Subject: z2p5 notes #3 Date: Tue, 24 Mar 1998 15:11:38 +0100 From: Bob Fosbury Organization: ST-ECF To: Andrea Cimatti , Bob Fosbury , Bob Goodrich , Joel Vernet , Marshall Cohen , Montse Villar , Sperello di Serego Alighieri Comments on notes from Montse... Montse Villar-Martin wrote: > > > Dear Bob, thanks for the information. I'll contact Andrea to arrange > my accomodation in Florence (I have not arranged my flight yet). > > I've read your notes. I send you here a few 'fast thoughts'. > > I've been thinking of an explanation for the > broader and fainter Lya in objects with higher polarization. It is interesting. > Unless the obtects are broad line radio > galaxies (CIV should answer this) Lya should be broader than other lines > in objects where geometry is such that makes the escape difficult for > the Lya photons. I have thought of two possibilities: a) a back-cone dominated > geometry b) a closed geometry I have had a preliminary look at the data. It is not so easy because the line profiles are complex, but if I just consider the FWHM, Ly-a is always broader - by up to a factor of two - than CIV which is, in most cases, broader than HeII. There is no correlation between line width and continuum polarization. > > a) Back cone dominates the emission nb. In Montse's terminology, the 'back' cone is the one closest to the observer - where we see the unilluminated faces of the clouds. > > Imagine a simple situation in which the absorbing (neutral H) > and emitting (ionized H) gas are at the same redshift. Let's forget about > dust now. The resonant scattering of the Lya photons by neutral H > produces a broadening of the line: the photons are scattered in space, but > also in frequency (they are papers published about this). A photon in > the core of Lya will be able to escape through the neutral gas when it > is scattered in frequency towards the wings of the line, where the > absortpion efficiency is much lower. More photons in the wings and less > in the core means that the line profile should be broader than other lines > and with absorption features. This can happen also from the 'front' (far) cone if there is neutral gas outside the cone. > > Therefore, broader Lya could be due to a geometry such that > the photons are scattered once and again till they manage to escape > through the neutral H. This could be the case of an object where the back > cone dominates the emission. This would also explain the fainter Lya > (many of the core photons escape in opposite direction to the observer, > towards the front, before they are scattered to the wings). > > In summary, the broadening+faintness of Lya could be explained > if the back cone dominates the line emission (something perfectly plausible, > as the asymmetries observed by McCarthy suggest). No dust is needed, only > neutral H. Dust is not rejected. > > The observed polarization is the reason to believe dust to be present. > Back dominated geometry would explain the higher polarization in objects with > faint Lya, if forward scattering is more efficient. We only get higher polarization in this case because the scattered component is boosted in intensity with respect to other continuum sources. The degree of polarization, however, decreases as we depart from 90 deg scattering doesn't it? > > b) the NLR dominates the emission you mean a small region near the nucleus? > > This possibility is more difficult for me to accept, although I still > think that the strenght of NV in the objects with higher polarization > suggests that they might be dominated by the nuclear emission > (NV is unresolved in TX02...) instead of the EELR. In this > case we might be dealing with a rather closed geometry, which could also > explain the broad Lya if there was no dust. In this case, the broadening > effect would be even more efficient, since there is no way for the photons > to escape unless the are scattered to the wings. I find difficult > to explain the faintness of Lya IN THE ABSCENCE OF DUST, unless > it is redistributed due the scattering by neutral H in a larger spatial region > which is not included totally in the slit (we miss flux, therefore, compared > to the other lines). In TXS02... is more extended than NV. We need to look at the spatial distribution of the different lines in our spectra and also at the HST images (which are mostly continuum). > > If dust exists (as polarization suggests) I think that the > broadening would not take place, since the photons would be destroyed > before they are scattered into the wings. Who knows... maybe the stronger > absorption features on Lya in this objects will produce an apparantely > broader profile. Yes, I think this may be the explanation for what we see. > Another problem I see with the > closed geometry is to explain the higher polarization since a large number > of opening angles is included in the slit due to the fact that we are in > the inner parts of the cones. This would dilute (I think) the polarization > unless a beamed component is present, which complicates things quite a lot ... > > I'll think also about the question of the the similar > continuum curves for the different objects. That result puzzles me. I'll > think about the explanation you propose. > > This is all for now... All the best > > Montse > > -- R A E Fosbury (Bob) Space Telescope - European Coordinating Facility rfosbury@eso.org, http://ecf.hq.eso.org/~rfosbury/ Tel:+49 89 320 06 235 (o) +49 89 609 9650 (h) Fax:+49 89 320 06 480