Line identification studies using traditional techniques and wavelength coincidence statistics
Author(s) -
Charles R. Cowley,
Saul J. Adelman
Publication year - 1990
Publication title -
publications of the astronomical society of the pacific
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.294
H-Index - 172
eISSN - 1538-3873
pISSN - 0004-6280
DOI - 10.1086/132736
Subject(s) - spurious relationship , computer science , coincidence , identification (biology) , stars , statistics , line (geometry) , physics , astrophysics , mathematics , machine learning , medicine , botany , alternative medicine , geometry , pathology , biology
Traditional line identification techniques result in the assignment of individual lines to an atomic or ionic species. These methods may be supplemented by wavelength coincidence statistics (WCS). The strength and weakness of these methods are discussed using spectra of a number of normal and peculiar B and A stars that have been studied independently by both methods. The present results support the overall findings of some earlier studies. WCS would be most useful in a first survey, before traditional methods have been applied. WCS can quickly make a global search for all species and in this way may enable identifications of an unexpected spectrum that could easily be omitted entirely from a traditional study. This is illustrated by O I. WCS is a subject to well known weakness of any statistical technique, for example, a predictable number of spurious results are to be expected. The danger of small number statistics are illustrated. WCS is at its best relative to traditional methods in finding a line-rich atomic species that is only weakly present in a complicated stellar spectrum.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom