The Square Kilometre Array Observatory (SKAO) will explore the radio sky to new depths in order to conduct transformational science. SKAO data products made available to astronomers will be correspondingly large and complex, requiring the application of advanced analysis techniques to extract key science findings. To this end, SKAO is conducting a series of Science Data Challenges, each designed to familiarise the scientific community with SKAO data and to drive the development of new analysis techniques. We present the results from Science Data Challenge 2 (SDC2), which invited participants to find and characterise 233245 neutral hydrogen (Hi) sources in a simulated data product representing a 2000~h SKA MID spectral line observation from redshifts 0.25 to 0.5. Through the generous support of eight international supercomputing facilities, participants were able to undertake the Challenge using dedicated computational resources. Alongside the main challenge, `reproducibility awards' were made in recognition of those pipelines which demonstrated Open Science best practice. The Challenge saw over 100 participants develop a range of new and existing techniques, with results that highlight the strengths of multidisciplinary and collaborative effort. The winning strategy -- which combined predictions from two independent machine learning techniques to yield a 20 percent improvement in overall performance -- underscores one of the main Challenge outcomes: that of method complementarity. It is likely that the combination of methods in a so-called ensemble approach will be key to exploiting very large astronomical datasets.

ABSTRACT We present sofia 2, the fully automated 3D source finding pipeline for the WALLABY extragalactic H i survey with the Australian SKA Pathfinder (ASKAP). sofia 2 is a reimplementation of parts of the original sofia pipeline in the c programming language and makes use of OpenMP for multithreading of the most time-critical algorithms. In addition, we have developed a parallel framework called sofia-X that allows the processing of large data cubes to be split across multiple computing nodes. As a result of these efforts, sofia 2 is substantially faster and comes with a much reduced memory footprint compared to its predecessor, thus allowing the large WALLABY data volumes of hundreds of gigabytes of imaging data per epoch to be processed in real time. The source code has been made publicly available to the entire community under an open-source licence. Performance tests using mock galaxies injected into genuine ASKAP data suggest that in the absence of significant imaging artefacts sofia 2 is capable of achieving near-100 per cent completeness and reliability above an integrated signal-to-noise ratio (SNR) of about 5–6. We also demonstrate that sofia 2 generally recovers the location, integrated flux, and w20 line width of galaxies with high accuracy. Other parameters, including the peak flux density and w50 line width, are more strongly biased due to the influence of the noise on the measurement. In addition, very faint galaxies below an integrated SNR of about 10 may get broken up into multiple components, thus requiring a strategy to identify fragmented sources and ensure that they do not affect the integrity of any scientific analysis based on the sofia 2 output.

AbstractThe Murchison Widefield Array (MWA) is an open access telescope dedicated to studying the low-frequency (80–300 MHz) southern sky. Since beginning operations in mid-2013, the MWA has opened a new observational window in the southern hemisphere enabling many science areas. The driving science objectives of the original design were to observe 21 cm radiation from the Epoch of Reionisation (EoR), explore the radio time domain, perform Galactic and extragalactic surveys, and monitor solar, heliospheric, and ionospheric phenomena. All together$60+$programs recorded 20 000 h producing 146 papers to date. In 2016, the telescope underwent a major upgrade resulting in alternating compact and extended configurations. Other upgrades, including digital back-ends and a rapid-response triggering system, have been developed since the original array was commissioned. In this paper, we review the major results from the prior operation of the MWA and then discuss the new science paths enabled by the improved capabilities. We group these science opportunities by the four original science themes but also include ideas for directions outside these categories.