Frequency-dependent time-delays for strong outbursts in selected blazars from the Metsähovi and the University of Michigan Radio Astronomy Observatory monitoring databases – I.

Abstract

The combined data of the University of Michigan Radio Astronomy Observatory and Metsahovi Radio Observatory provide us with radio light curves for active galactic nuclei monitored by both observatories from 4.8 to 37 GHz covering time-intervals up to ∼25 yr. We consider here such composite light curves for four gamma-ray blazars that have been nearly continuously monitored at both observatories: 0458−020, 0528+134, 1730−130 and 2230+114. We have decomposed the most prominent outbursts in the light curves of these four blazars into indi- vidual components using Gaussian model fitting, and estimated the epochs, amplitudes, and half-widths of these components as functions of frequency. We attempt to distinguish 'core outbursts', which show frequency-dependent time-delays and are associated with brightening of the core, from 'jet outbursts', which appear nearly synchronous at all frequencies and are accompanied by the emergence of new jet components and their subsequent evolution. The outbursts in 0528+134 and 2230+114 display fine structure and consist of individual sub- outbursts. Available 43-GHz Very Long Baseline Array images allow us to identify only one pure core outburst (in 2230+114) and one pure jet outburst (0458−020). Most of the outbursts analysed are mixed, in the sense that they display frequency-dependent time-delays (i.e. they are optically thick) and are associated with the eventual emergence of new jet components. The maxima of the jet and mixed outbursts probably correspond to epochs when newly ejected components become fully optically thin. These epochs are also marked by a significant in- crease in the angular velocities of the ejected components. There is evidence that the outbursts in 2230+114 repeat every 8.0 ± 0.3 yr, with the positions of individual suboutbursts being preserved from one quasi-periodic eight-year cycle to another, even though their amplitudes vary by more than a factor of 2. Preliminary estimates of the total durations of possible activity cycles based on an analysis of total flux-density variations and all available very long baseline interferometry data are given for the remaining sources.