There is a broad understanding of the processes in blazars that give rise to the high energy component that is seen. Current models ascribe this high energy emission to production by inverse Compton scattering of low energy photons by the relativistic jet. These soft photons may be either the synchrotron photons themselves (the SSC model) or photons produced in the disc or broad line region.
The peak frequency of this Compton component is then determined by the position of the lower energy synchrotron peak. A simple model has recently been introduced to account for the phenomenology of gamma-ray bright blazars (Fossati et al. 1998 , Ghisellini et al. 1998). This model predicts that
Stecker et al. (1996) have made predictions of the VHE flux from these objects, using a simple model for the VHE emission and taking into account absorption on the IR background. They use simple arguments to relate the VHE fluxes to the X-ray flux, assuming that the emission is similar to that observed for Mrk 421.
It should be emphasised that the models of Fossati et al. (1998), Ghisellini et al. (1998) and Stecker et al. (1996) are descriptive of classes of blazars, but do not provide accurate predictions for individual objects. Naturally, the models are consistent with the detailed TeV spectra available for Mrk 421 and Mrk 501, but it is far from clear that all blazars can be fully characterized by the synchrotron luminosity and peak energy flux frequency alone. A further caveat when using descriptive models to predict the flux for an individual object arises from the fact that the compiled SEDs are not derived from simultaneous measurements and may be subject to significant observing biases. Many EGRET blazars are only seen in outburst, whereas the radio measurements may be dominated by quiescent emission. Even with these reservations, it is still helpful to consider the order of magnitude predictions generated by applying these models to these objects.
PKS 2005-489 shows very different temporal behaviour in outburst to that of either Mrk 421 or Mrk 501. The latter two objects both show short timescale ( 15 min) variations, whereas the X-ray light curve of PKS 2005-489 during the 1998 outburst shows no evidence for any such variations. This suggests that a detailed model of this particular object must invoke factors that explain this difference and which may also affect the TeV emission. The VHE observations ended well before the peak of the outburst, so the upper limits given do not constrain correlation models as strongly as would be the case if the outburst had occurred three weeks earlier. Further VHE observations, even at the same sensitivity, would therefore be useful in the context of a multi-wavelength campaign.
In Fig. 3 we show the broadband spectral energy distributions (SEDs) for PKS 2005-489 and PKS 0548-322. Data points at other wavelengths are taken from the compilation of Fossati et al. (1998) which are obtained by averaging all available data in wavelength bands. In plotting points from the present work we assume a spectral index of -1.6. Since our VHE points are derived from exposures over a period of years, it is appropriate to add our data to these compilations. We note that our new limits to VHE emission provide a much stronger constraint to the SEDs than previously reported results.
We can use these SEDs to test whether our measurements suggest that these objects conform to the predictions of the model of Fossati et al. (1998). For PKS 0548-322, the synchrotron peak is seen to occur at a frequency of Hz. This in turn implies that the Compton peak should occur at Hz or TeV. The radio flux at 5 GHz is measured to be (Kuhr et al. 1981 , Stickel et al. 1991 , Giommi et al. 1995) which leads to an expected flux at the Compton peak (0.6 TeV) of .
Similarly, for PKS 2005-489, the SED (Fig. 3) indicates that the synchrotron peak occurs at about Hz. The model of Fossati et al. 1998then predicts that the Compton peak will occur at about Hz, i.e. GeV. The radio flux at 5 GHz is measured to be (Sambruna et al. 1996) which leads to an expected flux at the Compton peak (20 GeV) of . Neither of these predictions is in conflict with the limits to VHE emission from these objects.
Using the Stecker et al. (1996) model we predict values of for the VHE flux from PKS 2005-489 and for PKS 0548-322. Our experimental upper limits are not in conflict with these predictions.
Using an inhomogeneous jet model (due to Ghisellini et al. 1985) Sambruna et al. (1996) have modelled the expected VHE flux from PKS 2005-489. They predict a sharp cut-off in the Compton peak at about 100 GeV. Again, this is not in conflict with the present results.
© European Southern Observatory (ESO) 2000
Online publication: January 29, 2001