Pulsar Timing Phenomenology an overview George Hobbs Australia

Pulsar Timing Phenomenology … an overview…. George Hobbs Australia Telescope National Facility

Sorry for being late!

Introduction • Timing residuals for ‘normal’ pulsars (based on Jodrell Bank Observatory data) • Timing residuals for millisecond pulsars (based on Parkes/Arecibo/GB/Nancay data) Summary: • Timing noise is wide-spread in pulsars • Timing noise in millisecond pulsars is similar in structure to that seen in young pulsars • The predicted amplitude of timing noise in some millisecond pulsars is low and provides hope for the detection of GWs

Pulsar timing: The basics Model for pulsar spin down Obtain pulse arrival times at observatory Improve timing model Form timing residuals – how good is the timing model at predicting the arrival times

Pulsar post-fit timing residuals Fit for rotational period and its derivative (quadratic term)

PSR B 1900+06

PSR B 1828 -11 Timing noise explained as free-precession due to periodicities and correlated pulse shape changes

Characterising the residuals (mag. dipole rad. ) ‘n’ ranges between – 2. 6 x 108 and 2. 5 x 108 46% of F 2 measurements are negative • Strong correlation between amplitude of “timing noise” and first and second derivative of rotational frequency (also age) • No correlation found so far with timescale of timing noise and any pulsar parameter. • On average the timing residuals show sharper local maxima than local minima 30 yr

Evolution of the characterisation 1 year 5 year 6 year 11 year 35 year

Red noise simulations

Disproven theories of timing noise • • Off-line software Observatories/receiver systems … Frequency-dependent noise Timing noise is not correlated with “height above the Galactic plane, luminosity or pulse shape changes” – Cordes & Helfand (1980)

The cause of these structures in the timing residuals • • • Unmodelled binary companions Clouds of particles Post-Newtonian orbital effects Free-precession of the neutron star Vortex creep Accretion onto surface Magnetospheric effects Irregularities in terrestrial time standards Inaccuracies in planetary ephemeris Effects of gravitational waves

How do you tell? • Expect (pseudo)-sinusoidal features for orbital/precessional effects • Expect glitch-like phenomena in vortex creep models • Expect particular power-spectrum for magnetospheric/phase noise/slowing-down noise • Expect particular correlations between pulsars for GWs/time or solar system inaccuracies • Theory provides expected amplitudes and timescales

Timing noise in the millisecond pulsars

PSRs B 1855+09 and B 1937+21 Jodrell versus Arecibo residuals for B 1937+21

Microglitch in B 1821 -24 (M 24) • Cognard, Backer (2004)

PSR B 1744 -24 A (Nice, Arzoumanian, Thorsett), Terzan 5 Discussed possibilities: Timing noise intrinsic to the pulsar (but many times larger than other millisecond pulsars) Changes in the viewing geometry of the emission region (precession) A “lumpy” disk around the binary system (precursor to planet formation) Torques on the pulsar due to infalling matter

J 0437 -4715 and 35 ns result van Straten (2001), Nature: arrival times averaged in 40 phase bins – rms residual of 35 ns Current result: 450 ns with 5 minute integrations

Some systematic effect at ~100 ns Instrumental effects? Pulsar instabilities? 0437 -4715 1909 -3715 thesis, Splaver (2004) - arecibo Hotan: Parkes observations

PSR J 1909 -3715 • 5 minute integrations, rms = 200 ns, 5 minute integrations

Predicted amplitudes for the recycled pulsars only Jodrell data 3 yr spans Backer (2005) – Aspen meeting 8 yr spans where the spin frequency and its second derivative are measured over a 108 s interval.

Thanks to K. J. Lee sigmaz at 10 yr

Conclusion • Timing irregularities seen in both normal and millisecond pulsars • “Amount” of timing noise correlated with Pdot (and age) • Jodrell Bank observatory contains an archive of ~400 pulsars with data spanning up to 35 years • Many theories of timing noise … how can we disprove some of these models?