4. Implications for the models
There are several observational aspects of the fades that models have to explain:
Some of these features are probably connected, e.g., the missing 1998 radio fade is most probably connected to the low flux level since 1996.
Could we explain the 1998 observations, and the reoccurence patterns in the context of the possible scenarios presented by Takalo et al. (1990)? The obscuring dust cloud theory would have to presume a series of clouds with just the right properties to get the observed light curves in optical and radio. A binary black hole could achieve the variations by distorting the accretion disk of the primary in a "switching-off" model. The jet could also be misaligned in a way to produce the observed features. But all these suggestions lack a detailed model that gives quantitative answers and they do not address the missing radio fade. The binary black hole models do however explain the reoccurence.
In the binary black hole model by Lehto & Valtonen (1996) the companion is thought to be at such a high inclination that it goes very close to the jet. As the radio emission regions in the jet are thought to be further in the jet than the optical regions, the companion could travel between the two regions which could explain why there was no radio fade in 1998 (Valtonen et al. 1999). The reason for this is that the 1989 fade, or the eclipse according to the model, happened very close to the apocenter, but in 1998 the orbit has precessed so much that the eclipse occurs closer to the primary black hole. When the companion approaches the jet of the primary black hole, it must have an effect on the direction and other properties of the jet. The slow approach could cause the observed steady decline, which ends when the companion is too close to the jet. Then the jet changes direction rapidly and at some stage points towards us causing the sharp peak. Then the accretion disk of the companion blocks the optical emission region, which is seen as the deep minimum. The radio emission is unaffected, because the radio emission region is further from the primary than the companion (Valtonen et al. 1999). The companion leaves the jet in tangled manner, which could be seen as the variations after the minimum and it should also affect the degree of polarization. Our polarization data does not cover that period. After the minimum, the flux of OJ 287 should gradually stabilize. As the companion is still moving away from the primary black hole, the brightness of OJ 287 should stay at low values for some years. The next predicted crossing of the disk in 2006 occurs far in the disk and should not produce large flares (Pietilä 1998).
Pietilä (1998) has studied the parameter space of the Lehto & Valtonen (1996) model. The implications of the timing of the 1998 fade for the range of acceptable parameters can be found by looking at those solutions that have the required eclipse time. The solutions that have an eclipse time between are presented in Table 3. The parameters can vary in a narrower range, but they are similar to the original values of Pietilä (1998). The only bigger difference is in the range of the mass accretion rate. Without the restriction above, the mass accretion rate was not much limited; the range was . But the late eclipse time seems to favor only those solutions that have very small accretion rates: . This eclipse time also rules out small viscosity coefficients: .
Table 3. The range of parameters for acceptable solution in the Lehto & Valtonen (1996) model.
© European Southern Observatory (ESO) 1999
Online publication: April 28, 1999