Parkes Multibeam Pulsar Survey Research Articles

Discovery of 37 new pulsars through GPU-accelerated reprocessing of archival data of the Parkes multibeam pulsar survey

ABSTRACT We present the discovery of 37 pulsars from ∼ 20 yr old archival data of the Parkes Multibeam Pulsar Survey using a new FFT-based search pipeline optimized for discovering narrow-duty cycle pulsars. When developing our pulsar search pipeline, we noticed that the signal-to-noise ratios of folded and optimized pulsars often exceeded that achieved in the spectral domain by a factor of two or greater, in particular for narrow duty cycle ones. Based on simulations, we verified that this is a feature of search codes that sum harmonics incoherently and found that many promising pulsar candidates are revealed when hundreds of candidates per beam even with modest spectral signal-to-noise ratios of S/N∼5–6 in higher-harmonic folds (up to 32 harmonics) are folded. Of these candidates, 37 were confirmed as new pulsars and a further 37 would have been new discoveries if our search strategies had been used at the time of their initial analysis. While 19 of these newly discovered pulsars have also been independently discovered in more recent pulsar surveys, 18 are exclusive to only the Parkes Multibeam Pulsar Survey data. Some of the notable discoveries include: PSRs J1635−47 and J1739−31, which show pronounced high-frequency emission; PSRs J1655−40 and J1843−08 belong to the nulling/intermittent class of pulsars; and PSR J1636−51 is an interesting binary system in a ∼0.75 d orbit and shows hints of eclipsing behaviour – unusual given the 340 ms rotation period of the pulsar. Our results highlight the importance of reprocessing archival pulsar surveys and using refined search techniques to increase the normal pulsar population.

A GPU based single-pulse search pipeline (GSP) with database and its application to the Commensal Radio Astronomy FAST Survey (CRAFTS)

We developed a GPU based single-pulse search pipeline (GSP) with a candidate-archiving database. Largely based upon the infrastructure of the open source PulsaR Exploration and Search Toolkit (PRESTO), GSP implements GPU acceleration of the de-dispersion and integrates a candidate-archiving database. We applied GSP to the data streams from the Commensal Radio Astronomy FAST Survey (CRAFTS), which resulted in quasi-real-time processing. The integrated candidate database facilitates synergistic usage of multiple machine-learning tools and thus improves efficient identification of radio pulsars such as rotating radio transients (RRATs) and fast radio bursts (FRBs). We first tested GSP on pilot CRAFTS observations with the FAST Ultra-Wide Band (UWB) receiver. GSP detected all pulsars known from the the Parkes multibeam pulsar survey in the corresponding sky area covered by the FAST-UWB. GSP also discovered 13 new pulsars. We measured the computational efficiency of GSP to be ∼120 times faster than the original PRESTO and ∼60 times faster than an MPI-parallelized version of PRESTO.

A study of single pulses in the Parkes Multibeam Pulsar Survey

We reprocessed the Parkes Multibeam Pulsar Survey, searching for single pulses out to a DM of 5000 pc cm$^{-3}$ with widths of up to one second. We recorded single pulses from 264 known pulsars and 14 Rotating Radio Transients. We produced amplitude distributions for each pulsar which we fit with log-normal distributions, power-law tails, and a power-law function divided by an exponential function, finding that some pulsars show a deviation from a log-normal distribution in the form of an excess of high-energy pulses. We found that a function consisting of a power-law divided by an exponential fit the distributions of most pulsars better than either log-normal or power-law functions. For pulsars that were detected in a periodicity search, we computed the ratio of their single-pulse signal-to-noise ratios to their signal-to-noise ratios from a Fourier transform and looked for correlations between this ratio and physical parameters of the pulsars. The only correlation found is the expected relationship between this ratio and the spin period. Fitting log-normal distributions to the amplitudes of pulses from RRATs showed similar behaviour for most RRATs. Here, however, there seem to be two distinct distributions of pulses, with the lower-energy distribution being consistent with noise. Pulse-energy distributions for two of the RRATS processed were consistent with those found for normal pulsars, suggesting that pulsars and RRATs have a common emission mechanism, but other factors influence the specific emission properties of each source class.

ARECIBO PULSAR SURVEY USING ALFA. III. PRECURSOR SURVEY AND POPULATION SYNTHESIS

The Pulsar Arecibo L-band Feed Array (PALFA) Survey uses the ALFA 7-beam receiver to search both inner and outer Galactic sectors visible from Arecibo (32° ≲ ℓ ≲ 77° and 168° ≲ ℓ ≲ 214°) close to the Galactic plane (|b| ≲ 5°) for pulsars. The PALFA survey is sensitive to sources fainter and more distant than have previously been seen because of Arecibo's unrivaled sensitivity. In this paper we detail a precursor survey of this region with PALFA, which observed a subset of the full region (slightly more restrictive in ℓ and |b| ≲ 1°) and detected 45 pulsars. Detections included 1 known millisecond pulsar and 11 previously unknown, long-period pulsars. In the surveyed part of the sky that overlaps with the Parkes Multibeam Pulsar Survey (36° ≲ ℓ ≲ 50°), PALFA is probing deeper than the Parkes survey, with four discoveries in this region. For both Galactic millisecond and normal pulsar populations, we compare the survey's detections with simulations to model these populations and, in particular, to estimate the number of observable pulsars in the Galaxy. We place 95% confidence intervals of 82,000 to 143,000 on the number of detectable normal pulsars and 9000 to 100,000 on the number of detectable millisecond pulsars in the Galactic disk. These are consistent with previous estimates. Given the most likely population size in each case (107,000 and 15,000 for normal and millisecond pulsars, respectively), we extend survey detection simulations to predict that, when complete, the full PALFA survey should have detected normal pulsars and millisecond pulsars. Identical estimation techniques predict that normal pulsars and millisecond pulsars would be detected by the beginning of 2014; at the time, the PALFA survey had detected 283 normal pulsars and 31 millisecond pulsars, respectively. We attribute the deficiency in normal pulsar detections predominantly to the radio frequency interference environment at Arecibo and perhaps also scintillation—both effects that are currently not accounted for in population simulation models.

EINSTEIN@HOMEDISCOVERY OF 24 PULSARS IN THE PARKES MULTI-BEAM PULSAR SURVEY

We have conducted a new search for radio pulsars in compact binary systems in the Parkes multi-beam pulsar survey (PMPS) data, employing novel methods to remove the Doppler modulation from binary motion. This has yielded unparalleled sensitivity to pulsars in compact binaries. The required computation time of approximately 17000 CPU core years was provided by the distributed volunteer computing project Einstein@Home, which has a sustained computing power of about 1 PFlop/s. We discovered 24 new pulsars in our search, of which 18 were isolated pulsars, and six were members of binary systems. Despite the wide filterbank channels and relatively slow sampling time of the PMPS data, we found pulsars with very large ratios of dispersion measure (DM) to spin period. Among those is PSR J1748-3009, the millisecond pulsar with the highest known DM (approximately 420 pc/cc). We also discovered PSR J1840-0643, which is in a binary system with an orbital period of 937 days, the fourth largest known. The new pulsar J1750-2536 likely belongs to the rare class of intermediate-mass binary pulsars. Three of the isolated pulsars show long-term nulling or intermittency in their emission, further increasing this growing family. Our discoveries demonstrate the value of distributed volunteer computing for data-driven astronomy and the importance of applying new analysis methods to extensively searched data.

A coherent acceleration search of the Parkes multibeam pulsar survey – techniques and the discovery and timing of 16 pulsars

A fully coherent acceleration search algorithm has been applied to the Parkes multibeam pulsar survey of the Galactic plane to search for previously undiscovered relativistic binary pulsars. The search has resulted in the discovery of 16 pulsars including a binary millisecond pulsar and an intermittent pulsar. Despite a number of promising candidates there have been no new discoveries of relativistic binary pulsars. Here, we detail the acceleration search performed in our analysis and present coherent timing solutions for each of the pulsars discovered. In light of the lack of discoveries of relativistic binary pulsars, we also discuss the technique of acceleration searching and its effectiveness in finding these systems.

DISCOVERY OF FIVE NEW PULSARS IN ARCHIVAL DATA

Reprocessing of the Parkes Multibeam Pulsar Survey has resulted in the discovery of five previously unknown pulsars and several as-yet-unconfirmed candidates. PSR J0922−52 has a period of 9.68 ms and a dispersion measure (DM) of 122.4 pc cm−3. PSR J1147−66 has a period of 3.72 ms and a DM of 133.8 pc cm−3. PSR J1227−6208 has a period of 34.53 ms, a DM of 362.6 pc cm−3, is in a 6.7 day binary orbit, and was independently detected in an ongoing high-resolution Parkes survey by Thornton et al. and also in independent processing by Einstein@Home volunteers. PSR J1546−59 has a period of 7.80 ms and a DM of 168.3 pc cm−3. PSR J1725−3853 is an isolated 4.79 ms pulsar with a DM of 158.2 pc cm−3. These pulsars were likely missed in earlier processing efforts due to the fact that they have both high DMs and short periods, and also due to the large number of candidates that needed to be looked through. These discoveries suggest that further pulsars are awaiting discovery in the multibeam survey data.

RADIO AND X-RAY OBSERVATIONS OF THE INTERMITTENT PULSAR J1832+0029

We report on radio and X-ray observations of PSR 1832+0029, a 533-ms radio pulsar discovered in the Parkes Multibeam Pulsar Survey. From radio observations taken with the Parkes, Lovell and Arecibo telescopes, we show that this pulsar exhibits two spindown states akin to PSRs B1931+24 reported by Kramer et al. and J1841-0500 reported by Camilo et al. Unlike PSR B1931+24, which switches between "on" and "off" states on a 30--40 day time-scale, PSR 1832+0029 is similar to PSR J1841-0500 in that it spends a much longer period of time in the off-state. So far, we have fully sampled two off-states. The first one lasted between 560 and 640 days and the second one lasted between 810 and 835 days. From our radio timing observations, the ratio of on/off spindown rates is $1.77 \pm 0.03$. Chandra observations carried out during both the on- and off-states of this pulsar failed to detect any emission. Our results challenge but do not rule out models involving accretion onto the neutron star from a low-mass stellar companion. In spite of the small number of intermittent pulsars currently known, difficulties in discovering them and in quantifying their behavior imply that their total population could be substantial.

On the origin of a highly dispersed coherent radio burst

Abstract We discuss the possible source of a highly dispersed radio transient discovered in the Parkes Multibeam Pulsar Survey (PMPS). The pulse has a dispersion measure of 746 cm−3 pc, a peak flux density of 400 mJy for the observed pulse width of 7.8 ms and a flat spectrum across a 288-MHz band centred on 1374 MHz. The flat spectrum suggests that the pulse did not originate from a pulsar, but is consistent with radio-emitting magnetar spectra. The non-detection of subsequent bursts constrains any possible pulsar period to ≳1 s, and the pulse energy distribution to being much flatter than typical giant pulse emitting pulsars. The burst is also consistent with the radio signal theorized from an annihilating mini black hole. Extrapolating the PMPS detection rate provides a limit of ΩBH ≲ 5 × 10−14 on the density of these objects. We investigate the consistency of these two scenarios, plus several other possible solutions, as potential explanations to the origin of the pulse, as well as for another transient with similar properties: the Lorimer burst.

A search for dispersed radio bursts in archival Parkes Multibeam Pulsar Survey data

A number of different classes of potentially extra-terrestrial bursts of radio emission have been observed in surveys with the Parkes 64m radio telescope, including "Rotating Radio Transients", the "Lorimer burst" and "perytons". Rotating Radio Transients are radio pulsars which are best detectable in single-pulse searches. The Lorimer burst is a highly dispersed isolated radio burst with properties suggestive of extragalactic origin. Perytons share the frequency-swept nature of the Rotating Radio Transients and Lorimer burst, but unlike these events appear in all thirteen beams of the Parkes Multibeam receiver and are probably a form of peculiar radio frequency interference. In order to constrain these and other radio source populations further, we searched the archival Parkes Multibeam Pulsar Survey data for events similar to any of these. We did not find any new Rotating Radio Transients or bursts like the Lorimer burst. We did, however, discover four peryton-like events. Similar to the perytons, these four bursts are highly dispersed, detected in all thirteen beams of the Parkes multibeam receiver, and have pulse widths between 20--30 ms. Unlike perytons, these bursts are not associated with atmospheric events like rain or lightning. These facts may indicate that lightning was not responsible for the peryton phenomenon. Moreover, the lack of highly dispersed celestial signals is the evidence that the Lorimer burst is unlikely to belong to a cosmological source population.

Radio properties of rotating radio transients - I. Searches for periodicities and randomness in pulse arrival times

We have analysed the long- and short-term time dependence of the pulse arrival times and the pulse detection rates for eight rotating radio transient (RRAT) sources from the Parkes Multibeam Pulsar Survey. We find significant periodicities in the individual pulse arrival times from six RRATs. These periodicities range from ∼30 min to 2100 d, and from one to 16 independent (i.e. non-harmonically related) periodicities are detected for each RRAT. In addition, we find that pulse emission is a random (i.e. Poisson) process on short (hour-long) time-scales but that most of the objects exhibit longer term (months to years) non-random behaviour. We find that PSR J1819−1458 and PSR J1317−5759 emit more doublets (two consecutive pulses) and triplets (three consecutive pulses) than is expected in random pulse distributions. No evidence for such an excess is found for the other RRATs. There are several different models for RRAT emission depending on both extrinsic and intrinsic factors which are consistent with these properties.

Rotating Radio Transients: new discoveries, timing solutions and musings

We describe observations of Rotating RAdio Transients (RRATs) that were discovered in a re-analysis of the Parkes Multi-beam Pulsar Survey (PMPS). The sources have now been monitored for sufficiently long to obtain seven new coherent timing solutions, to make a total of 14 now known. Furthermore we announce the discovery of 7 new transient sources, one of which may be extragalactic in origin (with $z\sim0.1$) and would then be a second example of the so-called `Lorimer burst'. The timing solutions allow us to infer neutron star characteristics such as energy-loss rate, magnetic field strength and evolutionary timescales, as well as facilitating multi-wavelength followup by providing accurate astrometry. All of this enables us to consider the question of whether or not RRATs are in any way special, i.e. a distinct and separate population of neutron stars, as has been previously suggested. We see no reason to consider 'RRAT' as anything other than a detection label, the subject of a selection effect in the parameter space searched. However, single-pulse searches can be utilised to great effect to identify pulsars difficult, or impossible, to find by other means, in particular those with long-periods (half of the PMPS RRATs have periods greater than 4 seconds), high-magnetic field strengths ($B\gtrsim 10^{13}$ G) and pulsars approaching the 'death valley'. The detailed nulling properties of such pulsars are unknown but the mounting evidence suggests a broad range of behaviour in the pulsar population. The group of RRATs fit in to the picture where pulsar magnetospheres switch between stable configurations.

Selection of radio pulsar candidates using artificial neural networks

Radio pulsar surveys are producing many more pulsar candidates than can be inspected by human experts in a practical length of time. Here we present a technique to automatically identify credible pulsar candidates from pulsar surveys using an artificial neural network. The technique has been applied to candidates from a recent re-analysis of the Parkes multi-beam pulsar survey resulting in the discovery of a previously unidentified pulsar.

Further searches for Rotating Radio Transients in the Parkes Multi-beam Pulsar Survey

We describe the steps involved in performing searches for sources of transient radio emission such as Rotating Radio Transients (RRATs), and present 10 new transient radio sources discovered in a re-analysis of the Parkes Multi-beam Pulsar Survey. Followup observations of each new source as well as one previously known source are also presented. The new sources suggest that the population of transient radio-emitting neutron stars, and hence the neutron star population in general, may be even larger than initially predicted. We highlight the importance of radio frequency interference excision for single-pulse searches. Also, we discuss some interesting properties of individual sources and consider the difficulties involved in precisely defining a RRAT and determining where they fit in with the other known classes of neutron stars.

Timing observations of rotating radio transients

We present radio timing measurements of six rotating radio transient (RRAT) sources discovered in the Parkes Multibeam Pulsar Survey. These provide four new phase-connected timing solutions and two updated ones, making a total of seven of the original 11 reported RRATs now with high-precision rotational and astrometric parameters. Three of these seven RRATs have magnetic fields greater than 10^13 G, with spin-down properties similar to those of the magnetars and X-ray detected isolated neutron stars. Another two of these RRATs have long periods and large characteristic ages, and lie near the `death-line' for radio pulsar emission. The remaining two RRATs with timing solutions have properties typical of the bulk of the pulsar population. The new solutions offer insights into what might be responsible for the unusual emission properties. We demonstrate that the RRATs have significantly longer periods and higher magnetic fields than normal radio pulsars, and find no correlation with other spin-down parameters. These solutions also provide precise positions, which will facilitate follow-up studies at high energies, crucial for relating these sources with other neutron star populations.

An interference removal technique for radio pulsar searches

Searches for radio pulsars are becoming increasingly difficult because of a rise in impulsive man-made terrestrial radio-frequency interference. Here, we present a new technique, zero dispersion measure filtering, which can significantly reduce the effects of such signals in pulsar search data. The technique has already been applied to a small portion of the data from the Parkes multi-beam pulsar survey, resulting in the discovery of four new pulsars, so illustrating its efficacy.

Discovery of 28 pulsars using new techniques for sorting pulsar candidates

Modern pulsar surveys produce many millions of candidate pulsars, far more than can be individually inspected. Traditional methods for filtering these candidates, based upon the signal-to-noise ratio of the detection, cannot easily distinguish between interference signals and pulsars. We have developed a new method of scoring candidates using a series of heuristics which test for pulsar-like properties of the signal. This significantly increases the sensitivity to weak pulsars and pulsars with periods close to interference signals. By applying this and other techniques for ranking candidates from a previous processing of the Parkes Multi-beam Pulsar Survey, 28 previously unknown pulsars have been discovered. These include an eccentric binary system and a young pulsar which is spatially coincident with a known supernova remnant.

PSR J1753−2240: a mildly recycled pulsar in an eccentric binary system

We report the discovery of PSR J1753−2240 in the Parkes Multibeam Pulsar Survey data base. This 95-ms pulsar is in an eccentric binary system with a 13.6-d orbital period. Period derivative measurements imply a characteristic age in excess of 1 Gyr, suggesting that the pulsar has undergone an episode of accretion-induced spin-up. The eccentricity and spin period are indicative of the companion being a second neutron star, so that the system is similar to that of PSR J1811−1736, although other companion types cannot be ruled out at this time. The companion mass is constrained by geometry to lie above 0.48 solar masses, although long-term timing observations will give additional constraints. If the companion is a white dwarf or a main-sequence star, optical observations may yield a direct detection of the companion. If the system is indeed one of the few known double neutron star systems, it would lie significantly far from the recently proposed spin-period/eccentricity relationship.

Timing observations of Rotating Radio Transient J1819−1458 at Urumqi observatory

We report the timing-analysis results obtained for Rotating Radio Transient (RRAT) J1819−1458 from regular timing observations at 1.54 GHz using the Urumqi 25-m radio telescope between 2007 April and 2008 March. RRAT J1819−1458 is a relatively young and highly magnetized neutron star discovered by its sporadic short bursts in the Parkes Multibeam Pulsar Survey data. In 94 h of observation data, we detected a total of 162 dispersed bursts of RRAT J1819−1458 with the signal-to-noise ratios (S/Ns) above 5 σ threshold. Among them, five bursts clearly show two-component structure. The S/N of the strongest burst is 13.3. The source's DM measured through our data is 196.0 ± 0.4 pc cm−3. The timing position, frequency and its first derivative were determined using standard pulsar timing techniques on the arrival times of these individual bursts. The accuracy of the solved rotating parameters is improved comparing with that in previous publication. Our timing position with 2σ error is consistent with the position of its X-ray counterpart CXOU J181934.1−145804. The effect of timing noise and the phase fluctuation of the individual short bursts on the timing residuals is briefly discussed. The distribution of the timing residuals is bimodal, which cannot be explained readily by timing irregularity.

Discovery of Three Wide‐Orbit Binary Pulsars: Implications for Binary Evolution and Equivalence Principles

We report the discovery of three binary millisecond pulsars during the Parkes Multibeam Pulsar Survey of the Galactic plane. The objects are highly recycled and are in orbits of many tens of days about low-mass white dwarf companions. The eccentricity of one object, PSR J1853+1303, is more than an order of magnitude lower than predicted by the theory of convective fluctuations during tidal circularization. We demonstrate that under the assumption that the systems are randomly oriented, current theoretical models of the core-mass-orbital-period relation for the progenitors of these systems likely overestimate the white dwarf masses, strengthening previous concerns about the match of these models to the data. The new objects allow us to update the limits on the violation of relativistic equivalence principles to 95% confidence upper limits of 5.6 ? 10-3 for the strong equivalence principle parameter |?| and 4.0 ? 10-20 for the Lorentz-invariance/momentum-conservation parameter |3|.