Astron. Astrophys. 350, 997-1006 (1999)

4. A historical event?

A supernova going off at a distance of 200 pc should have been a spectacular sight for the contemporaries in the 13th or 14th century. (If we take into account the full band of our age determination also the 12th and the 15th century should not be excluded.) Just how spectacular this event was depends on the absolute visual magnitude . For a bright supernova of type Ia with -19.0 the burst of light would have been as bright as the full moon. For a type II, Ib or Ic which are intrinsically fainter the light output can be orders of magnitude less. For the sub-luminous or ultra-dim SNe recently discussed (Schaefer 1996, Hatano et al. 1997, Woltjer 1997) = -13.0 2. Examples quoted are SN 1181 with = -12.68 1.41 and Cas A with 13.03 2.69. Taking a somwewhat extreme position with = -11.5 the SN associated with RX J0852.0-4622 would still have been brighter than Venus. Taking the other extreme position of a type Ia the SN is a candidate to be recognized even in daylight. In any case it should have been seen.

Records of astronomical events including the epoch of the supernova proposed by us were taken by the far-east astronomers of China, Japan and Corea (Clark & Stevenson 1977, Ho Peng Yoke 1962). Their observatories were typically located at a geographical latitude of 35^o north (Clark & Stevenson 1977), within 5^o north or south, so that the SN of RX J0852.0-4622 would have risen above the horizon after sun-set by up to 11^o from middle of December to end of March. If the SN of RX J0852.0-4622 would have exploded say in late March it would have re-appeared in the northern hemisphere during night after more than 250 days with significantly lower brightness. The light curve of the sub-luminous SN 1997D (Turatto et al. 1998) provides an estimate of 5 at 250 days after outburst, which appears rather little compared with other SNe which show 8 like SN 1994W (Sollerman et al. 1998). With = -11.5, = 5 and a distance modulus of 6.5 the SN of RX J0852.0-4622 then had V = 0 and it is not unlikely to have escaped the attention of the medieval far-east observers. Even with the of SN 1181 the chances to miss it would not have been low. Furthermore, within this scenario, that only the tail of the light curve had been caught, the SN may have not been noticed as a "guest star" because the change was only against a pattern observed more than 200 days before. The light curve of SN 1997D also demonstrates that even if the SN went off in the December-March time frame the detection might have been prevented by a rather short peak/plateau period, and sky visibility conditions become important. For SN 1997D this period probably lasted for 60 days (Turatto et al.), over which the brightness decreased by 3.

The above exercise demonstrates with realistic data that it is indeed possible that the SN of RX J0852.0-4622 was not bright enough, despite its proximity. This sort of physical explanation requires a sub-luminous SN. The peak luminosity and the early lightcurve are determined by the ejecta mass, E₀, pre-SN radius, the structure of the outer layers, and its distribution. As shown by Chugai & Utrobin (1999) in their model for SN 1997D it appears that is rather low for this class of sub-luminous SNe, and it remains to be seen whether these SNe can actually produce enough ⁴⁴Ti (but c.f. Figs. 2, 3 for the minimum amount). We stress that we cannot exclude that the SN was indeed much brighter and even a daylight object. Observers located much further south, like the people of the Incas, the Aztecs or in Middle- and South-Africa, should have had better visibility and their traditions are recommended to be searched for an event pointing to a SN. In this context it is interesting to note that the records of the far-east observers as published by Ho Peng Yoke (1962) appear to be incomplete. The compilation shows three gaps, which are suspiciously long and statistically inconsistent with the mean rate of entries. These periods include the years of 773-814, 1245-1264 and 1277-1293, the latter two of which are relevant for RX J0852.0-4622.

Finally, we point out that there is a chance that the progenitor star of RX J0852.0-4622 is shown in ancient star charts if it happened to be a massive star. Up to now just one progenitor star of a supernova has been identified, which is the progenitor of SN 1987A, the B3 Ia blue supergiant Sanduleak with M_V = -6.8 (West et al. 1987). At a distance of 200 pc the apparent unreddened visual magnitude would have been V = -0.3, which would have made the star the brightest star in the Vela constellation located between Vel and Vel. A Wolf-Rayet type progenitor star would have been less bright with +2.5, but still comparable with the other bright stars in Vela. This opens up an interesting explanation for the apparent absence of a historical record, which admittedly is a speculation. If the progenitor star had been so bright the supernova might not have been noted down as a "guest star". The existing star would just have become brighter. And if the short peak of the outburst had been missed of whatsoever reason and only the tail of the supernova lightcurve has been observed the change of brightness might not have been spectacular, and the star would have disappeared slowly over a couple of years.

© European Southern Observatory (ESO) 1999

Online publication: October 14, 1999
helpdesk.link@springer.de