Marcin Sokolowski

We present AAVS3, the latest prototype station of the SKA-Low Aperture Array Verification System, and its role in mitigating design risks for the SKA-Low telescope. AAVS3 replaces the pseudo-random AAVS2 layout with the Vogel “sunflower” pattern to improve spatial uniformity and azimuthal diversity. Interferometric visibility calibration procedures were developed using CASA, including a basic Sun-based method and an advanced technique combining embedded element patterns with the global sky model. Electromagnetic simulations conducted during construction predicted a sensitivity depression near zenith at 125 MHz due to geometric regularities inherent in the Vogel layout, and on-sky measurements confirmed this behavior. The Vogel layout was therefore not adopted for the SKA-Low array. The calibration methods and measurements presented here provide technical guidance for SKA-Low prototyping and inform commissioning activities for the first SKA-Low stations.

Low-frequency pulsar observations are crucial for understanding pulsar emission spectra and population physics, as well as for probing the interstellar medium (ISM) and Earth’s ionosphere. We report the largest low-frequency pulsar census conducted in the Southern Hemisphere, covering 50–250 MHz, using the EDA2, an SKA-Low prototype station. In this survey, we detected 120 pulsars, including 23 first-time detections below 150 MHz and 5 below 100 MHz. For each source, we provide integrated pulse profiles and flux-density measurements across five subbands spanning 50–250 MHz. We also obtained improved dispersion measure (DM) values for 110 pulsars, with a median absolute DM correction of about 0.1 pc cm ⁻³ . We measured significant Faraday rotation for 40 pulsars with improved rotation measure (RM) values for four pulsars, as well as phase-resolved RM variation in J1453-6413. Full-polarimetric pulse profiles are provided for all these pulsars, with multifrequency polarimetric data for 20 of them. These results will enhance future SKA-Low science: refining pulsar population models, informing survey strategies, and advancing characterization of both the ISM and the ionosphere through low-frequency pulsar monitoring.

Petabytes of archival high time resolution observations have been captured with the Murchison Widefield Array. The search for Fast Radio Bursts within these using established software has been limited by its inability to scale on supercomputing infrastructure, necessary to meet the associated computational and memory requirements. Hence, past searches used a coarse integration time, in the scale of seconds, or analysed an insufficient number of hours of observations. This paper introduces BLINK, a novel radio interferometry imaging software for low-frequency FRB searches to be run on modern supercomputers. It is implemented as a suite of software libraries executing all computations on GPU, supporting both AMD and NVIDIA hardware vendors. These libraries are designed to interface with each other and to define the BLINK imaging pipeline as a single executable program. Expensive I/O operations between imaging stages are not necessary because the stages now share the same memory space and data representation. BLINK is the first imaging pipeline implementation able to fully run on GPUs as a single process, further supporting AMD hardware and enabling Australian researchers to take advantage of Pawsey’s Setonix supercomputer. In the millisecond-scale time resolution imaging test case illustrated in this paper, representative of what is required for FRB searches, the BLINK imaging pipeline achieves a 3 687x speedup compared to a traditional MWA imaging pipeline employing WSClean.

Precise polarimetric measurements of radio signals from the sky are crucial for achieving many scientific goals of modern and future radio telescopes. A key figure of merit (FoM) used to quantify the polarization purity of a radio polarimeter is the intrinsic cross-polarization ratio (IXR, introduced by Carozzi and Woan¹). This FoM is independent of the coordinate systems and is commonly used to assess the polarimetric performance of low-frequency phased aperture arrays, such as low frequency aperture array (LOFAR),² murchison widefield array (MWA),³ long wavelenght array (LWA),⁴ and the upcoming low-frequency component of the Square Kilometre Array (SKA-Low).⁵ Polarization calibration of such arrays is challenging due to their wide fields of view and the scarcity of polarized calibrators at these frequencies. In the literature, the various works investigating polarization performance using IXR are mostly relying on electromagnetic antenna simulations [e.g., aperture tile in focus (APERTIF),⁶ MWA,⁷ and SKA-Low⁸]. More recently, all-sky observations from the LOFAR-LBA Swedish station have been used to estimate the instrumental Stokes parameters across the sky and enable the computation of its IXR.⁹

It has been proposed that coherent radio emission could be emitted during or shortly following a gamma-ray burst (GRB). Here we present a low-frequency (170–200 MHz) search for radio pulses associated with long-duration GRBs using the Murchison Widefield Array (MWA). The MWA, with its rapid-response system, is capable of performing GRB follow-up observations within approximately 30 s. Our single pulse search, with temporal and spectral resolutions of 100 μs and 10 kHz, covers dispersion measures up to 5000 pc cm⁻³. Two single pulse candidates are identified with significance greater than 6σ, surviving a friends-of-friends analysis. We rule out random fluctuations as their origin at a confidence level of 97% (2.2σ). We caution that radio frequency interference from digital TV (DTV) is most likely the origin of these pulses since the DTV frequency bands almost cover the entire observing frequency band. If they are astrophysical signals, we estimate the peak flux densities for our pulse candidates of 3.6 ± 0.6 and 10.5 ± 1.5 Jy, with corresponding fluences of 431 ± 74 and 211 ± 37 Jy ms, respectively. Based on these observations and the assumption of a magnetar origin for the pulse, we constrain the radio emission efficiency as ϵ _r ∼ 10⁻³ for both candidates, which is consistent with pulsar observations. Our results highlight the promising potential of new-generation radio telescopes such as the MWA to probe the central engines of GRBs.