New results from an old friend - the Vela pulsar

There are many known pulsars, currently around 1,700 have been identified, and this multitude has made many contributions to our understanding of their average properties. There are, however, a few that have provided the bulk of the insights into the structure and environment of pulsars. Vela could well claim pre-eminence amongst these. The Vela pulsar was the first discovered in the southern hemisphere (Large et al., 1968) and in an exciting rush of papers was shown to be slowing down, having a polarization sweep across the pulse and to glitch (Radhakrishnan et al., 1969; Radhakrishnan and Manchester, 1969) confirming the rotating solid body as suggested by Gold (1968), and most importantly, identifying the emission regions as the magnetic poles (Radhakrishnan and Cooke, 1969), as opposed to the pulsating-star model.

This old friend has recently provided us with even more information, with observations from a selection of Australian, and orbital, telescopes. This current series of observations was sparked by the dedicated pulsar telescope operated by the University of Tasmania for over twenty years now. Following the last glitch (January 2000, Dodson et al., 2002) Chandra produced a stunning high-resolution image of the local X-ray emission (Pavlov et al., 2000; Helfand et al., 2001) showing a double cross-bow structure that many authors have modelled (Helfand et al., 2001; Radhakrishnan and Deshpande, 2001). Whilst the Vela X nebula had been imaged in great detail at 20 cm by Bock et al. (1998; 2002) and the VLA had imaged the region around the pulsar at 6 cm (Bietenholz et al., 1991) no arcsecond observations close to Vela had been undertaken at the ATCA. Therefore we observed at 6 cm and 3 cm using the pulsar-binning mode, with the pulsar as our phase reference.

With improved low surface-brightness sensitivity, higher frequency and longer integration times, we uncovered the radio shell around the X-ray structure as shown in Figure 1. The X-ray emission (data for which was downloaded from the Chandra archives) in grey-scale is overlaid with the 5-GHz radio contours. This highly polarized emission has a bright northern lobe (as seen by Bietenholz et al., 1991) and a weaker, more diffuse southern lobe.

Figure 1: The Chandra observation of the Vela pulsar-wind nebula (grey-scale) and the 5-GHz radio contours (-1,1,2,3,4,5 mJy/beam with 20-arcsecond beam). The de-rotated magnetic field lines are overlaid, with a 1-mJy bar at the bottom left, below the restoring-beam size. The proper motion vector shows the distance travelled in 1000 years, and ends with the 3-sigma error ellipse.

With the ATCA high-quality polarization measurements were possible. These provide the best evidence that the two lobes are related, as the de-rotated polarization angles are smooth and symmetrical around the pulsar. More details can be found in Dodson et al., (2003b).

In the 1.4-GHz observations of Bock et al. (1998) these lobes are visible, as are many other features in the region. However the polarized emission shows a similar termination to the south. We have shown that this is not due to RM depolarization, and appears to be a genuine termination, as opposed to those associated with the filaments (Bock et al., 2002).

It is worth noting that this is not a very luminous object, possibly explaining the small numbers of pulsar-wind nebulae found in various searches (e.g. Stappers et al., 1999).

Finally the LBA observations of the Vela parallax have been completed (Legge, 2002; Dodson et al., 2003a). The proper motion agrees with that of Bailes et al. (1989) and the parallax with that of Caraveo et al. (2001), with significantly reduced errors. The proper motion vector is shown overlaid in Figure 1, with the extent representing the distance travelled in the last 1000 years. The interest lies in the alignment of the projection of the proper-motion vector with the spin axis deduced from X-ray nebula. This is of great theoretical significance as several authors (e.g. Spruit and Phinney, 1998) predict just such an alignment if the initial impulse of the supernova was averaged over many rotations of the progenitor star. The alignment is very good, but not perfect, allowing the estimation of the duration of the supernova impulse.

Further LBA observations of several other pulsars, using the pulsar-binning mode, are being made. These, we hope, will be able to tell us as much as our all time favourite astronomical object.

References

Bailes, M., Manchester, R. N., Kesteven, M. J., Norris, R. P., and Reynolds, J., 1989, ApJL 343, L53
Bietenholz, M. F., Frail, D. A., and Hankins, T. H., 1991, ApJL 376, L41
Bock, D. C.-J., Frail, D. A., Sault, R. J., Green, A. J., and Milne, D. K., 1998, Memorie della Societa Astronomica Italiana 69, 919
Bock, D. C.-J., Sault, R. J., Milne, D. K., and Green, A. J., 2002, in "Neutron Stars in Supernova Remnants" (ASP Series), p197
Caraveo, P. A., De Luca, A., Mignani, R. P., and Bignami, G. F., 2001, ApJ 561, 930
Dodson, R., Legge, D., Reynolds, J., and McCulloch, P., 2003a, ApJ, submitted
Dodson, R., Lewis, D., McConnell, D., and Deshpande, A., 2003b, MNRAS, in press
Dodson, R. G., McCulloch, P. M., and Lewis, D. R., 2002, ApJL 564, L85
Gold, T., 1968, Nature 218, 731
Helfand, D. J., Gotthelf, E. V., and Halpern, J. P., 2001, ApJ 556, 380
Large, M., Vaughan, A., and Mills, B., 1968, Nature 220, 340
Legge, D., 2002, PhD thesis, University of Tasmania
Pavlov, G. G., Sanwal, D., Garmire, G. P., Zavlin, V.E., Burwitz, V., and Dodson, R. G., 2000, AAS Meeting 196, 37.04
Radhakrishnan, V. and Cooke, D. J., 1969, ApJL 3, 225
Radhakrishnan, V., Cooke, D. J., Komesaroff, M., and Morris, D., 1969, Nature 221, 443
Radhakrishnan, V. and Deshpande, A. A., 2001, A & A 379, 551
Radhakrishnan, V. and Manchester, R., 1969, Nature 222, 228
Spruit, H. C. and Phinney, E. S., 1998, Nature 393, 139
Stappers, B. W., Gaensler, B. M., and Johnston, S., 1999, MNRAS 308, 609

D. Legge, D. Lewis, A. A. Deshpande, R.. Dodson, J. Reynolds, D. McConnell and P. McCulloch
(rdodson@kerr.phys.utas.edu.au)