ξ Per: X-ray line profiles

Background

The Chandra grating data presented here comprises Obs. ID 4512, for a total effective exposure time of 158.8 kiloseconds. This is the same reduction that Emma used in her fitting.

The line profile fits we present here include a +57 km/s shift to account for the radial velocity of the star. The value is taken from Hoogerwerf et al. (2001) (which uses an average from several other studies) and corrected for the spacecraft's motion around the solar system barycenter (using this tool; the correction was less than 2 km/s).

We expect the inclusion of the redshift to increase the taustar values, as moving the line center to the red will move the line peaks to the blue.

Entire HETGS spectrum

MEG
MEG: entire spectrum

HEG
HEG: entire spectrum
The vertical dotted lines indicate the (rest) wavelengths of emission lines we fit. Lines that are not indicated are too weak or too blended to fit. In addition to these lines, we also fit the Ne IX He-beta line at 10.239 A.

Si XIII He-alpha: 6.6479, 6.6866, 6.7403 Angstroms

MEG
Si XIII MEG

HEG
Si XIII HEG 		 
 
[6.58:6.81], λo = 6.6479, 6.6866, 6.7403
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 1.29e-4
q = 0
hinf = 0
shift = 1.3 mA
taustar = 0.00   +/- (0.00:0.03)
      (0.00:0.14) at 95% confidence
Ro = 1.47   +/- (1.40:1.56)
norm = 9.67e-6 +/- (8.95e-6:1.03e-5)
G = 1.26   +/- (1.07:1.50)
rejection probability = ???% (C = 297.19; N = 272)
 
compare to Emma's fit

8.4210 Angstroms: Mg XII

MEG
Mg XII 8.4210 MEG

HEG
Mg XII 8.4210 HEG
[8.34:8.50], λo = 8.4210
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 1.97e-4
q = 0
hinf = 0
shift = 1.6 mA
taustar = 0.00   +/- (0.00:0.20)
      (0.00:0.76) at 95% confidence
Ro = 1.54   +/- (1.39:1.92)
norm = 2.61e-6 +/- (2.18e-6:3.10e-6)
rejection probability = ???% (C = 145.88; N = 188)
 
compare to Emma's fit

Mg XI He-alpha: 9.1687, 9.2297, 9.3143 Angstroms

MEG
Mg XI MEG

HEG
Mg XI HEG 		 
 
[9.08:9.40], λo = 9.1687, 9.2297, 9.3143
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 2.60e-4
q = 0
hinf = 0
shift = 1.7 mA
taustar = 0.40   +/- (0.22:0.62)
Ro = 1.59   +/- (1.52:1.68)
norm = 2.42e-5 +/- (2.29e-5:2.53e-5)
G = 0.96   +/- (0.83:1.09)
rejection probability = ???% (C = 367.75; N = 380)
 
compare to Emma's fit

10.239 Angstroms: Ne X

MEG
Ne X 10.239 MEG
[10.12:10.34], λo = 10.239
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 1.97e-4
q = 0
hinf = 0
shift = 1.9 mA
taustar = 0.42   +/- (0.03:1.26)
Ro = 1.47   +/- (1.30:1.70)
norm = 1.51e-6 +/- (1.94e-6:2.98e-6)
rejection probability = ???% (C = 111.72; N = 86)
 
Emma did not fit this weak line.

11.544 Angstroms: Ne IX

MEG
Ne IX 11.544 MEG

HEG
Ne IX 11.544 HEG
[10.12:10.34], λo = 11.544
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 4.63e-4
q = 0
hinf = 0
shift = 2.2 mA
taustar = 0.15   +/- (0.00:0.50)
Ro = 1.50   +/- (1.36:1.75)
norm = 9.74e-6 +/- (7.09e-6:1.23e-5)
rejection probability = ???% (C = 178.86; N = 244)
 
compare to Emma's fit

12.134 Angstroms: Ne X

MEG
Ne X 12.134 MEG

HEG
Ne X 12.134 HEG
[12.01:12.18], λo = 12.1339
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 3.58e-4
q = 0
hinf = 0
shift = 2.3 mA
taustar = 0.73   +/- (0.50:0.99)
Ro = 1.55   +/- (1.47:1.67)
norm = 4.02e-5 +/- (3.80e-5:4.31e-5)
rejection probability = ???% (C = 185.28; N = 200)
 
compare to Emma's fit

Ne IX He-alpha: 13.4473, 13.5523, 13.6990 Angstroms

MEG
Ne IX MEG

HEG
Ne IX HEG 		 
 
[13.32:13.55], λo = 13.4473, 13.5523, 13.6990
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 6.19e-4
q = 0
hinf = 0
shift = 2.6 mA
taustar = 0.00   +/- (0.00:0.04)
      (0.00:0.14) at 95% confidence
Ro = 1.63   +/- (1.57:1.70)
norm = 1.17e-4 +/- (1.10e-4:1.25e-4)
G = 0.97   +/- (0.83:1.14)
rejection probability = ???% (C = 260.81; N = 272)
 
compare to Emma's fit

Note that this is one case where the shifted model led to a lower taustar.

15.014 Angstroms: Fe XVII

MEG
Fe XVII 15.014 MEG

HEG
Fe XVII 15.014 HEG
[14.88:15.08], λo = 15.014
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 8.75e-4
q = 0
hinf = 0
shift = 2.8 mA
taustar = 0.22   +/- (0.10:0.36)
Ro = 1.56   +/- (1.47:1.67)
norm = 9.92e-5 +/- (9.31e-5:1.06e-4)
rejection probability = ???% (C = 229.59; N = 236)
 
compare to Emma's fit

16.780 Angstroms: Fe XVII

MEG
Fe XVII 16.780 MEG
[16.63:16.91], λo = 16.780
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 5.65e-4
q = 0
hinf = 0
shift = 3.2 mA
taustar = 0.25   +/- (0.06:0.56)
Ro = 1.34   +/- (1.26:1.47)
norm = 3.95e-5 +/- (3.35e-5:4.40e-5)
rejection probability = ???% (C = 109.23; N = 110)
 
compare to Emma's fit

Emma fit this feature simultaneously with the Fe XVII complex at 17.05/17.10 A.

17.051, 17.096 Angstroms: Fe XVII

MEG
Fe XVII 17.051, 17.096 MEG
[16.91:17.25], λo = 17.051, 17.096
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 5.65e-4
q = 0
hinf = 0
shift = 3.2 mA
taustar = 1.03   +/- (0.61:2.31)
Ro = 1.74   +/- (1.50:2.00)
norm = 6.46e-5 +/- (5.99e-5:6.97e-5) for the 17.051 A line, and 0.6* for the other line
rejection probability = ???% (C = 128.83; N = 130)
 
compare to Emma's fit

Emma fit this feature simultaneously with the Fe XVII line at 16.780 A.

18.969 Angstroms: O VIII

MEG
O VIII 18.969 MEG
[18.80:19.13], λo = 18.9689
vinf = 2450
β = 1
powerlaw continuum, n = 2; norm = 4.70e-4
q = 0
hinf = 0
shift = 3.6 mA
taustar = 0.40   +/- (0.21:0.66)
Ro = 1.67   +/- (1.58:1.79)
norm = 1.91e-4 +/- (1.76e-5:2.05e-5)
rejection probability = ???% (C = 107.39; N = 130)
 
compare to Emma's fit


A log of the spectral fitting is available.

Τ★ results

Comparison of Emma's original taustar values with those derived above, assuming a +57 km/s shift for all lines.

taustar comparison

Emma's results with no shift assumed are the open squares, the results with the shift are the filled circles. Note that Emma omitted the two He-like complexes, Si XIII and Ne IX, near 6.7 and 13.5 Angstroms, and also combined the Fe XVII lines near 17 Angstroms into a single fit result, whereas I fit the 16.780 A line separately from the blended 17.051 and 17.096. Emma's results are shifted horizontally by 0.1 A to facilitate comparison between the two sets of results.

It's clear that including the (positive) shift tends to increase taustar for each line, though generally the results for any individual line agree within the error bars, in the aggregate, the difference is modest but real.

From each of these ensembles of optical depth values, we can derive a mass-loss rate. We use a solar abundance opacity model (CNO processing would not make a difference, as none of the lines are longward of the oxygen edge, so the overall opacities are just about the same; although a non-solar overall metallicity would alter the results) for the results shown below. We also compare the derived mass-loss rate's predicted taustar vs. wavelength to that expected from the literature value of 1.1 X 10-6 Msun/yr (Repolust et al. 2004). We assume a stellar radius of 25.5 Rsun (Haser 1995).

Emma's results that do not assume a velocity shift:

Mdot from taustars, no shift

Mdot = 2.76 X 10-7 Msun/yr, with a 68% confidence range of 1.82 X 10-7 to 3.68 X 10-7, and chi-square = 5.4 (for N = 7).

Results that do assume a velocity shift:

Mdot from taustars, with shift

Mdot = 3.28 X 10-7 Msun/yr, with a 68% confidence range of 2.42 X 10-7 to 4.17 X 10-7, and chi-square = 14.4 (for N = 10). Note that for this fit, we have excluded the Ne IX complex (as Emma did) for which blending with iron lines can be a problem. We have included the Si XIII complex, which Emma did not include. If we exclude that too, then the mass-loss rate goes up to Mdot = 4.12 X 10-7 Msun/yr. (On the other hand, if we include it and also include the Ne IX complex, we get Mdot = 1.16 X 10-7 Msun/yr.)

So, the fitted taustar values derived assuming the 57 km/s shift, give a higher mass-loss rate (marginally statistically significant if we exclude the Si XIII result, and not even statistically significant if we include it).

Note: we have to decide whether to accept the Si XIII result or not. The fit looks poor and C > N by quite a bit.

 

last modified: 31 October 2011