Where have all the novae gone?

WARNER: OLD NOVAE Where have all the novae gone? Naked-eye novae were discovered in antiquity, but appear remarkably absent from modern observations. Brian Warner wonders what has happenend to old novae. T ycho Brahe and Johannes Kepler were both involved in the discovery of what they thought to be novae (now known to be supernovae) but why didn’t they discover any ordinary novae? Come to think of it, why didn’t Flamsteed, Bradley, Halley, the Cassinis, Argelander, Piazzi, Schroeter, Lacaille, the Struves, William and John Herschel – all of whom observed the sky extensively – discover any novae? The novae found by Tycho and Kepler alerted astronomers to the mutability of the heavens, and added to the fame of the discoverers; one would have thought that later professionals would have been enthusiastic to add to the list. It should have been easy – in the late 19th and 20th centuries most of the bright novae were found by amateurs, and with the unaided eye. Yet, in Europe, only three novae were recognized in the 17th century after Kepler’s discovery, only one was recognized in the 18th century, and the first nova in the 19th century was not found until nearly mid-century – and that was found telescopically! In that century the first A&G • February 2006 • Vol. 47 ABSTRACT An overview is given of the discovery of the brightest naked-eye novae, from early Oriental to modern times. Amateur astronomers appear prominently in the list, but throughout the centuries most well known professional observational astronomers made no such discoveries. After a nova explosion the star usually returns to its pre-nova brightness for at least two centuries. There is increasing evidence that after that the system fades to a very faint state, for perhaps several millennia. The results are described of a high-speed photometric study of southern old novae, carried out over the past five years. Many of the nova remnants show the characteristic periodic modulation of brightness associated with orbital motion. This survey has doubled the number of known orbital periods for these systems. nova to be discovered with the unaided eye was T CrB in 1866, when two thirds of the century had already passed. The supernovae of 1572 and 1604 could hardly have been overlooked – they were visible even in the daytime and cast strong shadows at night. Yet, according to the account published by Thomas Dick (1859), neither Tycho nor Kepler had noticed them until the public pointed them out. Of the nova of 1572: “Its appearance was sudden and brilliant. Its phenomena were so striking that the sight of it determined the celebrated Tycho Brahe to become an astronomer. He did not see it at half an hour past five, when he was returning from his house to his laboratory; but returning about ten, he came to a crowd of country people who were staring at something behind him. Looking round he saw this wonderful object.” And of the nova of 1604, Dick wrote: “On the 30th [of September], the sudden breaking of the clouds afforded one of Kepler’s friends an opportunity of having a very short view of it… On the 2nd, 3rd, 4th and 6th of October, it was seen by several persons in different places. On account of cloudy weather at Prague, where Kepler resided, he did not see it until the 8th of that month.” No other novae as bright as those two supernovae has since occurred, but ordinary novae with apparent magnitudes between –1 and +2 (of which there were six from 1901 to 2001 – see below) were largely ignored from the 16th until the 20th century. In this review we will start by listing the brightest novae found until the end of the 20th century and the astronomers – amateur and professional – who found them. After eruption a nova sinks back to its preeruption brightness for a century or more. There is increasing belief that the system then “hibernates” – dropping to faint magnitudes for millennia. We look at the general statistical evidence for this and discuss the recent discovery of hibernating novae. All novae are close binaries and the frequency distribution of their orbital periods is a valuable desideratum for comparison with stellar evolution calculations and population statistics. With this in mind we have conducted a survey of southern old novae which has added another 11 to the number of known orbital periods, effectively doubling what had been known. Ancient novae We begin by noting the discoveries of novae prior to 1850. Chinese, Japanese and Korean records from ~1500 BC onwards contain many descriptions of new stars, most of which must have been very conspicuous, probably at least second magnitude, to have been noticed. A list of Oriental novae is given by Stephenson 1.29 1: A montage of light curves for eight old novae. The variable star identification and orbital period is given at the top left of each panel. All light curves use orbital phase as the abscissa; the ordinates are magnitudes, but a variety of scales has been used in order to produce uniform height. WARNER: OLD NOVAE Table 1: Early Oriental novae Duration 712 837 1163 1356 1399 1437 1592 1592 1592 1690 – 75 days – – – 14 days 15 months 4 months 3 months 2 days From Stephenson (1986) (1986) – see table 1. The best determined positions for these novae have errors ~0.2° but most are ~1°. For none of them have any bright remnants been found. For example, several cataclysmic variable candidates were found in the vicinity of a nova that occurred in Scorpius, recorded in Korean records in 1437 (Clark and Stephenson 1977), but none of them is brighter than B ~ 19 (Shara, Moffat and Potter 1990). The year 1592 was a record for discovery of novae – the assiduous observers in Korea found three novae in widely different parts of the sky within a few days of each other. Returning to European successes, an apparent nova at second magnitude was discovered in the constellation Vulpecula on 20 June 1670 by the Carthusian monk Père Dom Anthelme. The recent history of this object (now known as CK Vul) has shown that there is a remnant at V ~ 21 but the infrared properties do not match that of an old nova, so in 1670 it was more probably a thermally pulsing giant, like presentday V605 Aql (Evans et al. 2002). A possible nova in Puppis was detected by J Richer on 12 and 21 January 1673 as a third magnitude star that he observed with a mural quadrant set up in Cayenne during an ocean voyage. A modern spectroscopic search for the remnant has been unsuccessful (Shara, Moffat and Potter 1990). Nova Orionis 1678 (now V529 Ori) was discovered on 28 March 1678 by J Hevelius, at about sixth magnitude, while he was observing a lunar occultation of χ1 Orionis. Another, unidentifiable, star was seen to be occulted and later reappeared (Ashworth 1981). This has long been regarded as a possible but not certain nova (Duerbeck 1987), but a recent probable identification of the remnant, at 19th magnitude, has been made (Robertson et al. 2000). If correct, this is the most e (...truncated)