From - Fri Aug 27 12:03:18 1999 Received: from eso-wall-ext.hq.eso.org by web3.hq.eso.org with SMTP (8.8.5/ eso-5.3) id PAA11386; Thu, 26 Aug 1999 15:57:34 +0200 (MET DST) Return-Path: Received: by eso-wall-ext.hq.eso.org; id PAA09452; Thu, 26 Aug 1999 15:51:10 +0200 Received: from cairanne.iap.fr(194.167.0.198) by eso-wall-ext.hq.eso.org via smap (4.1) id xma009392; Thu, 26 Aug 99 15:50:40 +0200 Received: from pauillac.iap.fr (pauillac [194.167.0.134]) by cairanne.iap.fr (8.8.8/jtpda-5.2.9.2) with ESMTP id PAA09153 ; Thu, 26 Aug 1999 15:57:02 +0200 (MET DST) Received: from localhost (villard@localhost) by pauillac.iap.fr (8.8.8/jtpda-5.2.9.2) with ESMTP id PAA10917 ; Thu, 26 Aug 1999 15:57:02 +0200 (MET DST) Date: Thu, 26 Aug 1999 15:57:01 +0200 (MET DST) From: Montserrat Villard-Martin To: rfosbury@eso.org, jvernet@eso.org Subject: Depletion, some more results Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Status: RO X-Mozilla-Status: 8001 X-Mozilla-Status2: 00000000 X-UIDL: f661057ff119943e74e2902f570851c7 Bob, Joel, I have done some more work on the depletion models. I thought that it would be good to produce more realisitc models. The first thing I did was to check whether there was an U sequence that could explain the correlation (i.e. do we need variation in metallicity?). I run several U sequences with different mu and Z (I tried even different densities). I have not found a U sequence that can explain the correlation, unless I vary the metallicity. All the new models have the same input parameters, except metallicity and dust content, that vary from object to object. I varied the abundances quadratically for N and linearly for the other metals and considered that the mass of dust to gas increases linearly as well with metallicity: mu=0.4, Z=0.4 solar, N=0.16 solar mu=1, Z=1 solar, N=1 solar mu=2, Z=2 solar, N=4 solar mu=3, Z=3 solar, N=9 solar This considers the case such that objects with higher metallicities have more dust as well. The results I get are: - Low density models (n=100) underpredict NV/HeII and CIV/HeII (for most objects). Low densities do not work - We need densities n>5e4, this points towards NLR (rather than EELR) densities - A sequence such that mu and Z increase from 0.4 to ~2 and take depletion into account can explain rather well the correlation in the NV diagram. - Depending on the selected U, the metallicity range can vary, but I think that it will be in the range [1-3] or [0.4-2]. I do not think the results will be extremely different. - This sequence can reproduce the C/He ratios as well, but it produces CIV/CIII almost constant what might suggest a certain U effect. - I tried also sequences in U for fixed metallicities and dust contents. They can explain the range in but CIV/CIII but not the correlation. Do you want that I send the plots and have a look? I think these models are more realistic. They do not contradict the previous results, except for the fact that we need high densities. If you think it is useful, I can refine a bit more the models (a density a bit higher will probably fit everything better). Cheers Montse Subject: Re: Depletion, some more results Date: Fri, 27 Aug 1999 12:12:13 +0200 From: Bob Fosbury Organization: ST-ECF To: Montserrat Villard-Martin CC: jvernet@eso.org References: 1 Montserrat Villard-Martin wrote: > > - Low density models (n=100) underpredict NV/HeII > and CIV/HeII (for most objects). Low densities do not work > > - We need densities n>5e4, this points towards NLR (rather > than EELR) densities What is the physical effect of density on the models? Not sure I understand exactly how this works. If we really need high densities it means that we must be seeing close to the BLR. How can this be reconciled with extended line emission? > > - A sequence such that mu and Z increase from 0.4 to ~2 and take > depletion into account can explain rather well the correlation in the > NV diagram. > How different are these results from the previous models which did not include the depletion effects? > Thanks Montse, we really need to undertstand this - especially the need for high densities. Can our correlations be explained by varying the weighting between an extended, low density and a nuclear, high density gas? This could mean geometry rather than chemical evolution. We have mentioned this before but... Bob... --