An optical and near-IR spectroscopic study of the extreme P Cygni-type supergiant HDE 316285

A detailed study of the Galactic P Cygni-type supergiant HDE 316285, based on high quality optical (AAT, MSG, CTIO) and near-IR (UKIRT, CFHT, CTIO) sp ectroscopy, is presented. As has been noted previously, its spectrum is dom inated by H, He I, and Fe II P Cygni profiles. Emission lines due to N I, N II, [N II], OI, Na I, Mg II, Al II, Ca II, Si II, Si III, Fe II and [Fe II ] can also be readily identified. Many of the metal lines are produced by c ontinuum fluorescence. The rich N spectrum, the paucity of the O spectrum ( only 2 O lines can be identified), and the apparent absence of emission due to C, strongly suggest that the atmosphere of the star is contaminated by CNO processed material. A comparison of the spectrum of HDE 316285 with P C ygni and He 3-519 is presented. From a spectral analysis using the non-LTE atmosphere code of Hillier (1991 ), and assuming a distance of 1.85 kpc, our preferred model for HDE 316285 has the following parameters: T-* = 15 kK, log L-*/L. = 5.44, (M) over dot = 2.4 x 10(-4) M. yr(-1), v(infinity) = 410 km s(-1), EB-v=1.81 mag, and H/ HeN similar to 1.5 by number. Due to the low degree of He ionization the de rived H/He abundance ratio and mass-loss rate are strongly coupled. Models with H/He=10 to 0.5 are equally capable of explaining the H and He I spectr um provided the mass-loss rate is scaled according to the approximate formu la (M) over dot = 9.1 + 26.3(He/H - 0.1) x 10(-5) M. yr(-1). Preliminary wo rk, however, indicates that a solar H/He ratio can be ruled out on the basi s of line strengths of other species - particularly N, Mg, Al. The stellar wind from HDE316285 is more extreme than P Cygni with a perform ance number (= ratio of wind momentum to radiative momentum) 30 times great er. The low H/He abundance patio and high N/He abundance ratio confirms tha t HDE 316285 is evolved. Although we find no evidence in the literature for photometric variability, we find strong evidence for significant spectral variability. Because of t he spectral variability, and because the stellar properties and chemical co ntent of HDE 316285 are similar to known luminous blue variables (LBVs), we suggest that it is a LBV. Support for this contention comes from the detec tion by McGregor et al. (1988) of a cold circumstellar dust shell associate d with HDE 316285. However HDE 316285, like P Cygni, could currently be in a relatively quiescent phase of its LBV life, exhibiting significant spectr al variations but not undergoing major photometric outbursts similar to AG Car. The mass loss of HDE 316285 is prodigious. In less than 10(5) years it will lose over 20 M.. Even if HDE 316285 is not an LBV, it is obviously in an e volutionary phase of short duration.