Ne X and Ne IX blending with nearby iron lines

Ne X: 

12.134 (emissivity weighted mean wavelength of the two doublet components) (12.1321, 12.1375)

Fe XVII: 12.1240 (4C: peaks at 0.544 keV, same as Ne X)
Fe XVII: 12.2660 (4D: peaks at 0.544 keV, same as Ne X) - about 85% as strong as the previous line
Fe XXIII: 12.1610 (peaks at l.366 keV)


Ne IX (peaks at 0.343 keV): 

13.4473
13.5523 (13.5503,13.5531)
13.6990


Fe XIX: 13.4620 (peaks at 0.685 keV; 0.185 strength of Ne IX res.)
Fe XIX: 13.4970 (peaks at 0.685 keV; 0.325 strength of Ne IX res.)
Fe XXI: 13.5070 (peaks at 0.862 keV; 0.397 strength of Ne IX res.)
Fe XIX: 13.5180 (peaks at 0.685 keV; 0.717 strength of Ne IX res.)
Fe XIX: 13.6450 (peaks at 0.685 keV; 0.115 strength of Ne IX res.)
Fe XIX: 13.7315 (peaks at 0.685 keV; 0.129 strength of Ne IX res.)
Fe XIX: 13.7458 (peaks at 0.685 keV; 0.112 strength of Ne IX res.)
Fe XX:  13.7670 (peaks at 0.862 keV; 0.191 strength of Ne IX res.)
Fe XIX: 13.7950 (peaks at 0.685 keV; 0.401 strength of Ne IX res.)
Fe XVII: 13.8250 (peaks at 0.544 keV; 0.521 strength of Ne IX res.)




From Maurice: 

David,

here's my opinion on Walborn et al, regarding the Ne X / Ne IX ratio: for most O stars, Ne X and Fe XVII are observed. The  4C and 4D lines of Fe XVII definitely need to be accounted for in any assessment of the strength of Ne X Ly alpha - I think the 4C line has a strength of order 50% of the Ne X Ly alpha line for most astrophysical sources which have Ne / Fe in roughly solar proportion.

On the other hand, it's hard to say how important Fe XIX and higher charge states are in many O star spectra. The relatively weak Fe XVIII emission, as well as the weakness of S XV, Si XIV, and Mg XII, all suggest that Fe XIX is not too strong. Of course, it would be best to be quantitative, and it's hard to be sure of one's assessment of Ne X / Ne IX without carefully dealing with this problem.

But when it comes to profile fitting, you need to be even more sure that you understand any possible Fe XIX contamination, because a small flux in Fe XIX lines that might have a small effect on the Ne X / Ne IX ratio might have a large effect on line profiles, if you don't account for it properly.

So I would say that 1.) we should throw out Ne IX, or else present results that we "ghost out" in the tau_* vs lambda plots, along with an explanation of the blending issue; 2.) what Walborn et al. did is probably not all that wrong, at least in terms of the meaningfulness of their Ne X / Ne IX results, but their assessment of the errors associated with their method is not sufficient.

Maurice