It is widely held that pulsars are born in the Galactic plane from
massive Population I stars (
~solar masses ) and that their large scale
heights can be attributed to a birth velocity of a few hundred km/s.
Proper motion measurements are the observational support
for the inference that pulsars are a high velocity population born in
the plane. Such measurements, when used together with the distance
and age of a pulsar, give the magnitude and direction of the
transverse velocity of the pulsar. Not only does this demonstrate
that the majority of pulsars are migrating out of the plane, but in
some individual cases it can be used to trace the pulsar back to its
origin, be that an OB association and/or a supernova remnant.
The distribution of pulsar velocities is another important area of research. Important questions can be answered on the origin of these velocities (binary disruption and/or random ``kicks") simply by knowing the true shape of the distribution. To date, proper motion measurements have been made for less than 15% of the known pulsars, producing a sample which is seriously limited by selection effects (i.e., bright, nearby pulsars). Indeed, attempts to correct for these selection effects have suggested that the mean pulsar velocity may be larger than was commonly thought by a factor of 2-3.
By necessity, astrometric observations have been made at centimeter
wavelengths to avoid ionospheric effects. However, as pulsars are
steep spectrum (
) objects, observations at
10-20cm require that we gate the correlator on pulse for
increased signal-to-noise ratio. The current small sample of pulsars
with proper motion measurements reflects the twin limitations of
sensitivity and positional accuracy. The current VLA yields superior
sensitivity but is limited in positional accuracy (
milli-arcsec/yr). Conversely, the VLBA has excellent positional
accuracy but is limited to phase reference observations on the
brighter pulsars (>5-10 mJy). The enhanced VLA, particularly the
A+ configuration with the wider bandwidth 1.4 GHz system and the new
2.4 GHz system, will greatly increase the number of objects for which
we can do accurate astrometry.
The VLA Development Plan benefits pulsar studies in two ways: through
improved signal-to-noise ratio and positional accuracy. All IF
improvements (receiver, fiber optics and correlator) provide a
signal-to-noise increase. The 2.4 GHz receiver and the increased
bandwidth of the 1.4 GHz receiver gain a factor of 2 in
signal-to-noise. However, the greatest improvements come from the A+
configuration. In the best case we should improve positional accuracy
by 
20 at 1.4 GHz and at 2.4 GHz. To achieve these improvements
for the rapidly rotating and high-dispersion-measure pulsars will
require that the new correlator have a frequency dependent gate so
that the pulse phase can be accurately tracked across the full
available bandwidth.
With the current VLA we can observe the 40 closest and brightest of 700 known pulsars. With the proposed improvements to VLA sensitivity and the A+ configuration we can do sub-milli-arc-second astrometry for mJy pulsars, with which a significant fraction (>30%) of the known pulsar population can be studied. The expected improvements, normalized to our current capabilities with the current VLA at 20 cm, are illustrated graphically in Fig. 3.5. We estimate that the number of pulsars with proper motion measurements would double, and that a trigonometric parallax could be measured for pulsars out to a distance of 1 kpc. Serious investigations of pulsar velocities could begin, unhampered by selection effects.
Figure 3.5: A comparison of the relative capabilities of several
VLA configurations for pulsar astrometry. Shown are the current VLA
and VLA+VLBA capabilities in the 1.4 GHz band, with and without the
gate, as well the predicted capabilities of the enhanced VLA,
with and without the A+ configuration. The two axes are gain in
sensitivity and gain in positional accuracy, and the numbers are gains
relative to the current VLA without gating. Displayed above the plot
are the number of pulsars available to each configuration.