Subject: Re: f line fluxes From: David Cohen Date: Mon, 19 Nov 2007 23:54:33 -0500 To: Skinner Steve CC: "Gagne, Marc" , David Cohen Steve and Marc, Here's the fit to the Mg XI complex based on the numbers in the table of individual line fits that Steve sent on 16 November: http://astro.swarthmore.edu/~cohen/projects/sigOri/mgxi_steve_fit.png I've placed my own continuum under the three Gaussians. It's very low (6e-6 ph/s/A) and using something higher can only decrease our estimate of the f flux (and thus of f/i). I'll note that this model gives C=102 for 102 bins, whereas my best-fit model (somewhat narrower lines; centroids fixed at lab values; f line flux effectively zero) has C=83 for the same bins (i.e. very significantly better). I'd venture to say that the above model doesn't look like a plausible fit (because of the f-line). I think it matters quite a bit what the upper limit on the forbidden flux is, because there's a big difference between saying that the x-rays are coming from 2 Rstar vs. saying they must be closer in than, say, 1.25 Rstar. As you can see from the model figure I sent this morning: http://astro.swarthmore.edu/~cohen/projects/sigOri/sOriA_mgxi_model.png if f/i is ~0.2, then r ~ 2 Rstar, but if f/i is ~0.1, then r < 1.5 Rstar. I don't know what constitutes "close in" in the context of this star, but I certainly don't view sigma Ori A's X-rays as being anything like theta1 Ori C's. David Skinner Steve wrote: > > Marc- > > I agree that it does make a difference whether we > quote an upper limit for the f-line, or an actual > value - at least when we try to infer the line formation > radius (or an upper limit on the formation radius). > > It's probably safer (or at least more conservative) > to quote an upper limit because even if there is > some f-line flux there (and I think there could be), > it is certainly faint. > > Based on David's initial estimage, Mg XI is forming > very close in - but we already had some idea this > would be the case based on analysis of other young > O stars, > > Ultimately we are limited by the S/N of our data > and obviously it would be better if we had 10,000 > MEG1 counts rather than ~1900 . > > If you want to weigh in on whether there is any f-line > flux there using ISIS, then a third opinion would probably > be useful (but might be more like getting three philosophers > to agree on the time of day). Do you have > the spectra/rmfs/arfs? As I recall I only sent the .tar > files to David, but I can put them in your /incoming area > if you have one, or in my anon ftp area, > > > Steve > > > > > ******************************************************* > * * > * Steve Skinner * > * * > * Center for Astrophysics and Space Astronomy (CASA) * > * 389 UCB * > * University of Colorado * > * Boulder, CO 80309-0389 USA * > * * > * Tel. 303-492-4202 * > * Fax. 303-492-4052 * > * E-mail skinners@casa.colorado.edu * > * Web: http://casa.colorado.edu/~skinners/ss.html * > * * > ******************************************************* > On Mon, 19 Nov 2007, Gagne, Marc wrote: > >> Steve, >> >> I've only seen the figures that David made available in his last message. From those, it's hard to tell what the fuzz is, but it looks like there are essentially no counts at the wavelength of the Mg XI f-line. >> >> The disparity between David's upper limit of 0.2 and your detection at 0.33 is important because the two results put different constraints on the formation radius. Does the Mg XI f/i ratio provide the best constraint on R? If so, I'm glad we're trying to get to the bottom of this. >> >> The disparity is understandable, I think, because you've determined the excess flux in the bins around the line and David has asked, what is the maximum flux a Gaussian could have at the (exact) wavelength of the f line? >> >> Although both approaches are valid as long we explain what we've done, I think I'm more comfortable with David's upper limit because the excess flux method assumes that all the excess flux around the f line is actually from the f-line. >> >> The modified ISIS code I use could attack this issue a third way if you're interested. >> >> Marc >> >> Marc Gagné >> Department of Geology & Astronomy >> West Chester University >> (610) 436-3014 >> >> >> -----Original Message----- >> From: Skinner Steve [mailto:skinners@Colorado.EDU] >> Sent: Monday, November 19, 2007 3:13 PM >> To: David Cohen >> Cc: Gagne, Marc; Steve Skinner >> Subject: Re: f line fluxes >> >> >> David: >> >> I took a second look at the Mg XI f-line. >> I am inclined to leave the Table the way it >> is for now, i.e. with a f-line flux 0.13e-5 ph/cm2/s >> (as opposed to an upper limit). >> >> I estimated the continuum flux over the E-range >> 1.4 - 1.8 keV and got F(contin) = 0.8 +- 0.1 e-5 ph/cm2/s >> (1.4 - 1.8 keV, or delta_E = 0.4 keV). >> I used a power law model (range: 1.4 - 1.8 keV) and a >> (global) bremss fit to the spectrum to get this estimate. >> Both of these models will overestimate >> the continuum due to line contriubtions, so the above value >> is really an upper limit to the continuum. >> >> A 3-component Gaussian model of the Mg XI triplet gives >> a total (line + cont) flux F(1.3291 - 1.3331 keV) = 0.135e-5. >> Based on the above continuum estimate (or rather the estimate >> of the upper limit on the continuum), I would >> expect no more than >> about 8e-8 ph/cm2/s of continuum in the 1.3291 - 1.3331 keV >> range. So, it looks like the region around the f line >> is well above the continuum, and I think we should thus quote a >> line flux (0.13e-5) rather than an upper limit. >> >> Note that the energy range I used to measure the line flux >> 1.3291 - 1.3331 keV (9.3005 - 9.3285 Ang.) has an E-width >> of 4e-3 keV, which is generous. It is about 33\% wider than >> the FWHM of the r line (FWHM ~ 3e-3 keV). >> So, I think >> we can safely say the f/i flux ratio is no greater than >> f/i = 0.13e-5/0.39e-5 = 0.333 for Mg XI. >> This is well below the theoretical low density limit, >> which I believe is something like R = 2.7 for Mg XI. >> >> I'll take a second look at Ne IX as well, but it is more difficult >> because of all the Fe line contamination. >> >> Steve >> >> >> ******************************************************* >> * * >> * Steve Skinner * >> * * >> * Center for Astrophysics and Space Astronomy (CASA) * >> * 389 UCB * >> * University of Colorado * >> * Boulder, CO 80309-0389 USA * >> * * >> * Tel. 303-492-4202 * >> * Fax. 303-492-4052 * >> * E-mail skinners@casa.colorado.edu * >> * Web: http://casa.colorado.edu/~skinners/ss.html * >> * * >> ******************************************************* >>