The International Pulsar Timing Array project combines observations of pulsars from both Northern and Southern hemisphere observatories with the main aim of detecting ultra-low frequency (∼ 10 −9 − 10 −8 Hz) gravitational waves. Here we introduce the project, review the methods used to search for gravitational waves emitted from coalescing supermassive binary black-hole systems in the centres of merging galaxies and discuss the status of the project.

The highly stable spin of neutron stars can be exploited for a variety of (astro-)physical investigations. In particular arrays of pulsars with rotational periods of the order of milliseconds can be used to detect correlated signals such as those caused by gravitational waves. Three such "Pulsar Timing Arrays" (PTAs) have been set up around the world over the past decades and collectively form the "International" PTA (IPTA). In this paper, we describe the first joint analysis of the data from the three regional PTAs, i.e. of the first IPTA data set. We describe the available PTA data, the approach presently followed for its combination and suggest improvements for future PTA research. Particular attention is paid to subtle details (such as underestimation of measurement uncertainty and long-period noise) that have often been ignored but which become important in this unprecedentedly large and inhomogeneous data set. We identify and describe in detail several factors that complicate IPTA research and provide recommendations for future pulsar timing efforts. The first IPTA data release presented here (and available online) is used to demonstrate the IPTA's potential of improving upon gravitational-wave limits placed by individual PTAs by a factor of ∼ 2 and provides a 2 − σ limit on the dimensionless amplitude of a stochastic GWB of 1.7 × 10 −15 at a frequency of 1 yr −1 . This is 1.7 times less constraining than the limit placed by , due mostly to the more recent, high-quality data they used. c 2015 RAS c 2015 RAS, MNRAS 000, 1-25 First IPTA Data Release 3 σJitter ∝ fJW eff 1 + m 2 I Np ,with fJ the jitter parameter, which needs to be determined experimentally (Liu et al. 2012;Shannon et al. 2014); W eff the pulse width; mI = σE/µE the modulation index, defined by the mean (µE) and standard deviation (σE) of the pulseenergy distribution; and Np = tint/P the number of pulses in the observation, which equals the total observing time divided by the pulse period. Consequently, the highest-precision timing efforts ideally require rapidly rotating pulsars (P 0.03 s) with high relatively flux densities (S1.4 GHz 0.5 mJy) and narrow pulses (δ 20%) are observed at sensitive (A eff /Tsys) telescopes with wide-bandwidth receivers (∆f ) and for long integration times (tint 30 min).

The analysis of high-precision timing observations of an array of ∼20 millisecond pulsars (a so-called 'timing array') may ultimately result in the detection of a stochastic gravitationalwave background. The feasibility of such a detection and the required duration of this type of experiment are determined by the achievable rms of the timing residuals and the timing stability of the pulsars involved. We present results of the first long-term, high-precision timing campaign on a large sample of millisecond pulsars used in gravitational-wave detection projects. We show that the timing residuals of most pulsars in our sample do not contain significant low-frequency noise that could limit the use of these pulsars for decade-long gravitational-wave detection efforts. For our most precisely timed pulsars, intrinsic instabilities of the pulsars or the observing system are shown to contribute to timing irregularities on a 5-year time-scale below the 100 ns level. Based on those results, realistic sensitivity curves for planned and ongoing timing array efforts are determined. We conclude that prospects for detection of a gravitational-wave background through pulsar timing array efforts within 5 years to a decade are good.

We present timing models for 20 millisecond pulsars in the Parkes Pulsar Timing Array. The precision of the parameter measurements in these models has been improved over earlier results by using longer data sets and modelling the non-stationary noise. We describe a new noise modelling procedure and demonstrate its effectiveness using simulated data. Our methodology includes the addition of annual dispersion measure (DM) variations to the timing models of some pulsars. We present the first significant parallax measurements for PSRs J1024−0719, J1045−4509, J1600−3053, J1603−7202, and J1730−2304, as well as the first significant measurements of some post-Keplerian orbital parameters in six binary pulsars, caused by kinematic effects. Improved Shapiro delay measurements have resulted in much improved pulsar mass measurements, particularly for PSRs J0437−4715 and J1909−3744 with M p = 1.44 ± 0.07 M and M p = 1.47 ± 0.03 M respectively. The improved orbital period-derivative measurement for PSR J0437−4715 results in a derived distance measurement at the 0.16% level of precision, D = 156.79 ± 0.25 pc, one of the most fractionally precise distance measurements of any star to date.

A 'pulsar timing array' (PTA), in which observations of a large sample of pulsars spread across the celestial sphere are combined, allows investigation of 'global' phenomena such as a background of gravitational waves or instabilities in atomic timescales that produce correlated timing residuals in the pulsars of the array. The Parkes Pulsar Timing Array (PPTA) is an implementation of the PTA concept based on observations with the Parkes 64-m radio telescope. A sample of 20 ms pulsars is being observed at three radio-frequency bands, 50 cm (ß700 MHz), 20 cm (ß1400 MHz), and 10 cm (ß3100 MHz), with observations at intervals of two to three weeks. Regular observations commenced in early 2005. This paper describes the systems used for the PPTA observations and data processing, including calibration and timing analysis. The strategy behind the choice of pulsars, observing parameters, and analysis methods is discussed. Results are presented for PPTA data in the three bands taken between 2005 March and 2011 March. For 10 of the 20 pulsars, rms timing residuals are less than 1 μs for the best band after fitting for pulse frequency and its first time derivative. Significant 'red' timing noise is detected in about half of the sample. We discuss the implications of these results on future projects including the International Pulsar Timing Array and a PTA based on the Square Kilometre Array. We also present an 'extended PPTA' data set that combines PPTA data with earlier Parkes timing data for these pulsars.