DEPARTMENT OF PHYSICS AND ASTRONOMY

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

FUV Line Identification

David BoyceDavid BoyceMartin Barstow, Paul Martin Barstow, Paul DobbieDobbie, Anne Forbes, Anne ForbesXX--ray and Observational Astronomy Groupray and Observational Astronomy Group

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

FUV lines in white dwarf spectraFUV lines in white dwarf spectra

Can we distinguish between Can we distinguish between photosphericphotospheric lines and ISM lines (without modelling)?lines and ISM lines (without modelling)?

Should we rely on modelling alone to identify Should we rely on modelling alone to identify stellar content?stellar content?

Why do we not detect the same line richness Why do we not detect the same line richness with FUSE as we do with HST?with FUSE as we do with HST?

Have we identified everything? Have we identified everything?

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Introduction Introduction --

The white dwarf spectrum in The white dwarf spectrum in the FUVthe FUV

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Velocity analysis of FUV lines in G191-b2b

0

5

10

15

20

25

30

35

40

900 950 1000 1050 1100 1150 1200

Wavelength/Angstroms

Vel

ocity

/km

s-1

Photospheric LinesISM lines

Introduction Introduction --

Why some white dwarfs are Why some white dwarfs are better than others for absorption workbetter than others for absorption work

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Processing and candidate selectionProcessing and candidate selection

90 white dwarfs from FUSE Multimission 90 white dwarfs from FUSE Multimission Archive, hosted by the Space Telescope Archive, hosted by the Space Telescope InstituteInstitute

The data has been calibrated using the calfuse pipeline version 2.0.5 or later, resulting in several spectra (covering different wavelength segments) per FUSE exposure. The exposures for each segment are then co-added, weighting each according to exposure time, and finally the segments are combined, weighted by their signal-to-noise ratio, to produce a single spectrum

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

High velocity component separation High velocity component separation candidatescandidates

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

RE0230z

G191-b2b

wd0621-376

wd2211-495

Finding the spectra forest in the FUVFinding the spectra forest in the FUV

CoCo--addition of addition of multiple multiple observations observations reveals spectral reveals spectral forest forest

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Velocity Velocity componentcomponent

identificationidentification

vcc+

=λλ0

From well known lines we know an average velocity for the star and the ISM in that direction

Pho

tosp

heric

ISM

Pho

tosp

heric

ISM

Line=photospheric

Pho

tosp

heric

ISM

Line=ISM

Graph of predicted rest wavelength

Using multiple stars we can differentiate between an unknown line being Photospheric or ISM without even knowing what it is.

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

G19

1-b2

bw

d221

1-49

5w

d062

1-37

5Calculate predicted Lab wavelength for all the stars that contain the line

Line Identification using multiple stars of Line Identification using multiple stars of different different VVphotphot

Reality Check5000 likely photospheric species in line list (Such as Ni IV, Fe V)A likely line should occur every 0.05 Angstrom

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Distance to nearest likely line (Data from wd2211-495 and wd0621-376)

-0.1

-0.05

0

0.05

0.1

0.15

920 940 960 980 1000 1020 1040

Wavelength/Angstroms

Diff

eren

ce b

etw

een

pred

ictio

n an

d la

b w

avel

engt

h/An

gstro

ms

In 98% of cases a line is found before one is expected.

A line is normally found within 0.02 Angstroms of where one is predicted

Difference between predicted rest Difference between predicted rest wavelength and first likely line wavelength and first likely line

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Velocity plot distinguishes Photospheric and ISM lines in this star

Wavelength against velocity Wavelength against velocity

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Some spread in velocity as the lines become comparable in size to noise.

Line strength against velocity for one starLine strength against velocity for one star

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007Velocity of 1031.912 & 1037.613 O VI lines suggest photospheric origin

Line strength against normalised velocity Line strength against normalised velocity for all starsfor all stars

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Averaged velocity across high radial velocity star reveals constantly outlying lines.

Line by line normalised velocity averageLine by line normalised velocity average

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Outlying line velocities indicative of

1.) Merged lines

2.) Lines with dual contributions

3.) Incorrect identificationse.g. C III at 927.02 contains ISM and photospheric component. Apparent velocity therefore a blend of the two

Outlying line feedback processOutlying line feedback process

DEPARTMENT OF PHYSICS AND ASTRONOMY

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

O IO I O I O I O I O I O I O IO I O I

O I

O I

O I

N I

O I

N I

Ar I

S VI

S VI

S IV

C III

O IVO III O III

O V O V

O VIO III O IV O V O IV

O IVO IV

P IV

P IV

Si II Si V

Si III Si IV N II

N IV N IV

N IVN IVN IV N IV

N III N I

C IIC II P IV

Al V

Ni V

Ni V

Ni IVNi IV

Ni V Ni IVNi IV

Ni V

Ni IV Ni IV Ni IVNi IV

Ni IV

Ni V

Ni IV Ni IV Ni V Ni III Ni IV

Ni IIINi VI

Ni IV Ni IV

Ni V Ni VI Ni III Ni III Ni III

Fe V Fe III Fe III

Fe III Fe III

Fe II

Fe III

Fe III

Fe IV Fe IV

Fe III

Fe III

Fe IV

Fe III

Fe III Fe IIIFe IV

Fe III

Fe IV Fe IVNi V

Fe III

Fe IV

Fe IIIFe IV

Fe IIIFe IV

Fe III

Fe III

Fe IVFe IV

Fe IVFe III

Fe IVFe III

Fe IV Fe V

Fe III

Fe III

Fe III

Fe V Fe V

N INi VI

Fe IIIFe V

Fe III

DEPARTMENT OF PHYSICS AND ASTRONOMY

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Results from velocity analysisResults from velocity analysis

Spectral richness Spectral richness explainable with explainable with established chemical established chemical content. content.

No exotics requiredNo exotics required

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Spectral synthesisSpectral synthesis

Forest modelled more accurately by synspec model 13,000K cooler than the stars effective temperature deduced from Lyman line fitting

Iron Nickel model Teff 62,000K Teff =49,000K

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Conclusions Conclusions ––

What this means for White What this means for White DwarfersDwarfers

Method for identifying linesMethod for identifying lines

No exotic elements requiredNo exotic elements required

Spectral Synthesis of the spectral forest Spectral Synthesis of the spectral forest suggests stratification suggests stratification –– discrepancy of discrepancy of temperature (and therefore depth) between line temperature (and therefore depth) between line forming regionsforming regions

PhD Recruitment Day PhD Recruitment Day –– 3131stst Jan 2007Jan 2007

Conclusions Conclusions ––

What this means for ISM What this means for ISM studiesstudies

Detected O VI is Detected O VI is photosphericphotospheric

Method for distinguishing which Method for distinguishing which velocity component an unidentified velocity component an unidentified line belongs toline belongs to

No modelling of background source No modelling of background source required. required.

DAVID BOYCE db139@star.le.ac.uk