Supernova Observation Page

Overview

The brilliant stars visible in the night sky were long thought to be eternal and unchanging. Occasionally, astronomers witnessed a brilliant new object flare up in a region previously occupied by faint stars. These objects were named novae, which is Latin for "new star". Of course, we now know that these objects do not represent the ignition of new stars, but are instead the cataclysmic deaths throes of old stars. These massive explosions are both spectacular and extremely rare-there have been only five supernovae visible to the naked eye in the last millennium. (Murdin 1) These occurred in 1006, 1054, 1181, 1572, and 1604. Modern observations of supernovae give us fundamental clues about the nature of the universe.

Historical Supernova Observations

One of the most well documented supernovae of antiquity was the supernova of 1054. Precise observations by Chinese astronomers pin the date of the event at July 4, 1054. According to records, the supernovae, or "guest star", as the Chinese astronomers referred to it, was as bright as the planet Venus. It was visible in the night sky for two years, and visible during the day for three weeks. A medieval Chinese historian wrote that Yang Wei-te, Director of the Astronomical Bureau, prostrated himself before the Emperor and begged forgiveness for not having predicted its appearance in advance. His fear of retribution may not have been unfounded, for according to legend, the court astronomers Ho and Hsi were beheaded for failing to predict the solar eclipse of 2137 BC. (Murdin 7) Curiously, there is no record of this supernovae in Europe, despite the fact that it was visible in the daylight for two weeks in what is now part of the constellation Taurus.

The two most celebrated supernovae documented by European observations were the supernovae of 1572 and 1604. The former was observed by the great Danish astronomer, Tycho Brahe. Brahe, considered the greatest naked eye astronomer of all time, was at the beginning of his career. His personal papers state that his observations of the supernovae inspired him to devote the remainder of his lifetime to astronomical observations. Brahe's former assistant, Johannes Kepler, was present to observe the next supernovae in 1604, the last to be visible to the naked eye for the next four centuries. Modern observations of the remnants of these supernovae confirm that both of these were Type I SN. (Murdin 91)

Of interesting historical note is that observations of supernovae prompted astronomers like Kepler to question the geocentric view of the universe, held since the time of Aristotle. Kepler disagreed with the notion the stars existed on fixed spheres. Instead, he held the opinion that there was material scattered throughout space that had an inherent ability to coalesce and self-ignite, an idea quite advanced for its time. Unfortunately, in support of his conclusion he pointed out the "fact" that simple life forms clearly appear spontaneously, like maggots on rotting meat. (Murdin 33) Despite such obvious mistakes in his reasoning, Kepler did eventually show that all of the planets moved in elliptical orbits around the sun.

Modern Observations

Modern observation of supernovae are, of course, far more quantitative than the observations from antiquity. One of the most fundamental observations of supernovae is the temporal evolution of their bolometric luminosity. The bolometric luminosity is ideally the sum of all contributions from gamma rays to microwave, however in common practice only the ultraviolet, optical, and infrared flux (0.35 to 5.0 m*10^-6) are utilized. This luminoisty is sometimes referred to as the "uvior" (from ultraviolet-optical-infrared) luminosity or Luvior. The observed Luvior (Suntzeff, 3) can be obtained from the integration of the flux distribution. The two types of supernovae yield distinctly different light curves. Type I supernovae are much brighter than Type II, but also decay much faster.

Type II Supernova Observations

The Large Magellanic Cloud, the galaxy where SN1987a was located. Note the Tarantula Nebula up and to the left of the galactic center.

This is a picture of the Tarantula Nebula with SN1987a down and to the right of it.

The brightest (and closest) Type II supernovae to occur in modern times was 1987A, which occurred in the Large Magellanic Cloud at a distance of 50 kpc. The bolometric light curve of 1987A was divided up into a number of distinct phases by Whitelock et al. The first phase, which lasted 7 days, marked the initial decline in brightness following the ultraviolet flash. This is due to the rapid adiabatic cooling of the photosphere following the passage of the shock wave. After the first week, the luminosity began to increase (phase 2) as the luminosity increase due to photospheric expansion begin to override the luminosity decrease due to the drop in temperature. The maximum luminosity was attained on day 88, after which the luminosity steeply declined. This phase of decline preceding the linear decline is known as phase 3.

Current theory suggests that in the first 4 weeks, the intensity of the light curve is dominated by the energy imparted to the outer envelope by the initial shock wave. The shape of the light curve during phases 2 and 3 is influenced by the hydrogen recombination front propagating inwards into the nebula, whose temperature is maintained by the radioactive decay of 56Ni and 56Co. (Suntzeff, 8). When the energy generated by the decay of cobalt is balanced by the outward radiation of that energy, the light curve begins to decay exponentially (phase 4). The radioactive decay of cobalt remains a significant source of energy up to at least day 850. (Whitelock 17) The presence of energy released by radioactive decay inflates the luminosity curve, creating a plateau. The degree of pronouncement of this plateau shows considerable variance, but the presence or absence of the plateau provides the basis of one the main subdivision of type II supernovae, namely type II-P (those with a plateau), and type II-L (those without a plateau).

The slope of the luminosity curve during this phase 4 was found to be consistent with that expected from cobalt decay, thus providing direct evidence that type- II supernovae are powered by this type of radioactive decay. Phase 4 also marked the first detection of x-rays and gamma rays, indicating that mixing and clumping must have occurred to some degree in the ejecta. (Whitelock 17) The rate of decline of the bolometric light curve increased after day 460 as the light curve entered phase 5. The energy emitted during this phase fell below that expected for cobalt decay, a fact which has been attributed to a decrease in optical depth to gamma rays. The difference between the observed decline rate and the cobalt decline rate was accounted for by increasing the flux of high energy photons. (Whitelock 18)

An interesting event occurred in phases 5 and 6, when the rate of decline again increased. This time the deficit could not be accounted for by adding the x-ray and gamma ray contribution to the Luvior as in phase 4. An analysis of the mid- and far-infrared wavelengths indicates the discrepancy is due to the presence of dust. There is some debate, however, whether the discrepancy is due to either the delayed arrival of thermal emission from pre-existing dust around the SN or the onset of dust formation in the ejecta. (Dwek 54) The origin of the dust is critical, for if the dust was already in place at the time of the explosion, it merely provided an echo that can be neglected from luminosity considerations. If, on the other hand, the dust formed with the ejecta, its energy cannot be neglected. Whitelock et al. (who coincidentally neglected the dust) state that the dust was clearly pre-existing (19), while Dwek (64) concludes that the dust definitely formed in the cooling ejecta. It is difficult to say who is correct, but Dwek's interpretation seems to provide a better explanation of the evolution of the IR spectrum.

One other interesting feature was found by the Hubble Space Telescope. The supernova remenant, when originally observed from earth, appeared to have an hourglass shape, which was similar to the structure found in a number of Type II supernova remenants. However, the pictures from the HST clearly show that there is no hourglass, only two rings. There is no general consensus as to the cause of these rings.

Type I Supernovae Observations

Type I supernovae are identified by the absence of hydrogen lines in the maximum light spectra. A further subclassification of SN I is based on the presence of absence of helium lines, specifically SN Ia (no helium), SN Ib (helium rich), and SN Ic (helium poor).

An example of a recent Type Ia Supernova is SN1994d, the large bright circle down and to the left of the galactic center. Click here for more pictures!

This page was written by Brad Thomson. Please direct any questions about the content of this page to him. HTML coding and formatting by Jamey Minnis and Brad Thomson.

Dwek, E. 'Infrared Emission from SN 1987A: Light Echoes or Dust Formation?' Supernovae: The Tenth Santa Cruz Summer Workshop in Astronomy and Astrophysics. pp. 54-65 Pub. by Springer-Verlag, 1991. 789 pgs.

Murdin, P and Murdin L. Supernovae. Pub by Cambridge UP, 1985. 185 pgs.