Ian Heywood

Abstract:

We present simulated images of Supernova 1993J at 8.4 GHz using Very Long Baseline Interferometry (VLBI) techniques. A spherically symmetric source model is convolved with realistic uv-plane distributions, together with standard imaging procedures, to assess the extent of instrumental effects on the recovered brightness distribution. In order to facilitate direct comparisons between the simulations and published VLBI images of SN1993J, the observed uv-coverage is determined from actual VLBI observations made in the years following its discovery. The underlying source model only exhibits radial variation in its density profile, with no azimuthal dependence and, even though this model is morphologically simple, the simulated VLBI observations qualitatively reproduce many of the azimuthal features of the reported VLBI observations, such as appearance and evolution of complex azimuthal structure and apparent rotation of the shell. We demonstrate that such features are inexorably coupled to the uv-plane sampling. The brightness contrast between the peaks and the surrounding shell material are not as prominent in the simulations (which of course assume no antenna- or baseline-based amplitude or phase errors, meaning no self-calibration procedures will have incorporated any such features in models). It is conclusive that incomplete uv-plane sampling has a drastic effect on the final images for observations of this nature. Difference imaging reveals residual emission up to the 8 sigma level. Extreme care should be taken when using interferometric observations to directly infer the structure of objects such as supernovae.

Abstract:

We present deep, multi-VLA-configuration radio images for a set of 18 quasars, having redshifts between 0.36 and 2.5, from the 7C quasar survey. Approximately one quarter of these quasars have FRI-type twin-jet structures and the remainder are a broad range of wide angle tail, fat double, classical double, core-jet and hybrid sources. These images demonstrate that FRI quasars are prevalent in the universe, rather than non-existent as had been suggested in the literature prior to the serendipitous discovery of the first FRI quasar a few years ago, the optically powerful "radio quiet" quasar E1821+643. Some of the FRI quasars have radio luminosities exceeding the traditional FRI / FRII break luminosity, however we find no evidence for FRII quasars with luminosities significantly below the break. We consider whether the existence of such high luminosity FRI structures is due to the increasingly inhomogeneous environments in the higher redshift universe.

Abstract:

MERLIN observations of the unusually slow nova V723 Cas are presented. Nine epochs of 6-cm data between 1996 and 2001 are mapped, showing the initial expansion and brightening of the radio remnant, the development of structure and the final decline. A radio light curve is presented and fitted by the standard Hubble flow model for radio emission from novae in order to determine the values of various physical parameters for the shell. The model is consistent with the overall development of the radio emission. Assuming a distance of 2.39 (+/-0.38) kpc and a shell temperature of 17000 K, the model yields values for expansion velocity of 414 +/- 0.1 km s^-1 and shell mass of 1.13 +/- 0.04 * 10^-4 Msolar. These values are consistent with those derived from other observations although the ejected masses are rather higher than theoretical predictions. The structure of the shell is resolved by MERLIN and shows that the assumption of spherical symmetry in the standard model is unlikely to be correct.

Abstract:

We compare the predictions of simple models for the radio emission from classical novae with the MERLIN radio observations of nova V723 Cas (Nova Cas 1995). Spherically symmetric and ellipsoidal radiative transfer models are implemented in order to generate synthetic emission maps. These are then convolved with an accurate representation of the uv coverage of MERLIN. The parameters and geometry of the shell model are based on those returned by fitting models to the observed light curve. This allows direct comparison of the model images with the nine 6-cm MERLIN images of V723 Cas. It is found that the seemingly complex structure (clumping, apparent rotation) evident in the observations can actually be reproduced with a simple spherical emission model. The simulations show that a 24-h track greatly reduces the instrumental effects and the synthetic radio map is a closer representation of the true (model) sky brightness distribution. It is clear that interferometric arrays with sparse uv coverage (e.g. MERLIN, VLBA) will be more prone to these instrumental effects especially when imaging ring-like objects with time-dependent structure variations. A modelling approach such as that adopted here is essential when interpreting observations. © 2007 RAS.