XUV-DISK) GALAXIES in the LOCAL UNIVERSE David A
Total Page:16
File Type:pdf, Size:1020Kb
The Astrophysical Journal Supplement Series, 173:538Y571, 2007 December A # 2007. The American Astronomical Society. All rights reserved. Printed in U.S.A. A SEARCH FOR EXTENDED ULTRAVIOLET DISK (XUV-DISK) GALAXIES IN THE LOCAL UNIVERSE David A. Thilker,1 Luciana Bianchi,1 Gerhardt Meurer,1 Armando Gil de Paz,2 Samuel Boissier,3 Barry F. Madore,4 Alessandro Boselli,3 Annette M. N. Ferguson,5 Juan Carlos Mun´oz-Mateos,2 Greg J. Madsen,6 Salman Hameed,7 Roderik A. Overzier,1 Karl Forster,8 Peter G. Friedman,8 D. Christopher Martin,8 Patrick Morrissey,8 Susan G. Neff,9 David Schiminovich,10 Mark Seibert,8 Todd Small,8 Ted K. Wyder,8 Jose´ Donas,3 Timothy M. Heckman,11 Young-Wook Lee,12 Bruno Milliard,3 R. Michael Rich,13 Alex S. Szalay,11 Barry Y. Welsh,14 and Sukyoung K. Yi12 Received 2007 April 30; accepted 2007 September 11 ABSTRACT We have initiated a search for extended ultraviolet disk (XUV-disk) galaxies in the local universe. Here we com- pare GALEX UVand visibleYNIR images of 189 nearby (D < 40 Mpc) S0YSm galaxies included in the GALEX Atlas of Nearby Galaxies and present the first catalog of XUV-disk galaxies. We find that XUV-disk galaxies are surpris- ingly common but have varied relative (UV/optical) extent and morphology. Type 1 objects (k20% incidence) have structured, UV-bright/optically faint emission features in the outer disk, beyond the traditional star formation thresh- old. Type 2 XUV-disk galaxies (10% incidence) exhibit an exceptionally large, UV-bright/optically low surface bright- ness (LSB) zone having blue UVÀKs outside the effective extent of the inner, older stellar population, but not reaching extreme galactocentric distance. If the activity occurring in XUV-disks is episodic, a higher fraction of present-day spirals could be influenced by such outer disk star formation. Type 1 disks are associated with spirals of all types, whereas Type 2 XUV-disks are predominantly found in late-type spirals. Type 2 XUV-disks are forming stars quickly enough to double their (currently low) stellar mass in the next Gyr (assuming a constant star formation rate). XUV-disk galaxies of both types are systematically more gas-rich than the general galaxy population. Minor external perturba- tion may stimulate XUV-disk incidence, at least for Type 1 objects. XUV-disks are the most actively evolving galax- ies growing via inside-out disk formation in the current epoch, and may constitute a segment of the galaxy population experiencing significant, continued gas accretion from the intergalactic medium or neighboring objects. Subject headinggs: galaxies: evolution — ultraviolet: galaxies Online material: machine-readable table, extended figure set 1. INTRODUCTION Kruit & Searle 1981; Florido et al. 2001), but rather are best rep- resented as a complex broken exponential (Pohlen & Trujillo Imprinted within the disk component of any spiral galaxy is a 2006; Erwin et al. 2005; Pohlen et al. 2002), sometimes even flat- wealth of information regarding its formation, and clues concern- tening in the outer disk. The color of the disk is also a function of ing its continued evolution. For instance, radial variations in the galactocentric distance (e.g., de Jong 1996; MacArthur 2006; average surface brightness of the stellar disk are not purely expo- MacArthur et al. 2004; Mun´oz-Mateos et al. 2007). Such com- nential with a sharp outer truncation, as once thought (e.g., van der plex surface brightness and color profiles trace changes in the disk stellar population, and reflect the assembly history of the galaxy, 1 Center for Astrophysical Sciences, The Johns Hopkins University, 3400 modulated by the combined effects of radial motion (of secular North Charles Street, Baltimore, MD 21218; [email protected]. or external origin) and extinction. The chemical evolution of a gal- 2 Departamento de Astrofı´sica, Universidad Complutense de Madrid, Madrid axyisalsodrivenbythestarformation history (Boissier & Prantzos 28040, Spain. 3 2000; Chiappini et al. 2003). Abundance gradients typically reveal Laboratoire d’Astrophysique de Marseille, BP 8, Traverse du Siphon, 13376 a comparatively enriched inner disk (Zaritsky et al. 1994; Henry Marseille Cedex 12, France. 4 Observatories of the Carnegie Institution of Washington, 813 Santa Barbara & Worthey 1999), although barred spirals are less steep in gen- Street, Pasadena, CA 91101. eral. The overall picture which emerges is that the outer disk is 5 Institute for Astronomy, University of Edinburgh, Royal Observatory often the locale of most significant recent change, although it re- Edinburgh, Edinburgh, UK. mains considerably less evolved, as expected for inside-out disk 6 School of Physics A29, University of Sydney, NSW 2006, Australia. 7 Five College Astronomy Department, Hampshire College, Amherst, formation (White & Frenk 1991; Mo et al. 1998; Boissier & MA 01003. Prantzos 1999; Brook et al. 2006; Mun´oz-Mateos et al. 2007). 8 California Institute of Technology, MC 405-47, 1200 East California Boulevard, Quantifying the recent star formation rate (SFR) across the entire Pasadena, CA 91125. 9 disk in a large galaxy sample will provide an anchor for the inter- Laboratory for Astronomy and Solar Physics, NASA Goddard Space Flight pretation of enrichment and disk-building processes. Center, Greenbelt, MD 20771. 10 Department of Astronomy, Columbia University, New York, NY 10027. In a broader sense, understanding the typical star formation 11 Department of Physics and Astronomy, The Johns Hopkins University, activity of outer disks is crucial because it has a bearing on diverse Homewood Campus, Baltimore, MD 21218. topics, including absorption-line diagnostics of metal abundance 12 Center for Space Astrophysics, Yonsei University, Seoul 120-749, Korea. in the circumgalactic medium, efforts to determine how (and at 13 Department of Physics and Astronomy, University of California, Los Angeles, CA 90095. what rate) gas accretes onto seemingly evolved galaxies, and how 14 Space Sciences Laboratory, University of California at Berkeley, 601 star formation is regulated (or inhibited) in low-density regions Campbell Hall, Berkeley, CA 94720. perhaps similar to protogalactic environments in the early universe. 538 XUV-DISK GALAXIES IN THE LOCAL UNIVERSE 539 It is known from deep H imaging that some galaxies possess Section 3 outlines our data analysis and XUV-disk galaxy clas- very extended star-forming disks (e.g., Ferguson et al. 1998b). sification scheme. We also examine global galaxy parameters and However, the incidence and underlying cause of this behavior are environmental properties of the XUV-disk class in comparison to still unknown. The new ultraviolet imaging surveys from NASA’s the entire sample. Section 4 presents an extensive discussion and GALEX (Galaxy Evolution Explorer) mission greatly facilitate the our main conclusions are listed in x 5. In Appendix A, we pro- study of low-intensity, outer disk star formation in a statistically vide a description and multiwavelength imaging for each galaxy significant number of galaxies (Gil de Paz et al. 2007a). GALEX in our XUV-disk catalog. Appendix B presents a short summary imaging benefits from a very low sky background, high sensitiv- of the inside-out disk formation models used to compare with ob- ity, and a large field of view (1.25 diameter), enabling efficient served galaxy characteristics. detection of otherwise elusive activity. By observing in the UV, GALEX probes a population of OB stars (rather than only the 2. SAMPLE SELECTION AND OBSERVATIONS higher temperature O stars predominantly ionizing H ii regions 2.1. The Galaxy Sample traced by H emission), and thus can catch galaxy evolution pro- cesses occurring at very low SFR surface densities (ÆSFR ). Our survey sample (Table 1) comprises 189 disk galaxies Thilker et al. (2005a) and Gil de Paz et al. (2005) discovered (À0:5 T 9:5, consequently S0YSm) within 40 Mpc included extended UV (XUV) emission in the extreme outer disk environ- in the GALEX Atlas of Nearby Galaxies (Gil de Paz et al. 2007a) 00 ment of M83 and NGC 4625. Both these studies established that and having i 80 , D25 > 90 , in addition to modest Galactic the amount and spatial extent of star formation in the outskirts of extinction, AV 0:5. By selecting only objects from the Atlas, we a galaxy can be underestimated by looking for H ii regions alone ensure that all galaxies were observed with homogeneous (deep) (as traced by H emission). However, a multitude of unanswered sensitivity. M31 and M33 have been excluded, as their large questions follow from the XUV-disk discovery. For instance, are angular size significantly complicates our analysis. The upper dis- XUV-disks rare or relatively common (e.g., Zaritsky & Christlein tance limit ensures that spatial resolution of our GALEX observa- 2007)? Are certain galaxy types more likely to host spatially ex- tions is 1 kpc or less. tended star formation? What is the characteristic morphology of Galaxy distances were assigned following Gil de Paz et al. XUVemission? Is the SF episodic, or continuous? M83 and NGC (2007a), who established a ranking of reliability for various pos- 4625 are rather dissimilar in their extended UV morphology. Does sible indicators (e.g., Cepheids, tip of the red giant branch, etc.) À1 À1 that suggest that the outer disk has a different origin or star for- with use of the Hubble constant (H0 ¼ 70 km s Mpc )adopted mation history (SFH), or have we simply caught these particular only as a fallback method. XUV-disks at different stages in their evolutionary progression? LEDAwas employed to extract ancillary quantities, such as ro- Outer disk star formation does have an impact on the course of tation velocities, mean surface brightness, H i mass, and MH i /LB. chemical evolution. Gil de Paz et al. (2007b) show that the met- LEDA contains MH i for almost all (>90%) of our targets, even if allicity of XUV-disks (M83, NGC 4625) is very low, although not some of these (single-dish) H i measurements could be underes- primordial.