2006A&A...456...97F

We report on VLBA observations of a γ-ray bright blazar NRAO 530 at multiple frequencies (5, 8, 15, 22, 39, 43, and 45GHz) in 1997 and 1999. These multi-epoch multi-frequency high-resolution VLBI images exhibit a consistent core-dominated morphology with a bending jet to the north of the core. The quasi-simultaneous data observed at five frequencies (5, 8, 15, 22, and 43GHz) in February 1997 enabled us to estimate the spectra of compact VLBI components in this highly variable source. Flat spectra are seen in the two central components (A and B), and the most compact component A that has the flattest spectral index at the south end is identified as the core. Based on the synchrotron cooling timescale argument, it is suggested that the observed inverted spectrum of component C is caused by the free-free absorption (FFA), although the synchrotron self-absorption (SSA) model cannot be ruled out definitely. While the SSA probably exists in component B, it is likely that the same FFA would produce the spectral turnover toward component B since the fitted FFA coefficients in both B and C components are almost the same. If so, the projected size of such an absorbing medium is at least 25pc. By adding our new measurements to previous data, we obtain apparent velocities of two components, B of 10.2c and E of 14.5c. These are consistent with the emergence of VLBI component associated with the flux density outburst; i.e. components B and E are related to strong γ-ray flares in 1994.2-1994.6 and 1995.4-1995.5, respectively. We investigate the spectral variability further by making use of the single-dish measurements covering a complete outburst profile from mid-1994 to mid-1998. It shows a continuous increase in the turnover frequency during the rising phase, and a gradual decrease after passing the peak of the flare. Finally, we discuss the equipartition Doppler-factor (δ_eq) based on analysis of the magnetic field and obtain δ_eqs of 3.7, 7.2, and 0.8 for components A, B, and C, respectively, which are all consistent with a larger flux density in component B, the non-detection of proper motion in component C, and a bent jet.