Search

High-precision timing is an exciting field of pulsar research that holds the promise of direct gravitational wave detection. This goal is at the limit of current technology, and requires the near-seamless combination of data from multiple pulsars observed with multiple telescopes. Accuracy in the recording of pulse profiles and the measurement of their times of arrival (TOAs) is key. In order to time a pulsar, a template is needed that is as close as possible to an ideal version of the pulse profile. Four pulsars were timed using analytic, noise-free templates composed of Gaussian components, and their system parameters were measured to high accuracy. These templates were found to be usable at different telescopes and were adaptable across more than 100 MHz in observing frequency without loss of TOA alignment. The same approach was used to investigate profile variation in the pulsar PSR J1022+1001, which is a promising member of the arrays used to search for gravitational waves. Variation with time was found that is almost equal in magnitude to typical measurement noise. The timing of the pulsar could not be improved using adaptive templates which were allowed to vary with the profile to a limited extent. If the variation is due to instrumental error, then its removal would improve timing accuracy significantly for this pulsar. If it is intrinsic, then it is an interesting and unusual phenomenon. PSRJ1022+1001 was compared to two similar pulsars, one of which showed a lesser degree of variation and one of which did not exhibit significant change.Timing of PSR J1022+1001 was used to calculate upper limits on the amplitude of the stochastic gravitational wave background. A reasonable limit was estimated to be Amax = 1.7 x 10⁻¹⁴, which is stringent for a single pulsar timed alone. However, there was evidence that the timing residuals were somewhat correlated in time, which can produce an artificially low limit. Nevertheless, PSR J1022+1001 has the potential to make a valuable contribution to gravitational wave detection. Investigation into its variation highlights the fact that the timing of a number of pulsars may be crucially improved by the next generation of processing instruments.

The publications from Appendix 6 (p. A52 to end) have been removed from the electronic version of this thesis due to third party copyright restrictions. The unabridged version can be viewed at the University of Leicester Library.