Spectral Properties and Variability of BIS objects
S. Gaudenzi, R. Nesci, C. Rossi, S. Sclavi, K. S. Gigoyan, A. M. Mickaelian
aa r X i v : . [ a s t r o - ph . S R ] J u l Manuscript for
Revista Mexicana de Astronom´ıa y Astrof´ısica (2007)
SPECTRAL PROPERTIES AND VARIABILITY OFBIS OBJECTS. S. Gaudenzi, R. Nesci, C. Rossi,
S. Sclavi, K. S. Gigoyan, and A. M.Mickaelian, Draft version: September 28, 2018
RESUMENUn gran n´umero de entradas del cat´alogo de fuentes puntuales
IRAS tienentodav´ıa una clasificaci´on aproximada. A trav´es del an´alisis y la interpretaci´onde nuevos datos obtenidos y datos de la literatura tuvimos como objetivoclarificar la naturaleza de las pobremente investigadas fuentes puntuales de
IRAS clasificadas como estrellas de tipo tard´ıo, pertenecientes al cat´alogo deestrellas Byurakan
IRAS . En 95 estrellas realizamos espectroscop´ıa de medi-ana resoluci´on para mejorar la clasificaci´on espectral que previamente se hab´ıabasado solo en ´ındices de color en el ´optico y/o espectros de baja resoluci´on.Recolectamos datos fotom´etricos de la literatura y obtuvimos nuevas obser-vaciones fotom´etricas para estudiar su variabilidad; asignamos o revisamosla clasificaci´on de variabilidad de nuestros objetos. Revisamos los tipos es-pectrales y encontramos diferencias sustanciales con respecto al cat´alogo BISoriginal en 45% de las estrellas; adem´as el 30% no ten´ıan clasificaci´on pre-via. Para las estrellas variables regulares determinamos sus distancias, susmagnitudes absolutas y sus tasas de p´erdida de masa. De las otras estrellas,estimamos sus distancias a partir de sus tipos espectrales y sus magnitudesaparentes: Se encontraron distancias desde 1.3 a 10 kpc, con una distanciamedia desde el plano Gal´actico de 2.8 kpc, lo cual indica que las estrellas delcat´alogo BIS en su mayor´ıa, pertenecen a la poblaci´on del halo de la Galaxia.Un gran n´umero de estas estrellas se encontrar´ıan al alcance de una medici´ondirecta de GAIA. Los espectros muestran que nueve estrellas son de tipo es-pectral G-K, cuatro estrellas de Carb´on N se encuentran en la rama giganteasint´otica, las otras son del tipo espectral M.ABSTRACTThrough the analysis and interpretation of newly obtained and of literaturedata we have clarified the nature of poorly investigated
IRAS point sourcesclassified as late type stars, belonging to the Byurakan
IRAS
Stars catalog.From medium resolution spectroscopy of 95 stars we have strongly revised 47spectral types and newly classified 31 sources. Nine stars are of G or K types, Observations collected with the Cassini Telescope at Loiano station of theINAF − Bologna Astronomical Observatory, with the Copernico Telescope of theINAF − Padova Astronomical Observatory and with the TACOR Telescope of the Universit`a“La Sapienza”, Roma. INAF/IAPS, Roma Italy. INAF/ Osservatorio di Monte Porzio, Roma, Italy. Universit`a La Sapienza Roma, Italy. Ambartsumian Byurakan Astrophysical Observatory (BAO). − type stars. From literature and new photometric observations we havestudied their variability behaviour. For the regular variables we determineddistances, absolute magnitudes and mass loss rates. For the other stars weestimated the distances, ranging between 1.3 and 10 kpc with a median of2.8 kpc from the galactic plane, indicating that BIS stars mostly belong tothe halo population. Key Words: — Stars: late type — Stars: variables — Stars: Mass loss1. INTRODUCTION: BYURAKAN –
IRAS
STARS CATALOGAsymptotic Giant Branch (AGB) stars are very luminous objects, in afast evolutionary phase which mix up the atomic nuclei produced by nuclearreaction in the inner part of the star into the external envelope and theninto the interstellar space through the stellar wind and the final formationof a Planetary Nebula. A better census of the AGB stars population andgood knowledge of their physical characteristics and variability is a basic re-quirement to perform satisfactory checks of our models of stellar evolutionand to study the chemical evolution of our Galaxy (Lebzelter et al. 2012;Bladh et al. 2013; Nowotny et al. 2013). The forthcoming direct distancemeasures by the GAIA mission of the stars within several kpc from the Sunwill allow a substantial improvement in this field. With this aim in mind,we performed a study of a sample of AGB candidates collected in the Byu-rakan Infrared Stars (BIS) catalog (Mickaelian & Gigoyan 2006), based onthe spectra of galactic sources visible in the objective prism plates of the FirstByurakan Survey (FBS) (Markarian et al. 1989) and on the
IRAS
PointSource Catalog IRAS PSC (1988).In 1983, the Infra Red Astronomical Satellite (
IRAS ) surveyed about96% of the sky in bands centred at 12, 25, 60, and 100 µ m. More than245000 point sources were detected and their fluxes and positions were listedin the IRAS
PSC. A recent work by Abrahamian et al. (2015) made across-check of the
IRAS
Point Source Catalog (PSC) and Faint Sources Cat-alog (FSC) Moshir et al. (1989) improving their positions and correlationwith infrared sources detected by more recent missions (
W ISE (Cutri et al.2012) ,
AKARI (Ishihara et al. 2010), 2MASS (Cutri et al. 2003)). The
IRAS archive also includes several spectra in the Low − Resolution Spectracatalog (IRAS LRS 1988) in the range 7.7 − µ m providing useful indi-cations on the chemical composition of the dust shells around the stars (seeOlnon & Raimond 1986, and the IRAS Explanatory Supplement 1988).The BIS catalog contains data for a final census of 276 IR sources as be-ing potentially stars of late spectral types. The stars were selected on thebasis of their low resolution spectra of FBS and of the Dearbon AstronomicalObservatory (Lee et al. 1947). Images of the Palomar Observatory Sky Sur-vey (DSS, Abell 1959 ) were also used to check the identification. The mostrecent citations of literature spectral types are to be found in Skiff (2016).IS SOURCES : OPTICAL RESULTS. 3The identifications on the FBS plates were carried out in the region with δ > +61 o and galactic latitude b > +15 ◦ , covering a surface of 1504 deg .For each object several information are given, such as accurate optical posi-tions for two epochs (B1950 and J2000), photometric data after cross − correlationwith MAPS (Cabanela et al. 2003), USNO − B1.0 (Monet et al. 2003) and2MASS catalogs, proper motions (PM) and a classification based on the Dig-itized First Byurakan Survey spectra (DFBS, Mickaelian et al. 2007) acces-sible online from the webpage http://ia2.oats.inaf.it/ . The stars have inter-mediate values of galactic coordinates, ranging in longitude between 90 and151 degrees, and in latitude between 14 and 45 degrees, with average value of30 degrees: they are therefore outside the solar circle and are likely membersof the halo / thick − disk population.Aiming at clarifying the nature of the stars included in the BIS catalog wemade a systematic collection of all the information available in literature andperformed the acquisition of new photometric and spectroscopic data.This paper is devoted to the optical properties of a subsample of 95 starsrandomly selected, while a companion paper will be dedicated to the infraredcharacteristics of all the stars of the BIS catalog using data collected frompublic archives. §
2, describes our targets; §
3, and 4, are devoted to theobservations and data analysis; §
5, is devoted to estimate and discussion ofthe parameters of the IR sources; our conclusions are summarised in § − type and Semi − Regular (SR) variables, OH and SiO sources,N − type carbon stars and unknown sources surrounded by thick circumstellarshells. Most of our targets are poorly studied both from the photometric andspectroscopic point of view and only a few of them are classified as variablesin the General Catalog of Variable stars (GCVS, Samus et al. 2017) or in theVariable Star Index (VSX, Watson et al. 2016) catalogs. A preliminary clueof variability was given in Mickaelian & Gigoyan (2006) using the differencesB1 − B2 and R1 − R2 of the two epochs of the DSS as reported in the USNO-B1.0 catalog.Among the 276 stars of the BIS catalog, 13 have color index 1.5 ≤ B − R ≤ − R ≥ − bandphotometry and medium resolution spectra of a sample of 95 stars: we haverandomly selected 9 with color index ≤ R filter in the pe-riod July − November 2011 with the TACOR telescope of the Department ofPhysics of the University “La Sapienza” in Rome equipped with an ApogeeU2 CCD. The data were reduced by means of standard IRAF procedures .Table 1 presents the journal of observations; the columns have the follow-ing meaning: 1 - BIS number in the catalog; 2 - IRAS
FSC designation of theobjects; 3 - date of observations; 4 - spectral type according to the classifica-tion from our CCD spectra; 5 - Red magnitude at the date of observation andobservatory: Loiano; Cima Ekar; all the other data were obtained withthe TACOR telescope. 6 - magnitude range from the archive of the NorthernSky Variability Survey (NSVS, Wozniak et al. 2004a; see Section 4.2); 7 -our variability classification defined in Section 4.2.A large number of spectroscopic standards (from K7 to M9 giants anddwarfs) were observed with the same instrumental configuration as our targetstars and used as the basis for our classification. We have selected the tem-plates of M standards from several catalogs and Spectral Libraries (Kirkpatrick et al.1991; Torres-Dodgen & Weaver 1993; Fluks et al. 1994; Allen & Strom 1995;Gray & Corbally 2009). Spectra of M standards were also downloaded fromthe site http: // kellecruz.com / M standards / . For the carbon stars we usedBarnbaum et al. (1996) and Totten & Irwin (1998).4. DATA ANALYSISIn the following sections we describe the general characteristics of thespectra and the optical light curves for individual stars, when available.4.1 . General characteristics for spectral classification Spectral types were derived by overlapping the spectral tracing of the tar-gets with those of the reference stars, overplotting them with IRAF/splot.Classification was made independently by three of us and the typical uncer-tainty is one subtype. We have checked the self-consistency of the referencestars in the same way. Most of the reference stars are nearby. The spectralresolution used allows to discriminate luminosity classes but not the metalcontent. TACOR = Telescopio A COntrollo Remoto IRAF is distributed by the NOAO, which is operated by AURA, under contract withNSF.
IS SOURCES : OPTICAL RESULTS. 5The spectra of almost all our targets are typical for M − type stars. Inthese stars the most prominent absorptions belong to the TiO bands at 4761,4954, 5167, 5448, 5862, 6159, 6700, 7055 and 7600 ˚A. In some cases theVO bands of the red system are also present, with several band heads in therange 7334 − − b triplet 5167, 5173, 5184 ˚A; the NaD doublet5890, 5896 ˚A; the CaOH diffuse bands centered at 5550 and 6230 ˚A, theMgH bands at 4780, 5211 ˚A; the CaH at 6382, 6908, 6946 ˚A (Mould 1976;Giovagnoli & Mould 1994). Few atomic lines belonging to Fe and Ti are alsopresent in some spectra.From their spectral characteristics four targets appears to be N − type AGBcarbon stars. Three of them are embedded in a dense envelope with the blueregion strongly underexposed.Two stars, BIS 104 and BIS 106, were erroneously classified as carbon starson the basis of the IRAS spectrum (Guglielmo et al. 1997) while our spectraclearly show the typical features of intermediate M-type stars.The nine stars in our sample with B − R ≤ ≥ at 6867 and 7594 ˚A, and of H O at 7186 ˚A, are notremoved. In these figures the ordinates are relative intensities corrected foratmospheric extinction, normalised with the maximum set to 100.4.2 . Photometric variability
Most of our targets are present in the NSVS database, collected between1997 and 2001 from the ROTSE − I (Robotic Optical Transient Search Experi-ment I) experiment (Wozniak et al. 2004a). The observations spanning up toone year gave us reliable indications about the photometric variability of oursample. We have downloaded all the available light curves from the NSVS website to check the photometric behaviour and compare the ROTSE magnitudes(R r ) with our new data. Some objects are not present in the NSVS archive,being fainter than ROTSE − I detection limit (15.5 mag). Few stars have been GAUDENZI, NESCI & AL.observed by the Catalina Real − time Transient Survey (CRTS ). Magnitudesfrom other catalogs have been only considered as indicative, being obtainedfrom the DSS plates where many stars are saturated.It is worth to remember that the NSVS data were obtained with an unfil-tered CCD, so that the quantum efficiency of the sensor makes the effectiveband most comparable to the Johnson R band (Wozniak et al. 2004a), orbetter a mix of V and R colors, which is a function of the spectral type ofthe star. To inter − calibrate the NSVS and our magnitudes we used the starsin our sample with a very stable NSVS light curve. Our photometry is tiedto the R magnitude scale of the GSC2.3.2 catalog (Lasker et al. 2008), anda good calibration would require a bigger set of non variable stars in theM0 − M8 spectral types range to define a reliable color correction. In fact, wehave verified that for M8 stars, which emit most the photons in the IR tailof the unfiltered detector, the ROTSE instrumental magnitude are generallybrighter than for M1 stars of similar R magnitude. In any case, even withthis caveat , our data have been useful to confirm the variability/stability ofthe stars of our sample.We assign three main variability indices on the basis of the light curves: (1) regular, large amplitude variables, larger than 1.1 mag; (2) irregular, large amplitude variables, between 0.5 and 1.1 mag; (3) small amplitude variables stars ( up to 0.5 mag) or non − variable ( up to0.3 mag).A few stars show small amplitude either irregular or quasi regular variability.To these stars we assigned intermediate variability classes (2/3, 2/1), reportedin Table 1 only. Here below we grouped stars with similar characteristics toavoid useless repetitions. For a number of stars we add spectroscopic and/orphotometric details. Variability class (1)
BIS 007; BIS 116; BIS 133 (IY Dra); BIS 196; BIS 267 :BIS 007: For this star Wozniak et al. (2004b) report an estimated periodof 341 days. The light curve is compatible with that of a Mira − variable star,but the photometric behaviour deserves a long term observational program toimprove the accuracy of the period. The star is erroneously classified in BIScatalog as an N − type carbon star. No other spectroscopic information wasfound in literature. We observed BIS 007 in July 2007 and July 2008. Theenergy distribution is similar in the two epochs as well as the spectral features;in both epochs H δ and H γ are in emission, other hydrogen lines being hiddenby the strong molecular absorptions. We classify this object as an S-type star,similar to HD56567 (S5/6 subtype). In Fig. 1 the 2008 spectrum is presented.BIS 116: During the ROTSE monitoring the star showed a continuousmodulation with period of 160 days, in agreement with Nicholson & Watkins(2006); our photometric data are in agreement with the expected values.Wozniak et al. (2004b) classified this star as a Semi − Regular variable but IS SOURCES : OPTICAL RESULTS. 7
Fig. 1. Selected optical spectra of stars with emission lines. From bottom totop: BIS 207 (M5, Semi − Regular); BIS 007 (Stype, Mira − type); BIS 116 (M2 − M4,Short − Period Mira − type). The blue part of the spectrum is replotted enlarged forbetter visibility above each star. GAUDENZI, NESCI & AL.
Fig. 2. Selected optical spectra. From bottom to top: BIS 034 (M2, very stable);BIS 264 (M4+dust); BIS 122 (M7, Semi − Regular with emission lines). The bluepart of the spectrum is replotted enlarged for better visibility above each star.
IS SOURCES : OPTICAL RESULTS. 9
Fig. 3. Spectral evolution of the two carbon stars BIS 036 and BIS 184 normalisedto 100. The spectra at different dates are vertically shifted for better visibility. − Period Mira. Wehave obtained three spectra, in 2007, 2008 and 2015. The continuum andthe intensity of molecular absorptions are variable. In the three spectra thehydrogen Balmer lines are in emission, though with different intensities. InFig.1 the 2008 spectrum is presented.BIS 133 ( IY Dra ): This is a Mira − type variable star whose photo-graphic magnitudes at minimum and maximum luminosities are presented inKazarovets et al. (2000). A lower limit of 351 days for the period is given byWozniak et al. (2004b). We carefully verified our R magnitudes which areall at the faint limits of the ROTSE magnitudes even taking into account thecolor correction. The star is strongly saturated in the POSS red plates. TheSloan Digital Sky Survey (SDSS ) reports r = 14 .
83 mag, corresponding toan R magnitude in the range 13.80 − R r R magnitude( 9.75 ) corresponds to a brighter object. This implies that the period shouldbe longer than the 255 days reported by the automatic calculation of NSVS.From the spectroscopic point of view BIS 267 is an interesting object withthe higher lines of the hydrogen Balmer series in emission ( extremely strongH γ and H δ , no H α , no H β ). The absorption spectrum is remarkable for thestrength of the TiO molecular bandsBIS 196: The ROTSE light curve, one year long, shows strong varia-tions with continuous modulation between R r IRAS color correction. The photometric behavior of this giantdeserves a long term observational program to compute the exact period.
Variability class (2e)
BIS 002; BIS 122; BIS 207; BIS 219; BIS 264 :Emission lines of the hydrogen Balmer series characterise the spectra ofthese stars (see Figures 1 and 2); the NSVS archive classify BIS 122, BIS 207and BIS 264 as Semi − Regular variable with a period of 147, 181 and 154 days,respectively, but modulations with shorter periodicity and small magnitude IS SOURCES : OPTICAL RESULTS. 11oscillations are also present in the light curves. Our observations show vari-ability inside the ROTSE range. Spectral features are typical for middle tolate − type giants. Variability class (2) Semi Regular
BIS 043 (KP Cam); BIS 198; BIS 276 ; BIS 001; BIS 006; BIS 014; BIS 032;BIS 037; BIS 038; BIS 088 ; BIS 103 ; BIS 104; BIS 106; BIS 120 ; BIS 123;BIS 132; BIS 138; BIS 168 ; BIS 173; BIS 200; BIS 211; BIS 213; BIS 214;BIS 271 :These stars have been classified by Wozniak et al. (2004b) as large am-plitude, Semi-Regular variables. The light curves show a semi-regular patternwith continuous modulation and large amplitude ( > − M7 giant stars.BIS 043 is also present as a Semi − Regular variable in 76th list of variablestars (Kazarovets et al. 2001). The visual inspection of its light curve doesnot support convincing evidence for the quasi periodicity of about 150 daysautomatically calculated by NSVS.BIS 138 is the only other star (besides the carbon BIS 184 quoted below),to have an
IRAS
LRS infrared spectrum classified as 24 (star with ”not toothick oxigen-rich envelope” ).BIS 276 shows two minima at the same magnitude at the beginning andthe end of the ROTSE monitoring, with a period automatically calculated of436 days. At variance with expectation, in all our 6 observations we foundthe source always nearly at the same luminosity, inside the ROTSE range.
Variability class (2) Irregular . BIS 003; BIS 004; BIS 015; BIS 039; BIS 126; BIS 136; BIS 142; BIS 145;BIS 154; BIS 156; BIS 167; BIS 170; BIS 209; BIS 212; BIS 216; BIS 226:
During the ROTSE monitoring the light curves showed irregular variabilitywith maximum amplitude of 0.5 mag. Our measures are generally slightlyfainter than the ROTSE values, in agreement with the expected differencedue to the color correction for the spectral type of the stars, and in betteragreement with the R values from literature. All these stars have spectraltypes between M5 and M8.The light curve of BIS 226 shows irregular variability with two minima at R r about 11.7 while our R magnitude 12.85 is much fainter in agreement withother VIZIER catalogs. Variability class (2/3)
BIS 010; BIS 107; BIS 113; BIS 172; BIS 174; BIS 201; BIS 224; BIS 255;BIS 275; BIS 285:
During the ROTSE monitoring these stars showed random variability withmaximum excursion of 0.3 magnitudes. The R magnitudes from our observa-tions are also in agreement with the ROTSE range. Most of these stars havean early M spectrum.We included the M5 star BIS 010 in this variability class in spite of the2 GAUDENZI, NESCI & AL.fact that this star was neither monitored by ROTSE because of its faintnessnor by the Catalina survey which does not cover its declination zone. Thebrightness of this star is probably overestimated in every optical catalog beingthe northern component of an apparent close binary (3 arcsec ). The southerncomponent almost disappears in the POSS2 IR plate and we found that itsspectrum is early G. From the comparison of the images of DSS1 with DSS2 nomagnitude variations of BIS 010 are evident. Our observations in 2013, 2016,2017 did not reveal spectroscopic or photometric variation. Its extremely redinfrared magnitudes (see our companion paper) cannot be justified by theimmediate explanation of a long period variable. A very long monitoringshould be anyhow recommendable. Variability class (3) . BIS 034; BIS 044; BIS 067; BIS 087; BIS 099; BIS 102; BIS 110; BIS 137;BIS 143; BIS 155; BIS 197; BIS 199; BIS 203; BIS 210; BIS 215 BIS 228; BIS 247*;BIS 248; BIS 256; BIS 258; BIS 260 :All these stars are very stable during the ROTSE monitoring with a maxi-mum variability of 0.1 magnitude. The photometric data available in literatureand our values indicate small fluctuations compatible with ROTSE or Catalinasurvey data. No spectroscopic information was found in the literature exceptthose by Mickaelian & Gigoyan (2006). Our revised spectral classificationturned out to be early M − type for all stars. Fig. 2 shows the spectrum ofBIS 034 which can be better classified as an S type star of subtype S2/3,very similar to the prototypes HD 49368 and HD22649. May be interestingis the very stable BIS 247 showing emission lines in the spectrum indicating acircumstellar envelope, whose origin would deserve further studies. Carbon stars
BIS 036; BIS 184 (HP Cam); BIS 222; BIS 194 :BIS 194 is stable while the other stars are Irregular variables. BIS 036 andBIS 222 have been identified as R Coronae Borealis candidates by differentauthors. These stars could not be observed by ROTSE being too faint. Wemade repeated observations of these stars; the first spectra obtained were pre-sented in Rossi et al. (2016). Here we will only report the spectral evolutionof BIS 036 and BIS 184 shown in Fig. 3. BIS 194 did not change and BIS 222progressively faded making impossible the acquisition of new spectra with ourinstruments.For BIS 036 we get the light curve from the Catalina Survey where the Vmagnitude ranges from 13.5 to 16.5. Actually we verified that on the POSS1plate taken on February 5, 1954 the star was quite luminous, brighter than thenearby star having RA(2000) 05:28:56.5 and DEC(2000) +69:20:31, while onthe POSS2 plate taken on October 31, 1994 it appears much fainter, likely afteran episode of mass ejection. On the FBS plate 0138a taken on November 21,1969 the star appears as a very short faint segment, consistent with R ≤
16. Onthe basis of the Catalina light curve and of the energy distribution, recentlyLee (2015) included this star in a list of R Coronae Borealis candidates,IS SOURCES : OPTICAL RESULTS. 13although the strong photometric variations, typical for R CrB stars have neverbeen reported. We have observed this star in 2008, 2016 and 2017 detectingchanges in the photometry and spectrum. The spectrum of BIS 036 is typicalfor a very late carbon star : in addition to the CN bands, the (0,1) transitionat λ molecule is barely visible. A deepNaD absorption is present at λ α in emission: thisfeature, added to the photometric behavior, while compatible with a longperiod variable star rules out the R CrB hypothesis.BIS 184: This N type carbon star is a known Semi − Regular variable star(HP Cam), listed in the General Catalog of Variable Stars (Samus et al.2017). A photometric variability with a period of 296 days is reported by(Wozniak et al. 2004b). The Low Resolution Spectrum from
IRAS is theonly spectroscopic reference (Kwok et al. 1997) where this object is classifiedas a carbon star based on the presence of the SiC emission feature at 11.2 µ m(LRS classification is 44, according to Little-Marenin et al. 1987). This clas-sification is confirmed by our optical spectra which also showed variations, asexpected from its classification. A substantial strengthening of H α occurredbetween January and March 2016. In Fig. 3 we do not show the February2017 spectrum, which is practically the same as January 2016.The variability of BIS 222, was ascertained from the comparison of thehistorical plates of the POSS and the FBS: it was as strong as the two nearbystars in the Red POSS1 plate taken in 1955, when its red magnitude was about17.3. It is barely visible in the FBS plate No. 1332 (1975): taken into accountthat 17.5 mag is the limit in the photographic band for the FBS plates, thiscould be a clue for the magnitude at that epochs. The star is invisible inthe POSS2 red plate (1996) and very strong in the POSS2 photographic IR(1997) plate; on January 2010 we have measured an R magnitude of 18.5.For all these reasons we assigned to BIS 222 the variability class (2) . Ourspectrum is consistent with that of a dust − enshrouded carbon star of a verylate (N8 − N9) subtype, similar to the well known IRC +10216 (CW Leo),an archetype of post AGB star and Pre − Planetary Nebula object. The onlyspectral features visible in the optical spectrum are the red and Near IR bandsof CN molecule at λ α is present in emission( see Fig.1 of Rossi et al. 2016 ). Note that Tisserand (2012) included thisstar among the R CrB candidates on the basis of the infrared colors and theenergy distribution ( star No. 1542 ), but the presence of H α could makequestionable this classification. Being the star fainter during our more recentobservations, we could obtain photometric data only. BIS 194 : the short monitoring (120 days only) by ROTSE indicates avery small modulation between 9.4 and 9.6 mag. Our R magnitude, 9.3 ± (3) taking into account the ROTSE monitoring4 GAUDENZI, NESCI & AL.and our photometry only. No spectroscopic information were found in theliterature. Our spectrum is that of an N type carbon star ( see Fig.1 ofRossi et al. 2016 ) with well expressed absorption of C molecule (Swansystem) with band heads at λλ λλ − − averaged IR magnitudes. Tothis purpose we have used 2MASS and IRAS catalogs. Being aware thatresults from sparse observations must be considered with care, but, havinga limited phase coverage of the measurements, obtained at random phases,we used the data as if they were the mean values of the magnitudes and, asuncertainties, we adopted the typical amplitude.
Carbon stars
To estimate the absolute luminosity and distance of the “naked” carbonstar BIS 194, we used the relation between the color index J − K and theabsolute K magnitude M K obtained by Demers & Battinelli (2007), validfor J − K in the range between 1.4 and 2.3 magnitudes.We obtained M K = − .
88 mag and d=3.8 kpc.Concerning the N − type dusty carbon stars BIS 036, BIS 184 and BIS 222,we applied various empirical relations and models available in literature. Theresults are in general agreement except for the mass loss rates, where the modeldependence is strong and the differences are significant from one method toanother. Our procedures are described below and the results are summarizedin Table 2 where the columns have the following meaning: 1 - BIS number;2 - absolute K magnitude M K ; 3 - range of distances d; 4 derived distanceto the galactic plane Z 5 - apparent bolometric magnitude m bol ; 6 - absolutebolometric magnitude M bol ; 7 - mass loss from Le Bertre & Winters (1998);8 - mass loss from Whitelock et al. (2006).IS SOURCES : OPTICAL RESULTS. 15TABLE 2PHYSICAL PARAMETERS AND DISTANCES FOR THE DUSTYCARBON STARS.BIS M K d Z m bol M bol Log ˙M Log ˙MNo. mag kpc kpc mag mag LW-98 Wh-06036 -6.9 : -7.3 10.7 : 13.0 3.4 : 4.0 10.6 : 11.0 -4.2 : -4.9 -5.6 -5.1184 -7.5 : -7.8 1.5 : 1.8 0.6 : 0.7 6.8 : 6.9 -4.1 : -4.3 -6.3 -5.8222 -7.0 : -7.4 9.8 : 12.6 3.0 : 3.8 9.9 : 11.0 -4.4 : -5.5 -5.3 -4.9TABLE 3PHYSICAL PARAMETERS AND DISTANCES OF THE O − RICHMIRAS.BIS P M K m bol M bol d Z Log ˙MNo. days mag mag mag kpc kpc007 341 -7.87 7.3 -4.80 2.65 ± .40 0.6 : 0.8 -7.92 ± ± .55 3.2 : 3.9 -7.43 ± >
351 -7.90 9.1 -4.84 5.90 ± .85 2.3 : 3.0 -7.26 ± ± .45 1.6 : 1.8 -7.35 ± >
255 -7.40 9.1 -4.45 5.20 ± .85 2.5 : 3.5 -8.2 ± K magnitudes using the relationsbetween M( K s ) and ( J − K s ) o from the calibration by Mauron (2008). Fromthese values we then computed the range of distances to the stars.A cross check of the estimated absolute K magnitude can be made in thecase of BIS 184 using the cited period of 296 days.From the relation by Knapp et al. (2003):M K = − .
34 LogP( d ) − . −→ M K = − . ± . K = − . − .
69 (LogP − . −→ M K = − . ± . m bol = K + BC K using the calibrations to the bolometric corrections byWhitelock et al. (2006). We obtained a range of values inside the limitsreported in column 4 of Table 2. The results obtained using the calibrationsby Le Bertre et al. (2001) are in good agreement with those listed in thetable.From the derived distances and from m bol we could infer the range ofabsolute bolometric magnitudes.For BIS 184 M bol can also be obtained in an independent way, using theperiod − luminosity relation by Feast et al. (2006) (see their Fig. 2 and eq. 1with c=2.06). The result, M bol = − ± − dependent models, with fixed assumptions on theparameters of the circumstellar shells; for the stars in common with a paper byWhitelock et al. (1994) they obtained a satisfactory agreement, though withlower values of ˙M. A good correlation between K − [12] color and the totalmass loss rate was derived by Whitelock et al. (2006) refining the previouswork. Our results confirm the systematic difference of a factor of three betweenthe two methods. In the last two columns of Table 2 we report the results (inM ⊙ /year), the uncertainties are about ± O − rich Miras For the M − type Mira variables a number of relations involving period andIR luminosities (discussed in our companion paper) can be applied. Theresults are summarised in Table 3 where the meaning of the columns is thefollowing: 1 - BIS number; 2 - variability period P reported by NSVS; 3 -absolute K magnitude M K ; 4 - apparent bolometric magnitude m bol ; 5 -absolute bolometric magnitude M bol ; 6 - distances and corresponding errors;7 derived distance to the galactic plane; 8 - mass loss.The results for BIS 133 reported in the table are lower limits, being theperiod longer than 351 days. Similarly we have assigned class 1 to BIS 267although this star is not known to be a Mira type but shows a regular pat-tern in the light curve of ROTSE experiment, as described above. At theopposite side, BIS 116 has the shortest period, the smallest amplitude andthe earliest spectral type: its position in almost all color − color diagrams isaligned with the long-period Miras. Whitelock et al. (2000) divided the Mirawith period below 225 days in two groups, “Short Period − Blue” and “ShortPeriod − Red” depending on their infrared colors and average spectral types.All spectroscopic and photometric characteristics of BIS 116 lead us to placethis star in the group of the “Short Period − Blue” Miras.We have computed the absolute K and bolometric magnitudes using therelation between period and magnitude given in Whitelock (2012) and refer-ences therein:IS SOURCES : OPTICAL RESULTS. 17M K = -3.69* (Log(P) − − bol = -3.00* Log(P) + 2.8For the uncertainty on the period, after accurate inspection of the NSVSlight curves, we assumed five days for BIS 116 and ten days for the other stars.By propagating the errors we obtained ∆M ∼ ∼ bol and M K .We have also computed the apparent bolometric magnitude with BC K = 3.15 mag for BIS 007, BIS 133, BIS 196, BC K = 2.9 mag for BIS 267, andBC K = 2.8 mag for BIS 116, derived from the calibrations by Whitelock et al.(2000). As uncertainties we assumed the typical amplitude of the K magni-tude that is 0.4 mag for the late − type Miras and 0.2 mag for BIS 116 andBIS 267.We could then infer a crude estimate of the distances and of the mass lossrate. The distances derived from the bolometric and the K magnitudes agreevery well within the errors. To compute the mass loss we used calibrationbetween ˙M and K − [12] color index obtained by Le Bertre & Winters (1998).Here the contribution to the uncertainties is also due to the non − simultaneousobservations of the different sets of IR data. Semi − Regular variables
Our sample includes several Semi − Regular variables, but only five, namelyBIS 122, BIS 198, BIS 207, BIS 209 and BIS 276 have well sampled light curvein the NSVS. We could derive a range of absolute magnitudes and distances byapplying the relations found by Knapp et al. (2003) and by Barthes et al.(1999). These last authors found different relations P − M K for different kine-matics characteristics and obtained significant results by dividing their datainto four groups, representative of four different populations. Group 1 and2 have kinematics characteristics corresponding to old disk stars; group 3has kinematics indicating a younger population, group 4 contains only highvelocity stars. Only group 1 contains long periods.We report the results in Table 4 where the columns have the followingmeaning: 1 - BIS number; 2 - variability period P; 3 - K magnitude from2MASS; 4 - absolute K magnitude M K from Knapp et al. (2003); 5 - rangeof M K from Barthes et al. (1999), obtained applying the relation giving thebest agreement with Knapp et al. (2003); 6 - corresponding group numberfollowing Barthes et al. (1999); 7 - range of distances from the minimum andmaximum of columns 4 and 5. 8 derived distance to the galactic plane. Non variable stars
All the non − variable or small amplitude variable stars are giant of M0-M4sub-classes. For these stars we derived a range of distances between 1.0and 3.3 kpc, by adopting absolute visual magnitudes − ≤ M V ≤ − − − REGULAR VARIABLE STARS.BIS P K s M K M K gr Dist K Zdays mag K03 B99 B99 kpc kpc122 147 6.63 -7.40 -7.35:-7.44 2/3 6.3 : 6.5 3.5 : 3.6198 161 5.71 -7.45 -7.42:-7.53 2/3 4.2 : 4.4 2.5 : 2.6207 181 6.42 -7.52 -7.52:-7.65 2/3 6.1 : 6.5 3.9 : 4.1209 296 5.00 -7.80 -7.95:-8.12 2/3 3.6 : 4.2 2.3: 2.6276 436 6.16 -8.00 -7.94:-8.18 1 6.5 : 7.4 3.4 : 3.9kpc, Z = 1 . , σ = 0 .
45 kpc). Regarding the supergiant BIS 137 (M0 I), byadopting an absolute visual magnitude M V ∼ ∼ ∼ − type stars, nine are earlier than M, the others are M − type giants.We have divided our stars into three main variability classes: regular vari-able, irregular variable, photometrically stable. About 60% of the stars showlarge amplitude, irregular or semi − regular light curves; in this category wehave included the three dust enshrouded carbon stars. All the early M − typestars and the naked carbon star BIS 194 were found to be stable. Only 5 starsare Mira variables. Our study pointed out some peculiar stars which deservemore detailed studies.Our spectral classification together with the data collected from literatureallowed us to estimate absolute magnitudes, mass loss and distances for anumber of targets, finding a good agreement between results obtained fromdifferent methods.Knowing the distances to the Sun and the galactic latitude, we can inferthat the median distance of the variable stars from the galactic plane is 3.0 kpc,( σ =1.5) none being farther than 5.4 kpc. Our spectral resolution does notallow to investigate for chemical differences between thick disk and halo stars.IS SOURCES : OPTICAL RESULTS. 19Anyhow, given the distance from the galactic plane, most of these stars arelikely to be halo members.For the non − variable stars we derived an average distance of 1.03 kpc fromthe galactic plane, suggestive of a mixed population of thick − disk and halo.Regarding the evolutionary status of our sample, most of the stars havean absolute magnitude appropriate for being in the AGB phase.In a companion study (Gaudenzi et al., submitted) we have analysed theInfrared properties of all the stars of the BIS catalog. To this purpose we havemade use of near-IR (2MASS), mid-IR ( W ISE ), and far- IR (
IRAS and
AKARI ) photometric data to investigate their behaviour on various color-magnitude and color-color diagrams and graphically distinguish various typesof sources.
Acknowledgements.
This research has made use of the SIMBAD database,operated at CDS, Strasbourg, France. This publication has made use ofdata products from: the Two Micron All − Sky Survey database, which isa joint project of the University of Massachusetts and the Infrared Processingand Analysis Center/California Institute of Technology; the Wide − field In-frared Survey Explorer, which is a joint project of the University of California,Los Angeles, and the Jet Propulsion Laboratory/California Institute of Tech-nology, funded by the National Aeronautics and Space Administration; theNorthern Sky Variability Survey (NSVS) created jointly by the Los AlamosNational Laboratory and University of Michigan; the NASA/IPAC Extra-galactic Database (NED) which is operated by the Jet Propulsion Laboratory(JPL), California Institute of Technology, under contract with the NationalAeronautics and Space Administration; the International Variable Star Index(VSX) database, operated at AAVSO, Cambridge, Massachusetts, USA. TheUniversity “La Sapienza” of Rome, Italy, supported the project with fundsfrom MIUR. REFERENCES Abell G.O. 1959 ASPL 8, 121Abrahamyan H. V., Mickaelian A. M., Knyazyan A. V.2015, Astronomy and Com-puting, 10, 99Allen L.E., Strom K.M. 1995, AJ, 109, 1379,VizieR On-line Data Catalog: J/AJ/109/1379Barnbaum C., Stone R.P.S., Keenan P.C. 1996, ApJS, 105, 419Barthes D., Luri, X., Alvarez A., Mennessier M.O. 1999, A&AS, 140, 55Bladh S., H¨ofner, S., Nowotny, W., Aringer B., Eriksson, K. 2013, A&A, 553, A20Cabanela,J.E., Humpheys, R., Aldering,J., et al. 2003, PASP, 115, 837Cutri R.M., Skrutskie M.F., van Dyk S., et al. 2003, The 2MASS All-Sky PointSource Catalog, University of Massachusetts and Infrared Processing and Anal-ysis Center (IPAC/California Institute of Technology)VizieR On-line Data Catalog: II/246WISE Vizier On-line Data Catalog: II/311Demers S., Battinelli P. 2007, A&A,473, 143
Feast M.W., Whitelock P., Menzies J.W. 2006, MNRAS 369, 791Fluks M.A., Plez B., The P.S., et al. 1994, A&AS, 105, 311Giovagnoli A., Mould J. 1994, PASP, 106, 376Gray R.O., Corbally C.J. 2009, Stellar Spectral Classification, (Princeton UniversityPress)Guglielmo F., Epchtein N., Arditti, F., Sevre F. 1997, A&AS, 122,489IRAS Catalogs and Atlases vol. 1, Explanatory Supplement, 1988, (C. Beichman, etal., NASA RP-1190)VizieR On-line Data Catalog: II/274Joint IRAS Science Working Group, 1987, IRAS Low Resolution Spectra (LRS),(NASA RP-1190 (1987))VizieR On-line Data Catalog: III/197Joint IRAS Science Working Group, 1988, IRAS Catalog of Point Sources, Version2.0 (IPAC 1986)VizieR On-line Data Catalog: II/125Ishihara, D., Onaka, T., Kataza, H., et al. 2010, A&A, 514A, 1AKARI/IRC Mid-Infrared All-Sky Survey, Vizier On-line Data Catalog II/297Johnson H.R., Querci F.R., Jordan S., et al. 1986, The M − Type Stars (NASA-SP492)Kazarovets E.V., Samus N.N., Durlevich O. 2000, IBVS No 4870, 1Kazarovets E.V., Samus N.N., Durlevich O. 2001, IBVS No 5135, 1Kirkpatrick D.J., Henry T.J., McCarthy D.W. Jr. 1991, ApJS, 77, 417Knapp G.R., Pourbaix D., Platais I., Jorissen A. 2003, A&A, 403, 993Kwok S., Volk K., Bidelman W.P. 1997, ApJS, 112, 557Lasker B., Lattanzi M.G., McLean B.J., et al. 2008, AJ, 136, 735VizieR On-line Data Catalog: I/305Le Bertre T., Winters J.M. 1998, A&A, 334, 173Le Bertre T., Matsuura M., Winters J.M., et al. 2001, A&A, 376, 997Lebzelter T., Heiter U., Abia C., et al. 2012, A&A 547 A,108Lee O.J., Gore G.D., Baldwin T.J. 1947, Annals of Dearborn Observatory, vol. V,Dearborn Observatory Catalogue of faint red stars, part 1CVizier on-line catalog II/68Lee, C.H. 2015, A&A, 575, A2.Little-Marenin I.R., Ramsay M.E., Stephenson C.B., Little S.J., Price S.D. 1987,AJ, 93, 663Markarian B.E., Lipovetsky V.A., Stepanian J.A., Erastova L.K., Shapovalova A.I.1989, Commun. Special Astrophys. Obs., 62, 5VizieR On-line Data Catalog: VII/172Mauron N. 2008, A&A, 482, 151Mickaelian A.M., Gigoyan K.S. 2006, A&A, 455, 765VizieR On-line Data Catalog: III/237AMickaelian A.M., Nesci R., Rossi C., et al. 2007, A&A, 464, 1177Monet D., Levine S.E., Canzian B., et al. 2003, AJ, 125, 984;VizieR On-line Data Catalog: I/284Mould J.R. 1976, ApJ, 207, 535Moshir et al. 1989 IRAS Faint Source Catalogue, version 2.0, 0Vizier On-line Catalog II/156Nicholson M., Watkins P. 2006, OEJVS, 31, 1Nowotny W., Aringer B., H¨ofner S., Eriksson K., 2013, A&A, 552, A20
IS SOURCES : OPTICAL RESULTS. 21
Olnon F.M., Raimond E., Neugebauer, G., et al. 1986, A&AS, 65, 607Rossi C., Gaudenzi S., Frasca A., et al. 2010, MemSAIS 14, 127Rossi, C., Dell’Agli F., Di Paola A., Gigoyan, K.S., Nesci, R. 2016, MNRAS, 456,2550Samus N.N., Kazarovets E.V., Durlevich O.V., Kireeva N.N., Pastukhova E.N. 2017,General Catalogue of Variable Stars: Version GCVS 5.1, Astronomy Reports, vol.61, No. 1, pp. 80-88VizieR On-line Data Catalog: B/gcvsSharples R.M., Whitelock P.A., Feast M.W. 1995, NMRAS, 272, 139Skiff B.A. 2016, B/mk Catalogue of Stellar Spectral Classifications, Lowell Obser-vatorySparke L.S., Gallagher J.S. III 2000, Galaxies In the Universe: An Introduction,(Cambridge University Press)Stephenson C.B. 1986, ApJ, 301, 927The P.S., Thomas D., Christensen C.G., Westerlund B.E. 1990, PASP, 102, 565Tisserand,P. 2012, A&A, 539, 51Torres-Dodgen A.V., Weaver A.W. 1993, PASP, 105, 693VizieR On-line Data Catalog: III/181Totten E.J., Irwin, M.J. 1998, MNRAS, 294, 1Watson, C., Henden, A. A., Price, A. 2016, AAVSO International Variable StarIndex VSX, yCat, 102027Whitelock P.A., Menzies J., Feast M.W., et al. 1994, MNRAS, 267, 711Whitelock P.A., Marang F., Feast M.W. 2000, MNRAS, 319, 728Whitelock P.A., Feast M.W, Marang F., Groenewegen M.A.T. 2006, MNRAS, 369,751Whitelock P.A. 2012, Ap&SS, 341, 123Wozniak P.R., Vestrand W.T., Akerlof C.W., et al. 2004a, AJ, 127, 2436http://skydot.lanl.gov/nsvs/nsvs.pht/Wozniak P.R., Williams S.J., Vestrand W.T., Gupta V.2004b, AJ, 128, 2965,VizieR On-line Data Catalog: J/AJ/128/2965Zacharias,N., Finch, C.T., Girard,T.M., et al. 2012, The fourth US Naval Observa-tory CCD Astrograph Catalog (UCAC4).VizieR On-line Data Catalog: I/322a
S.Gaudenzi and R.Nesci: INAF/IAPS, via Fosso del Cavaliere 100, 00133Roma, Italy ([email protected]); ([email protected]).C. Rossi: INAF-Osservatorio Astronomico di Roma, Via Frascati 33, 00040,Monte Porzio Catone (RM), Italy ([email protected]).S. Sclavi and C. Rossi: Dipartimento di Fisica, Universit`a La Sapienza, Piaz-zale Aldo Moro 3, 00185 Roma, Italy.K.S.Gigoyan and A.M. Mickaelian: V. A. Ambartsumian Byurakan Astrophys-ical Observatory (BAO) and Isaac Newton Institute of Chile, ArmenianBranch, Byurakan 0213, Aragatzotn province, Armenia.2 GAUDENZI, NESCI & AL.TABLE 1LOG OF OBSERVATIONS, SPECTRAL CLASSIFICATION,MAGNITUDES AND VARIABILITY CLASS. a BIS
IRAS
FSC Date Sp. type R mag R r Var. class001 F03503+6918 29 Nov 13 M7 III 10.17 ± ± ±
13 Jan 16 8.7 ± I mag003 F03558+7007 16 Dec 13 M6 III 9.78 ± ± ± ± − ± ±
14 Feb 17 14.01 ± I mag.014 F04173+7232 18 Dec 13 M6 III 10.93 ± ± ± ± − ±
213 Jan 16 15.53 ±
13 Jan 16 18.27 ± V mag18 Mar 16 16.1 ±
14 Feb 17 15.30 ±
037 F05286+7105 18 Dec 13 M7 III 12.72 ± ± − ± − ± − ±
2a 1
Loiano Observatory; Cima Ekar Observatory; all the other data wereobtained with the TACOR telescope.
IS SOURCES : OPTICAL RESULTS. 23TABLE 1LOG OF OBSERVATIONS, SPECTRAL CLASSIFICATION,MAGNITUDES AND VARIABILITY CLASS. Continue
BIS
IRAS
FSC Date Sp. type R mag R r Var. class044 F05468+7300 17 Dec 13 M4 III 9.12 ± ± − ± ± ± ± ± ± ± ± ± ± ± − ±
06 Jul 11 10.9 ± ± ± ± ±
120 F16359+6439 12 Jul 07 M6 III 11.20 ± − ± ± ± ± ± − ±
10 Jul 11 12.30 ± ± ± ± BIS
IRAS
FSC Date Sp. type R mag R r Var. class123 F17305+6432 03 Aug 07 M5 III 10.8 ± − ± ± ± ± ± − ± ± ± ± ± − ± ± ± ± ± − ± ± ± ± ± − ± ± ± ± ± ± − ± ± ± ± ± ± no data16 Dec 13 8.60 ± IS SOURCES : OPTICAL RESULTS. 25TABLE 1LOG OF OBSERVATIONS, SPECTRAL CLASSIFICATION,MAGNITUDES AND VARIABILITY CLASS. Continue
BIS
IRAS
FSC Date Sp. type R mag R r Var. class154 F12234+6915 20 May 1 4 M7 III 10.04 ± ± ± ± − ± ± ± ± ± − ± ± ± ± ± ± − ± ± ± ± ± ± ± ± ± ±
14 Feb 17 9.35 ±
194 F07003+6815 22 Jan 08 N 9.3 ± − ±
196 F07497+6526 22 Jan 08 M8 III 13.40 ± − ±
197 F08034+6612 18 Dec 13 M5 III 10.14 ± ± − BIS
IRAS
FSC Date Sp. type R mag R r Var. class199 F08520+6724 22 Jan 08 M1 III 8.64 ± − ± ± ± ± ± − ± ± ± ± ± ± − ±
06 Jul 11 10.41 ± ± ± ± ± ± − ± ± ± ± − ±
09 Jul 11 10.0 ± ± ± ± ± − ± ± ± ± ± ± IS SOURCES : OPTICAL RESULTS. 27TABLE 1LOG OF OBSERVATIONS, SPECTRAL CLASSIFICATION,MAGNITUDES AND VARIABILITY CLASS. Continue
BIS
IRAS
FSC Date Sp. type R mag R r Var. class215 F03328+7717 15 Jan 16 M3 III 12.54 ± ± ± ±
216 Dec 13 20.6 ±
13 Jan 16 20.9 ±
15 Feb 17 20.5 ±
15 Feb 17 17.60 ± I mag224 F04423+7314 17 Dec 13 M5 III 11.01 ± ± ± ± ± − ± ± ± ± ±
17 Mar 12 8.56 ±
17 Mar 12 9.35 ±
17 Mar 12 11.2 ±
258 F11594+7309 20 May 14 M0 III 8.64 ± ± − ± − ± ± ± ± ± BIS
IRAS
FSC Date Sp. type R mag R r Var. class271 F15152+7303 20 June 15 M4 III 10.5 ± ± − ± ± ± ± ± − ±
07 Jul 11 10.73 ± ± ± ± ± − ± ± ± ± ±±