Parkes and Fermi Pulsar Science


In the coming years we can expect a revolution in our understanding of the high energy emission from pulsars. The Fermi satellite, launched in 2008, makes a deep survey of the gamma-ray sky with angular resolution, spectral resolution and sensitivity that is an order of magnitude better than any previous all-sky survey in gamma-rays. One very important goal of this mission is to discover many gamma-ray pulsars. Prior to the Fermi launch there are only seven pulsars that are known to emit gamma-rays, so our knowledge about the gamma-ray properties of pulsars is very limited. It is difficult to estimate how many pulsars will be discovered by Fermi, but it could be of the order of 150 or more.

During the first phase of the Fermi mission, the satellite will be surveying the whole sky on a regular basis. Because of the high spatial resolution of the LAT (Large Area Telescope) on board Fermi (30 arcseconds to 5arcminutes), we can tell if a photon comes from a known radio pulsar. This will be a huge step forward compared with previous missions, for which it is very difficult to prove any association with known sources. In principle, each gamma-ray photon can be tagged with a pulse phase and hence, over the lifetime of the mission, produce pulse profiles. However, this will only be possible when there is an accurate enough pulsar ephemeris that the rotational phase of the pulsar can be obtained with high enough precision. This ephemeris can only be derived from pulsar timing in the radio band.

The majority of the presently known pulsars that emit gamma-rays are young and energetic. These are also the pulsars that are known to be relatively unstable because of timing noise and glitches. This means that it is essential to time the pulsars that can be expected to be discovered by Fermi on a very regular basis to provide an ephemeris that is good enough to fold the gamma-ray data. This ephemeris should be accurate during the whole mission time, or else the sharp features that are often observed in gamma-ray profiles will be smeared out and valuable information will be lost.

The radio timing community have therefore undertaken to time a large number of pulsars in support of the Fermi mission. An initial list of pulsars was drawn up, largely consisting of objects with spin down energies greater than 10^34 erg/s. These were allocated to programs carried out at Parkes (Australia), Jodrell Bank (UK), Nancay (France), Green Bank (USA) and Arecibo (USA). The table of who is observing which pulsar can be found by clicking here.


A description of our observing setup can be found in Weltevrede et al. 2010, PASA, 27, 64. (Abstract from astro-ph) (Paper from ADS)

We started collecting data as part of the Fermi collaboration in February 2007 and since then have been observing every month. The basic idea behind the data taking is:

  • Minimum 2 minutes per pulsar
  • Maximum time set by need to reach s/n of ~5
  • Observe pulsed cal every 30 minutes
  • Use the DFBs to obtain full Stokes data. DFBs automatically fold and dedisperse the data. We set the sub-integration time to 30 seconds.
  • The majority of data are taken at 1300 MHz with 1024 bins per period and 1024 channels across 256 MHz bandwidth.
  • Every 6 months we collect data using the 10/50 cm receiver. At 3100 MHz, we use 1024 MHz of bandwidth. At 700 MHz, we use about 40 MHz of effective bandwidth

    Data reduction uses the publicly available psrchive package and the timing is done with tempo2. In brief this involves

  • Interference removal
  • Gain and Polarization calibration using the cal observations
  • Time and Frequency scrunching to produce a full Stokes profile
  • Deriving an arrival time (ToA)
  • Timing

    The complete database of all the observations can be found by following this link. The table is self-explanatory. It contains a listing of all the pulsars observed and the number of observations for each pulsar. Clicking on a pulsar name takes you to the pulsar page which contains a full listing of every observation. Individual calibrated pulse profiles can be seen by following the link, or the entire collection at wavelenghts of 50cm, 20cm and 10cm can be looked at.


    A full list of the publications from the collaboration can be found here.

    Data Access

    The standard ATNF Data Access policy states "Data obtained from the ATNF facilities is made available to other users on request 18 months after the date of observations, unless a special case for extended proprietary rights is accepted by the Director or Time Allocation Committee. The ATNF Director may override the release of data at his discretion.

    Currently the pulsar data from Parkes is available on-line through the data portal. Request for data from this project can be made by emailing Simon.Johnston AT If large amounts of data are requested, users will need to supply their own storage medium.

    We can provide the following data and data products and software

  • Original (raw) data as output from the DFB. The names of the files can be found via the database.
  • PSRCHIVE can be used to analyse these data. You can download PSRCHIVE from sourceforge.
  • TEMPO2 can be used for timing of these data. You can download TEMPO2 from sourceforge. See also the tempo2 pages at ATNF.
  • The timing solutions for all the pulsars valid up to August 2009 can be found on the Fermi website.

Staff space