2021ApJ...911...14K

We examine a sample of 21 gamma-ray burst (GRB) afterglow light curves at radio frequencies and compare them to the X-ray and/or optical properties of the afterglows and to the predictions of the standard jet/fireball model. Our sample includes every Swift GRB with an X-ray light curve indicating a jet break and with a published radio light curve, as well as several other targets with observed X-ray or and/optical jet breaks. We examine the late-time decline of each burst and attempt to fit an analytical model based on the standard GRB afterglow equations to each data set. We show that most of the events in our Swift GRB sample are incompatible with the radio light-curve behavior predicted by conventional afterglow theory. Many exhibit a late-time radio decline incompatible with the post-break X-ray or optical afterglow. Only one radio afterglow in this sample, at any time, shows the eventually expected decline of ∼t^–2, although two others show it in their millimeter light curve. Several others remain consistent with the standard model if such a decline began after the observations. The radio behavior alone does not, however, indicate whether a GRB can be fit by our modeling code. Indeed, several of the well-fit GRBs may only appear so due to a lack of multiwavelength data. While a second source of emission can account for some of the anomalous radio behavior, our tests indicate this is often not the case unless the main jet component is simultaneously suppressed.