Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://star.arm.ac.uk/~csj/lines/lte/lines.lte.ps Äàòà èçìåíåíèÿ: Tue Nov 13 16:40:40 2001 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 07:46:31 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: unidentified sources

LTE LINES --- Atomic data for LTE calculations
C.S.Jeffery, Dept of Physics and Astronomy, University of St Andrews, North Haugh, St Andrews, Fife
KY16 9SS, Scotland
Abstract
An `end­user' oriented database containing atomic data suitable for the LTE analysis of early­type stellar
atmospheres has been extended and updated using modern atomic data. Data for each line comprise wave­
length, oscillator strength, radiative and electron damping widths, lower­level excitation potential, ion and
multiplet identification and source citations. Whilst much remains to be done, this paper reports on work
carried out, and invites reaction from potential users and other database compilers.
1 Introduction
Analyses of astronomical spectra depend critically on reliable sources of atomic and molecular data. Amongst
the CCP's, the large scale computation and collation of these data remains principally the domain of Collab­
orative Computational Projects Nos. 2 (Continuum States of Atoms and Molecules) and 6 (Heavy Particle
Dynamics). However, CCP7 is concerned with making these data accessible for astronomical applications
and has, over the years and thanks to the generosity of their authors, acquired a number of databases.
One such database attempts to collate and maintain `recommended' data for LTE analyses of early­
type stars. It has been extended with, in particular, the end user in view and may find practical uses in a
variety of applications such as (eg) the spectrum­synthesis in non­LTE calculations.
2 Available linelists
The CCP7 library (Jeffery 1990) currently contains the Kurucz & Peytremann (1975) list of semi­empirical
gf­values, the Kurucz (1990) list of semi­empirical gf­values for iron­group elements, and the Bell (1973) list
of gf­vales for atomic and molecular transitions.
A substantial body of new atomic data, gf­values in particular, has been calculated as a result of
recent endeavours such as the Opacity Project (Seaton 1987 and subsequent publications). Owing to the
number of methods currently used in atomic calculations, the cited data for a given transition may vary
substantially according to author. Serious attempts to evaluate the data objectively have been undertaken
in a few cases (eg CIII, NIV, OV, Allard et al. 1991). These are invaluable to the spectroscopist whose
results depend on reliable data but who cannot afford to frequently reassess the extensive literature.
Dufton and coworkers at Queen's University of Belfast use a linelist containing a restricted number of
data for LTE calculations of transitions used frequently in the analysis of main­sequence B­star atmospheres.
It forms a part of their LTE line­analysis package (SPECTRUM). Significant features of this linelist are its
format, which is sufficiently simple to allow easy identification of all data (including sources), and its small
size, which allows for convenient maintenance. The data included were wavelengths, oscillator strengths,
electron and radiative damping constants and excitation potentials. Most data pertain to transitions of light
ions in the blue­visible.
This linelist was identified as a valuable starting point for establishing a resource of modern atomic
data appropriate for LTE analyses of stellar atmospheres, and possibly with more extended applications.
3 LTE LINES
A project to make spectroscopic abundance analyses of a number of hydrogen­deficient B­type supergiants
(eg Jeffery & Heber 1991) demonstrated a number of areas for improvement, particularly with respect to
transition identification, source identification and the number of transitions treated.
Therefore a substantial revision to the Belfast linelist was undertaken. Firstly line identification has
been improved by the addition of a (Moore) multiplet identification. The number of transitions treated has
1

been increased to include weaker lines (frequently measurable in supergiants and non­rotating B stars) and
more ions. The citation of sources has been generalized. A substantial number of data have been updated
from Opacity Project data (Yan et al. 1987), recent Munich publications (Becker & Butler 1988, 1989, 1990,
Kilian et al. 1991) and from the critical compilation of Allard et al. (1991).
Data for the individual ions are stored separately. For the original application (SPECTRUM), the
separate ions are simply combined as required to form a linelist suitable for the spectrum being analysed.
Normally, the expanded citation list is appended to the concatenated linelist to ensure the integrity of source
identification.
The database is held on­line in the CCP7 data/program library (Jeffery 1990) on the STARLINK
microVAX system at St Andrews. Files may be found in the VMS directory:
(STARLINK/SPAN) SASTAR::CCP7DISK:[CCP7.LTE—LINES]
(Internet/Janet) UK.AC.ST­AND.STAR::CCP7DISK:[CCP7.LTE—LINES]
The ion data are stored in files with a simple naming convention such that lines due to CII (ie C + )
are in C2.DAT, CIII in C3.DAT, SiIV in SI4.DAT, etc. By way of example, the appendices show data for some
low excitation lines in the spectrum of CII (file C2.DAT). An explanation of the columns and an abstract
from the file which amplifies the source citations are also shown. With varying degrees of completeness and
mainly in the blue visible, data have been compiled for several ions listed in the appendix.
4 Data quality and completeness
It is intended that the database should contain `recommended' data for LTE analyses of stellar atmospheres.
This does not necessarily mean that they are the `best' data, or are based on the most accurate atomic
calculations. It does mean that where more than one value for a datum is available, an assessment of data
quality should have occurred. At present, the criteria for selection is as follows:
1. Objectively evaluated data (eg Allard et al. 1991)
2. Opacity Project oscillator strengths (where available)
3. Modern theoretical oscillator strengths
4. Experimental lifetimes
5. Other data (eg semi­empirical values)
Oscillator strengths are given for about 1080 lines, mostly in the blue­visible (–4000 \Gamma 5000 š A). Calculated
damping widths are available for ¸ 20%.
One objective is to contain the size of the database for maintainability and not to duplicate the efforts
of data producers. Therefore completeness over a large wavelength range would be counter­productive, since
(eg) for B stars, many lines in the red show large departures from LTE. In general, data should only be
incorporated if they are useful for the analysis of astronomical spectra. Semi­empirical oscillator strengths
are only useful as a last resort for estimating abundances.
5 Future
At present, a number of additional developments are being considered. Utilities for restricting and sorting
the lines in wavelength or by ion and multiplet will become available. One simple format change would allow
greater precision and higher data density, whilst it has been suggested that ion damping widths should be
tabulated separately. Further identification of the atomic transitions (ie electron configurations) would be
helpful. A high priority, from the author's point of view, is greater completeness for ultraviolet transitions
and for damping widths in the blue­visible.
At present, the linelist format is compatible with the Belfast LTE line formation code SPECTRUM.
In order to maintain this compatibility, format changes will have to be reviewed carefully. However it may be
desirable to work towards a format which is common across many codes and institutions in order, ultimately,
to reduce the effort of maintaining this (or other) databases.
Comments from potential users or contributors would be welcomed by the author. Since it is likely
that this effort is being duplicated in many institutions, interaction with other data compilers would be
valuable.
2

Acknowledgments
I am grateful to Philip Dufton for being so understanding when I first rearranged his linelists, and to Uli
Heber, Keith Butler and Mike Seaton, who have transmitted many of the data currently stored.
References
Allard M., Artru M.­C., Lanz T., & Le Dourneuf M., 1990. A&AS 84, 563
Becker S.R. & Butler K. 1988. A&A 201, 232
Becker S.R. & Butler K. 1989. A&A 209, 244
Becker S.R. & Butler K. 1990. A&A 235, 326
Bell R.A. 1973. MNRAS 164, 197
Jeffery C.S. 1990. CCP7 Newsletter No. 14, 43
Jeffery C.S. & Heber U. 1991. A&A (submitted)
Kilian J., Montenbruck O., & Nissen P.E. 1991. A&AS 88, 101
Kurucz R.L. 1990. Trans. IAU XXB, 168
Kurucz R.L. & Peytremann E. 1975. SAO Spec.Rep.No. 362
Seaton M.J. 1987. J.Phys.B. 20, 6363
Yan Y., Taylor K.T. & Seaton M.J. 1987. J.Phys.B. 20, 6399
3

A Database contents
A.1 Example data file
File: C2.DAT
6 2 1323.91 1.32E+00 7.30E­04 7.30E­05 9.290 ­1.00 Tob 84/ /
6 2 1334.53 5.25E­01 2.30E­04 2.30E­05 0.000 ­1.00 Tob 84/ /
6 2 6578.10 9.13E­01 2.64E­03 5.11E­04 14.445 2.00 Yan 87/Jon 71/Nus 81
6 2 6582.90 4.57E­01 2.64E­03 5.11E­04 14.445 2.00 Yan 87/Jon 71/Nus 81
6 2 3918.98 2.85E­01 3.70E­03 2.50E­04 16.333 4.00 Yan 87/Gol 82/Nus 81
6 2 3920.69 5.70E­01 3.70E­03 2.50E­04 16.334 4.00 Yan 87/Gol 82/Nus 81
6 2 4267.02 3.62E+00 9.97E­03 1.51E­03 18.047 6.00 Yan 87/Kon 76/Jon 71
6 2 4267.27 5.42E+00 9.97E­03 1.51E­03 18.047 6.00 Yan 87/Kon 76/Jon 71
6 2 4637.63 5.80E­02 1.00E­02 1.50E­04 21.150 12.01 Yan 87/ /
6 2 4638.91 1.16E­02 1.00E­02 1.50E­04 21.150 12.01 Yan 87/ /
6 2 4306.33 2.05E­02 1.00E­02 1.50E­04 21.150 12.02 Yan 87/ /
6 2 4307.59 4.11E­02 1.00E­02 1.50E­04 21.150 12.02 Yan 87/ /
6 2 4802.70 3.88E­01 2.85E­01 1.69E­04 22.000 17.08 Yan 87/ /Yan 87
6 2 4313.10 4.19E­01 3.36E­03 3.41E­04 23.120 28.00 Yan 87/Sah 69/Yan 87
6 2 4317.26 9.79E­01 3.36E­03 3.41E­04 23.120 28.00 Yan 87/Sah 69/Yan 87
6 2 4318.60 3.90E­01 3.36E­03 3.41E­04 23.120 28.00 Yan 87/Sah 69/Yan 87
6 2 4321.65 1.25E­01 3.36E­03 3.41E­04 23.120 28.00 Yan 87/Sah 69/Yan 87
6 2 4323.10 7.83E­02 3.36E­03 3.41E­04 23.120 28.00 Yan 87/Sah 69/Yan 87
6 2 4325.83 4.22E­01 3.36E­03 3.41E­04 23.120 28.00 Yan 87/Sah 69/Yan 87
6 2 4326.16 3.91E­01 3.36E­03 3.41E­04 23.120 28.00 Yan 87/Sah 69/Yan 87
...etc.
A.2 Column explanation
File: COLS.TXT
Column Explanation Example
1 Atomic number 6 Carbon
2 Ionisation stage 2 Singly ionised
3 Wavelength (Angstroms) Moo 48, etc. 4267.2
4 Oscillator strength g.f
5 Electron damping width:
FWHM/PI/4 * WL**2/C * 1.E15
WL is wavelength in Angstroms
C is speed of light (Angstroms/second)
6 Radiative damping width:
FWHM/PI/4 * WL**2/C
7 Excitation energy of lower level (electron Volts)
8 Multiplet identification Moo 48, etc
­ve numbers indicate a UV multiplet
9 References for 4, 5 and 6.
4

A.3 Source citation
File: REFS.DAT
All wavelengths, ionization potentials and energy levels are taken from:
Wiese et al. NBS publications.
C.E.Moore , Selected tables of atomic spectra : NSRDS ­ NBS3 sections 1­10.
Effects of ion broadening also included in the electron width
Major compilations are shown by a '+'
All 90. + Allard,M., Artru,M.­C., Lanz,T., ``& Le Dourneuf,M., 1990.
Astr.Astrophys.Supp. 84, 563.
Art 81. Artru,M.C., Jamar,C., Petrini,D., & Praderie,F., 1981.
Astr.Astrophys.Supp. 44, 171.
...
Bec 88. Becker,S.R., & Butler,K., 1988. Astr.Astrophys. 201, 232. O II
Bec 89. Becker,S.R., & Butler,K., 1989. Astr.Astrophys. 209, 244. N II
Bec 90. Becker,S.R., & Butler,K., 1990. Astr.Astrophys. 235, 326. Si II/III/IV
...
Gol 82. Goly & Weniger 1982. J.Quant.Spectrosc.Radiat.Transfer, 28,389.
...
Jon 71. Jones,W.M., Benett,S.M., & Griem,H.R., 1971,
University of Maryland, Tech.report #71­128.
...
Kon 76. Konjevic,N., & Wiese,W.L., 1976. J.Phys.chem.ref.Data, 5,(No.2),259.
...
Nus 81. Nussbaumer,H., & Storey,P.J., 1981. Astr.Astrophys., 96,91.
...
Sah 69. Sahal­Brechot,S., 1969. Astr.Astrophys., 2, 322.
...
Tob 84. Tobin & Kaufmann 1984. Mon.Not.R.astr.Soc., 207, 369.
...
Yan 87. Yan,Y., Taylor,K.T. & Seaton,M.J., 1987. J.Phys.B. 20, 6399.
...
A.4 Ions with data available
File: IONS.TXT
LI1.DAT Li I
C1.DAT C2.DAT C3.DAT C4.DAT C I/II/III/IV
N2.DAT N3.DAT N II/III
O1.DAT O2.DAT O3.DAT O I/II/III
NE1.DAT Ne I
MG2.DAT Mg II
AL2.DAT AL3.DAT Al II/III
SI2.DAT SI3.DAT SI4.DAT Si II/III/IV
P2.DAT P3.DAT P II/III
S2.DAT S3.DAT S II/III
CL2.DAT Cl II
A2.DAT A II
CA2.DAT Ca II
TI2.DAT Ti II
CR2.DAT Cr II
FE1.DAT FE2.DAT FE3.DAT Fe I/II/III
NI2.DAT Ni II
5