Enabling wide-field, high-spatial-resolution fast transient searches on modern interferometry

Abstract

Fast transients reveal the energetic universe to us. They are usually the products of rapid and enormous energy releases. To study the physical nature behind them, we must know the subtle temporal structure of the signal and where precisely they come from. Radio observation has been an indispensable part of the study of fast transients, thanks to their daylight-insusceptibility and weather-tolerance. The ever powerful single-dish telescopes have been used in pulsar observations and fast radio transient surveys, yielding substantial results. However, some sporadic or even one-off fast radio transients have brought challenges to the single-dish observations. Other than high time resolution, efficient fast transient surveys require a wide field of view and spatial resolution. While larger single-dish telescopes are harder and much expensive to build, interferometry provides a practical alternative. The high spatial resolution, scalability, flexibility and cost-efficiency have made interferometry wildly used in modern radio astronomy. With the beamforming techniques, interferometric arrays are able to carry out wide-field, high resolution surveys for fast radio transients. <br /> In this thesis, I start the introduction with an instrument-wise history of radio observation and its importance in astronomy. Following that is the introduction of the fast radio transients, including various types of pulsars and fast radio bursts. Then I present radio interferometry as a solution to tackle the challenges in the fast radio transient surveys. A brief history of radio interferometry and its achievements are also provided. <br /> The second chapter presents the various requirements for fast radio surveys with interferometry, such as high time and frequency resolution, large field of view and high spatial resolution. I then explain how beamformed observation with interferometric arrays can meet these requirements. A comparison between the raw-voltage beamforming and visibility beamforming is presented in the perspective of computing complexity. At the end of the chapter, I briefly outline the configuration of the MeerKAT telescope, its various backends and planned science projects. The overview of the transient search system of MeerKAT based on the abovementioned beamforming technique is presented. The reason for MeerKAT using the raw-voltage beamforming method is explained. <br /> The third chapter describes the general beamforming technique in detail and provides its implementation. Following that, I provide the solutions to the challenges that come with this technique, such as the characterization of volatile beam shapes, the generation of efficient tiling of beams and the prediction of the evolution of the tiling through time. To provide a more realistic perspective, the integration of these techniques in the MeerKAT telescope is illustrated and the corresponding capacity and statistics are provided. As an evaluation, a real beamformed observation on 47 Tucanae using the said techniques is presented and the localization capability with multiple beams is demonstrated. The development and deployment of this technique have been proven successful and effective. <br /> The performance of this beamforming platform has been further tested in numerous observations and surveys which are detailed in the fourth chapter. For example, the TRAPUM project consisting of but is not limited to surveys of globular clusters, nearby galaxies and targeted pulsar searches of unidentified FERMI gamma-ray sources. Another important survey is the MGPS project which searches pulsars in the Galactic plane at both L-band and S-band. From both projects, we have discovered a total of 66 pulsars at the time of writing. <br /> I report a search for giant pulses in selected pulsars in the fifth chapter. Many models try to explain the emission mechanism behind giant pulses from pulsars, some of which invoke the re-connection events near the light cylinder. In the meantime, a population of giant pulse emitters shares similar properties, such as strong magnetic fields near the light cylinder and high energy emission. I select a couple of pulsars with these properties as candidates. They were observed using the Effelsberg 100-meter telescope. I created a pipeline based on Heimdall to search giant pulses on the data and find no credible detection. The conclusion for the non-detection is that these pulsars do not emit giant pulses which are detectable by our observation, or the condition of high magnetic field strength near the light cylinder and high energy emission may not be a sufficient condition for the occurrence of giant pulses.