A great deal has been learnt about supernovae over the last 500 days, or so. The major deficiencies in theory highlighted by the case of SN 1987A are briefly reviewed, as are the sucesses.

The story of the discovery of SN 1987A by Ian Shelton and Oscar Duhalde is well known, but many other events around the time of discovery are largely unknown. I shall describe these events and some of the confusion generated on 24 February 1987.

Several critical issues recently raised by observations of SN 1987A are addressed. These include: 1) the evolution of the pre-supernova star, why it was blue, what its composition and core structure were; 2) the detailed isotopic composition of the ejecta; 3) why and to what extent the supernova composition was mixed in velocity space; 4) the interpretation of recently observed infrared lines, especially their profiles and the existence of red-shifted ‘wings’; and 5) what has become of the neutron star.

The theory of neutron star formation is addressed in the light of the detected neutrino burst from SN 1987A. A brief review of how supernova neutrino theory has evolved over the last 30 years and a general analysis of the SN 1987A detections is presented.

Following the ‘prompt’ radio outburst seen soon after the neutrino emission in SN 1987A (Turtle et al. 1987), we initiated a program to monitor the supernova at the Tidbinbilla Deep Space Communications Complex at 8.4 GHz in a search for radio emission from the expanding remnant. No radio emission has been detected to date (DOY 151, May 30 1988).

The interpretation of the radio burst from SN 1987A as an expanding synchrotron source is briefly reviewed.

Observational and theoretical constraints which apply to models of the precursor star before and at the time of explosion are presented. Stellar evolution sequences for the precursor star produced by various authors are described, and an attempt is made to define what these sequences tell us about the evolution of star of mass ∼ 20 M⊙.

The IUE observations of SN 1987A are presented and the most important results are briefly summarised. The photometric observations of SN 1987A are discussed in some detail in the context of the supernova energetics. Adding the information from spectroscopy and neutrino observations, it is concluded that SN 1987A is a ‘normal’ Type II explosion (i.e. core collapse of a massive star) with an unusually compact initial configuration, just as expected for the progenitor Sk −69°202.

Some special issues to which IUE observations have provided crucial insight are presented. The arguments leading to the definitive identification of Sk−69°202 as the progenitor of the LMC supernova are reviewed, and a preliminary determination of the reddening toward SN 1987A, based on the IUE spectra of stars 2 and 3, is discussed. Recent observations at the predicted position of the SN ultraviolet light echo are reported, and evidence, albeit circumstantial, is provided that a UV echo has been detected. Some consequences of this result are explored. The relevance of this observation lies in the fact that a UV echo, if really detected, would give direct information of the spectrum of the SN at the time of the shock breakout.

The extensive body of broad-band photometric data for SN 1987A obtained at SAAO has been used to derive the bolometric light curve up to day 450 after the outburst. The salient features of the curve are discussed in this paper.

Spectropolarimetric observations of SN 1987A obtained with the AAT are presented and discussed. The foreground interstellar polarization is estimated to have a value of 0.8% at position angle 40°. In the early stages polarization features on the absorption troughs of the P-Cygni profiles are observed. At later stages the polarization is dominated by the effects of electron scattering. After correction for the effects of interstellar polarization the position angle of the electron scattering polarization is 110°. The spectrum of the polarized radiation shows broad redshifted features corresponding to the emission lines which can be understood as the consequence of electron scattering in an expanding atmosphere. The position angle observed is consistent with the elongation of the supernova image seen by speckle inteferometry.

Results from optical and infrared observations of SN 1987A obtained at Cerro Tololo Inter-American Observatory since maximum light (100-400 days after core collapse) are reviewed.

The following are brief summaries of recent observational investigations of SN 1987A by ESO staff and their collaborators, which were reported at the Conference.

Mt John Observatory photographic spectra of SN 1987A have been obtained on 111 nights during the first year at resolutions of 1.1 Å (blue) and 1.6 Å (red). The early spectra are dominated by broad P Cygni profiles from neutral or singly ionised species. Ba absorption lines may be present. After six months nebular emission features emerged, including [OI] 630, 636 nm and two unidentified UV lines (367, 375 nm) in the previously dark region below 380 nm. Absorption line radial velocities show steep declines in the first month, but are almost constant or only slowly decline after 100 days. For the Hα absorption minimum the initial decline rate was 690 ± 70 km s−1d−1 and the initial velocity was −20.2±0.5 Mm s−1 (LMC frame). By 1988 Feb the slowest material in absorption was at −2.2 Mm s−1.

The emission maxima of Hα and NaID show anomalous redshifts of about 1.0 Mm s−1. The [OI] lines show no such redshift. Recent [OI] and Hα spectra at higher resolution show ‘fine structure’ in the profiles, indicating inhomogeneity in the ejecta. Hβ, Hγ and Hδ (but not Hα) were all weaker or absent from 1987 late March to early May, but strong thereafter. From about 1987 Mar 18 to Apr 17 Hα showed a bump on the blue side of the emission at 647 nm and a double peak (658 and 668 nm). Absorption bumps in the Hα profile in early spectra (1987 Feb) may be due to circumstellar water vapour.

Spectra computed from hydrodynamical models for SN 1987A at very early times (t < 2 days) are presented and discussed.

Model calculations for the photospheres of SN 1987A are presented. Spherical symmetry and density profiles are assumed to be given by the homologous expansion of the stellar structure of a B3I supergiant. For later stages the density slopes are determined. For a number of elements (H, He, C, N, O, Ne, Na, Mg, K, Ca, Ba) detailed atomic models are used. The statistical equations and the radiation transport are treated consistently as well for the lines as for the continua. In addition, line blanketing due to other elements are included under the assumption of pure scattering. The importance of a sophisticated NLTE-treatment is demonstrated. Chemical metal abundances are derived by comparing the observed and synthetic spectra. Good agreement is found if a third of solar abundance for all elements with the exception of the s-process elements is assumed. An overabundance of He by a factor of 1-2 is indicated by the models corresponding to the time before June 87. No gradient in the chemical abundances of the hydrogen rich layers has to be assumed. Up to October 87 the He abundance at the photosphere is increased to 5 times solar, i.e. the continua are formed in helium rich layers. The chemical profile indicates strong mixing processes of the different layers of the progenitor during the explosion. Additionally, pecularities of hydrogen line profiles are discussed. Mainly, they can be understood by selective line blanketing effects. The total mass of the hydrogen rich shell is estimated to amount to be 9-11 M⊙.

Supernovae of type II (SN II) are found to exhibit a correlation (r = 0.45 ± 0.15) between M(max) and parent galaxy luminosity. This correlation might be understood by assuming that the metal-rich red supergiant progenitors of SN II in luminous galaxies have more extended photospheres than do those of the more metal-deficient evolved progenitors of SN II on low-luminosity galaxies. For h = 0.5 it is found that SN 1987A was ∼1.5 mag fainter at maximum in blue light than any other well-observed SN II. For h = 1.0 SN 1948B, SN 1973R, SN 1972Q and SN 1987A exhibit comparable MB(max). It would be important to undertake systematic searches for sub-luminous SN II in dwarf galaxies and in the metal-poor extreme outer regions of giant- and supergiant spirals.

Spectrophotometric optical fluxes of SN 1987A have been measured at high accuracy with the 61cm telescope equipped with a rapid spectrum scanner. This highly homogeneous data set covers the wavelength range 3200-8700Å at 10Å resolution. Selected wavelength regions, e.g. around Hα, have also been measured at 3Å resolution. The Hα line shows an undisturbed P Cygni-type profile until March 15, superimposed by variable fine-structure thereafter, developing within a few days into a blue-shifted, peak-like flux excess and a red-shifted flux deficiency. This fine-structure is intrinsic to Hα. Similar fine-structure is also observed in Na I-D, in Paschen-α and possibly in Hβ. We discuss our observations in terms of the ‘mystery spot’, and of models involving a jet and interactions with surrounding pre-existing material, and we propose density inhomogeneities, or occupation number deficits due to NLTE effects, within the inner regions of the envelope as origin.

High quality near-infrared (1-5 μm) spectra of SN 1987A have been obtained at MSSSO, AAO, ESO, CTIO and SAAO with good temporal coverage. These near-infrared observations are reviewed with particular reference to spectra obtained by the MSSSO group on the ANU 2.3m telescope at Siding Spring Observatory.