Astron. Astrophys. 355, 915-921 (2000)

3. Discussion

The periodic light curve of OJ287 was explained by the beaming of the optically thin radiation originating in a relativistic precessing jet as it approaches the line of sight. The work of Katz (1997) demonstrated the possibility of having precession periods of only a few years in a binary black hole system when the accretion disk of the primary does not coincide with the orbital plane. Pairs of black holes can be formed as a consequence of interactions and eventually merging of galaxies. The difference in position between the BL Lac object OJ287 and the center of the faint underlying nebulosity could be considered as observational evidence of this merging phenomena (Heidt et al. 1999).

A moderate value of was found for the jet, in contrast to the value of 20 estimated by Katz (1997). The discrepancy is due to a misinterpretation of the relation between the duration of the periodic optical flares and the orbital phase, since it was not taken into account the fact that the precession velocity is not constant in the observer reference frame (Abraham & Carrara 1997).

The projected angle of the jet in the plane of the sky vary between and . The kpc scale jet is initialy directed along (Perlman & Stocke 1994), well within the allowed jet directions. Gabuzda & Cawthorne (1996) found two new features (K4 and K5) in the VLBI map of OJ287 for epoch 1990.47. Their position angles are compatible with the precessing jet model, allowing for the uncertainties introduced by the poor coverage of the uv plane in the north south direction. The predicted velocity would be , not very different from the velocities of the other features. K5 can be identified with feature A in the 43 GHz map at epoch 1991.27 (Tateyama et al. 1996) and in the maps of Marscher & Marchenko (1997) as the features furthest from the core in the first four maps.

Close to the epoch of the 1994 flare, it was found a periodicity in the formation of superluminal features of about one year in the reference frame center at the source, which corresponds to about in the precessing jet phase. The superposition of closely spaced trajectories at the epoch of maximum approach to the line of sight can account for the second flare, observed both at radio and optical wavelengths, about one year after the periodic flares.

The superluminal features found by Tateyama et al. (1999) in the analysis of Geodetic VLBI data at 8.3 GHz have velocities larger than those found by Gabuzda & Cawthorne (1996) at earlier epoch, which were used in the fitting of our model. We belive that the discrepancy is due to the small number of components allowed in the model fitting and the incomplete coverage of the time series in the Geodetic data.

The angle between the jet and the line of sight varies between and , the corresponding Doppler factors, for are, respectively, 12.8 (during the outbursts) and 1.1 (most of the time). Madejski & Schwartz (1983) calculated the beaming factor from the ratio of the measured X-ray fluxes and those expected from the synchrotron self-Compton process and obtained the values of 0.47 and 1.7, depending on the choice of the synchrotron self absorption frequency. Madau et al. (1987), on the other hand, reported a value of 8.7 using similar data, the difference being the choice of the geometrical parameters. Considering the large uncertainty in the values estimated for , the prediction of the precessing jet model seems reasonable, it accounts for boosting and fast variability during the strong flares and it can probably accommodate the observed X-ray emission when no beaming is present.

The model can also explain the high degree of polarization detected in component K2 by Roberts et al. (1987), if the shock moves with a speed in the reference frame of the jet. The Doppler factor introduced by this velocity is very close to unity and does not affect the boosting properties of the model.

© European Southern Observatory (ESO) 2000

Online publication: March 21, 2000
helpdesk.link@springer.de