SCIENCE WITH THE ARMENIAN VIRTUAL OBSERVATORY (ArVO)

Transcription

1 SCIENCE WITH THE ARMENIAN VIRTUAL OBSERVATORY (ArVO) A.M. MICKAELIAN 1, L.A. SARGSYAN 1, K.S. GIGOYAN 1, L.K. ERASTOVA 1, P.K. SINAMYAN 1, L.R. HOVHANNISYAN 1, E. MASSARO 2, R. NESCI 2, C. ROSSI 2, S. GAUDENZI 2, S. SCLAVI 2, G. CIRIMELE 3, D. WEEDMAN 4, J. HOUCK 4, D. BARRY 4, A. SARKISSIAN 5, W. THUILLOT 6, J. BERTHIER 6, P. PRUGNIEL 7, I. KOCHIASHVILI 8, G.A. MIKAYELYAN 9 1 Byurakan Astrophysical Observatory (BAO); 2 La Sapienza Università di Roma, Italy; 3 M.I.G.G., Italy; 4 Cornell University, Ithaca, N.Y., USA; 5 Service d Aéronomie, IPSL, France; 6 Institut de Mécanique Céleste et de Calcul des Éphémérides (IMCCE), Observatoire de Paris, France; 7 Observatoire de Lyon, CRAL, France; 8 Tbilisi State University and National Astrophysical Observatory, Georgia; 9 Yerevan State University (YSU), Armenia Abstract. The main goal of the Armenian Virtual Observatory is to develop efficient methods for science projects based on the digitized famous Markarian survey (Digitized First Byurakan Survey, DFBS) and other large astronomical databases, both Armenian and international. Two groups of projects are especially productive: search for new interesting objects of definite types by low-dispersion template spectra, and optical identifications of new gamma, X-ray, IR and radio sources. The first one is based on modeling of spectra for a number of types of objects: QSOs, Seyfert galaxies, white dwarfs, subdwarfs, cataclysmic variables, planetary nebulae, C stars, etc. Each kind of object appears in the DFBS with its typical SED and spectral lines (for objects having broad lines only). The search criteria define how many objects will be found for further study, and may restrict these numbers leaving with the best candidates. At present, a number of science projects of search for new objects have been started: search for blue stellar objects, search for extremely red objects, search for variable objects, etc. Optical identifications have been proven to be rather efficient for IR sources from IRAS PSC and FSC. Tests have been carried out for X-ray and radio sources as well. Key words: virtual observatories surveys databases QSOs white dwarfs cataclysmic variables carbon stars asteroids optical identification. 1. INTRODUCTION The Armenian Virtual Observatory (ArVO) ( is based on the Digitized First Byurakan Survey (DFBS) (Mickaelian et al. 2007; a collaboration project between the Byurakan Astrophysical Observatory (Armenia), La Sapienza Università di Roma (Italy), Cornell University (USA), and VO-France. ArVO provides a large area spectroscopic database where search and studies of new objects and optical identification of X-ray, IR, and radio sources are especially efficient. ArVO is a part of the International Virtual Observatories Alliance (IVOA) ( an organization which coordinates VO projects and organizes the development of VO standards. A number of science projects based

2 on ArVO have already been started using the DFBS low-dispersion spectra and all existing astronomical databases. The joint usage and comparative analysis of all existing data leads to new exciting results and discoveries. 2. LARGE AREA SPECTROSCOPIC DATABASES During the recent decades, a few large area spectroscopic surveys have been carried out. Among them the most important are the First Byurakan Survey (FBS, Markarian et al. 1989), the Second Byurakan Survey (SBS, Stepanian 2005), Case Low Dispersion Northern Sky Survey (Pesch et al and references therein), Hamburg Quasar Survey (HQS, Hagen et al. 1999), Hamburg-ESO Survey (HES, Wisotzki et al. 2000), and Sloan Digital Sky Survey (SDSS, Adelman-McCarthy et al. 2007). By its accuracy and sensitivity, SDSS has a number of advantages compared to the previous photographic surveys. However, some of them (FBS, HQS, and HES) have been digitized and provide low-dispersion spectra over large area of sky. DFBS is the largest area spectroscopic survey covering 17,000 sq. degrees and providing data for more than 20,000,000 objects at high-galactic latitudes. Moreover, DFBS may be a reliable complement for SDSS at brighter magnitudes where SDSS images are saturated, and may be used for search for new bright QSOs and Seyfert galaxies, as well as a number of other types of objects. ArVO is a project of the Byurakan Astrophysical Observatory (BAO) aimed at construction of a modern system for astronomical data archiving, extraction, acquisition, reduction, use and publication as for the DFBS data, so as for all existing information in other astronomical databases. 2. THE DIGITIZED FIRST BYURAKAN SURVEY DFBS The DFBS is the digitized version of the Markarian survey (the First Byurakan Survey, FBS; Markarian et al. 1989). The FBS has been carried out by Markarian, Lipovetski and Stepanian in with the Byurakan 1m Schmidt telescope and 1.5 prism Kodak IIaF plates in 1142 fields have been taken, so that there is a good-quality plate in each field. FBS covers more than 17,000 deg 2 at d>- 15 and b >15. The limiting magnitude for the majority of plates is 17.5 m -18 m. The dispersion is 1800 Ǻ/mm near Hg (mean spectral resolution being about 50Ǻ). The spectral range is ǺǺ, and there is a sensitivity gap near 5300Ǻ. Thus, it is possible to compare the red and blue parts of the spectrum (easily distinguishing red and blue objects), follow the spectral energy distribution (SED), notice some emission and absorption lines, and have some understanding about the nature of objects. Each FBS plate contains low-dispersion spectra of some 15,000-

3 20,000 objects, and there are ~ 40,000,000 spectra for ~ 20,000,000 objects in the whole survey. Markarian survey was the first systematic objective-prism survey in the world. The DFBS provides all FBS spectra in the digital format and is available through its web portal at Rome ( (Fig. 1). Fig 1. The DFBS web portal at Rome. The picture shows a piece of a DFBS plate with a given central position and an extracted spectrum on the right with its available data measured from the DFBS, including the position, R and B magnitudes, rough classification, and the length in pixels. The FBS was conducted originally for search for galaxies with UV-excess, which was a new method of search for AGN. Markarian survey led to a sample of 1515 UV-excess (UVX) galaxies (later called Markarian galaxies). The study of Markarian galaxies brought to discovery of many new Seyferts and spectral classification of this type of objects and first definition of starburst galaxies. The catalog of Markarian galaxies is available at CDS (Markarian et al. 1997). The second part of the FBS was devoted to the discovery and study of the blue stellar objects (BSOs) (Abrahamian et al. 1999; Mickaelian 2000). A survey on late-type stars on the FBS plates is being carried out since 1987 (Gigoyan et al and references therein). Another project was optical identification of 1577 previously

4 unidentified sources from the IRAS Point Source Catalog (PSC) (IRAS 1988) and construction of the Byurakan-IRAS galaxy (BIG) and Byurakan-IRAS Stars (BIS) samples (Mickaelian & Sargsyan 2004; Mickaelian & Gigoyan 2006). The DFBS project was conducted to make available the FBS plates for the astronomical community, and provide the database for future efficient use. The DFBS included the digitization (scanning) of the FBS plates, a high-accuracy astrometric solution, extraction software, wavelength and flux calibration, creation of templates and automated classification of low-dispersion spectra, the DFBS catalog and database, and creation of the DFBS web page and user interface. 3. CLASSIFICATION OF THE DFBS SPECTRA The accurate classification of the DFBS spectra is especially important for the correct recognition and distinguishing of the objects for various purposes. In any case, the classification is approximate, as only some broad absorption and emission lines are seen and the spectral energy distribution (SED) is the main criterion. Fig 2. Low-dispersion 2D spectra and their extracted 1D counterparts. First row: A-type star, AGN (Mrk 266), QSO, and WD. Second row: CV, PN, M star, and N-type carbon star (from left to right). Two opposite approaches are applied for the classification. The first is based on modeling template spectra for different types of known objects. A search for these objects is being made in the DFBS, and their low-dispersion spectra are being extracted (Fig. 2). Then we search for similar new objects (QSO, BLL, Sy, CV,

5 WD, sd, M, C stars, etc.) among the FBS low-dispersion spectra. The second approach is based on making a numerical classification scheme for all FBS spectra. The classification principles are based on the relation of magnitudes and widths of the spectra (for separation of stellar and diffuse objects), SED, colors, length of the spectrum, presence or absence of broad spectral lines, etc. The third possible method is based on modeling of theoretical spectra for a number of types of objects: QSOs, Seyfert galaxies, white dwarfs, subdwarfs, cataclysmic variables, planetary nebulae, C stars, etc. The SED, emulsion response curve, calibration, and some other effects must be taken into account, which makes this method rather tricky and not practical. 4. SCIENCE PROJECTS WITH THE ARMENIAN VIRTUAL OBSERVATORY Two groups of projects based on DFBS are especially productive: search for new interesting objects of definite types by low-dispersion template spectra, and optical identifications of new gamma, X-ray, IR and radio sources. The first one is based on making templates or modeling of spectra for a number of types of objects: QSOs, Seyfert galaxies, white dwarfs, subdwarfs, cataclysmic variables, planetary nebulae, C stars, etc. Each kind of object appears in the DFBS with its typical SED and spectral lines when available, however affected also by its brightness, so that each template works for definite range of magnitudes. The search criteria define how many objects will be found for further study, and may restrict these numbers leaving with the best candidates. At present, several projects of search for new objects have been started: search for blue stellar objects (including bright QSOs and Seyferts), search for extremely red objects, search for variable objects, search and study of asteroids in DFBS, etc. Optical identifications have been proven to be rather efficient for IR sources from IRAS PSC and FSC, and SST, as well as for X- ray and radio sources SEARCH FOR BLUE STELLAR OBJECTS One of the main projects based on the DFBS is the continuation of the survey for Blue Stellar Objects (BSOs). QSOs and Seyferts are the most interesting group among them. Though there are more than 140,000 quasars known so far (Véron- Cetty & Véron 2006; Adelman-McCarthy et al. 2007), the number of bright quasars (Schmidt & Green 1983) is still not complete (Mickaelian et al. 2001). Therefore, search for new bright quasars missed in SDSS are very important. They must be very nearby and allow study of the Local Universe. Compared to faint ones, they give possibility to study them in details (spectra, morphology, etc.). Bright QSOs are probable strong sources of g, X-ray, IR, radio, and therefore,

6 space telescope targets. Among the 69 very high luminosity objects (M abs -30.0) almost all have been discovered from the surveys like HQS, SBS, CSO, etc., and only 4, from the SDSS, which means that the bright quasar surveys must be used for further searches of such exotic objects. Finally, the AGN classification is based on bright objects, and more objects would be extremely useful. Other interesting types among the BSOs are cataclysmic variables (CV), planetary nebulae nuclei (PNN) white dwarfs (WD), hot subdwarfs, horizontal branch B (HBB) stars, etc. The same criteria as previously used during the eye selection of such objects are being used. In addition, the DFBS numerical data (colors and the length of the spectrum) allow us keep the selection homogeneously and exclude erroneous objects included in the previous lists. Using the dedicated BSpec software written by one of the authors (GC), we have obtained a list of DFBS stars, their positions, B and R magnitudes, and preliminarily classification for DFBS zones with central d=+39 and d=+43. The spectral length l>90pix (compared to the total length l=107pix) was used as a criterion to search for UV excess objects, as this corresponds to the criteria used during the 2nd part of the FBS. However, the spectra of objects with B<13 always occupy the full length, and they were excluded from the lists. On the other hand, for the fainter objects (near the plate limit), we weaken the criteria of selection (l>80pix), as their spectra are shorter. An additional point for the UV excess object classification is the following: the spectra of the UV excess objects are divided into two parts by a sensitivity gap in green; the red-yellow part of the spectra must be weaker and the blue-ultraviolet part must be brighter and more extended. We started the project in the DFBS zone d=+39 and d=+43 to compare the results with those obtained before during the 2nd part of the FBS. Later on, cross-correlations with available catalogs and multiwavelength analysis was made for the found objects SEARCH FOR EXTREMELY RED OBJECTS Among the extremely red objects we expect mainly stars of late M8-M9 sub-types or very late N-type Carbon stars close to the FBS detection limit. However, some high-redshift QSOs may be found as well. The criteria for searching of such objects in the DFBS are based on spectral length (l<40pix, compared to the total length of l=107pix). To avoid miss-selection, we have excluded objects near the plate limit, as due to the low sensitivity of the photographic emulsion in blue, all such spectra are short. On the other hand, very bright stars show longer spectra even when their spectral type is late, so the criteria must be weakened to l<50pix. We obtained the first list of candidate objects for the DFBS zone with central d=+39 with their positions, B and R magnitudes, and other spectral data. Our purpose is to complete the selection procedure for such objects, based on all FBS sky area coverage and

7 investigation of these objects at high Galactic latitudes. Cross-correlations with X- ray, optical, IR, and radio catalogs have been made. The DFBS spectrum of a highredshift QSO B is given in Fig. 3. Fig 3. The high-redshift QSO B with the Lyman break clearly seen in the DFBS spectrum SEARCH FOR VARIABLE OBJECTS Using BSpec program for DFBS central d=+39 and d=+43, we have got a list of DFBS stars, their positions, B and R magnitudes, and preliminarily classifications. Later on, B and R magnitudes were cross-correlated with their corresponding B and R magnitudes from other plates in the same field in order to find out variability between two or more repeated plates stars. First, the systematic shift between the two plates measurements was eliminated, and then the rms was calculated for objects grouped by their B magnitudes (to avoid changes due to non-linearity). For variability selection 3s criteria for B and R magnitudes were used. The selected variables were checked with their multiwavelength data for their variability confirmation. After analysis of all available data, the objects were grouped into candidates of blazars (those having radio identifications in NVSS (Condon et al. 1998) and FIRST (Becker et al. 2003)), cataclysmic variables (blue objects), Mira type variables (late type stars), etc. Blazars are the most important subsample among these objects and space observations are going to be organized with the X- ray telescopes SEARCH AND STUDY OF ASTEROIDS This project was started on the initiative of W. Thuillot and J. Berthier (IMCCE, Paris) and is aimed at discovery and study of known and new asteroids in the

8 DFBS fields (Thuillot et al. 2007). Aladin and SkyBote VO tools have been used. First, the brightest asteroids (<15 m -16 m ), which may be visible in the DFBS plates were selected from the catalog. Then they were divided into two groups of fast and slow asteroids with a division parameter, estimated as the motion of 3² during 20 min (the typical exposure time of a DFBS plate), which meant that the spectra were extended or stellar. All asteroid spectra found in DFBS by SkyBote were extracted. In addition to the study of known asteroids to improve their ephemeris, another task was the modeling of template spectra of asteroids by means of the star-like spectra and further search for new candidate asteroids by similar spectra and comparison with DSS1/DSS2 fields for elimination of the stars. Spectral analysis of the asteroid spectra to get some physical parameters will later be made. Fig. 4 gives spectra for four known asteroids (three fast and one slow) extracted from DFBS. Fig. 4. Low-dispersion spectra of four asteroids extracted from the DFBS plates. From left to right: 104 Klymene, 288 Glauke, 627 Charis, and 1323 Tugela. Tugela is a slow asteroid and its spectrum is not extended during the 20 min exposure OPTICAL IDENTIFICATION OF X-RAY, IR, AND RADIO SOURCES The DFBS fields and the low-dispersion spectra are ideal tool for optical identification of non-optical sources, like X-ray (ROSAT (Voges et al. 1999; 2000), etc.), IR (IRAS (IRAS 1988; Moshir et al. 1990), SST ( etc.), and radio (NVSS (Condon et al. 1998), FIRST (Becker et al. 2003), etc.) ones. While the direct images only give an understanding about the position, brightness and extension of objects, the spectra prompt about the nature of objects based on the spectral energy distribution, color, spectral lines, etc. Fig. 5 clearly shows how the procedure works and how easy the selection of the correct optical counterpart would be.

9 Fig. 5. Automatic optical identification procedure for X-ray, IR and radio sources in the corresponding error circles from DSS and DFBS containing direct and low-dispersion spectral images respectively. One of the identification projects is the study of stellar objects among the Spitzer Space Telescope (SST) IR sources in the Boötes region (RA=14:32:05.71, DEC=+34:16:47.5, 3x3 degree field) (Hovhannisyan et al. 2008). Along with the DFBS spectra, DSS1/DSS2 (McGlynn et al. 1994; Lasker et al. 1996), SDSS, USNO-B1.0 (Monet et al. 2003), MAPS (Cabanela et al. 2003), 2MASS (Cutri et al. 2003), and other multiwavelength databases have been used as well providing accurate positions, multiwavelength photometry, proper motions, variability information, etc. 136 stellar objects have been identified, including a few with 24 μm IR excesses referred to their circumstellar envelopes and disks. 5. SUMMARY Together with the DFBS project, some other digitization projects in Byurakan (DSBS, 2.6m spectra, Coma region plates, etc.) are active and will provide data for the Armenian Virtual Observatory. It provides a modern environment for use of large astronomical databases and efficient work with them. A number of science projects based on the ArVO are being developed, including most important automatic search for new objects (particularly, AGN), and optical identifications of radio/ir/x-ray sources using the low-dispersion spectra. DFBS has been extremely useful for a number of science projects based on the selection and classification of the low-dispersion spectra. DFBS opens a new time domain important for proper motion and variable objects. DFBS should be used together with other multiwavelength data. A global spectroscopic database will be rather useful: a combination of DFBS, DSBS, HQS, HES, etc. VO standards are being developed for the DFBS and ArVO. E.g., Simple Spectral Access (SSA) has been made by

10 one of the authors (PP) for the DFBS spectra giving possibility to use the DFBS in the VO software such as VOSpec and VOSED. Acknowledgments. AMM acknowledges the support from UNESCO-ROSTE for his participation in the SREAC meeting in Athens and the organizers for their hospitality. The authors acknowledge ANSEF grant in 2007, CRDF grant ARP YE-06, and ISTC grant A-1451, as well as support from La Sapienza Università di Roma, Cornell University, and ANR (France). REFERENCES Adelman-McCarthy J.K. et al.: 2008, Astrophys. J. Suppl. Ser., in press. Abrahamian, H. V., Mickaelian, A. M., Lipovetsky, V. A., Stepanian, J. A.: 1999, CDS, Strasbourg, Catalog II/223. Becker R.H., Helfand D.J., White R.L., et al.: 2003, The FIRST Survey Catalog, Version 2003Apr11, 1997, Astron. J. 475, 479. Cabanela, J. E., Humphreys, R. M., Aldering, G., et al.: 2003, Publ. Astron. Soc. Pacific, 115, 837. Condon, J. J., Cotton, W. D., Greisen, E. W., et al.: 1998, Astron. J., 115, Cutri, R. M., Skrutskie, M. F., Van Dyk, S., et al.: 2003, The 2MASS All-Sky Catalog. Final Release, University of Massachusetts and IPAC/California Institute of Technology. Gigoyan K.S., Abrahamyan H.V., Azzopardi M., et al.: 2003, Astrophysics, 46, 475. H.-J Hagen, D.Engels, D. Reimers: 1999, Astron. Astrophys. Suppl. Ser. 134, 483. Hovhannisyan L.R., Weedman D., Mickaelian A.M., et al.: 2008, Astron J., in press. IRAS Catalogs and Atlases: 1988, 2. The Point Source Catalog, NASA. Lasker, B. M., Doggett, J., McLean, B., et al., ASP Conf. Ser. 101, 88, Markarian, B. E., Lipovetski, V. A., Stepanian, J. A., Erastova, L. K., Shapovalova, A. I.: 1989, Commun. Spec. Astrophys. Obs. 62, 5. Markarian, B. E., Lipovetski, V. A., Stepanian, J. A.: 1997, CDS, Strasbourg, Catalog VII/172. McGlynn, T., White, N. E., Scollick, K., ASP Conf. Ser., 61, 34, Mickaelian, A. M.: 2000, Astron. Astrophys. Transactions, 18, 557. Mickaelian, A. M., Gonçalves A. C., Véron-Cetty, M.-P., Véron, P.: 2001, Astrophysics, 44, 14. Mickaelian, A. M., Sarsgyan, L. A.: 2004, Astrophysics, 47, 213. Mickaelian A.M., Gigoyan K.S.: 2006, Astron. Astrophysics, 455, 765. Mickaelian A.M., Nesci R., Rossi C., et al.: 2007, Astron. Astrophysics, 464, Monet, D. G., Levine, S. E., Casian, B., et al.: 2003, Astrophys. J. 125, 984. Moshir, M., Kopan, G., Conrow, T., et al., Infrared Astronomical Satellite Catalogs, The Faint Source Catalog, Version 2.0, NASA, Pesch P., Stephenson C.B., MacConnell D.J.: 1995, Astrophys. J. Suppl. Ser. 98, 41. Schmidt, M., Green, R. F.: 1983, Astrophys. J., 269, 352. Stepanian, J. A.: 2005, Rev. Mex. Astron. Astrofis. 41, 155. Thuillot W., Berthier J., Sarkissian A., et al.: 2007, Hi. Astron., 14, Cambridge Univ. Press, 616. Véron-Cetty, M.-P., Véron, P.: 2006, Astron. Astrophys. 455, 773. Voges, W., Aschenbach, B., Boller, T., et al., The ROSAT all-sky survey Bright source catalogue, Astron. Astrophys. 349, 389, Voges, W., Aschenbach, B., Boller, Th., et al.: 2000, ROSAT All-Sky Survey Faint source Catalogue, Max-Planck-Institut fur extraterrestrische Physik, Garching. Wisotzki L., Christlieb N., Bade N., et al.: 2000, Astron. Astrophys. 358, 77. Received on 30 December 2007.