Statistical search for angular non-stationarities of long gamma-ray burst jets using <i>Swift</i> data

ABSTRACT In a previous article, we argued that angular non-stationarities of gamma-ray burst (GRB) jets can result in a statistical connection between the angle values deduced from jet break times and the variabilities of prompt light curves. The connection should be an anticorrelation if luminosity densities of jets follow a power-law or a uniform profile, and a correlation if they have a Gaussian profile. In this follow-up paper, we search for the connection by measuring Spearman’s rank correlation coefficient in a sample of 19 long GRBs observed by the Swift satellite. Using 16 of the GRBs with well-defined angle measurements, we find $\rho = -0.38_{-0.1}^{+0.1}$ and $p = 0.15_{-0.09}^{+0.14}$. Adding three more GRBs to the sample, each with a pair of equally possible angle values, can strengthen the anticorrelation to $\rho =-0.46_{-0.08}^{+0.09}$ and $p=0.05_{-0.03}^{+0.07}$. We show that these results are incompatible with non-stationary jets having Gaussian profiles, and that ≳100 GRBs with observed afterglows would be needed to confirm the potential existence of the angle-variability anticorrelation with 3σ significance. If the connection is real, GRB jet angles would be constrainable from prompt gamma light curves, without the need of afterglow observations.