Jesse K. Hill

Star Formation Triggering Mechanisms in Dwarf Galaxies: The Far-Ultraviolet, Hα, and H I Morphology of Holmberg II

Far-ultraviolet (FUV), Ha, and H I observations of dwarf galaxy Holmberg II are used to investigate the means by which star formation propagates in galaxies lacking global internal triggering mechanisms such as spiral density waves. The observations trace the interaction between sites of massive star forma- tion and the neutral and ionized components of the surrounding ISM in this intrinsically simple system. Both local and large-scale triggering mechanisms related to massive star formation are seen, suggesting that feedback from massive stars is a microscopic process operating in all galaxies to a certain degree. The data emphasize the importance of local conditions in regulating star formation from evidence such as massive stars inside ionized shells, compact H II regions surrounding aging clusters, and stars formed in chains of progressing age. Surface brightness pro—les show that current activity correlates with the time-averaged level of past star formation at a given radius demonstrating a reliance on local conditions. Large-scale triggering by H I shells is supported by observations of progenitor populations as well as secondary sites of star formation associated with their dense rims. Analysis of the energy available from massive stars inside H I shells indicates that energy deposited into the ISM from supernovae and stellar winds is sufficient to account for the H I morphology. Ages of individual star-forming regions are derived using B ,H a, and FUV photometry and show both older, diUuse FUV regions and younger, compact H II regions. The distribution of ages is reconciled with the H I morphology, showing a clear preference of young regions for areas of dense H I and old regions for H I voids. Global kinematical properties may also play a role in the star formation process since diUerences in the rotation characteristics of the neutral gas disk correlate with diUerences in triggering mechanisms. Large-scale feedback from massive stars is shown to operate in regions that lack diUerential shear in the gas disk. Subject headings: galaxies: dwarfgalaxies: ISMgalaxies: individual (DDO 50) ¨ stars: formationultraviolet: galaxies

THE ULTRAVIOLET IMAGING TELESCOPE: INSTRUMENT AND DATA CHARACTERISTICS

The Ultraviolet Imaging Telescope (\UIT) was flown as part of the \AstroMiss\ observatory on the Space Shuttle Columbia in December 1990 and again on the Space Shuttle Endeavor in March 1995. Ultraviolet (1200-3300A) images of a variety of astronomical objects, with a 40\arcmin\ field of view and a resolution of about 3\arcsec, were recorded on photographic film. The data recorded during the first flight is available to the astronomical community through the National Space Science Data Center (NSSDC); the data recorded during the second flight will soon be available as well. This paper discusses in detail the design, operation, data reduction, and calibration of \UIT, providing the user of the data with information for understanding and using the data. It also provides guidelines for analyzing other astronomical imagery made with image intensifiers and photographic film.

Ultraviolet Imaging Telescope Observations of the Magellanic Clouds

We present wide-field far-ultraviolet (FUV; 1300–1800 A) images of the Large and Small Magellanic Clouds (LMC, SMC). These data were obtained by the Ultraviolet Imaging Telescope (UIT) during the Astro-1 (1990 December 1–10) and Astro-2 (1995 March 2–18) missions; the images provide an extensive FUV mosaic of the SMC and contain numerous regions in the LMC, covering a wide range of stellar densities and current star formation activity. A total of 47 LMC/Lucke-Hodge and 37 SMC/Hodge OB associations are completely or partially included in the observed fields. FUV data can identify the hottest OB stars more easily than can optical photometry, and these stars dominate the ionizing flux, which is correlated to the observed Hα flux of the associated H II regions. Of the H II regions in the catalog of Davies, Elliott, & Meaburn (DEM), the UIT fields completely or partially include 102 DEM regions in the LMC and 74 DEM regions in the SMC. We present a catalog of FUV magnitudes derived from point-spread function photometry for 37,333 stars in the LMC (the UIT FUV magnitudes for 11,306 stars in the SMC were presented recently by Cornett et al.), with a completeness limit of mUV ≈ 15 mag and a detection limit of mUV ≈ 17.5. The average uncertainty in the photometry is ~0.1 mag. The full catalog with astrometric positions, photometry, and other information is also available from publicly accessible astronomical data archives. We divide the catalog into field stars and stars that are in DEM regions. We analyze each of these two sets of stars independently, comparing the composite UV luminosity function of our data with UV magnitudes derived from stellar evolution and atmosphere models in order to derive the underlying stellar formation parameters. We find a most probable initial mass function (IMF) slope for the LMC field stars of Γ = -1.80 ± 0.09. The statistical significance of this single slope for the LMC field stars is extremely high, though we also find some evidence for a field star IMF slope of Γ ~ -1.4, roughly equal to the Salpeter slope. However, in the case of the stars in the DEM regions (the stars in all the regions were analyzed together as a single group), we find three IMF slopes of roughly equal likelihood: Γ = -1.0, -1.6, and -2.0. No typical age for the field stars is found in our data for time periods up to a continuous star formation age of 500 Myr, which is the maximum age consistent with the completeness limit magnitude of the catalogs luminosity function. The best age for the collection of cluster stars was found to be t0 = 3.4 ± 1.9 Myr; this is consistent with the age expected for a collection of OB stars from many different clusters.

The Ultraviolet Imaging Telescope - Design and performance

The instrumental configuration, calibration, and operations during the first flight of the Ultraviolet Imaging Telescope on the Astro-1 mission, December 2-10, 1990, are described. The UV images of a wide variety of astronomical objects were recorded with a 40-arcmin diameter field of view. Images of targets as faint as magnitude 21 (UV) were secured with a resolution of about 3 arcsec. The optics, light baffling, and image motion compensation system are summarized, and detectors and electronic subsystems are described.

Ultraviolet imaging of old population in nearby galaxies

We analyze UV imagery of two Sb bulges and two E galaxies obtained with the Ultraviolet Imaging Telescope during the Astro-1 mission. The UV brightness of these systems is not produced by recent massive star formation. Instead it must originate in relatively low-luminosity objects, probably low-mass, post-giant-branch stars. We find extended, large-amplitude UV color gradients, which are probably related to abundance gradients within the galaxies. M32 has a color gradient opposite to the other three objects, possibly because of an intermediate-age population.

Ultraviolet Colors and Extinctions of H II Regions in the Whirlpool Galaxy (M51)

Far-UV (wavelength 1520 A), U, Hα, and R images of the interacting Sbc spiral galaxy M51 were obtained by the Ultraviolet Imaging Telescope (UIT) during the Astro-2 Spacelab mission of 1995 March and at Mount Laguna Observatory. The μ152 - μU radial gradient of over 1 mag, becoming bluer with increasing radius, is attributed primarily to a corresponding radial extinction gradient. Magnitudes in both UV bands and Hα fluxes are reported for 28 H II regions. Optical extinctions for the 28 corresponding UV sources are computed from the measured m152 - U colors by fitting to the optical extinctions of Nakai & Kuno. The estimated normalized far-UV extinction A152/E(B-V) increases with increasing Galactocentric distance (decreasing metallicity), from 5.99 to 6.54, compared with the Galactic value 8.33. The best-fit m152 - U color for no extinction, -3.07, is the color of a model solar metallicity starburst of age ~2.5 Myr with IMF slope -1.0. H II regions show decreasing observed Hα fluxes with decreasing radius, relative to the Hα fluxes predicted from the observed f152 for age 2.5 Myr, after the Hα and f152 are corrected for extinction. We attribute the increasing fraction of missing Hα flux with decreasing radius to increasing extinction in the Lyman continuum. The increasing extinction-corrected far-UV flux of the H II regions with decreasing distance to the nucleus is probably a result of the corresponding increasing column density of the interstellar gas resulting in larger mass OB associations. The estimated dust-absorbed Lyman continuum energy flux is ~0.6 times the far-infrared energy flux of M51 observed by IRAS.

Images in the rocket ultraviolet - The starburst in the nucleus of M83

Ultraviolet images of the SAB(s)c I-II galaxy M83 (NGC 5236) obtained with a rocketborne telescope in broad bandpasses centered at 1540 A and 2360 A show a bright resolved nuclear source which accounts for approximately 20 percent of the flux of the galaxy in both bandpasses. Low-resolution International Ultraviolet Explorer spectra of this source reveal an energy distribution similar to that of the starburst nucleus of NGC 7714. Strong blueshifted absorption lines can be interpreted as evidence for a nuclear wind powered by supernovae. Observations from UV, X-ray, optical, and far-infrared bandpasses are consistent with a starburst approximately one-sixth as strong as that in M82. A scaling of the M82 models of Rieke et al. (1980) predicts that the nucleus of M83 contains 10 to the 6th - 3 x 10 to the 7th solar masses in young stars and has a supernova rate of approximately 0.01 per yr.

Simulated optical images of galaxies at z about 1 using ultraviolet images of nearby galaxies

Ultraviolet sounding rocket images of several nearby galaxies are used to simulate the appearance in optical bandpasses of similar systems at redshifts z = 0.5-2.67, as observed by the WFPC on HST and by ground-based instruments. Since the morphology of galaxies is a strong function of the wavelength, the appearance of galaxies at large redshifts is subjected to a large k-correction effect. The strong dependence of monochromatic surface brightness on redshift also implies that observed morphology of distant systems will be crucially dependent on the limiting surface brightness set by the sky background. Although the angle subtended by sources depends only weakly on z, the fraction of a galaxy whose surface brightness is above the detection threshold varies strongly with z. Morphological distinctions of spiral from elliptical, spiral from irregular, and barred from unbarred types become more difficult, as does accurate evaluation of interactions or of the local environment. 32 refs.

Ultraviolet imaging telescope and optical emission-line observations of H II regions in M81

Images of the type Sab spiral galaxy M81 were obtained in far-UV and near-UV bands by the Ultraviolet Imaging Telescope (UIT) during the Astro-1 Spacelab mission of 1990 December. Magnitudes in the two UV bands are determined for 52 H II regions from the catalog of Petit, Sivan, & Karachentsev (1988). Fluxes of the H-alpha and H-beta emission lines are determined from CCD images. Extinctions for the brightest H II regions are determined from observed Balmer decrements. Fainter H II regions are assigned the average of published radio-H-alpha extinctions for several bright H II regions. The radiative transfer models of Witt, Thronson, & Capuano (1992) are shown to predict a relationship between Balmer Decrement and H-alpha extinction consistent with observed line and radio fluxes for the brightest 7 H II regions and are used to estimate the UV extinction. Ratios of Lyman continuum with ratios predicted by model spectra computed for initial mass function (IMF) slope equal to -1.0 and stellar masses ranging from 5 to 120 solar mass. Ages and masses are estimated by comparing the H-alpha and far-UV fluxes and their ratio with the models. The total of the estimated stellar masses for the 52 H II regions is 1.4 x 10(exp 5) solar mass. The star-formation rate inferred for M81 from the observed UV and H-alpha fluxes is low for a spiral galaxy at approximately 0.13 solar mass/yr, but consistent with the low star-formation rates obtained by Kennicutt (1983) and Caldwell et al. (1991) for early-type spirals.

Ultraviolet Imaging Telescope ultraviolet images - Large-scale structure, H II regions, and extinction in M81

The study employs UV images of M81 obtained by the Ultraviolet Imaging Telescope (UIT) during the December 1990 Astro-1 spacelab mission to determine 2490- and 1520-A fluxes from 46 H II regions and global surface brightness profiles. Comparison photometry in the V band is obtained from a ground-based CCD image. UV radial profiles show bulge and exponential disk components, with a local decrease in disk surface brightness inside the inner Lindblad Resonance about 4 arcmin from the nucleus. The V profile shows typical bulge plus exponential disk structure, with no local maximum in the disk. There is little change of UV color across the disk, although there is a strong gradient in the bulge. Observed m152-V colors of the H II regions are consistent with model spectra for young clusters, after dereddening using Av determined from m249-V and the Galactic extinction curve. The value of Av, so determined, is 0.4 mag greater on the average than Av derived from radio continuum and H-alpha fluxes.