MeerTime - the MeerKAT Key Science Program on Pulsar Timing
M. Bailes, E. Barr, N. D. R. Bhat, J. Brink, S. Buchner, M. Burgay, F. Camilo, D. J. Champion, J. Hessels, G. H. Janssen, A. Jameson, S. Johnston, A. Karastergiou, R. Karuppusamy, V. Kaspi, M. J. Keith, M. Kramer, M. A. McLaughlin, K. Moodley, S. Oslowski, A. Possenti, S. M. Ransom, F. A. Rasio, J. Sievers, M. Serylak, B. W. Stappers, I. H. Stairs, G. Theureau, W. van Straten, P. Weltevrede, N. Wex
aa r X i v : . [ a s t r o - ph . I M ] M a r MeerTime - the MeerKAT Key Science Program onPulsar Timing
M. Bailes † a , E. Barr b , N. D. R. Bhat c , J. Brink d , S. Buchner e , M. Burgay f , F. Camilo e ,D. J. Champion b , J. Hessels g , G. H. Janssen g , h , A. Jameson a , S. Johnston i , A.Karastergiou j , R. Karuppusamy b , V. Kaspi k , M. J. Keith l , M. Kramer b , M. A.McLaughlin m , K. Moodley n , S. Oslowski a , A. Possenti f , S. M. Ransom o , F. A. Rasio p ,J. Sievers q , M. Serylak e , B. W. Stappers l , I. H. Stairs r , G. Theureau stu , W. vanStraten v , P. Weltevrede l , N. Wex b c (cid:13) Copyright owned by the author(s) under the terms of the Creative CommonsAttribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). https://pos.sissa.it/ - Speaker. a OzGrav, Centre for Astrophysics and Supercomputing, Swinburne University of Technology, H11PO Box 218 Hawthorn, Vic, 3122, Australia b Max-Planck-Institut für Radioastronomie Auf dem Hügel 69, D-53121 Bonn, Germany c International Centre for Radio Astronomy Research, Curtin University Bentley, WA 6102,Australia d Independent African P.O. Box 17633, Bainsvlei, South Africa, 9338 e Square Kilometer Array South Africa, The Park, Park Road, Pinelands, Cape Town 7405 f INAF - Osservatorio Astronomico di Cagliari via della Scienza 5, 09047 Selargius (CA), Italy g ASTRON, the Netherlands Institute for Radio Astronomy, Postbus 2, 7990 AA, Dwingeloo, TheNetherlands. h Department of Astrophysics/IMAPP, Radboud University, P.O. Box 9010, 6500 GL Nijmegen,The Netherlands. i CSIRO Astronomy and Space Science PO BOX 76, NSW 1710, Australia. j Astrophysics, University of Oxford, Denys Wilkinson Building, Keble Road, Oxford OX1 3RH,UK k McGill University, 3600 University St., Montreal, QC H3A 2T8, Canada l Jodrell Bank Centre for Astrophysics, School of Physics , Astronomy, University of ManchesterAlan Turing Building, Oxford Road, Manchester, M13 9PL, UK m WVU, Center for Gravitational Waves and Cosmology, Chestnut Ridge Research Building,Morgantown, WV 26506 n Astrophysics and Cosmology Research Unit, School of Mathematics, Statistics and ComputerScience, University of KwaZulu-Natal, Durban, 4041, South Africa o NRAO, 520 Edgemont Rd., Charlottesville, VA, 22903, USA q Astrophysics , Cosmology Research Unit, School of Mathematics, Statistics , Computer Science,University of KwaZulu-Natal, Durban, 4041, South Africa r Dept. of Physics and Astronomy, University of British Columbia, 6224 Agricultural Road,Vancouver, BC V6T 1Z1 Canada p CIERA, Northwestern University, Evanston, IL, USA r Dept. of Physics , Astronomy, University of British Columbia, 6224 Agricultural Road,Vancouver, BC V6T 1Z1 Canada s Laboratoire de Physique et Chimie de l’Environnement et de l’Espace LPC2E CNRS-Universitéd’Orléans, F-45071 Orléans, France t Station de radioastronomie de Nançay, Observatoire de Paris, PSL Research University,CNRS/INSU F-18330 Nançay, France u Laboratoire Univers et Théories LUTh, Observatoire de Paris, PSL Research University,CNRS/INSU, Université Paris Diderot, 5 place Jules Janssen, 92190 Meudon, France v Institute for Radio Astronomy & Space Research, Auckland University of Technology, PrivateBag 92006, Auckl, 1142, New Zealand, he MeerKAT telescope represents an outstanding opportunity for radio pulsar timing sciencewith its unique combination of a large collecting area and aperture efficiency (effective area ∼ ), system temperature ( T < ∼ − ◦ ) and ability to formup to four sub-arrays. The MeerTime project is a five-year program on the MeerKAT array byan international consortium that will regularly time over 1000 radio pulsars to perform tests ofrelativistic gravity, search for the gravitational wave signature induced by supermassive blackhole binaries in the timing residuals of millisecond pulsars, explore the interiors of neutron starsthrough a pulsar glitch monitoring programme, explore the origin and evolution of binary pul-sars, monitor the swarms of pulsars that inhabit globular clusters and monitor radio magnetars.MeerTime will complement the TRAPUM project and time pulsars TRAPUM discovers in sur-veys of the galactic plane, globular clusters and the galactic centre. In addition to these primaryprogrammes, over 1000 pulsars will have their arrival times monitored and the data made imme-diately public. The MeerTime pulsar backend comprises two server-class machines each of whichpossess four Graphics Processing Units. Up to four pulsars can be coherently dedispersed simul-taneously up to dispersion measures of over 1000 pc cm − . All data will be provided in psrfitsformat. The MeerTime backend will be capable of producing coherently dedispersed filterbankdata for timing multiple pulsars in the cores of globular clusters that is useful for pulsar searchesof tied array beams. The first real-time pulsar profiles have been obtained as part of the MeerKATcommissioning process, and useful scientific data will start to come online through 2017. AllMeerTime data will ultimately be made available for public use, and any published results willinclude the arrival times and profiles used in the results. MeerKAT Science: On the Pathway to the SKA25-27 May, 2016Stellenbosch, South AfricaeerTime
1. Introduction
Fifty years after their discovery[1], radio pulsars remain a fertile area of astronomical re-search. In the standard model[2], radio pulsars are highly-magnetised neutron stars with magneticfield strengths of 10 -10 G, some 10 km in radius and with masses of between ∼ ⊙ . Mag-netic dipole radiation removes energy from the pulsar, and a light-house beam of radio emissionsweeps through space and results in a train of regular pulses that can be routinely detected by radiotelescopes of sufficient aperture. The energy loss results in a slow spin-down of the pulsar, andby accurately measuring pulse times of arrival at the observatory, a large number of observableparameters become available to pulsar astronomers.The MeerKAT Key Science Project on Pulsar Timing originated in 2010 when a strong sciencecase was put to an international time allocation committee based upon the then draft sensitivity forthe MeerKAT telescope of 220 m /K and bandwidth of 850 MHz. The project received a draftallocation of 7800h and the Key Science Project is now referred to as “MeerTime”. Up until 1974 all known pulsars were solitary objects with spin periods P >
33 ms with lim-ited astrophysical applications. Two major discoveries rocked the pulsar community when im-proved pulsar survey instrumentation enabled the discovery of faster pulsars. In 1974 the binaryPSR B1913+16 was found using the Arecibo telescope[3]. It was a 59 ms pulsar orbiting anotherneutron star every 7.75 hours. Its timing ultimately led to the Nobel prize in physics for verifyingthe existence of gravitational waves. Then in 1981 the celebrated 1.55 millisecond pulsar, PSRB1937+21 was discovered [4] . Today the pulsar catalogue[5] lists over 250 binary pulsars (almost10% of the population) and 350 millisecond pulsars. These pulsars are ideal relativistic laborato-ries and probes of stellar evolution. Their inherent stability give them applications in the search forgravitational waves. The majority of known pulsars reside in the southern hemisphere and 95% ofall pulsars are accessible to MeerKAT with its Southern location and 75 degree elevation limit.
MeerTime has three high-priority science themes: • Relativistic and Binary Pulsars:
Pulsars that possess compact binary companions in tightrelativistic orbits allow tests of General Relativity and its alternatives that are impossible toperform elsewhere. Binary pulsars are also fossil records of stellar evolution and accretionphysics. In addition General Relativity can be used to determine the masses of millisecondpulsars that inform us about the equation of state of nuclear matter. • Millisecond Pulsar Timing and Gravitational Wave Detection:
Millisecond pulsars canbe used to search for the signatures of gravitational waves generated by supermassive blackhole binaries and/or cosmic strings in the early Universe. • Globular Cluster Pulsar Timing:
Globular clusters are breeding grounds for millisecondpulsars and via partner exchange in their dense cores produce exotic systems that permitexperiments otherwise impossible to perform that inform us about General Relativity andalso the equation of state of nuclear matter via neutron star spins and masses.1 eerTime
MeerKAT’s design (comprising of 64 independent elements) means that it is possible to formup to four sub-arrays, allowing for some novel approaches to pulsar timing. This permits sub-arraymodes that enable the simultaneous production of pulse arrival times from four different pulsars indifferent regions of the sky.The MeerKAT system design originally forecast a sensitivity of some 220 m /K, but it nowappears that the aperture efficiency and system temperature may enable a remarkable sensitivity inexcess of 400 m /K, virtually quadrupling its efficiency for pulsar timing. For reference the otherSouthern hemisphere radio telescopes are all less than 70 m /K, although the new ultra-widebandreceiver (700 MHz-4 GHz) for the Parkes 64 m radio telescope still make it a highly-effective pulsartelescope, particularly for dispersion measure determinations.Further classes of pulsar were part of the original science case: • Young, Glitching and Highly-magnetised pulsars:
These objects allow probes of neutronstar interiors where physics is at its most extreme. • The Thousand Pulsar Array:
Most pulsars are not timed regularly because of the timecommitment required on otherwise over-subscribed instruments. It can be demonstrated thatMeerKAT can time over 1000 pulsars a day, and that for a modest increase in MeerTime’stime request a legacy dataset could be created that monitored most of the known pulsarpopulation. This would answer questions like, do “normal” pulsars also glitch? Are theperiod derivatives stable? What are the proper motions of pulsars? Are pulsar pulse profilesreally stable? If we monitor a thousand years of pulsar rotation history every year, whatunexpected phenomena will we see? • RRATs (Rotating RAdio Transients):
In 2010 the newly discovered population of “RRATs”was a topic of much interest. These objects appeared to be pulsars that only pulsed sporad-ically. Their birthrate appeared extremely high, yet little was known about the population.This population was the final science theme of the original proposal.The construction of the MeerKAT telescope is a great opportunity to advance radio pulsarscience. The vast majority of pulsars ( >
2. The MeerKAT as a pulsar telescope
Although individual pulses from a pulsar are irregular in their amplitude, shape and polari-sation, their mean profile is often remarkably stable. This fact permits extremely accurate pulsearrival times (ToAs) to be derived by integrating a pulsar’s pulses for many 100s-thousands of peri-ods and comparing the time-tagged profile against a standard. The error in a pulse ToA σ depends2 eerTime upon the time derivative of the profile and the signal-to-noise ratio snr . A good rule of thumb isthat the error in a ToA σ can be approximated by σ ∼ w / ( × snr ) (2.1)where w is the half width of the profile. In reality, this relation is only an approximation andassumes that every pulse is identical. A more correct approach would use the derivative of thepulse profile, and make some allowance for the fact that not every pulse is identical.The radiometer equation for a radio telescope describes the snr expected from a source of fluxdensity S as snr = SG p ( BN p t ) T rec + T sky r P − ww (2.2)where G is the gain of the telescope in K Jy − , B is the processed bandwidth (in Hz), N p is thenumber of orthogonal polarisations (maximum of 2), t is the integration time in seconds, T rec and T sky are the receiver and system temperature respectively (in K), P is the pulse period and w is thewidth of the pulse.The first determination of the MeerKAT’s technical specifications are extremely encouraging.The dishes are an offset gregorian and have an effective diameter of 13.965 m and an apertureefficiency of 0.71-0.81. If we adopt 0.76 as a mean efficiency this gives each dish a gain G of G = . π ( . / ) k = .
042 K Jy − (2.3)where k is Boltzmann’s constant. If added coherently, the 64-dish instrument has a total gain of G = G = . − .In array release 3, the telescope will possess a total usable bandwidth of 770 MHz of bandwidthin two orthogonal polarisations. Early tests of the receivers suggests that T rec will be less than 20K,possibly as low as ∼ µ s for just 15m, the expected snr ∼ T rec = ∼
3. The MeerKAT pulsar backend system
In the mid-late 1990s pulsar astronomers recognised the increases in timing precision thatcould be achieved by coherently dedispersing the voltages induced in the receiver of a single dishby the deconvolution of the signal with an appropriate filter. Coherent dedispersion used to repre-sent a major challenge to pulsar astronomy, requiring custom boards to digitise the voltages at therequisite rate (1/ B ) where B is the bandwidth of the backend and capture them in clusters of comput-ers. It was originally thought that a computing cluster might require a large cluster of workstations3 eerTime to achieve the necessary signal processing to coherently dedisperse the radio frequency signalsfrom MeerKAT, but since 2010 there have been several important developments in the creation ofcoherent dedispersion pulsar processors or “backends”. • The community has developed open-source software that has been largely adopted providingincreased rigour and testing. Most of the pulsar community now uses and contributes to thepsrchive suite of tools to process and manipulate folded pulsar profile data [6] . • A software library (psrdada) that both captures data from UDP streams and moves it tocomputer’s memory (RAM) is available from sourceforge.com that permits the quick de-velopment of capture engines. A library (dspsr) that transforms voltage data to coherently-dedispersed folded profiles is also available as an open-source software library [7] Theseopen-source libraries greatly facilitate the creation of pulsar processing backends. • Large-N Fourier transforms used to be extremely expensive to compute, requiring vast arraysof CPUs and multiplexing of the data because of the time taken to compute each one. Thetechnological breakthrough of the graphics processing unit (GPU) in consumer games cardshas reduced the cost of the necessary computations by more than an order of magnitude since2010. Almost all pulsar processors now utilise GPUs as the Fourier transform engine. • Back in 2010, 10-Gb ethernet represented the peak performance one could aspire to whentrying to perform lossless data capture. This meant that with 8-bit sampling, several O(4)computers would be required to capture a single stream from MeerKAT’s tied array beams.Now it is possible to have a single machine capture over 54 Gb/s of data without loss usingdual 40 Gb Network Interface Cards (NICs). • The use of interferometers for pulsar timing has been steadily increasing. The Westerborkarray has been joined by the LEAP project that ties the major European VLBI telescopesinto a single coherent beam, and the Very Large Array, LOFAR and the UTMOST project inAustralia are all examples of interferometers observing pulsars.MeerKAT’s pulsar timing hardware comprises two machines that were used to prototype thepulsar processor for the SKA at Swinburne University of Technology. These machines can co-herently dedisperse two parallel 850 MHz dual-polarisation streams simultaneously and one iscurrently at the MeerKAT site.Progress towards regular pulsar timing is continuing. A major breakthrough occurred in Q22016 when the first pulsar profile from the Vela pulsar (PSR J0835-4510) was produced from asingle beam with data written to disk. This observation confirmed that MeerKAT single disheswere producing very high quality data with system temperatures near the published specification(<20K). Shortly afterwards, the first tied array beam was created on the bright millisecond pulsarPSR J0437–4715 and processed in software from voltages recorded to disk. Although satellitetransmissions are present in the band, over 75% of it is “useable”. In October the first real-timepulsar profiles were produced validating that the pulsar processor can capture data at the requisiterate and process them. Currently the polyphase filterbank data cannot be correctly coherentlydedispersed because of digitally-induced artefacts in the polyphase filterbank frequency channels.4 eerTime
Coherent dedispersion will be required to achieve the ultimate timing precision promised by thesystem and should be possible when the beam-former and polyphase filterbanks are moved to theSKARAB boards in mid 2017.
High dispersion pulsars suffer from multi-path propagation effects that limit their use in pulsartiming experiments. The effects scale as the wavelength to the power 4.4, and hence some pulsarsdemand the use of the highest frequencies to perform the best science. Much beyond 3 GHz thereare relatively few pulsars that still retain enough flux to make this worthwhile, as many pulsarshave steep spectral indices (usually between –1 and –3).Professor Michael Kramer (MPIfR) has been leading a project to upgrade MeerKAT to operatebeyond 1.7 GHz with the deployment of 64 ×
4. The MeerTime Science Case
In MeerTime’s original 2010 science case it was anticipated that MeerKAT could indepen-dently detect a gravitational wave background after 5 years if the dimensional amplitude exceeded2 × − based upon limits in vogue at the time [9]. The current best millisecond pulsar for timingaccuracy (PSR J1909–3744) already suggests that an amplitude of this magnitude is ruled out[8]and that gravitational wave detection from pulsars will require international coordination and co-operation. MeerKAT can dramatically increase the pool of MSPs from which a gravitational wavebackground or individual binaries can be searched for with its unique combination of sensitivity,geographical location, ability to sub-array and the speed at which it can traverse the sky. Reardonet al. (2016) [10] recently reported on the timing of 20 MSPs from Parkes and based upon hisresiduals and the relative sensitivity of the two telescopes, the MeerKAT should increase the num-ber of pulsars with sub-us residuals from 5 to 16 objects if sensitivity was the only improvementfactor, but MeerKAT’s ability to subarray and seek out those MSPs that are experiencing scintilla-tion maxima gives us hope that it can do much better than a simple scaling of sensitivities mightsuggest.Several studies [11,12,13,14] have demonstrated that pulsar timing precision is ultimately lim-ited by the stochastic wideband impulse-modulated self-noise (SWIMS, also known as jitter andsingle-pulse variability) that is intrinsic to the pulsar emission. Consequently, optimal use of fullarray sensitivity requires the ability to divide it into sub-arrays; furthermore, because it is impera-tive to account for this noise in high-precision pulsar timing data analysis, the instrumentation for5 eerTime pulsar timing must be updated to produce additional statistical information. It has been demon-strated[12,13] that arrival time estimation bias can be mitigated by measuring the periodic correla-tions of the Stokes parameters and Shannon et al. (2014) [14] have described how jitter noise canbe characterised and incorporated in estimates of arrival time precision. Ongoing research by ourteam will combine these approaches using generalised least squares estimation to simultaneouslyreduce bias, accurately estimate uncertainty, and increase the sensitivity of experiments such aspulsar timing arrays.Gravitational waves are just one of the exciting science cases to be realised by timing an arrayof millisecond pulsars. Timing residuals also contain a wealth of information about the parametersof the parent binary, useful for studies of stellar evolution, the IGM and even our own planetaryephemerides. Since our original proposal the Fermi satellite has unveiled a tremendous populationof millisecond pulsars (now up to 350) waiting for an instrument capable of producing accuratearrival times to capitalise on them. MeerKAT is such an instrument. Pulsars are remarkable laboratories for the study of gravitation. In both the highly relativis-tic interior and the vicinity of a pulsar (and its binary companion, in case of double neutron-starsystems or potential pulsar-black hole system) space-time may significantly deviate from the pre-dictions of General Relativity (GR) [15]. Pulsar timing therefore provides a unique tool for probinggravity in the strong field regime, enabling high-precision tests of GR or other theories of gravity.Double-neutron-star systems such as the Double Pulsar [16] provide unrivalled probes for testingmost aspects of GR. Binary pulsars with a white dwarf companion and hence large mass dipole canset interesting constraints on alternative theories that predict, for instance, the existence of gravita-tional dipole radiation [17]. Resolving tight binary orbits to investigate effects such as the Shapirodelay requires short-spaced observations with high sensitivity. Meanwhile, identifying the weaksignatures of subtle relativistic effects needs long-term monitoring with good cadence. MeerKAT’sexcellent sensitivity (surpassing even our high expectations in 2010) and good frequency coveragewill make it the premier telescope for studying Southern-sky pulsars. This will not only improveexisting GR tests but will allow us to measure new effects to probe new physics. This is bestdemonstrated with the unique Double Pulsar, where the precision of tests of gravity will go beyondthe current best weak-field tests in the solar system. As has been shown (Kehl 2015, Masters the-sis, University of Bonn), we expect to measure the moment-of-inertia of the J0737-3039A in theDouble Pulsar for the first time, providing a handle on the equation-of-state of super-dense matter.With the sensitivity provided by MeerTime, we will also determine masses for both pulsarsand their companions. These can be used to test theories of binary evolution [18] and to investi-gate the distribution of neutron-star masses. In particular, the discovery of massive neutron stars[19,17] suggests that high-mass population of neutron stars exists, even possibly resulting frombirth [20,21]. As mass statistics improve, we will get closer to identifying the maximum masspossible for a neutron star, itself a constraint on the equation of state.MeerTime will furthermore provide astrometry (distances, proper motions and hence veloc-ities) for millisecond and binary pulsars, allowing us to infer their birth velocities and constrainasymmetric supernova kicks, particularly in double-neutron-star systems[22].6 eerTime
Globular clusters are treasure troves of exotic millisecond pulsars, for a recent scientific overviewsee reference [23]. The cores of globular clusters have stellar densities 10 - 10 times greater thanin the Galactic field; this promotes the formation of binary systems in which a neutron star canbe recycled to millisecond rotation rates via the transfer of matter and angular momentum froma Roche-lobe-filling companion. This extreme stellar density can also lead to exchange interac-tions, which create bizarre pulsar systems, unlike anything so-far seen in the Galactic field. Cur-rently there are 146 pulsars known in 28 globular clusters - including the fastest-spinning pulsarknown[24], exotic eccentric binaries suitable for neutron star mass measurements [25] and a uniquetriple system with a planetary companion [26]..Literally all of these 146 pulsars are visible to MeerKAT and MeerTime plans a sensitive, andcomprehensive globular cluster pulsar timing campaign. Combining MeerKAT timing data with upto three decades of archival measurements from GBT, Arecibo, and Parkes, MeerTime will probethe spin, orbital, and proper motions of these pulsars in unprecedented detail and measure previ-ously inaccessible system parameters that will allow us to probe accretion physics, dense matter,gravitational theories, and the evolution and properties of the clusters themselves in exquisite de-tail. From a practical point of view, timing globular cluster pulsars also provides a great efficencybecause in some cases (e.g. M28, 47 Tucanae and Terzan 5) dozens of millisecond pulsars canbe observed simultaneously. MeerKAT will revolutionize searches of southern globular clustersvia TRAPUM and the long-term timing of these and existing pulsars via MeerTime. To achievesensitivity to 10 µ Jy pulsars in these clusters, MeerTime plans typically 1-hr timing sessions forthese clusters.
Not all pulsars pulse regularly. Since the connection between pulsar radio emission and timingproperties in the so-called intermittent pulsars was first pointed out [27] more pulsars exhibitingthese properties have been discovered, as well as, e.g., multi-wavelength moding pulsars where thepulse profile changes significantly between two states having different radio and X-ray properties[28]. Secular variations in previously thought stable pulse profiles were also seen for a large sampleof ordinary pulsars [29] with a clear connection between timing irregularities/noise and emissionproperties/profiles. In this context, progress is being made on understanding pulsar interiors andhow/if the observed pulse profile variations could be ascribed to long term free precession. Finally,the LOFAR telescope is revealing the complex imprints of the ISM on pulsar data [30]. The Thou-sand Pulsar Timing Array will provide an opportunity to study the breadth of pulsar phenomenol-ogy. That will result in new breakthroughs relating to the interiors, to the magnetosphere and tothe environment of pulsars, to ISM and Galactic magnetic field studies, and will lead to improvedpulsar timing. MeerKAT is an exceptionally sensitive telescope for this purpose. We estimate that,with the current full MeerKAT sensitivity, we can obtain a high signal to noise (>20) profile of apulsar with a duty cycle of 5% and a flux density of 0.30 mJy (of which more than 1000 are visiblefrom Meerkat) within a minute of observation. From a sensitivity perspective therefore, it is easy toaccommodate regular observations of 1000 pulsars within the requested 16 h per observing epoch.7 eerTime
Young and energetic pulsars are often associated with supernova remnants (SNRs), pulsar windnebulae (PWNe), and/or high-energy X-ray/gamma-ray point sources [31]. Timing of young pul-sars provides their spin-down rate, which then sets the energy budget powering the PWN and otherhigh-energy emission [32]. Long-term timing provides the proper motion, which is a key ingredi-ent for deciphering the morphology of SNRs and PWNe (including bow shocks). In some cases, itis also possible to measure the neutron star’s braking index, revealing the multi-pole nature of themagnetic field and perhaps also its evolution in time [33]. Many young pulsars also show glitches,which probe the neutron star interior in a unique way [34]. With MeerKAT’s large sensitivity, itmay also be possible to probe the nebulae surrounding some young pulsars through precise char-acterization of their scattering/dispersion/rotation measure with time or other propagation effects[35].
To date, 4 of 23 known magnetars have been detected at radio wavelengths [36]. They sharesome characteristics distinct from those of the normal pulsar population, such as flat radio spectra.They are also extremely variable; 2 of the 4 are currently no longer radio emitters [37] althoughthis could change, and new ones could be discovered. Through frequent timing observations ofthe active radio magnetars we aim to obtain a continuous record of their torque, which illuminatesthe continued release of magnetic energy in the neutron star. Our broader aims are to develop abetter understanding of the dynamical behaviour of magnetar magnetospheres, and to establish theconditions under which radio emission takes place therein. Very frequent observations are neededbecause the torque on a magnetar can change by 10% on weekly timescales.
These objects are no longer being observed as part of our timing project. Their poor positionsare not well-mapped to the small tied array beam of MeerTime and they don’t appear to be anythingexcept an extension of nulling pulsars.
5. Observing strategy and data products
To detect the stochastic gravitational wave background or individual sources requires the high-est precision possible. This calls for regular observing cadence, preferably of order 20 times peryear. These observations will make MeerKAT a critical contributor to the broader InternationalPulsar Timing Array (IPTA) effort. MeerKAT has a unique opportunity to contribute to the directdetection of gravitational waves. It will be the most sensitive telescope in the Southern hemisphere,and as mentioned before its ability to sub-array means it can employ novel techniques to fully ex-ploit MSP scintillation. It will be possible for sub-arrays to be searching for MSPs to time that areat scintillation maxima, whilst the majority of the antennas are conducting routine timing.The best observing strategy for binary and millisecond pulsars is to observe entire orbits wherepractical (this prevents unfortunate covariances between binary and other parameters) and to haveoccasional “campaigns” when the cusps in Shapiro delays are visible. MeerTime’s immediate focus8 eerTime is on systems in which gravitational wave emission is implied by monitoring their orbital periodderivatives and on those binaries where pulsar masses can be achieved with MeerKAT’s increasedsensitivity over existing facilities. Currently, orbital decays have been detected for 8 binary pulsarsystems while detections for a further 4-6 (some unpublished pulsars) systems can be expected withMeerKAT. In addition, mass measurements of binary neutron stars should be possible for at least10 more systems [38]. Full orbits on these systems are possible with MeerKAT but not telescopeslike Arecibo or FAST.Most pulsars in globular clusters are very stable timers. As one of the aimed measurables is theproper motion of clusters, many sessions throughout the year are necessary. Since globular clusterMSPs are in average fainter sources than the Galactic ones, long integrations are a necessity.To enable science it is essential that the pulse profiles produced by the backend are createdin a format accessible via public domain packages such as psrchive. The pulsar processor willcreate FITS-format folded archives and coherent filterbanks at a nominal dispersion measure, alsoin FITS. The arrival times will be in reference to the observatory clock, which will ultimately bereferenced to UTC-NIST.The dimension of the scientific data products is manageable (60 TB over five years). Softwarepipelines will be made open access and available on the project website, as will clock correctionfiles.Once the data are calibrated they can be fit with the standard pulsar timing packages tempo,tempo2 and PINT.It is rare that a single epoch of pulsar timing results in a publishable outcome. Instead pulsarparameters slowly become scientifically interesting as the time span increases. Pulsar positionsrapidly increase their precision once a year of data is obtained, proper motions usually take afew years to manifest themselves in timing residuals and the discovery of the gravitational wavebackground is likely to take a decade or so. MeerTime’s data release policy is as follows: Datafrom the 1000 pulsar array will be available immediately. The MSP, relativistic binary and globularpulsar data will be released 18 months after they are recorded. We intend to publish all times ofarrival on something like a annual basis in scheduled “data releases”. All publications will providethe arrival times and raw data from the observations that led to any claimed results.
Acknowledgements
M Bailes acknowledges the Australian Research Council grants CE170100004 (OzGrav) andFL150100148. A. P. and M. Bu. acknowledge the support of the Italian Ministry of ForeignAffairs and International Cooperation, Directorate General for the Country Promotion (BilateralGrant Agreement ZA14GR02 - Mapping the Universe on the Pathway to SKA). This research issupported by the Max-Planck-Society and by the ERC Synergy Grant “BlackHoleCam: Imagingthe Event Horizon of Black Holes” (Grant 610058). Pulsar research at UBC is supported by anNSERC Discovery Grant and by the Canadian Institute for Advanced Research. This research issupported by NSF IRES Award 7706217.
References [1] A. Hewish, S. J. Bell, J. D. H. Pilkington, P. F. Scott and R. A. Collins,
Observation of a RapidlyPulsating Radio Source , Nature , 709 (February 1968). eerTime [2] D. R. Lorimer and M. Kramer, Handbook of Pulsar Astronomy December 2004.[3] R. A. Hulse and J. H. Taylor, Discovery of a pulsar in a binary system , ApJL , L51 (January 1975).[4] D. C. Backer, S. R. Kulkarni, C. Heiles, M. M. Davis and W. M. Goss,
A millisecond pulsar , Nature , 615 (December 1982).[5] R. N. Manchester, G. B. Hobbs, A. Teoh and M. Hobbs,
The Australia Telescope National FacilityPulsar Catalogue , AJ , 1993 (April 2005).[6] A. W. Hotan, W. van Straten and R. N. Manchester, PSRCHIVE and PSRFITS: An Open Approach toRadio Pulsar Data Storage and Analysis , PASA , 302 (2004).[7] W. van Straten and M. Bailes, DSPSR: Digital Signal Processing Software for Pulsar Astronomy , PASA , 1 (January 2011).[8] R. M. Shannon, V. Ravi, L. T. Lentati, P. D. Lasky, G. Hobbs, M. Kerr, R. N. Manchester, W. A.Coles, Y. Levin, M. Bailes, N. D. R. Bhat, S. Burke-Spolaor, S. Dai, M. J. Keith, S. Osłowski, D. J.Reardon, W. van Straten, L. Toomey, J.-B. Wang, L. Wen, J. S. B. Wyithe and X.-J. Zhu, Gravitational waves from binary supermassive black holes missing in pulsar observations , Science , 1522 (September 2015).[9] J. P. W. Verbiest, M. Bailes, W. A. Coles, G. B. Hobbs, W. van Straten, D. J. Champion, F. A. Jenet,R. N. Manchester, N. D. R. Bhat, J. M. Sarkissian, D. Yardley, S. Burke-Spolaor, A. W. Hotan andX. P. You,
Timing stability of millisecond pulsars and prospects for gravitational-wave detection , MNRAS , 951 (December 2009).[10] D. J. Reardon, G. Hobbs, W. Coles, Y. Levin, M. J. Keith, M. Bailes, N. D. R. Bhat, S. Burke-Spolaor,S. Dai, M. Kerr, P. D. Lasky, R. N. Manchester, S. Osłowski, V. Ravi, R. M. Shannon, W. van Straten,L. Toomey, J. Wang, L. Wen, X. P. You and X.-J. Zhu,
Timing analysis for 20 millisecond pulsars inthe Parkes Pulsar Timing Array , MNRAS , 1751 (January 2016).[11] K. Liu, J. P. W. Verbiest, M. Kramer, B. W. Stappers, W. van Straten and J. M. Cordes,
Prospects forhigh-precision pulsar timing , MNRAS , 2916 (November 2011).[12] S. Osłowski, W. van Straten, G. B. Hobbs, M. Bailes and P. Demorest,
High signal-to-noise ratioobservations and the ultimate limits of precision pulsar timing , MNRAS , 1258 (December 2011).[13] S. Osłowski, W. van Straten, P. Demorest and M. Bailes,
Improving the precision of pulsar timingthrough polarization statistics , MNRAS , 416 (March 2013).[14] R. M. Shannon, S. Osłowski, S. Dai, M. Bailes, G. Hobbs, R. N. Manchester, W. van Straten, C. A.Raithel, V. Ravi, L. Toomey, N. D. R. Bhat, S. Burke-Spolaor, W. A. Coles, M. J. Keith, M. Kerr,Y. Levin, J. M. Sarkissian, J.-B. Wang, L. Wen and X.-J. Zhu,
Limitations in timing precision due tosingle-pulse shape variability in millisecond pulsars , MNRAS , 1463 (September 2014).[15] T. Damour and G. Esposito-Farèse,
Tensor-scalar gravity and binary-pulsar experiments , PhysRevD , 1474 (July 1996).[16] M. Kramer, I. H. Stairs, R. N. Manchester, M. A. McLaughlin, A. G. Lyne, R. D. Ferdman,M. Burgay, D. R. Lorimer, A. Possenti, N. D’Amico, J. M. Sarkissian, G. B. Hobbs, J. E. Reynolds,P. C. C. Freire and F. Camilo, Tests of General Relativity from Timing the Double Pulsar , Science ,97 (October 2006). eerTime [17] J. Antoniadis, P. C. C. Freire, N. Wex, T. M. Tauris, R. S. Lynch, M. H. van Kerkwijk, M. Kramer,C. Bassa, V. S. Dhillon, T. Driebe, J. W. T. Hessels, V. M. Kaspi, V. I. Kondratiev, N. Langer, T. R.Marsh, M. A. McLaughlin, T. T. Pennucci, S. M. Ransom, I. H. Stairs, J. van Leeuwen, J. P. W.Verbiest and D. G. Whelan, A Massive Pulsar in a Compact Relativistic Binary , Science , p. 448(April 2013).[18] T. M. Tauris and G. J. Savonije,
Formation of millisecond pulsars. I. Evolution of low-mass X-raybinaries with Porb > , A&A , 928 (October 1999).[19] P. B. Demorest, T. Pennucci, S. M. Ransom, M. S. E. Roberts and J. W. T. Hessels,
A two-solar-massneutron star measured using Shapiro delay , Nature , 1081 (October 2010).[20] T. M. Tauris, N. Langer and M. Kramer,
Formation of millisecond pulsars with CO white dwarfcompanions - I. PSR J1614-2230: evidence for a neutron star born massive , MNRAS , 2130(September 2011).[21] J. Antoniadis, D. L. Kaplan, K. Stovall, P. C. C. Freire, J. S. Deneva, D. Koester, F. Jenet and J. G.Martinez,
An Eccentric Binary Millisecond Pulsar with a Helium White Dwarf Companion in theGalactic field , ApJ , p. 36 (October 2016).[22] I. H. Stairs, S. E. Thorsett, R. J. Dewey, M. Kramer and C. A. McPhee,
The formation of the doublepulsar PSR J0737-3039A/B , MNRAS , L50 (November 2006).[23] J. Hessels, A. Possenti, M. Bailes, C. Bassa, P. C. C. Freire, D. R. Lorimer, R. Lynch, S. M. Ransomand I. H. Stairs,
Pulsars in Globular Clusters with the SKA , Advancing Astrophysics with the SquareKilometre Array (AASKA14) , p. 47 (April 2015).[24] J. W. T. Hessels, S. M. Ransom, I. H. Stairs, P. C. C. Freire, V. M. Kaspi and F. Camilo,
A RadioPulsar Spinning at 716 Hz , Science , 1901 (March 2006).[25] S. M. Ransom, J. W. T. Hessels, I. H. Stairs, P. C. C. Freire, F. Camilo, V. M. Kaspi and D. L. Kaplan,
Twenty-One Millisecond Pulsars in Terzan 5 Using the Green Bank Telescope , Science , 892(February 2005).[26] S. Sigurdsson, H. B. Richer, B. M. Hansen, I. H. Stairs and S. E. Thorsett,
A Young White DwarfCompanion to Pulsar B1620-26: Evidence for Early Planet Formation , Science , 193 (July 2003).[27] M. Kramer, A. G. Lyne, J. T. O’Brien, C. A. Jordan and D. R. Lorimer,
A Periodically Active PulsarGiving Insight into Magnetospheric Physics , Science , 549 (April 2006).[28] W. Hermsen, J. W. T. Hessels, L. Kuiper, J. van Leeuwen, D. Mitra, J. de Plaa, J. M. Rankin, B. W.Stappers, G. A. E. Wright, R. Basu, A. Alexov, T. Coenen, J.-M. Grießmeier, T. E. Hassall,A. Karastergiou, E. Keane, V. I. Kondratiev, M. Kramer, M. Kuniyoshi, A. Noutsos, M. Serylak,M. Pilia, C. Sobey, P. Weltevrede, K. Zagkouris, A. Asgekar, I. M. Avruch, F. Batejat, M. E. Bell,M. R. Bell, M. J. Bentum, G. Bernardi, P. Best, L. Bîrzan, A. Bonafede, F. Breitling, J. Broderick,M. Brüggen, H. R. Butcher, B. Ciardi, S. Duscha, J. Eislöffel, H. Falcke, R. Fender, C. Ferrari,W. Frieswijk, M. A. Garrett, F. de Gasperin, E. de Geus, A. W. Gunst, G. Heald, M. Hoeft,A. Horneffer, M. Iacobelli, G. Kuper, P. Maat, G. Macario, S. Markoff, J. P. McKean, M. Mevius,J. C. A. Miller-Jones, R. Morganti, H. Munk, E. Orrú, H. Paas, M. Pandey-Pommier, V. N. Pandey,R. Pizzo, A. G. Polatidis, S. Rawlings, W. Reich, H. Röttgering, A. M. M. Scaife, A. Schoenmakers,A. Shulevski, J. Sluman, M. Steinmetz, M. Tagger, Y. Tang, C. Tasse, S. ter Veen, R. Vermeulen,R. H. van de Brink, R. J. van Weeren, R. A. M. J. Wijers, M. W. Wise, O. Wucknitz, S. Yatawatta andP. Zarka,
Synchronous X-ray and Radio Mode Switches: A Rapid Global Transformation of the PulsarMagnetosphere , Science , p. 436 (January 2013). eerTime [29] A. Lyne, G. Hobbs, M. Kramer, I. Stairs and B. Stappers, Switched Magnetospheric Regulation ofPulsar Spin-Down , Science , p. 408 (July 2010).[30] A. M. Archibald, V. I. Kondratiev, J. W. T. Hessels and D. R. Stinebring,
Millisecond PulsarScintillation Studies with LOFAR: Initial Results , ApJL , p. L22 (August 2014).[31] M. S. E. Roberts, J. W. T. Hessels, S. M. Ransom, V. M. Kaspi, P. C. C. Freire, F. Crawford and D. R.Lorimer,
PSR J2021+3651: A Young Radio Pulsar Coincident with an Unidentified EGRET γ -RaySource , ApJL , L19 (September 2002).[32] F. Camilo, R. N. Manchester, B. M. Gaensler and D. R. Lorimer,
Heartbeat of the Mouse: A YoungRadio Pulsar Associated with the Axisymmetric Nebula G359.23-0.82 , ApJL , L25 (November2002).[33] M. A. Livingstone, V. M. Kaspi, F. P. Gavriil and R. N. Manchester,
21 Years of Timing PSRB1509-58 , ApJ , 1046 (February 2005).[34] C. M. Espinoza, A. G. Lyne, B. W. Stappers and M. Kramer,
A study of 315 glitches in the rotation of102 pulsars , MNRAS , 1679 (June 2011).[35] A. G. Lyne, R. S. Pritchard and F. Graham-Smith,
Pulsar reflections within the Crab Nebula , MNRAS , 67 (February 2001).[36] F. Camilo, S. M. Ransom, J. P. Halpern, J. Reynolds, D. J. Helfand, N. Zimmerman and J. Sarkissian,
Transient pulsed radio emission from a magnetar , Nature , 892 (August 2006).[37] F. Camilo, S. M. Ransom, J. P. Halpern, J. A. J. Alford, I. Cognard, J. E. Reynolds, S. Johnston,J. Sarkissian and W. van Straten,
Radio Disappearance of the Magnetar XTE J1810-197 andContinued X-ray Timing , ApJ , p. 110 (April 2016).[38] F. Özel and P. Freire,
Masses, Radii, and the Equation of State of Neutron Stars , Annual Review ofAstronomy & Astrophysics , 401 (September 2016)., 401 (September 2016).