Spectral classification of stars based on LAMOST spectra

In this work, we select spectra of stars with high signal-to-noise ratio from LAMOST data and map their MK classes to the spectral features. The equivalent widths of prominent spectral lines, which play a similar role as multi-color photometry, form a clean stellar locus well ordered by MK classes. The advantage of the stellar locus in line indices is that it gives a natural and continuous classification of stars consistent with either broadly used MK classes or stellar astrophysical parameters. We also employ an SVM-based classification algorithm to assign MK classes to LAMOST stellar spectra. We find that the completenesses of the classifications are up to 90% for A and G type stars, but they are down to about 50% for OB and K type stars. About 40% of the OB and K type stars are mis-classified as A and G type stars,respectively. This is likely due to the difference in the spectral features between late B type and early A type stars or between late G and early K type stars being very weak. The relatively poor performance of the automatic MK classification with SVM suggests that the direct use of line indices to classify stars is likely a more preferable choice.     

Prediction of Spectral and Luminosity Classes from Spectral Indices with Artificial Neural Networks

This paper deals with the application of artificial neural networks topredict the spectral types and luminosity classes from spectralindices. The precision reached in predicting spectralclasses is 2-subclasses for 90% of the test stars.The success rate in prediction of luminosity classes is more than 90\% for the classes III and V. The results providea new method to predict spectral and luminosity classes from the spectral indices, which will be useful when data from large surveylike 2dF and Sloan Digital Sky Survey will come into effect.     

Support Vector Machine combined with K-Nearest Neighbors for Solar Flare Forecasting

A method combining the support vector machine (SVM) the K-Nearest Neighbors (KNN), labelled the SVM-KNN method, is used to construct a solar flare forecasting model. Based on a proven relationship between SVM and KNN, the SVM-KNN method improves the SVM algorithm of classification by taking advantage of the KNN algorithm according to the distribution of test samples in a feature space. In our flare forecast study, sunspots and 10cm radio flux data observed during Solar Cycle 23 are taken as predictors, and whether an M class flare will occur for each active region within two days will be predicted. The SVM- KNN method is compared with the SVM and Neural networks-based method. The test results indicate that the rate of correct predictions from the SVM-KNN method is higher than that from the other two methods. This method shows promise as a practicable future forecasting model.     

New Features of Parenago’s Discontinuity from Gaia DR1 Data

The velocity field of main-sequence stars and red giants from the TGAS catalogue with heliocentric distances up to 1.5 kpc has been analyzed for various spectral types. To estimate the influence of a low accuracy of stellar parallax measurements on the results of a kinematic analysis of distant stars, first we have studied in detail how the kinematic parameters derived with 1/π distances are shifted when these distances are replaced by three other versions of distances from Astraatmadja et al. (2016b). We have obtained detailed tables in which the ranges of these shifts in the Ogorodnikov–Milne and Bottlinger model parameters are shown for the stars of each spectral type. We have the smallest shifts in the case of determining the Oort coefficients A and B, for which there are 10% shifts only for main-sequence stars of spectral type B. In the remaining cases, these shifts are 0–3%. For the remaining parameters the shifts do not exceed 30%. Thus, we have shown that using the 1/π distance scale in estimating the Ogorodnikov–Milne and Bottlingermodel parameters (except for the parameter Ω″₀) yields reliable results even when using parallaxes with large relative errors (up to 60%). To study Parenago’s discontinuity, we have investigated the dependence of the Ogorodnikov–Milne and Bottlinger model parameters on color for 1 260 071 mainsequence stars and 534 387 red giants. As far as we know, such a data set is used for the first time to investigate Parenago’s discontinuity. The main result is the detection of maximum points at B ? V = 0.75 after which the solar velocity component V and the Oort coefficient B decrease when moving from blue stars to red ones. This fact is a new feature of Parenago’s discontinuity, because the component V does not change in the classical case at B ? V >0.6. We have made an attempt to represent the well-known Parenago’s discontinuity as a special case of the more complex effect of a gradual change in a number of kinematic parameters as the mean age and composition of the group of stars under study changes.     

Automated spectral classification using template matching

An automated spectral classification technique for large sky surveys is pro-posed. We firstly perform spectral line matching to determine redshift candidates for an observed spectrum, and then estimate the spectral class by measuring the similarity be-tween the observed spectrum and the shifted templates for each redshift candidate. As a byproduct of this approach, the spectral redshift can also be obtained with high accuracy. Compared with some approaches based on computerized learning methods in the liter-ature, the proposed approach needs no training, which is time-consuming and sensitive to selection of the training set. Both simulated data and observed spectra are used to test the approach; the results show that the proposed method is efficient, and it can achieve a correct classification rate as high as 92.9%, 97.9% and 98.8% for stars, galaxies and quasars, respectively.     

A modified genetic algorithm for precise determination the geometrical orbital elements of binary stars

The paper presents a modified genetic algorithm called adapted genetic algorithm with adjusting population size (AGA-POP) for precise determination the orbital elements of binary stars. The proposed approach is a simple, robust way that can be considered to be a new member in the class of self organizing genetic algorithms. The proposed AGA-POP is applied on the star η Bootis of MK type G0 IV to find a set of optimal orbital elements. This leads to obtain the best fitting of Keplerian and phase curves. The modified method is compared with other different methods such as standard genetic algorithm, adapted genetic algorithm (AGA) and least square methods. Simulation results show the effectiveness of using AGA-POP compared with other different classic genetic algorithms in reducing the computation time. Also, better performances have been achieved when using the proposed technique.     

Main-sequence magnetic CP stars: II. Physical parameters and chemical composition of the atmosphere

This paper continues a series of reviews dedicated to magnetic CP stars. The occurrence frequency of CP stars among B5–F0-type main-sequence stars is shown to be equal to about 15–20%. The problems of identification and classification of these objects are addressed. We prefer the classification of Preston, which subdivides chemically peculiar stars into the following groups: Am, λ Boo, Ap/Bp, Hg-Mn, He-weak, and He-strong stars. The main characteristic features of objects of each group are briefly analyzed. The rotation velocities of CP stars are shown to be about three times lower than those of normal stars of the same spectral types (except for λ Boo and He-strong objects). The rotation periods of CP stars range from 0.5 to 100 days, however, there is also a small group of objects with especially long (up to several tens of years) variability periods. All kinds of peculiar stars can be found in visual binaries, with Am-and Hg-Mn-type stars occurring mostly in short-period binaries with P < 10 days, and the binary rate of these stars is close to normal. The percentage of binaries among magnetic stars (20%) is lower than among normal stars. A rather large fraction of CP1-and CP2-type stars was found to occur in young clusters (with ages smaller than 10⁷ years). Photometric and spectral variability of peculiar stars of various types is discussed, and it is shown that only objects possessing magnetic fields exhibit light and spectral variations. The chemical composition of the atmospheres of CP stars of various types is considered. The abundances of various elements are usually determined by comparing the line profiles in the observed spectrum with those of the synthetic spectra computed for various model atmospheres. Different mechanisms are shown to contribute to chemical inhomogeneity at the star’s surface, and the hypothesis of selective diffusion of atoms in a stable atmosphere is developed. Attention is also paid to the problems of the determination of local chemical composition including the stratification of elements. Some of the coolest SrCrEu peculiar stars are found to exhibit fast light variations with periods ranging from 6 to 15 min. These variations are unassociated with rotation, but are due to nonradial pulsations. The final part of the the review considers the fundamental parameters of CP stars. The effective temperatures, luminosities, radii, and masses of these objects are shown to agree with the corresponding physical parameters of normal main-sequence stars of the same spectral types.     

Optical spectroscopic classification and membership of young M dwarfs in star-forming regions

The spectral type is a key parameter in calibrating the temperature which is required to estimate the mass of young stars and brown dwarfs. We describe an approach developed to classify low-mass stars and brown dwarfs in the Trapezium Cluster using red optical spectra, which can be applied to other star-forming regions. The classification uses two methods for greater accuracy: the use of narrow-band spectral indices which rely on the variation of the strength of molecular lines with spectral type and a comparison with other previously classified young, low-mass objects in the Chamaeleon I star-forming region. We have investigated and compared many different molecular indices and have identified a small number of indices which work well for classifying M-type objects in nebular regions. The indices are calibrated for young, pre-main-sequence objects whose spectra are affected by their lower surface gravities compared with those on the main sequence. Spectral types obtained are essentially independent of both reddening and nebular emission lines.
Confirmation of candidate young stars and brown dwarfs as bona fide cluster members may be accomplished with moderate resolution spectra in the optical region by an analysis of the strength of the gravity-sensitive Na doublet. It has been established that this feature is much weaker in these very young objects than in field dwarfs. A sodium spectral index is used to estimate the surface gravity and to demonstrate quantitatively the difference between young (1–2 Myr) objects, and dwarf and giant field stars.     

Spectral indices in quantitative spectral classification from stellar libraries

With the use of modern detectors stellar spectral classification libraries have been extended from the photographic regime to the near ℝ at 11000 Å. We have defined new spectral indices within this extended wavelength-range that can be used to determine the luminosity classification for G-K-M stars. An advantage of the new indices, which sample the stellar flux in and out of selected spectral features, is that they are insensitive to catalog differences. This facilitates the use of many catalogs, with varying resolution, different reddening corrections, and calibrations, hence extending the total number of stellar standards available. Furthermore, we demonstrate that the indices can be used to infer absolute magnitudes with good accuracy. The indices should prove useful for analysis of spectra from distant clusters, galaxies, and in particular for problems involving spectral synthesis of stellar populations of galaxies.     

Clustering analysis of line indices for LAMOST spectra with AstroStat

The application of data mining in astronomical surveys,such as the Large Sky Area MultiObject Fiber Spectroscopic Telescope(LAMOST)survey,provides an effective approach to automatically analyze a large amount of complex survey data.Unsupervised clustering could help astronomers find the associations and outliers in a big data set.In this paper,we employ the k-means method to perform clustering for the line index of LAMOST spectra with the powerful software Astro Stat.Implementing the line index approach for analyzing astronomical spectra is an effective way to extract spectral features for low resolution spectra,which can represent the main spectral characteristics of stars.A total of 144 340 line indices for A type stars is analyzed through calculating their intra and inter distances between pairs of stars.For intra distance,we use the definition of Mahalanobis distance to explore the degree of clustering for each class,while for outlier detection,we define a local outlier factor for each spectrum.Astro Stat furnishes a set of visualization tools for illustrating the analysis results.Checking the spectra detected as outliers,we find that most of them are problematic data and only a few correspond to rare astronomical objects.We show two examples of these outliers,a spectrum with abnormal continuum and a spectrum with emission lines.Our work demonstrates that line index clustering is a good method for examining data quality and identifying rare objects.     

Photometric distances to nine dark globules

Distances to nine dark globules are determined by a method using optical ( VRI ) and near-infrared (near-IR) ( JHK ) photometry of stars projected towards the field containing the globules. In this method, we compute intrinsic colour indices of stars projected towards the direction of the globule by dereddening the observed colour indices using various trial values of extinction   A _V   and a standard extinction law. These computed intrinsic colour indices for each star are then compared with the intrinsic colour indices of normal main-sequence stars and a spectral type is assigned to the star for which the computed colour indices best match with the standard intrinsic colour indices. Distances ( d ) to the stars are determined using the   A _V   and absolute magnitude  ( M_V )  corresponding to the spectral types thus obtained. A distance versus extinction plot is made and the distance at which   A _V   undergoes a sharp rise is taken to be the distance to the globule. All the clouds studied in this work are in the distance range 160–400 pc. The estimated distances to dark globules LDN 544, LDN 549, LDN 567, LDN 543, LDN 1113, LDN 1031, LDN 1225, LDN 1252 and LDN 1257 are  180 ± 35, 200 ± 40, 180 ± 35, 160 ± 30, 350 ± 70, 200 ± 40, 400 ± 80, 250 ± 50  and 250 ± 50 pc, respectively. Using the distances determined, we have estimated the masses of the globules and the far-IR luminosity of the IRAS sources associated with them. The mass of the clouds studied are in the range  10–200 M_⊙  .     

Dynamical study of wide pairs of stars based on data from the WDS catalog

Using the method of apparent motion parameters, we have studied the relative motion of the components in 561 pairs of wide (ρ > 2″) and relatively nearby (Hipparcos parallaxes > 0.01″) visual double stars based on data from the WDS catalog. The minimum masses of the double stars have been calculated at given parallaxes. We have identified 358 optical pairs. For 11 stellar pairs, we have found the minimum mass to exceed the estimate corresponding to their spectral types and luminosities. This excess is 5–7 M _⊙ for two stars, ADS 7446 and 9701.     

The absence of apparent correlations between iras and optical spectral features of late-type stars

The spectral signatures of circumstellar dust shells (CS) of late-type stars cannot be tied in a simple manner to their optical spectral features. The spectral indices of featureless low-resolution spectrometer (LRS) spectra of K and early-M (M0-M5) stars are consistent with that of the Rayleigh-Jeans law and those of late-M (M6 and later) stars tend to have lower values showing CS radiation. There is no correlation observed between the optical sub-type and the 10 m emission strength for M stars. The absence of a simple relationship is also evident for C stars where the temperature and abundance indices show no correlation with the 11 m emission strength. The sample is confined to the IRC stars identified with accurate spectra observed by the LRS of IRAS. The IRC stars are the whole sky survey done with the 2 m band, where late-type stars are bright and interstellar extinction is one-tenth of the visual colour band.     

Spectral classification of Pleiades brown dwarf candidates

We report on the results of the spectroscopy of 10 objects previously classified as brown dwarf candidates via RIJHK colors by Eisenbeiss et al. (2009), who performed deep imaging observations on a ∼0.4 sq.deg. field at the edge of the Pleiades. We describe and judge on classification techniques in the region of M‐type stars. To classify and characterise the objects, visual and near infrared spectra have been obtained with VLT FORS and ISAAC. The spectral classification was performed using the shape of the spectra as well as spectral indices that are sensitive to the spectral type and luminosity class of M‐type stars and late M‐type brown dwarfs. Furthermore a spectrophotometric distance was calculated and compared the distance of the Pleiades to investigate the membership probability. As a second argument we analyzed the proper motion. The brown dwarf candidates were found not to be brown dwarfs, but late‐K to mid‐M‐type dwarf stars. Based on the obtained distance and tabulated proper motions we conclude that all objects are background dwarf stars (© 2011 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Effects of rotation on the colours and line indices of stars

The intrinsicuvby and HΒ indices of member stars of α-Persei, Pleiades and Scorpio-Centaurus association have been analysed in detail for rotation effects. These stars range in spectral type from B0 to F0 and the observed effects of rotation are found to be in agreement with photometric effects calculated by Collins & Sonneborn (1977) for rigidly rotating B0 to F0 stars On leave of absence from Assumption College, Changanacherry, Kerala.     

Spectral observations of red dwarfs II. New observations of flare stars in the Pleiades cluster region

We studied the spectra of 24 flare stars of the Pleiades cluster region obtained on the 2.6m telescope of the Byurakan observatory in the years 1986–1989. From the spectral indices of titanium oxide (TiO) and other chemical elements of the stars investigated we determined their spectral classes and luminosities (absolute visual stellar magnitudes). The agreement of these moduli of the stars with the distance modulus of the Pleiades cluster confirms that 21 of them belong to the cluster. Only three of them (FSP 169, FSP 377, and FSP 435) are possibly not members of the cluster. This indicates that the flare activity of a star is a reliable criterion for its membership in a nearby stellar system (cluster or association).Translated fromAstrofizika, Vol. 37, No. 4, 1994.     

Luminosities of FBS late type M stars

A catalog of 995 C and late M type stars is compiled from the First Byurakan Survey of the sky (FBS). The question of effectively separating giants and dwarfs among the stars of spectral class M is examined using data from this catalog, the 2MASS survey, and a number of other catalogs. This has made it possible to determine the luminosity class of 846 class M stars from the FBS survey. In the sample of FBS late M stars, 187 are dwarfs and 659 are giants. The sample of FBS late M stars is compared statistically with the set of known samples of type M stars. The subsample of FBS M-giants is proved to be similar to Stephenson’s standard sample of M-giants. A detailed statistical analysis of these two samples is carried out. __________ Translated from Astrofizika, Vol. 50, No. 4, pp. 589–602 (November 2007).     

E (B-V) determination from an UV-visual two-colour diagram: O and B stars in the catalogue of stellar ultraviolet fluxes

We present here, for the O and B type stars in the Catalogue of Stellar Ultraviolet Fluxes, an approach which does not require a precise knowledge of spectral type and luminosity class for derivingE(B-V) colour excesses. The method is based on the use of an UV-visual two-colour diagram; galactic variations in the interstellar extinction law are analyzed and fully taken into account. Our results have been compared with those derived by using the differences between observed and intrinsic colours for stars with known spectral classification. The very good agreement in a large number of cases (94 per cent) demonstrates that our approach permits the derivation of reliable colour excess values for early type stars even if only a rough spectral classification is available.     

Calculated atmospheric color refraction and observed stellar positions

Astrometric observations at different zenith distances have been performed in Dresden in an area centered atNGC 6791 where there are some stars with reliable color information (widely dispersed spectral types in the MK systemand color indices B_T – V_T) as well as with accurate positions from Tycho‐2 catalog. The results are used to estimate how significant improvements in stellar positions may be when accurate corrections for color refraction are taken into account. We have treated two cases for refraction calculations: (1) a photometric case for color indices and (2) a spectral case for spectral types and luminosity classes. To calculate refraction we use Stone's modified computer code (Malyuto & Meinel 2000). To treat the photometric case we have calculated the synthetic color indices for the spectral energy distributions of Sviderskiene (1988). The positional improvements due to including color refraction corrections are significant and slightly larger in the spectral case. An improvement of about 15% is reached at a zenith distance of 65°. Our basic conclusion is that color refraction should be taken into account for obtaining accurate stellar positions from ground based observations at larger zenith distances. Reliable refraction corrections may be calculated from spectral and/or photometric data.     

SCUBA photometry of candidate Vega-like sources

New SCUBA measurements at millimetre wavelengths are presented for a sample of Vega-like stars. Six stars were detected, while sensitive upper limits were obtained for a further 11 sources. Most of the sample selected from a recent catalogue of Vega-like stars have infrared excesses similar to those of the prototype Vega-like stars α Lyr and α PsA. Their IR–submm spectral indices are steep, indicating that the submm emission from the discs is dominated by grains which are smaller than the wavelength of observation and that only small grains exist in those dusty discs. HD 98800 has an IR–submillimetre spectral index of less than two, which suggests that grains have grown to more than 0.3 mm in size. Hipparcos parallax data for HD 42137 and HD 123160 suggest that these two stars are giants rather than dwarfs, similar to the situation previously found for HD 233517. Dust masses, or upper limits, were derived for the sample; these indicate that most of the sources do not have as much dust as Herbig Ae/Be or T Tauri stars, but are likely to have dust masses comparable to those of the prototype Vega-like stars.