aa r X i v : . [ a s t r o - ph . H E ] J a n
50 Years of Pulsars!
R. N. Manchester
CSIRO Astronomy and Space Science, PO Box 76, Epping NSW 1710, AustraliaE-mail: dick.manchester@csiro
Abstract.
A brief, personal, and very incomplete account of 50 years of pulsar astronomypresented at the Conference Dinner for “Physics of Neutron Stars – 2017 – 50 Years After”, heldin Saint Petersburg, July 2017.
1. Introduction and the discovery of pulsars
First, I would like to thank George Pavlov for the invitation to attend this meeting and to presentthis talk at the Conference Dinner. I have to say, I was a little surprised to be asked to givethe after-dinner talk as I am not well-known as a raconteur. I could only conclude that it wasbecause I am old!It is true that I have been involved with pulsars one way or another since their discovery, orat least the official announcement of it by Hewish et al. in the February 24, 1968, edition ofNature. I started work at the CSIRO Parkes Observatory just 12 days before the Nature paperwas published and soon became involved in pulsar research.But first back to the real beginning. In mid-1967, Jocelyn Bell was a graduate student atCambridge University helping to build the “Four-acre array”, an array of dipoles tuned to 81.5MHz that Antony Hewish had designed to investigate interplanetary scintillation of radio sourceswith the aim of identifying compact sources likely to be quasars. With the completion of thearray, Jocelyn was given the job of examining the chart recordings to find the rapidly fluctuatingsignals from compact sources that scintillated as a result of propagation through the solar wind.She did find many of these, but also began to notice some signals that were subtly different fromboth the scintillating sources and radio-frequency interference. The first recorded detection ofone of these “scruff” sources was on August 6, 1967, as shown in Figure 1. The break-through andthe point that marks the real discovery of pulsars was when she realised that this source returnedat the same sidereal time most days. This meant that it had to be outside the solar system andunrelated to terrestrial activity. A few months later, on November 28, she and Antony Hewishobtained the first high-speed recording to show that the scruff was in fact a train of pulses witha spacing of about 1.3 seconds.Follow-up observations with the Four-acre array and other instruments at Cambridge refinedthe periodicity of the pulsar and showed that the signals were dispersed by their passage throughthe interstellar medium. A paper reporting the discovery and follow-up observations of thepulsar, then named CP 1919, and an interpretation that favoured radial pulsations of either awhite dwarf or a neutron star was submitted to Nature by Hewish, Bell, et al. on February 9,1968, and published on February 24 [2].In 1974 Antony Hewish was awarded the Nobel Prize in Physics for the discovery of pulsars. igure 1.
Left: Chart recordings from the Four-acre array at Cambridge showing the firstdetected signals from CP 1919 and the first recording to show that the emission was pulsed.(Reprinted with permission from Reviews of Modern Physics [1], copyright (1975) by theAmerican Physical Society.) Right: Jocelyn Bell and Antony Hewish at IAU Symposium 95,Bonn, August 1980.
2. Pulsar observations at Parkes
Just two weeks after the publication of the Nature discovery paper a team from the CSIROand the University of Sydney installed a remarkable set of five coaxial feeds with correspondingreceivers at the focus of the Parkes 64-m radio telescope. On the morning of March 8, I wasin the control room as the telescope slewed around to the position of CP 1919 and saw thewonderful burst of pulses come through on the chart recorder seconds later. By good fortune,this first observation at 150 MHz just caught a large scintillation maximum. The pulsar was notseen again with such high signal-to-noise for the rest of the session. This image was recorded forposterity on the first Australian $50 note (Figure 2) and also published with what are still someof the best measurements of the individual-pulse spectra of any pulsar [3].Later on that year, having learnt something about measuring polarisation with the 18-cmreceiver while working with Brian Robinson and Miller Goss on observations of OH masers,I was asked to assist Radhakrishnan investigate the polarisation of the Vela pulsar. Theseobservations led to rotating-vector model (RVM) (or magnetic-pole model as Rad preferred tocall it) for pulsar polarisation (Figure 3) [4]. In follow-up observations in mid-March, 1969, weagain pointed the telescope to the Vela pulsar, recording the data with a signal-averaging systemthat was set up to fold the data at the expected topocentric period of the pulsar. To our surprise,rather than remaining a fixed phase on the oscilloscope display, the pulse was noticeably driftingto the left, indicating an error in the predicted period. By the evening, after a while trying tounderstand what was wrong, Rad said that he was going to bed and left me to sort out theproblem. I again checked all the instrumentation, looked at some other pulsars and eventuallyconcluded that it must be the pulsar that had changed. I left a note for Rad and went to bedmyself. igure 2.
The first Australian $50 note, featuring the Parkes radio telescope, a portrait of SirIan Clunies Ross (the first Chairman of CSIRO) and various images relating to Australian radioastronomy and biological sciences. On the left of the note is an image of the first pulsar recordingmade at Parkes, of CP 1919 and recorded on March 8, 1968.
Figure 3.
Left: Radhakrishnan in the year 1996 (image credit: Library, Raman ResearchInstitute). Right: plots illustrating the frequency-independence of the position-angle swing acrossthe Vela pulsar mean pulse profile and the interpretation of that in terms of the magnetic-pole orrotating-vector model (after Radhakrishnan & Cooke [4]). In this model, the radiation is emittedtangentially to field lines and is linearly polarised in the plane of field-line curvature as indicatedby the double-headed arrows in the lower plot. (Image credit: Peter Shternin)Observations over the next few days confirmed that it was a real decrease in the pulsar periodby about 3 parts in a million - the first detection of a pulsar glitch! Rad contacted Paul Reichleyand George Downs who we knew were observing the pulsar using the Goldstone antenna at the JetPropulsion Laboratory, Caltech, and found that they too had observed the glitch. Back-to-backpapers reporting the discovery were published in Nature on April 19, 1969 [5, 6]. . Searching for pulsars
Searches for pulsars are fundamentally important. Not only does the discovery of previouslyunknown pulsars increase the sample for all manner of statistical studies of pulsar propertiesand evolution, but almost every significant pulsar search has uncovered some unexpected andinteresting pulsar or class of pulsars. Examples are binary pulsars [7], pulsars in globular clusters[8], RRATs [9], pulsars with planets [10], pulsars with main-sequence binary companions [11],etc., etc..More than 2600 pulsars are now known. Figure 4 shows the rate of pulsar discovery since1968 by observatory (or in a couple of cases, pulsar class). Small searches (including the originalCambridge search) are grouped under “Other”. This figure highlights the major contributionsmade by the Molonglo radio telescope in the first decade – twice in this time, more than halfof the known pulsars were discovered at Molonglo – and the Parkes radio telescope. In the late2000’s, Parkes had found more than twice as many pulsars as the rest of the world’s telescopesput together. Even now, the Parkes share is more than 55% of the total. The Parkes MultibeamPulsar Survey [12] by itself has more than 830 pulsars to its credit, including recent discoveriesfrom reanalyses of the dataset, e.g., [13]. In recent years, the
Fermi
Gamma-ray Space Telescopehas been very successful in uncovering previously unknown pulsars, especially millisecond pulsars(MSPs) in so-called “redback” and “black widow” binary systems where the pulsar radio emissionis often obscured or eclipsed by plasma streams from the companion star. Most of these systemshave been found through radio searches of unidentified gamma-ray sources with properties knownto be characteristic of pulsars, e.g., [14].
Figure 4.
Rate of pulsar discovery, sorted by observatory or class. Each bar represents thenumber of pulsars discovered in the corresponding 2-year interval. Data from the ATNF PulsarCatalogue [15]People often ask me why Parkes has been so successful in pulsar discovery. There are threemain reasons: • The centre of our Galaxy passes almost overhead at Parkes and the southern part of theGalactic plane is far richer than that in the north.
At ATNF we have had and continue to have a very skilled group of engineers and excellentco-operation with the scientific staff. This led, for example, to the development of the Parkes13-beam receiver system, by far the most successful pulsar-finding machine ever. • We had very experienced teams of scientists working on the pulsar survey projects at Parkes(see, e.g., Figure 5). This led to the development of efficient signal-processing systems (bothhardware and software) for pulsar detection and confirmation.
Figure 5.
Andrew Lyne, Joe Taylor and Dick Manchester at IAU Colloquium 177 “PulsarAstronomy – 2000 and beyond”, Bonn, August, 1999.
4. Key discoveries
The discovery of the first binary pulsar, PSR B1913+16, at Arecibo in 1974 by Russell Hulseand Joe Taylor [7] is the prime example of an unexpected and exciting find in pulsar searches.Observations of this system over the next few years provided the first accurate determinations ofneutron-star masses, the first observational evidence for gravitational radiation and confirmationthat Einstein’s general theory of relativity gives an accurate description of motions in stronggravitational fields [16, 17]. These important results led to the award of the 1993 Nobel Prize inPhysics to Taylor and Hulse. As an aside, I note that I shared an office with Joe at the Universityof Massachusetts from 1971 to 1974, but unfortunately was not involved in the Arecibo search!Probably the next most important pulsar discovery was that of the first MSP at Arecibo byDon Backer and his colleagues [18]. This was a little different to the other examples of importantdiscoveries in that it was not found in a large-scale search, but rather in a directed investigation ofa highly unusual radio source, 4C21.53W. This source was known to have a compact component(it scintillated) and it is steep-spectrum and polarised, all properties consistent with it being apulsar. Yet efforts to detect a pulsar in this direction (including one at Parkes by Nichi D’Amicoand RNM) had been unsuccessful. This changed dramatically when Backer et al. employed a sub-millisecond sampling system that revealed the 1.558 ms periodicity. Figure 6 shows a photographof Don Backer, probably taken in the early 2000’s, and images from the discovery paper. Papersuggesting that the rapid spin of the pulsar originated in an earlier phase of accretion from acompanion star (since disappeared) were quickly published [19, 20]. The “recycling” idea hadpreviously been invoked to account for the short period of PSR B1913+16 [21].
Figure 6.
The discovery of the first MSP: Don Backer who led the team (image credit: BobRood), a 600-MHz image of 4C21.53W made with the Westerbork Synthesis Radio Telescope,and a signal-averaged recording from Arecibo showing the 1.56 ms periodicity. The cross marksthe position of the compact source in the Westerbork image. (Reprinted by permission fromMacmillan Publishers Ltd: Nature, [18], copyright 1982.)Finally, I must mention the Double Pulsar, PSR J0737 − . ◦ yr − , four times that of PSR B1913+16. Secondly, it was and remains the only DNSsystem where both stars have been observed as pulsars. The B star, although younger than theA star, has a much longer pulse period, about 2.77 s. Thirdly, the orbit is viewed almost exactlyedge-on, making the Shapiro delay easy to observe [24] and allowing eclipses of the A pulsar bythe magnetosphere of the B pulsar [25]. Recent observations of the system have resulted in thedetection of six relativistic effects, including a measurement of the orbital decay that is an orderof magnitude more precise than the PSR B1913+16 determination and fully consistent with theprediction from general relativity (Kramer et al., in preparation). It never ceases to amaze methat, despite the development of many alternative theories of relativistic gravity since generalrelativity, Einstein seems to have got it right the first time.
5. Future prospects
The past 50 years of pulsar astronomy have been more than interesting but there is much to lookforward to. Large new radio telescope such as the recently commissioned Five hundred metreAperture Spherical Telescope (FAST) in Guizhou province, China [26] will enable the discoveryand study of many more pulsars, as well as increasing the precision of observations of currentlynown pulsars. Further in the future, the Square Kilometre Array [27] will provide unrivalledsensitivity and resolution for essentially all radio astronomy applications. New X-ray telescopessuch as eRosita, to be launched next year [28], and Athena, due for launch in 2028 [29], willprovide wide-field spectroscopic imaging with unprecedented sensitivity and resolution. Existingradio telescopes will be enhanced with new wideband receivers and more sophisticated signal-processing systems. All of these and more will surely probe deeper into the secrets of pulsars,uncovering many things that are not even dreamt of yet. The future of pulsar astronomy andastrophysics is bright!
Bol~xoe spasibo!
Acknowledgments
It has been a privilege and a pleasure to have travelled through 50 years of pulsar astronomyand astrophysics and I thank all those colleagues and friends who have helped me along the way.
References [1] Hewish A 1975
Rev. Modern Physics Nature
Nature
Astrophys. Lett. Nature
Nature
ApJ
L51–L53[8] Lyne A G, Brinklow A, Middleditch J, Kulkarni S R, Backer D C and Clifton T R 1987
Nature
Nature
Nature
ApJ
L37–L41[12] Manchester R N, Lyne A G, Camilo F, Bell J F, Kaspi V M, D’Amico N, 2001
MNRAS
ApJ
ApJ AJ Nature
ApJ
Nature
Curr. Sci. Nature
ApJ
Nature
Science
Science
ApJ
L131–L134[26] Nan R, Li D, Jin C, Wang Q, Zhu L, Zhu W, Zhang H, Yue Y and Qian L 2011
Intnl J. of Mod. Phys. D astro-ph/0409274 [28] Predehl P, Andritschke R, Babyshkin V, Becker W, Bornemann W, Bräuninger H, et al. 2016 Space Telescopesand Instrumentation 2016: Ultraviolet to Gamma Ray ( SPIE vol 9905) p 99051K[29] Ayre M, Bavdaz M, Ferreira I, Wille E, Lumb D and Linder M 2015