Andrew Hopkins, Ron Ekers, Carole Jackson, Lawrence Cram, Anne Green, Dick Manchester, Lister Staveley-Smith and Ray Norris, PASA, 16 (2), in press.
Next Section: Masers and spectral lines
Title/Abstract Page: Summary of the ``Sub-microJansky
Previous Section: Radio stars - Lawrence
Pulsars - Dick Manchester
There are currently about 750 pulsars identified in the current literature. All but a handful of these are within our Galaxy, and most are within a few kpc of the Sun. Pulsars fall into two main classes: `normal' pulsars which are typically less than 107 years old and have periods of between 30 ms and several seconds, and `millisecond' pulsars, which typically have periods of between 1.5 and 20 ms and are believed to be `recycled' or spun-up by accretion in a binary system. Most known pulsars are in the Galactic disk, but a significant population of millisecond pulsars is found in globular clusters. Estimates based on previous large-scale searches show that the total number of potentially detectable normal pulsars in the Galaxy is at least 30,000. Surprisingly, the total number of potentially detectable millisecond pulsars is at least as large. Existing searches have had limited sensitivity to millisecond pulsars, so they form less than 10% of the known pulsar population.
Large-scale searches for pulsars are undertaken for several reasons. Most obviously, they increase the number of known pulsars, allowing better determinations of the underlying pulsar population, and a larger sample for various studies of, for example, the pulse emission process, associations with supernova remnants, and the interstellar medium. However, a major motivation for such searches is the likelihood of finding unusual or unexpected types of pulsars. Pulsar astronomy has a history of such finds, for example, the first binary pulsar, the first millisecond pulsar, the first pulsar with a main-sequence companion, the first extra-solar planetary system etc. Many of these finds have important consequences, none more so that the Hulse-Taylor binary PSR 1913+16 (Taylor & Weisberg 1982), which has been used to verify general relativity and give the first observational evidence for gravitational radiation. Future searches will undoubtedly uncover new and exciting objects.
Essentially all pulsar observations are sensitivity limited - confusion is not a problem with pulsars. With its large collecting area, the Square Kilometre Array will have a major impact on pulsar astronomy. For purposes of comparison, we scale the parameters of the current Parkes Multibeam survey. This has a frequency of 1.4GHz and a bandwidth of 288MHz in each of two polarisations. This frequency is about optimum for many pulsar studies as it avoids the worst of the interstellar effects such as dispersion and scattering, but is not so high that pulsar flux densities are too weak.
In a full tied-array mode, in one minute, the Square Kilometre Array will give a 
detection of a pulsar with a mean flux density of 
Jy. Even if only part of the array (e.g., a compact central portion) can be used in this mode, the sensitivity will be very high. This will form a powerful instrument for studies of known pulsars, including timing, polarisation, individual pulse studies, etc. To exploit it fully, the array should provide 10 or more beams within an area of at least 1 square degree, with full Stokes parameters, high time resolution (better than 
s) and good frequency resolution (at least 1000 frequency channels across the band).
Searching is probably best done by incoherently summing the output of a number of subarrays. This involves some loss of sensitivity, but at least for a disk population, this is exactly compensated for by the increased area searched. Furthermore, it is not possible to arbitrarily reduce the time spent on each beam area as the sensitivity is increased, as a large number (say at least 103) pulse periods must be observed to provide period discrimination. A good compromise would be to incoherently sum the outputs of 100 subarrays, each equivalent to a 110m dish and having a beam area of about 0.15 square degrees. With this system, a search along the Galactic Plane, similar to that of the Parkes Multibeam survey but with 10 times the sensitivity, would detect about 5000 pulsars with 4 min/beam or a total time of about 40 days. Depending on the back-end system, this survey could also detect a similar number of millisecond pulsars. Globular clusters would also be an attractive target; for example, in a few hours it should be possible to detect more than 100 millisecond pulsars in the core of 47 Tucanae!
It is clear that the Square Kilometre Array can be a superb pulsar machine. However, careful thought will have to be given to the optimal way to provide high time and frequency resolution in the various configurations. Polarisation properties are also crucial for pulsar applications.
Next Section: Masers and spectral lines
Title/Abstract Page: Summary of the ``Sub-microJansky
Previous Section: Radio stars - Lawrence
© Copyright Astronomical Society of Australia 1997