Printer Friendly

The exotic side of superhumps and accretion discs: the 2013 George Alcock Memorial Lecture, given at Burlington House, Piccadilly, London on 2013 May 29.


George Alcock (1912-2000) was without doubt one of the most accomplished visual observers of the twentieth century. His visual discoveries from his Peterborough home are legendary: five comets and five novae. I had the honour of meeting George on only one occasion, but it is etched on my memory. The event was the 25th Anniversary meeting of the JAS held at Holborn Library on 1978 April 29. George gave an inspiring talk with a particular focus on his comet observations.

During the tea break my school chums and I, who had been excused games for the afternoon to attend the meeting, had a chat with George who showed us many of his exquisitely detailed drawings of comets that he had brought along. He encouraged us to spend time observing and to record what we saw. We lapped up his advice and it was timely since we had just completed construction of an observatory and 14-inch reflector at school. Moreover my personal astronomical journey had already progressed from the obligatory 60mm Tasco refractor, via a 6-inch Fullerscopes Newtonian, to a 10-inch telescope and every spare moment was spent reading about, talking about or, weather permitting, actually doing observing.

Over the next two and half decades, astronomy was never far from the surface, although it often succumbed to the demands of career and family. I wouldn't have had it any other way, but when the opportunity arose in 2004 to settle down in the UK after several years overseas I realised my ambition of establishing a permanent observatory in my garden in Cheshire: the Bunbury Observatory (above). The dome currently houses a Celestron 11 mounted on a Gemini Telescope Design G41+ equatorial. Having learnt the basics of CCD imaging of deep sky objects, I was soon bitten by the variable star bug, recognising that valuable scientific work in this field could still be done by amateurs. Consequently since 2005 the telescope has mainly been used for CCD photometry of cataclysmic variable stars (CVs), particularly dwarf novae.

Dwarf novae

Dwarf novae are interacting binary stars in which a cool main sequence secondary star loses mass to a white dwarf primary. Since the material carries substantial angular momentum, it does not settle on the primary immediately, but forms an accretion disc around the white dwarf. As material builds up in the disc, a thermal instability is triggered that drives the disc into a hotter, brighter state causing an outburst in which the star apparently brightens by several magnitudes. The outbursts repeat quasi-periodically. One of the best known dwarf novae is SS Cygni, which spends most of its time in quiescence at 12th magnitude, but every couple of months suddenly brightens to 8th magnitude for a few days before gradually fading again.

Dwarf novae of the SU UMa family occasionally exhibit 'superoutbursts', which last several times longer than normal outbursts, usually a week or two, and may be up to a magnitude brighter. During a superoutburst the lightcurve of an SU UMa system is characterised by superhumps. These are saw-tooth modulations in the lightcurve, usually 0.1 to 0.3 mag in amplitude, the period of which is a few percent longer than the orbital period. They are thought to arise from the interaction of the secondary star orbit with a slowly precessing eccentric accretion disc. The eccentricity of the disc arises because a 3:1 resonance occurs between the secondary star orbit and the motion of matter in the outer accretion disc. (1)

I pursue two main observational programmes. One is to patrol for outbursts of poorly characterised CVs. These include targets on the VSS Recurrent Objects Programme, (2) and CV candidates identified by recent surveys such as the Sloan Digital Sky Survey (SDSS). Having observed an outburst, or perhaps responding to an outburst alert from another observer, I carry out time-resolved photometry. This simply means sitting on the target and taking a succession of images, usually of 30 to 60 seconds duration, for as long as possible. This is the relaxing bit as the telescope does its work and all I have to do is to shift the dome round periodically as the target tracks across the sky and check for rain (several times it has rained from an apparently clear sky!).

Monitoring an outburst is an activity which lends itself to cooperation, since to get as complete a light curve as possible observers should be distributed around the world. For me, working with observers across the globe, both amateur and professional, is one of the most interesting aspects of the hobby and has resulted in many friendships. In addition, some real science can be done by analysing the combined data.

So what information can be gleaned from observing a superoutburst? Well, measuring the superhump period, [], immediately gives a reasonable idea of the orbital period, [P.sub.orb], of the system, since there exists an empirical relationship between the two. Moreover in some systems a separate orbital hump, superimposed on the larger superhump profile, allows an independent measurement of [P.sub.orb]. Knowing the superhump period excess, ([] - [P.sub.orb])/[P.sub.orb], allows one to estimate the mass ratio of the secondary to the white dwarf primary. Already a picture of the binary system begins to emerge.

Eclipsing dwarf novae

If the binary system is highly inclined to the observer, eclipses will be observed. Measuring eclipse times of minimum allows an eclipse ephemeris to be drawn up, along with an accurate measurement of [P.sub.orb]. An example was the superoutburst of NZ Boo observed in 2009 July, during which deep eclipses of up to 2.1 magnitudes were seen. (3) The outburst lasted 16 days and its amplitude above mean quiescence was 3.9 magnitudes. By measuring eclipse times and supplementing these with photometry obtained with the 1.3m MDM telescope on Kitt Peak some years before when the system was quiescent by our co-workers, Prof Joe Patterson and Jonathan Kemp of the Center for Backyard Astrophysics, we determined [P.sub.orb] as 84.8min.

Careful analysis of individual eclipses can reveal even more information. Thus we found that the eclipse duration gradually decreased as the superoutburst progressed and the star faded, indicating that the accretion disc was shrinking from the outside inwards as material drained from it. Sometimes even more information can be extracted from an eclipse lightcurve, for example the location of the bright spot, where the accretion stream hits the disc, and the white dwarf itself can be mapped. Careful analysis of the Kitt Peak quiescence photometry of NZ Boo showed a feature which we interpreted as the end of the eclipse of the white dwarf.

High inclination systems are usually revealed by a doubling of the hydrogen spectral lines due to the Doppler effect. But this is not always the case. For example SDSS J081610.84+453010.2 has single spectral lines and therefore it was a surprise when our photometry during a superoutburst revealed shallow eclipses which we interpreted as grazing occultations of the edge of the accretion disc. (4) In the case of V1227 Her, tiny (0.08 mag) eclipses were only seen near superoutburst maximum when the disc was near its greatest extension. (5)

The evolution of outbursts

Often people stop observing a superoutburst after the first few days, once [] has been established and the excitement dies down. This is a pity as continued monitoring can reveal further secrets. For example, in most cases the value of [] changes during an outburst, as does the amplitude of the superhumps, and sometimes there appears to be a correlation between the two, as well as other gross changes in the lightcurve. The significance of these is not fully understood and definitely warrants further study. Moreover, although the lightcurve of different superoutbursts of a particular star is broadly similar, there are subtle variations, the significance of which is also not completely clear.

Another hotly debated topic is what actually initiates a superoutburst. In some SU UMa systems a normal outburst seems to trigger the subsequent superoutburst. Rather few precursor outbursts have been studied in detail, mainly due to the practical difficulty of actually catching them in the act--again this is where monitoring and prompt photometry by amateurs can be helpful. We did manage to observe a precursor outburst preceding a superoutburst of V342 Cam, during which we observed orbital humps which gave way to superhumps during the rise to the superoutburst itself. (6) Similarly a precursor outburst in NN Cam (7) was well observed. In the future a global study of precursor outbursts might be possible.

WZ Sagittae systems

SU UMa type dwarf novae have [P.sub.orb] from about 2 hours down to about 80 min. Some of the more exotic SU UMa systems are the WZ Sge stars, which lie at the lower end of this range. These highly evolved systems have relatively low mass transfer rates and consequently tend to go into outburst infrequently: sometimes after years or even decades.

When they do it's always exciting to catch one. One example is V358 Lyr which Gary Poyner and I found in outburst using the Bradford Robotic Telescope (BRT) in 2008 November, some 43 years after its only other confirmed outburst. (8) Then there was the morning in late 2006 October when I rose early to spend some time with the predawn variables and, alighting on the field of SS LMi, found it to be in outburst for the first time in 26 years. (9) As the sky was already brightening further photometry was impossible. There then followed an anxious two days before another observer, Tom Krajci in the US, confirmed it and further photometry showed superhumps revealing its SU UMa nature for the first time.

The lightcurves of WZ Sge systems are intriguing since they often show multiple rebrightenings, or echo outbursts, after the main event. This again is where continued monitoring after the initial excitement pays off. For example, the 2008 superoutburst of DY CMi lasted about 19 days and was followed by six remarkable echo outbursts. (10) At the other end of the orbital period distribution of dwarf novae lie systems such as V630 Cas, whose period is 2.6 days. Gary Poyner and I ran a campaign to observe its rare outburst in 2009, only the third on record, in which the rise to maximum took a leisurely 61 days with a decline of 43 days. (11)

AM CVn systems: the helium dwarf novae

An even more exotic class of CV are the AM CVn systems. These are also accreting binaries containing a white dwarf, but they are almost or entirely hydrogen deficient and the accretion disc is made of helium. These ultracompact binaries have very short orbital periods of 5 to 65 min.

Compared to their hydrogen-containing cousins, relatively few AM CVn systems are currently known: only about three dozen at the time of writing. With the advent of digitised sky surveys, the number is increasing rapidly, but further long-term studies are required to help characterise their behaviour. Again this is where the amateur comes in as some are sufficiently bright for small telescopes equipped with CCD to patrol for outbursts and to conduct time resolved photometry when one is detected. There are in fact three groups of AM CVns--those in a permanently bright state, those in a permanently faint state and those that undergo outbursts --and which group a star is in is related to the orbital period.

One of the AM CVn systems on my patrol programme is SDSS012940.05+384210.4, and I was fortunate in spotting an outburst in late 2009 using the BRT. (12) Follow-up photometry revealed tiny superhumps similar to those seen in SU UMa systems. We also observed six echo outbursts, leading us to suggest it is a helium analogue of a WZ Sge system. Detailed analysis allowed us to estimate its orbital period, hitherto unknown, as 37 minutes. Noting the similarity of its orbital period to another AM CVn system, SDSS J124058.03-015919.2, prompted us to wonder whether this too could be an outbursting system. Sure enough, interrogation of the online All Sky Automated Survey (ASAS3) and Catalina Real-Time Sky Survey (CRTS) databases revealed a previously unrecognised outburst in 2005.

Comparing the properties of helium and hydrogen accretion discs may provide a better understanding of accretion disc physics. AM CVn stars also hold great interest as mass transfer in these systems is believed to be driven by gravitational radiation. Understanding their population may help in the interpretation of gravitational wave detections by planned orbiting gravitational wave observatories. They also hold considerable interest as they are considered by many to be progenitors of some Type Ia supernovae.

The truly exotic: black holes and X-ray binaries

There is an even more exotic class of binary system in which the accretor is a neutron star or even a black hole. These systems emit X-rays and are sometimes called Low Mass X-Ray Binaries. Some systems also undergo outbursts similar to dwarf novae in which the accretion disc brightens and even shows superhumps.

One such system is V404 Cygni, which is believed to comprise a 17 solar mass black hole with a red giant in a 6.5 day orbit around it. Three outbursts of V404 Cyg have been observed: in 1938 ('Nova Cyg 1938'), 1956 and the last in 1989, when it was identified as the optical counterpart of an X-ray transient detected by the Ginga satellite. I monitor it every clear night, hoping that one day when I download an image of the field, the 18th magnitude star at its position will once again be replaced by a beacon shining at 11th magnitude!

Accretion is not restricted to binary systems. Discrete black holes also accrete material in their vicinity. Another example is accretion during star formation. In this process, as a star condenses out of a rotating interstellar cloud, the cloud gets flattened into a disc which begins to be accreted onto the new star, forming a T Tauri object. When most of the material has been accreted, the remnants still carry significant angular momentum. These eventually condense into planets as the particles in the disc collide and stick together. This is probably how our solar system formed.

Studying the accretion discs around black holes and in early stellar systems clearly poses severe observational challenges (and in the case of the solar system it is impossible!) However, CVs are laboratories from which we can learn much about the underlying physics of accretion. And the amateur astronomer continues to play an important role in revealing the secrets of these systems.

I have stressed the fact that observing CVs is an activity that lends itself to cooperation. I therefore would like to acknowledge the cooperation I have enjoyed with many astronomers, both amateur and professional, over the past eight years. Many have shared their hard-won photometric data with me, that have been essential to the analyses we have jointly published in the Journal and elsewhere.


(1) An excellent introductory book to CVs is: Hellier C., Cataclysmic Variable Stars--how and why they vary, Springer, 2001.

(2) Poyner G., http://www.garypoyner.pwp.

(3) Shears J. et al., J. Brit. Astron. Assoc., 121, 96-104 (2011)

(4) Shears J. et al., J. Brit. Astron. Assoc., 122, 237-241 (2012)

(5) Shears J. et al., J. Brit. Astron. Assoc., accepted for publication (2013)

(6) Shears J. et al., New Astronomy, 16, 311-316 (2011)

(7) Shears J. et al., J. Brit. Astron. Assoc., 121, 355-362 (2011)

(8) Shears J. et al., J. Brit. Astron. Assoc., 120, 43-48 (2010)

(9) Shears J. et al., J. Brit. Astron. Assoc., 118, 95-100 (2008)

(10) Shears J. et al., J. Brit. Astron. Assoc., 11 9, 340-347 (2009)

(11) Shears J. & Poyner G., J. Brit. Astron. Assoc., 120, 169-172 (2010)

(12) Shears J. et al., J. Brit. Astron. Assoc., 122, 49-53 (2012)
COPYRIGHT 2013 The British Astronomical Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Shears, Jeremy
Publication:Journal of the British Astronomical Association
Geographic Code:4EUUK
Date:Dec 1, 2013
Previous Article:Ordinary Meeting, 2013 May 29: held at the Royal Astronomical Society, Burlington House, Piccadilly, London WI.
Next Article:The 2013 October outburst of comet C/2012 X1 (LINEAR).

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |