Professor Rob Fender

Abstract:

The neutron star X-ray binary, EXO 0748−676, was observed regularly by the Rossi X-ray Timing Explorer (RXTE) and XMM–Newton during its first detected outburst (1985–2008). These observations captured hundreds of asymmetric, energy-dependent X-ray eclipses, influenced by the ongoing ablation of the companion star and numerous Type I thermonuclear X-ray bursts. Here, we present the light curves of 22 Type I X-ray bursts observed by RXTE that coincide, fully or partially, with an X-ray eclipse. We identify nine instances where the burst occurs entirely within totality, seven bursts split across an egress, and six cases interrupted by an ingress. All in-eclipse and split bursts occurred while the source was in the hard spectral state. We establish that we are not observing direct burst emission during eclipses since the companion star and the ablated outflow entirely obscure our view of the X-ray emitting region. We determine that the reflected flux from the outer accretion disc, even if maximally flared, is insufficient to explain all observations of in-eclipse X-ray bursts and instead explore scenarios whereby the emission arising from the X-ray bursts is scattered, either by a burst-induced rise in that provides extra material, an accretion disc wind or the ablated outflow, into our line of sight. However, the rarity of a burst and eclipse overlap makes it challenging to determine their origin.

Abstract:

<jats:title>ABSTRACT</jats:title> <jats:p>In this work, we explore the applicability of unsupervised machine learning algorithms to finding radio transients. Facilities such as the Square Kilometre Array (SKA) will provide huge volumes of data in which to detect rare transients; the challenge for astronomers is how to find them. We demonstrate the effectiveness of anomaly detection algorithms using 1.3 GHz light curves from the SKA precursor MeerKAT. We make use of three sets of descriptive parameters (‘feature sets’) as applied to two anomaly detection techniques in the astronomaly package and analyse our performance by comparison with citizen science labels on the same data set. Using transients found by volunteers as our ground truth, we demonstrate that anomaly detection techniques can recall over half of the radio transients in the 10 per cent of the data with the highest anomaly scores. We find that the choice of anomaly detection algorithm makes a minor difference, but that feature set choice is crucial, especially when considering available resources for human inspection and/or follow-up. Active learning, where human labels are given for just 2 per cent of the data, improves recall by up to 20 percentage points, depending on the combination of features and model used. The best-performing results produce a factor of 5 times fewer sources requiring vetting by experts. This is the first effort to apply anomaly detection techniques to finding radio transients and shows great promise for application to other data sets, and as a real-time transient detection system for upcoming large surveys.</jats:p>