Finding pulsars with MeerKAT and other multibeam systems

UoM administered thesis: Phd

  • Authors:
  • Vincent Morello


Pulsars are fascinating objects that enable a variety of high-precision experiments. A key aspect of pulsar science is the continued search for more of them, with an emphasis on unusual specimens that challenge and expand our understanding of the laws of physics. This thesis starts with a search of archival survey data acquired with the Parkes radio telescope (Chapter 2), using a highly optimised GPU-based search code and an accurate candidate classification algorithm, resulting in the discovery of 23 pulsars missed in previous searches. We empirically noticed that longer period sources were less easily detected by the standard method based on Fourier transforming the data, leading us to experiment with another technique: the fast folding algorithm (FFA). In early tests of the code, we found an underlying periodicity to a sequence of faint pulses recorded at Parkes; these turned out to be from a previously unknown pulsar, PSR J2251 − 3711, whose spin period of 12.1 seconds was the longest known at the time of discovery. An analysis of the unusual properties of this source is presented in Chapter 3. The idea that the FFA could fill the sensitivity gap of the standard method in the long-period regime is not new. In Chapter 4 we demonstrate analytically an even stronger result: the FFA is the most sensitive method to all periodic signals. Furthermore, we show that the sensitivity of some past surveys, an important ingredient to models of the Galactic pulsar population, has been overestimated; the population might thus be larger than previously thought. We also release our FFA search pipeline implementation, which is shown to be fast enough to process the data generated by modern surveys. Lastly, we contributed to building MeerTRAP, a backend for the MeerKAT radio telescope that continuously searches for radio transients and pulsars, with a wide field of view and sensitivity unmatched in the Southern hemisphere. The project is ambitious with many technical challenges; our contribution consisted of developing the high-level control system software for the MeerTRAP processing cluster, which is the subject of Chapter 5. The fast radio transient search pipeline is now functional, and once finished, a search for periodic signals will be added, eventually providing the FFA with an ideal hunting ground.


Original languageEnglish
Awarding Institution
Award date31 Dec 2020