Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.eso.org/~ppadovan/survey/dxrbs_paper/node17.html
Дата изменения: Wed Dec 3 16:31:47 2003
Дата индексирования: Sat Dec 22 22:52:53 2007
Кодировка:
Поисковые слова: m 45

The Browne & Marchã Effect

The BM effect causes low-luminosity blazars (particularly BL Lacs) to be missed in surveys because the apparent luminosity of the non-thermal nuclear source does not exceed that of the host galaxy by a large factor. X-ray and radio-faint surveys are the most heavily affected. The effect probably is more severe for LBLs than HBLs, since galaxies hosting a relatively weak non-thermal source of the LBL type (where the peak of the synchrotron emission is at energies lower than 4000 Å) might not qualify as a BL Lac simply because its nonthermal emission at 4000 Å  is already much reduced compared to its peak in the IR, and not strong enough to produce a Ca II H & K break less than 25%. Yet an HBL-type object with identical radio characteristics (flux, spectral shape, polarization, etc.) would be much more likely to be classified as a BL Lac since its synchrotron emission at 4000 Å  is much stronger and still growing.

We may gauge the impact of the BM effect upon our sample by considering the likely properties of such objects (low-luminosity blazars hidden within bright galaxies). In such a case, the optical spectrum would resemble that of a radio galaxy, either with broad, narrow, or no emission lines whatsoever (a similar argument, although for BL Lacs only, was given in Laurent-Muehleisen et al. 1997). We have attempted to eliminate this ambiguity for broad-lined sources by grouping all broad-line radio galaxies with the FSRQs (but see §7.2 below). It therefore remains for us to consider the narrow-lined objects which remain in our sample.

Three of our first 85 IDs (WGAJ0340.8-1814, WGAJ0500.1-3040, WGAJ2317.9-4213) meet this description and have herein been described as radio galaxies. A few other objects which we have tentatively identified as BL Lacs (Table 5) may fall into this classification when higher signal-to-noise spectra are taken (these have been individually discussed in §4.2). Ten of the previously identified sources also fall into this category as NLRGs. But given that these objects exhibit flat-spectrum radio sources, some of these objects may house low-luminosity blazars. This is particularly true of the ten previously identified sources, for which the NLRG classifications (taken from the literature) were made by older standards (usually - but not always - the classical definition mentioned in §3). Further observations should be made to further probe their nature.

We have plotted these objects in Figures 6 and 7. Inspection of Figure 6 reveals that these NLRGs are less luminous than BL Lacs on average, but not the least luminous objects in our sample. This result is consistent with the predictions made by Browne & Marchã (1993) and Marchã & Browne (1995). Perhaps more interesting is the fact that the large majority (10 of 13) of these NLRGs lie within the LBL region of the $(\alpha_{\rm ox},\alpha_{\rm ro})$ plot (Figure 7). This verifies our suspicion (above) that, since the synchrotron continuum produced by LBLs most often peaks in the near-to-mid infrared, and is already decreasing in the optical, the BM effect is stronger among LBLs than HBLs. Also noteworthy is the fact that the majority of these radio galaxies - 7 of 13 - are at low values of $\alpha_{\rm ro}$ and high values of $\alpha_{\rm ox}$. As with the BL Lacs which occupy this region of Figure 7, these objects are all at low z and therefore the optical fluxes are largely contaminated by the host galaxy.

In addition to these objects, a large fraction of the newly-identified BL Lacs in our sample (10 of 22; see Table 5) have been so classified only by virtue of our usage of the expanded Marchã et al. (1996) definition of the BL Lac class. These objects would not have been classified as BL Lacs under earlier, more restrictive definitions of the class (Stickel et al. 1991, Stocke et al. 1991, Perlman et al. 1996a), though they might have received some mention under such standards. The large fraction of objects falling in this category confirms the predictions of Marchã & Browne (1995) for low-flux-limit X-ray surveys such as this. Similar results were also found in the RGB BL Lac Survey (Laurent-Muehleisen et al. 1997), as well as the 200 mJy sample (Marchã et al. 1996). Note that Marchã et al. showed that at least some of the objects outside the ``classical'' BL Lac region of the $(C, W_\lambda)$ (but within the expanded Marchã et al. definition) also share the polarization properties of BL Lacs.

Returning to Figure 7, it is now important to point out that we suspect that the two ROSAT-based samples to which we compared the DXRBS BL Lacs in §6.2, those of Kock et al. (1996) and Nass et al. (1996), may contain a number ($\sim$ 20%, as in the EMSS; Stocke et al. 1997) of objects which could be classified as BL Lacs using the Marchã et al. (1996) redefinition of the BL Lac class. This is because both Kock et al. and Nass et al. used (somewhat unclearly defined) versions of the classical BL Lac-radio galaxy division to define their samples. It is also possible that these two samples may contain a few radio galaxies whose spectra should be more carefully scrutinized, as the regions of parameter space that they cover overlap significantly with the radio galaxies in our sample (Figure 7). Further evidence for this point can be seen in the recent findings of Laurent-Muehleisen et al. (1997, specifically their Fig. 3).