Search

SKA pathfinders (e.g., ASKAP and MeerKAT) are performing large HI surveys, surpassing previous-generation radio surveys in angular resolution, volume, and depth — going nearly two orders of magnitude deeper. The vast amount of data being produced by these telescopes creates a need for new tools and techniques, such as automated source-extraction. One particular science goal is to trace the large scale structure of galaxies in the local Universe behind the Zone of Avoidance (ZoA). This is difficult to do in the optical and infrared bands because of the thick dust and stellar crowding along the plane of the Milky Way. HI surveys are ideal because the 21 cm spectral line emission of neutral hydrogen (HI) atoms is unaffected by these features. This dissertation presents two HI surveys in the ZoA: the MeerKAT16 Early Science project, conducted before the completion of the full array, with 16-dish subarrays and the ROACH-32k correlator; and later, the Galactic Plane Legacy Survey (GPS), a larger MeerKAT survey project utilising 60-dish subarrays and the SKARAB-4k correlator. The motivation of the MeerKAT16 survey was two-fold: serving as a pilot project for the envisioned large MeerKAT64 Vela Supercluster (VSCL) survey; and a test-bed for optimising galaxy identification processes using automated pipelines. The survey aimed to map the rich galaxy cluster VC04, embedded in one of the prospective VSCL walls. We searched two mosaic cubes for HI sources by means of a deep visual search, and then using the Source Finding Application (SoFiA; version 1.3.2). The purpose of using both methods was to understand and optimise the SoFiA pipeline on real data, which is important for the development of source finding strategies for future large surveys. We catalogued 119 reliable galaxy detections (and an additional 37 candidates with lower certainty) within the early science data. SoFiA found galaxies all the way at the VSCL distance (V(hel) ≈ 18 000 km/s), where we detected hints of two walls, and HI deficiency in the centre of VC04. GPS surveyed a long narrow strip (|b| ≤ 2°) along the Galactic plane, with the aim of penetrating the most obscured part of the ZoA. A segment of the data spanning 302° ≤ l ≤ 332° was reduced, imaged, and analysed as part of this project. This region is interesting because it encompasses the Great Attractor (GA) — a massive overdensity highly influential to the local flow of galaxies. Aided by the newer version of SoFiA (v2.3.1), we found 477 galaxy candidates, most of which are new discoveries, and performed an in-depth comparison against the largest systematic HI survey covering the southern ZoA up to this point — the Parkes HIZOA survey. Additionally, we compared our results to simulations that follow the method used for the SKA HI science case. The GPS redshift distribution reveals a striking overdensity at the GA distance (V(hel) ≈ 4000 km/s), inconsistent with a uniform galaxy distribution, whereas the high-redshift end is more underdense than predicted by the simulation. These deep interferometric HI surveys provided a new glimpse of highly interesting structures crossing the Galactic plane, demonstrating that large scale structure can be mapped even in the deepest part of the ZoA. The success of MeerKAT16 and GPS is a preview of the HI science potential of the upcoming SKA, which will surpass MeerKAT in depth, sensitivity and resolution.