late-time spectra obtained with the Hubble Space Telescope ( 2000 days; Wang et al 1996), which shows that the dust was still present.

The optical spectra of SN 1987A provide considerable evidence for the formation of clumps and mixing of different layers in the ejecta, as had already been deduced from other studies (e.g. X-ray emission; Arnett et al 1989, and references therein). 1. Very early, Hanuschik & Dachs (1987; see also Phillips & Heathcote 1989) drew attention to the "Bochum event," an asymmetry in the H and other hydrogen-line profiles. One possibility is that a blob of Ni⁵⁶ was ejected asymmetrically (Chugai 1992, Utrobin et al 1995). 2. Stathakis et al (1991) showed that the [O I] 6300, 6364 profile was serrated in a manner similar to that found for SN 1985F by Filippenko & Sargent (1989), with FWHM typically 80 km s^-1 for the emission-line peaks. The interpretation is that the [O I]-emitting material is clumpy, probably owing to the formation of Rayleigh-Taylor instabilities at the boundary of the oxygen-rich and helium-rich layers. Chugai (1994a) estimated the mass of clumpy oxygen to be 1.2-1.5 M. 3. Hanuschik et al (1993) showed that the H profile exhibited peaks with a somewhat larger velocity scale: FWHM = 160-400 km s^-1. Subsequently, Spyromilio et al (1993) demonstrated that at least one of the H clumps was also visible in the [Ca II] 7291, 7324 and [Fe II] 7155 emission lines, directly demonstrating that small-scale mixing of hydrogen and radioactive products occurred in the ejecta.