We report on a VLA survey for late-time radio emission from 59 supernovae (SNe) of Type I b/c, which have been associated with long-duration gamma-ray bursts (GRBs). An off-axis GRB burst (i.e. whose relativistic jet points away from us) is expected
to have late-time radio emission even in the absence of significant prompt gamma-ray emission. From our sample, we detected only SN 2003gk with an 8.4-GHz flux density of $2260 pm 130 ,mu$Jy. Our subsequent VLBI observations of SN 2003gk, at an age of $sim$8 yr, allowed us to determine its radius to be $(2.4 pm 0.4) times 10^{17}$ cm, or $94 pm 15$ light days. This radius rules out relativistic expansion as expected for an off-axis GRB jet, and instead suggests an expansion speed of $sim 10:000$ km s$^{-1}$ typical for non-relativistic core-collapse supernovae. We attribute the late-onset radio emission to interaction of the ejecta with a dense shell caused by episodic mass-loss from the progenitor. In addition, we present new calculations for the expected radio lightcurves from GRB jets at various angles to the line of sight, and compare these to our observed limits on the flux densities of the remainder of our SN sample. From this comparison we can say that only a fraction of broadlined Type I b/c SNe have a radio-bright jet similar to those seen for GRB afterglows at cosmological distances. However, we also find that for a reasonable range of parameters, as might be representative of the actual population of GRB events rather than the detected bright ones, the radio emission from the GRB jets can be quite faint, and that at present, radio observations do not place strong constraints on off-axis GRB jets.
We present a 3D numerical simulation of the recently discovered cometary structure produced as Mira travels through the galactic ISM. In our simulation, we consider that Mira ejects a steady, latitude-dependent wind, which interacts with a homogeneou
s, streaming environment. The axisymmetry of the problem is broken by the lack of alignment between the direction of the relative motion of the environment and the polar axis of the latitude-dependent wind. With this model, we are able to produce a cometary head with a ``double bow shock which agrees well with the structure of the head of Miras comet. We therefore conclude that a time-dependence in the ejected wind is not required for reproducing the observed double bow shock.
