Subject: Porosity modified X-ray line profiles From: Stan Owocki Date: Thu, 12 Jan 2006 01:45:29 -0500 To: Cohen David CC: Townsend Rich , ud-Doula Asif , Cutright Dan , Gagne Marc David, Attached below are plots of the porosity-modified line profiles, using the kappa_eff/kappa=1/(1+taub) formalism. The basic parameters are Ro=1.5, qq=0, and taustar = 0.1, 1, 3, 5, 10 (black, blue, violet, red, green), with then porosity length scaling as h=hp*r, with hp=0, 0.1, 0.25, 0.5, 0.75, and 1, 1.5, 2, 5, 10 plotted in each of the panels below. Note that the line profiles are self-normalized to their own peak, but to keep the curves distinct, I've applied a scale factor 1.05,1, .95,.9, .85 for the black, blue, violet red, and green curves (for taustar = 0.1, 1, 3, 5, 10). The basic results are as we expected. The models with porosity length much smaller than the local radius, e,g. hp=0.1 and hp=0.25 (2nd and 3rd panels) are very similar to the non-porous model (1st panel). For hp of order unity, the profiles become more symmetric, but the taustar >= 1 still have notable asymmetry. Indeed, you don't recover the full optically thin case, i.e. similar to taustar = 0.1, until *very* large porosity slopes, i.e. hp= 5 or 10 (last two curves). It should be relatively straightforward to extend Roban Kramer's profile fitting codes to include this formulation of porosity to derive formal constraints on this porosity length slope. Clearly fitting the nearly symmetric profiles with models of expected taustar ~= 5-10 will require very large porosity slopes, i.e. hp >~ 1. As I noted, it seems that a radially constant porosity length does not yield an analytical optical depth integral, and so unfortunately it will be quite a bit more expensive to compute models for that case. Stan pastedGraphic.tiff Content-Type: image/tiff Content-Encoding: base64