The role of gas in galaxy evolution : infall, star formation, and internal structure

The story of a typical spiral galaxy like the Milky Way is a tale of the transformation of metal-poor hydrogen gas to heavier elements through nuclear burning in stars. This gas is thought to arrive in early times during the assembly phase of a galaxy and at late times through a combination of hot and cold “flows” representing external evolutionary processes that continue to the present. Through a somewhat still unclear mechanism, the atomic hydrogen is converted to molecules that collect into clouds, cool, condense, and form stars. At the end of these stars’ lives, much of their constituent gas is returned to the galaxy to participate in subsequent generations of star formation. In earlier times in the history of the universe, frequent and large galaxy mergers brought additional gas to further fuel this process. However, major merger activity began an ongoing decline several Gyr ago and star formation is now diminishing; the universe is in transitioning to an era in which the structural evolution of disk galaxies is dominated by slow, internal (“secular”) processes. In this evolutionary regime, stars and the gas from which they are formed participate in resonant gravitational interactions within disks to build ephemeral structures such as bars, rings, and small scale-height central bulges. This regime is expected to last far into the future in a galaxy like the Milky Way, punctuated by the periodic accretion of dwarf satellite galaxies but lacking in the “major” mergers that kinematically scramble disks into ellipticals. This thesis examines details of the story of gas from infall to structure-building in three major parts. The High- and Intermediate-Velocity Clouds (HVCs/IVCs) are clouds of H i gas at velocities incompatible with simple models of differential Galactic rotation. Proposed ideas explaining their observed properties and origins include (1) the infall of low-metallicity material from the Halo, possibly as cold flows along filaments of a putative “Cosmic Web”; (2) gas removed from dwarf satellite galaxies orbiting the Milky Way via some combination of ram pressure stripping and tidal disruption; and (3) the supply and return feeds of a “Galactic Fountain” cycling gas between the Disk and Halo. Numerical values of their observed properties depend strongly on the Clouds’ distances. In Chapter 2, we summarize results of an ongoing effort to obtain meaningful distances to a selection of HVCs and IVCs using the absorption-line bracketing method. We find the Clouds are not at cosmological distances, and with the exception of the Magellanic Stream, they are generally situated within a few kiloparsecs of the Disk. The strongest discriminator of the above origin scenarios are the heavy element abundances of the Clouds, but to date few reliable Cloud metal- licities have been published. We used archival UV spectroscopy, supplemented by new observations with the Cosmic Origins Spectrograph aboard the Hubble Space Telescope and H I 21 cm emission spectroscopy from a variety of sources to compute elemental abundances relative to hydrogen for 39 HVC/IVC components along 15 lines of sight. Many of these are previously unpublished. We find support for all three origin scenarios enumerated above while more than doubling the number of robust measurements of HVCs/IVCs in existence. The results of this work are detailed in Chapter 3. In Chapter 4 we present the results of a spectroscopic study of the high-mass protostellar object NGC 7538 IRS 9 made with the Texas Echelon Cross Echelle Spectrograph (TEXES), a sensitive, high spectral resolution, mid-infrared grating spectrometer and compare our observations to published data on the nearby object NGC 7538 IRS 1. Forty-six individual lines in vibrational modes of the molecules C₂H₂, CH₄, HCN, NH₃ and CO were detected, including two isotopologues (¹³CO, ¹²C¹⁸O) and one combination mode ([nu]₄+[nu]₅ C₂H₂). Fitting synthetic spectra to the data yielded the Doppler shift, excitation temperature, Doppler b parameter, column density and covering factor for each molecule observed; we also computed column density upper limits for lines and species not detected, such as HNCO and OCS. We find differences among spectra of the two objects likely attributable to their differing radiation and thermal environments. Temperatures and column densities for the two objects are generally consistent, while the larger line widths toward IRS 9 result in less saturated lines than those toward IRS 1. Finally, we compute an upper limit on the size of the continuum-emitting region (~2000 AU) and use this constraint and our spectroscopy results to construct a schematic model of IRS 9. In Chapters 5 and 6, we describe studies of the bright, nearby, edge-on spiral galaxies NGC 4565 and NGC 5746, both previously classified as type Sb spirals with measured bulge-to-total luminosity ratios B/T ≃ 0.4. These ratios indicate merger-built, “classical” bulges but in reality represent the photometric signatures of bars seen end-on. We performed 1-D photometric decompositions of archival Hubble Space Telescope, Spitzer Space Telescope, and Sloan Digital Sky Survey images spanning a range of wavelengths from the optical to near-infrared that penetrate the thick midplane dust in each galaxy. In both, we find high surface brightness, central stellar components that are clearly distinct from the boxy bar and from the disk; we interpret these structures as small scale height “pseudobulges” built from disk material via internal, resonant gravitational interactions among disk material − not classical bulges. The brightness profiles of the innermost component of each galaxy is well fitted by a Sersic function with major/minor axis Sersic indices of n = 1.55±0.07 and 1.33±0.12 for NGC 4565 and n = 0.99±0.08 and 1.17 ± 0.24 for NGC 5746. The true “bulge-to-total” ratios of these galaxies are considerably smaller than once believed: 0.061+0.009 and 0.136 ± 0.019, −0.008, respectively. Therefore, more galaxies than we thought contain little or no evidence of a merger-built classical bulge. We argue further that a classical bulge cannot hide behind the dust lane of either galaxy and that other structures built exclusively through secular evolution processes such as inner rings, both revealed through the infrared imagery, argue strongly against any merger violence in the recent past history of these objects. From a formation point of view, NGC 4565 and NGC 5746 are giant, pure-disk galaxies, and we do not understand how such galaxies form in a ΛCDM universe. This presents a challenge to our picture of galaxy formation by hierarchical clustering because it is difficult to grow galaxies as large as these without making big, classical bulges. We summarize the work presented in this thesis in Chapter 7 and conclude with speculations about the future direction of research in this field.