Morgan-Keenan Research Articles

A Blended Artificial Intelligence Approach for Spectral Classification of Stars in Massive Astronomical Surveys.

This paper analyzes and compares the sensitivity and suitability of several artificial intelligence techniques applied to the Morgan–Keenan (MK) system for the classification of stars. The MK system is based on a sequence of spectral prototypes that allows classifying stars according to their effective temperature and luminosity through the study of their optical stellar spectra. Here, we include the method description and the results achieved by the different intelligent models developed thus far in our ongoing stellar classification project: fuzzy knowledge-based systems, backpropagation, radial basis function (RBF) and Kohonen artificial neural networks. Since one of today’s major challenges in this area of astrophysics is the exploitation of large terrestrial and space databases, we propose a final hybrid system that integrates the best intelligent techniques, automatically collects the most important spectral features, and determines the spectral type and luminosity level of the stars according to the MK standard system. This hybrid approach truly emulates the behavior of human experts in this area, resulting in higher success rates than any of the individual implemented techniques. In the final classification system, the most suitable methods are selected for each individual spectrum, which implies a remarkable contribution to the automatic classification process.

Low-metallicity massive single stars with rotation

Context. Metal-poor massive stars are assumed to be progenitors of certain supernovae, gamma-ray bursts, and compact object mergers that might contribute to the early epochs of the Universe with their strong ionizing radiation. However, this assumption remains mainly theoretical because individual spectroscopic observations of such objects have rarely been carried out below the metallicity of the Small Magellanic Cloud. Aims. Here we explore the predictions of the state-of-the-art theories of stellar evolution combined with those of stellar atmospheres about a certain type of metal-poor (0.02 Z⊙) hot massive stars, the chemically homogeneously evolving stars that we call Transparent Wind Ultraviolet INtense (TWUIN) stars. Methods. We computed synthetic spectra corresponding to a broad range in masses (20−130 M⊙) and covering several evolutionary phases from the zero-age main-sequence up to the core helium-burning stage. We investigated the influence of mass loss and wind clumping on spectral appearance and classified the spectra according to the Morgan-Keenan (MK) system. Results. We find that TWUIN stars show almost no emission lines during most of their core hydrogen-burning lifetimes. Most metal lines are completely absent, including nitrogen. During their core helium-burning stage, lines switch to emission, and even some metal lines (oxygen and carbon, but still almost no nitrogen) are detected. Mass loss and clumping play a significant role in line formation in later evolutionary phases, particularly during core helium-burning. Most of our spectra are classified as an early-O type giant or supergiant, and we find Wolf–Rayet stars of type WO in the core helium-burning phase. Conclusions. An extremely hot, early-O type star observed in a low-metallicity galaxy could be the result of chemically homogeneous evolution and might therefore be the progenitor of a long-duration gamma-ray burst or a type Ic supernova. TWUIN stars may play an important role in reionizing the Universe because they are hot without showing prominent emission lines during most of their lifetime.

Towards an improvement in the spectral description of central stars of planetary nebulae

Context. There are more than 3000 known Galactic planetary nebulae, but only 492 central stars of Galactic planetary nebulae (CSPN) have known spectral types. It is vital to increase this number in order to have reliable statistics, which will lead to an increase of our understanding of these amazing objects. Aims. We aim to contribute to the knowledge of central stars of planetary nebulae and stellar evolution. Methods. This observational study is based on Gemini Multi-Object Spectrographs (GMOS) and with the Intermediate Dispersion Spectrograph (IDS) at the Isaac Newton Telescope spectra of 78 CSPN. The objects were selected because they did not have any previous classification, or the present classification is ambiguous. These new high quality spectra allowed us to identify the key stellar lines for determining spectral classification in the Morgan-Keenan (MK) system. Results. We have acquired optical spectra of a large sample of CSPN. From the observed targets, 50 are classified here for the first time while for 28 the existing classifications have been improved. In seven objects we have identified a P-Cygni profile at the He I lines. Six of these CSPN are late O-type. The vast majority of the stars in the sample exhibit an absorption-type spectrum, and in one case we have found wide emission lines typical of [WR] stars. We give a complementary, and preliminary, classification criterion to obtain the sub-type of the O(H)-type CSPN. Finally, we give a more realistic value of the proportion of CSPN that are rich or poor in hydrogen.

PEPSI deep spectra

Context. High-resolution échelle spectra confine many essential stellar parameters once the data reach a quality appropriate to constrain the various physical processes that form these spectra. Aim. We provide a homogeneous library of high-resolution, high-S/N spectra for 48 bright AFGKM stars, some of them approaching the quality of solar-flux spectra. Our sample includes the northern Gaia benchmark stars, some solar analogs, and some other bright Morgan-Keenan (M-K) spectral standards. Methods. Well-exposed deep spectra were created by average-combining individual exposures. The data-reduction process relies on adaptive selection of parameters by using statistical inference and robust estimators. We employed spectrum synthesis techniques and statistics tools in order to characterize the spectra and give a first quick look at some of the science cases possible. Results. With an average spectral resolution of R ≈ 220 000 (1.36 km s−1), a continuous wavelength coverage from 383 nm to 912 nm, and S/N of between 70:1 for the faintest star in the extreme blue and 6000:1 for the brightest star in the red, these spectra are now made public for further data mining and analysis. Preliminary results include new stellar parameters for 70 Vir and α Tau, the detection of the rare-earth element dysprosium and the heavy elements uranium, thorium and neodymium in several RGB stars, and the use of the 12C to 13C isotope ratio for age-related determinations. We also found Arcturus to exhibit few-percent Ca II H&K and Hα residual profile changes with respect to the KPNO atlas taken in 1999.

Improved spectral descriptions of planetary nebulae central stars

Context. At least 492 central stars of Galactic planetary nebulae (CSPNs) have been assigned spectral types. Since many CSPNs are faint, these classification efforts are frequently made at low spectral resolution. However, the stellar Balmer absorption lines are contaminated with nebular emission; therefore in many cases a low-resolution spectrum does not enable the determination of the H abundance in the CSPN photosphere. Whether or not the photosphere is H deficient is arguably the most important fact we should expect to extract from the CSPN spectrum, and should be the basis for an adequate spectral classification system. Aims. Our purpose is to provide accurate spectral classifications and contribute to the knowledge of central stars of planetary nebulae and stellar evolution. Methods. We have obtained and studied higher quality spectra of CSPNs described in the literature as weak emission-line star (WELS). We provide descriptions of 19 CSPN spectra. These stars had been previously classified at low spectral resolution. We used medium-resolution spectra taken with the Gemini Multi-Object Spectrograph (GMOS). We provide spectral types in the Morgan-Keenan (MK) system whenever possible. Results. Twelve stars in our sample appear to have normal H rich photospheric abundances, and five stars remain unclassified. The rest (two) are most probably H deficient. Of all central stars described by other authors as WELS, we find that at least 26% of them are, in fact, H rich O stars, and at least 3% are H deficient. This supports the suggestion that the denomination WELS should not be taken as a spectral type, because, as a WELS based on low-resolution spectra, it cannot provide enough information about the photospheric H abundance.

An MK-like system of spectral classification for hot subdwarfs

An MK (Morgan-Keenan)-like system of spectral classification for hot subdwarfs is presented. We find that a three-dimensional spectral type, consisting of a “spectral” class, a “luminosity” class, and a “helium” class, is necessary to classify the sdO (subdwarf O) and sdB (subdwarf B) stars. In addition, the helium-strong stars appear to form two parallel spectral sequences: one showing strong lines of CII, CIII, or CIV, and the other with these same lines weak or absent. We also give a preliminary calibration of the new spectral types in terms of e ective temperature, surface gravity, and surface helium-to-hydrogen abundance ratio, and show the relation between the new spectral types and the natural groups defined by the Palomar-Green (PG) survey.

ι Orionis—Evidence for a Capture Origin Binary

The star ι Ori (HD 37043, HR 1899) is a 29.1338 day period, double-lined O-type spectroscopic binary with a highly eccentric orbit. The classification of the primary (O9 III) is well established. The secondary, roughly 2 mag fainter, is generally acknowledged as an early-type B star, but claims about its luminosity class have been made only from its estimated light contribution and not from its spectrum. We have collected visual data with the 1 m Multiple-Telescope Telescope (MTT; at the Center for High Angular Resolution at Georgia State University) in four wave bands well distributed in orbital phase, with the periastron passage particularly well covered. We present reconstructed component spectra from our tomographic reconstruction technique. According to three Morgan-Keenan (MK) luminosity criteria, we can unambiguously characterize the secondary as a B1 III-IV giant (MK type or B0.8 III-IV interpolated type). We also have constructed three line-strength indices from eighteen observed individual lines for the secondary and a grid of comparison stars. These results confirm the typing from the MK luminosity criteria. Furthermore, we argue, from the mass ratio (q = 0.5), evolutionary considerations, the orbital scale, and the unremarkable CNO element equivalent widths of the secondary, that this system is not coevolved nor has it undergone significant mass transfer. We postulate an origin for the system in a binary-binary collision and suspect that the runaway stars μ Col and AE Aur may have been the original partners for the current ι Ori secondary and primary, respectively.

Observed Interstellar Extinction in the Ultraviolet

An Observational Program to measure stellar energy distributions - in the ultraviolet between 3000 Å and 1200 Å has recently been completed. The instrumentation used was similar to that reported in reference 1. A necessary preliminary for discussion of this material is some knowledge of the extinction due to interstellar absorption. The trend of this extinction was first reported in 1964. (See ref. 2.) Additional observational material has since been obtained and is presented herein.Five pairs of stars with similar Morgan-Keenan (MK) spectral classifications have been observed that are suitable for obtaining extinction: ζ Per and ρ Leo; ξ Per and ζ Pup; o Per and β Can Maj; σ Sco and π SCO; and δ Sco and τ SCO. The telemetry records were hand reduced by the author, each star was multiplied by the absolute calibration curve for the appropriate instrument, and the resulting flux curve was placed on a magnitude scale.