No Arabic abstract
The HI in disk galaxies frequently extends beyond the optical image, and can trace the dark matter there. I briefly highlight the history of high spatial resolution HI imaging, the contribution it made to the dark matter problem, and the current tension between several dynamical methods to break the disk-halo degeneracy. I then turn to the flaring problem, which could in principle probe the shape of the dark halo. Instead, however, a lot of attention is now devoted to understanding the role of gas accretion via galactic fountains. The current $rm Lambda$ cold dark matter theory has problems on galactic scales, such as the core-cusp problem, which can be addressed with HI observations of dwarf galaxies. For a similar range in rotation velocities, galaxies of type Sd have thin disks, while those of type Im are much thicker. After a few comments on modified Newtonian dynamics and on irregular galaxies, I close with statistics on the HI extent of galaxies.
The low column density gas at the outskirts of galaxies as traced by the 21 cm hydrogen line emission (HI) represents the interface between galaxies and the intergalactic medium, i.e., where galaxies are believed to get their supply of gas to fuel future episodes of star formation. Photoionization models predict a break in the radial profiles of HI at a column density of 5x10E+19 cm^-2 due to the lack of self-shielding against extragalactic ionizing photons. To investigate the prevalence of such breaks in galactic disks and to characterize what determines the potential edge of the HI disks, we study the azimuthally-averaged HI column density profiles of 17 nearby galaxies from The HI Nearby Galaxy Survey (THINGS) and supplemented in two cases with published Hydrogen Accretion in LOcal GAlaxieS (HALOGAS) data. To detect potential faint HI emission that would otherwise be undetected using conventional moment map analysis, we line up individual profiles to the same reference velocity and average them azimuthally to derive stacked radial profiles. To do so, we use model velocity fields created from a simple extrapolation of the rotation curves to align the profiles in velocity at radii beyond the extent probed with the sensitivity of traditional integrated HI maps. With this method, we improve our sensitivity to outer-disk HI emission by up to an order of magnitude. Except for a few disturbed galaxies, none show evidence for a sudden change in the slope of the HI radial profiles, the alleged signature of ionization by the extragalactic background.
In this short write-up, I will concentrate on a few topics of interest. In the 1970s I found very extended HI disks in galaxies such as NGC 5055 and NGC 2841, out to 2 - 2.5 times the Holmberg radius. Since these galaxies are warped, a tilted ring model allows rotation curves to be derived, and evidence for dark matter to be found. The evaluation of the amount of dark matter is hampered by a disk-halo degeneracy, which can possibly be broken by observations of velocity dispersions in both the MgI region and the CaII region.
We investigate the metallicity dependence of HI surface densities in star-forming regions along many lines of sight within 70 nearby galaxies, probing kpc to 50 pc scales. We employ HI, SFR, stellar mass, and metallicity (gradient) measurements from the literature, spanning a wide range (5 dex) in stellar and gas mass and (1.6 dex) in metallicity. We consider metallicities as observed, or rescaled to match the mass-metallicity relation determined for SDSS galaxies. At intermediate to high metallicities (0.3-2 times solar), we find that the HI surface densities saturate at sufficiently large total gas surface density. The maximal HI columns vary approximately inversely with metallicity, and show little variation with spatial resolution, galactocentric radius, or among galaxies. In the central parts of massive spiral galaxies the HI gas is depressed by factors of 2. The observed behavior is naturally reproduced by metallicity dependent shielding theories for the HI-to-H2 transitions in star-forming galaxies. We show that the inverse scaling of the maximal HI columns with metallicity suggests that the area filling fraction of atomic-molecular complexes in galaxies is of order unity, and weakly dependent on metallicity.
The HI and CO components of the interstellar medium (ISM) are usually used to derive the dynamical mass M_dyn of nearby galaxies. Both components become too faint to be used as a tracer in observations of high-redshift galaxies. In those cases, the 158 $mu$m line of atomic carbon [CII] may be the only way to derive M_dyn. As the distribution and kinematics of the ISM tracer affects the determination of M_dyn, it is important to quantify the relative distributions of HI, CO and [CII]. HI and CO are well-characterised observationally, however, for [CII] only very few measurements exist. Here we compare observations of CO, HI, and [CII] emission of a sample of nearby galaxies, drawn from the HERACLES, THINGS and KINGFISH surveys. We find that within R_25, the average [CII] exponential radial profile is slightly shallower than that of the CO, but much steeper than the HI distribution. This is also reflected in the integrated spectrum (global profile), where the [CII] spectrum looks more like that of the CO than that of the HI. For one galaxy, a spectrally resolved comparison of integrated spectra was possible; other comparisons were limited by the intrinsic line-widths of the galaxies and the coarse velocity resolution of the [CII] data. Using high-spectral-resolution SOFIA [CII] data of a number of star forming regions in two nearby galaxies, we find that their [CII] linewidths agree better with those of the CO than the HI. As the radial extent of a given ISM tracer is a key input in deriving M_dyn from spatially unresolved data, we conclude that the relevant length-scale to use in determining M_dyn based on [CII] data, is that of the well-characterised CO distribution. This length scale is similar to that of the optical disk.
The outskirts of galaxies offer extreme environments where we can test our understanding of the formation, evolution, and destruction of molecules and their relationship with star formation and galaxy evolution. We review the basic equations that are used in normal environments to estimate physical parameters like the molecular gas mass from CO line emission and dust continuum emission. Then we discuss how those estimates may be affected when applied to the outskirts, where the average gas density, metallicity, stellar radiation field, and temperature may be lower. We focus on observations of molecular gas in the outskirts of the Milky Way, extragalactic disk galaxies, early-type galaxies, groups, and clusters. The scientific results show the versatility of molecular gas, as it has been used to trace Milky Way spiral arms out to a galactocentric radius of 15 kpc, to study star formation in extended ultraviolet disk galaxies, to probe galaxy interactions in polar ring S0 galaxies, and to investigate ram pressure stripping in clusters. We highlight the physical stimuli that accelerate the formation of molecular gas, including internal processes such as spiral arm compression and external processes such as interactions.