Six remarkable northerly novae in 2020–’21

2022 August 7

Jeremy Shears

During late 2020 and 2021, six relatively bright novae appeared in the northern celestial hemisphere. The light
curves of V1391 Cas, V1112 Per, V1405 Cas, V1674 Her, V606 Vul and RS Oph are presented and discussed.
They illustrate the remarkable diversity of nova behaviour and include representatives of the major classes of
novae: fast, slow, dusty, and recurrent

Introduction

Novae represent a class of cataclysmic variable star, which are known to be interacting binaries where a cool secondary star loses mass to a white dwarf primary. Material from the secondary falls through the inner Lagrangian point, L1, and because it carries substantial angular momentum, it does not settle on the primary immediately, but instead forms an accretion disc around the primary. Material flowing through the accretion disc accumulates on the surface of the white dwarf and eventually sufficient material builds up to trigger a runaway thermonuclear reaction. The ensuing eruption causes the system to increase in brightness dramatically – perhaps 10 million-fold or more – in a matter of hours or days, blowing the outer layers of the white dwarf away into space as an expanding gas shell. With time, the gas cools and the once-bright nova begins to fade: the outburst is over.

Novae can be classified by the times taken for the brightness to decline by two or three magnitudes: t₂ and t₃. For example, according to the General Catalogue of Variable Stars, fast novae have t₃ < 100 days, while slow novae take more than 150 days. Fast and slow novae are generally referred to as classical novae. A further class is that of the ‘recurrent novae’, which undergo two or more outbursts, usually separated by many years.

During late 2020 and 2021, we were treated to six relatively bright novae in the northern hemisphere. They include representatives of the major classes: fast novae; slow, dusty novae; and a recurrent nova. This paper presents the light curves of these objects, which illustrate the remarkable diversity of nova behaviour. It also highlights the importance of amateur observations of these exciting transients, both in their discovery and in follow-up. It is a pleasure to showcase the hard work of the many dedicated observers that have followed these novae. Note that whilst some representative spectra are shown, they are only the ‘tip of the iceberg’ from the rich collection in the BAA spectroscopy database; these warrant further analysis but that is beyond the scope of this paper.

Nova Cas 2020 (V1391 Cas)

Nova Cas 2020 (Figure 1) was discovered on 2020 Jul 27 by Stanislav Korotkiy & Kirill Sokolovsky. Their ‘New Milky Way’ survey,¹ operating at the Ka-Dar Observatory in the Russian Caucasus at an altitude of 2,000m (Figure 2), makes use of off-the-shelf equipment to discover transient objects. The equipment would be familiar to many amateur observers: a Canon 135mm ƒ/2.0 lens with an SBIG ST-8300M CCD camera, mounted on a Sky-Watcher HEQ-5 Pro mount (Figure 3).

The discovery magnitude was 12.9 CV. Spectroscopy by Sokolovsky et al. (2020) suggested that this was an Fe II-type classical nova near maximum.² The expanding velocity of the nova ejecta was initially around –850 ±100km/s. A spectrum by David Boyd on 2020 Jul 30.905 (Figure 4) showed an apparent full width at half maximum for the H-alpha line of 508km/s, while there is evidence of weak P Cygni absorption on the blue side of H-alpha, extending to –800km/s relative to the peak of emission. There are also emission lines of Fe II 5018, Fe II 5169 and O I 5577. A low-resolution objective prism spectrum by Mike Harlow taken on Jul 31 is shown in Figure 5.

The light curve (Figure 6) shows that at its brightest the nova reached magnitude 10.6, on Aug 10. After maximum, the nova became fainter and fell to magnitude ∼13.2 around Aug 16, before becoming brighter again. It varied between magnitude 11.6 and 13.6 for around three months; such oscillations are typical of dusty novae. Then, in the second week of December, it faded rapidly and was below magnitude 21 in the late part of that month. The fade appears to have been due to dust absorption.³ There was then a recovery in brightness to magnitude ~16.4 by late 2021 April, presumably as the dust cleared. From this point, it exhibited a very gradual fade, reaching magnitude 17 by the end of 2021. Due to the complexity of the first part of the light curve, we did not determine t₂ and t₃.

The pronounced ‘dust dip’ is similar to that observed in DQ Her (Nova Her 1934) roughly four months after peak brightness. This was caused by dust forming as the ejected shell expanded and cooled. About 20% of novae show evidence of dust formation within months of the eruption. However, the precise mechanism responsible for dust formation is still a mystery. One hypothesis is that once the temperature in the nova shell drops to ~1,000– 2,000 K, atoms of carbon and silicon as well as molecular species condense into very small grains, typically 0.01 to 1 microns across. This ‘dust’ blocks the radiation from the nova and its luminosity drops. The onset of dust formation is usually rapid, since when a few grains form, they block the radiation, causing more grains to condense. As the shell continues to expand, the dust dissipates and the radiation can once again pass through, causing the nova to apparently brighten. Recently, however, Derdzinski et al. (2017) suggested that strong shocks in nova ejecta might precipitate dust formation.⁴ They proposed that dust formation occurs within the cool, dense shell behind the shock, where the density is high enough for rapid dust nucleation.

Spectroscopy of V1391 Cas by Fujii et al. (2021) revealed the presence of the diatomic molecules C₂ and CN on 2020 Aug 12,⁵ but these were only in evidence for about three days. The formation of C₂ and CN indicates that the nova envelope gas was carbon-rich. This is only the second example of a C₂/CN-forming nova, the other being V2676 Oph. This latter nova exhibited a late-phase grain-formation episode, which is also the likely explanation of the V1391 Cas fade in 2020 December, as described previously in this paper.

Dust formation and subsequent dissipation might also explain the oscillations in the light curve of V1391 Cas observed in the early part of the eruption, before the pronounced dust dip. According to this model, the dust formation leads to a fade, then if the dust traps enough energy to destroy the grains, the nova can rebrighten – and the cycle continues. However, such oscillations are poorly studied and even less well understood.

Nova Per 2020 (V1112 Per)

The first report of Nova Per 2020 came from Seiji Ueda (Kushiro, Hokkaido, Japan), who found it on 2020 Nov 25 at 19:22 UT. Ueda has dedicated himself to the detection of novae using relatively simple equipment: a Canon EOS 6D digital camera with a 200mm ƒ/3.2 lens. The Russian team of Korotkiy, Sokolovsky & Olga Smolyankina were pipped at the post, making an independent discovery at 20:15 UT on the same night.

The VS-Alert e-mail community was informed of Ueda’s magnitude 10.6 CV discovery in a report sent at 21:37 UT.⁶ This e-mail was picked up by the author a few minutes later, when he came inside for a break from routine observing. There being no charts available at the time, he produced his own using planetarium software. This allowed him to pick up the nova via CCD imaging at 21:55 UT, and it was brightening rapidly. Gary Poyner (Birmingham) observed it visually at 22:31 UT. This exciting evening showed how quickly discoveries can be communicated and followed up worldwide. An image of the field by Martin Mobberley is shown in Figure 7.

Spectroscopy on Nov 26.05 by Munari et al. (2020) showed it to be a classical nova.⁷ Banerjee et al. (2020) reported its transition to a typical Fe II nova.⁸ David Boyd was the first amateur to obtain a spectrum, on Nov 26.772 (Figure 8). It shows characteristic H Balmer and He I emission lines, with prominent P Cygni dips on their blue edges caused by absorption in the material expanding towards us from the nova explosion. David’s spectrum taken five days later (also shown in Figure 8) shows the dramatic change which took place shortly after the initial eruption. In addition to the general increase in the energy being emitted across the visual spectrum, the strength of the hydrogen Balmer emission lines and the depth of the P Cygni absorption dips on the blue side of the lines had both increased. The components of the singly-ionised iron Fe II multiplet, produced in the so-called ‘iron curtain’ which forms after the initial explosion, had become prominent.

The light curve (Figure 9) shows that the nova displayed significant variation during the first two weeks of the eruption (Figure 9b). It reached magnitude 8.2–8.3 during flares on Nov 29 and Dec 1, 2, 3, 4 & 7. A gradual decline began around 2020 Dec 11 and by 2021 Jan 4 it was magnitude ~11, at which point a rapid fade set in. This decline was accompanied by a marked reddening, suggesting the onset of dust formation. This was confirmed by Banerjee et al. (2021) using near-infrared JHK-band photometry on 2021 Jan 15.⁹ (The reddening was evident in many of the unfiltered CCD observations submitted during this period, which were 0.5 to 0.8 magnitudes brighter than V and visual data – it is for this reason that only V and visual data are included in Figure 9.)

This was thus a typical ‘dust fade’, which reached a minimum of magnitude 16 at the end of 2021 January. After that, the nova gradually brightened again, reaching magnitude 14.2 in early May before a gap in observations until mid-August of 2021. From this point there was a slow fade, from magnitude 14.6 to 15.5 by the end of 2021. As the year closed, the nova was still 4.5 magnitudes above quiescence.

Robin Leadbeater obtained a low-resolution spectrum at 280 days on 2021 Sep 2, by which time the nova had faded to magnitude 15 (Figure 10). It is a nebula-type spectrum with strong forbidden emission lines, particularly O III, and a very weak, almost undetectable continuum. Leadbeater estimated that 78% of the light in the V passband came from just the O III pair of lines at 4959/5007Å.

The measured values of t₂ and t₃, 19 and 30 days respectively (Table 1), are consistent with this being a fast nova.

(Login or click above to view the full illustrated article in PDF format)