Tidal dwarf galaxies in gas-rich groups

Abstract

I develop new methods for identifying and measuring tidal dwarf galaxies, using a sample of galaxies within Hi-rich groups that have no evidence of advanced major mergers. These groups are taken from the Survey of Ionization in Neutral Gas Galaxies (SINGG, Meurer et al., 2006), an optical follow-up survey to the HI Parkes All Sky Survey (HIPASS, Barnes et al., 2001). Fifteen of the fields contain four or more emission line galaxies and are named Choir groups. I detect new dwarf galaxies that are too small to be individually detectable in HIPASS; they are detectable in the SINGG narrow-band imaging because of their star formation and membership of these HI-rich groups. The Choir groups are compact, with a mean projected separation between the two brightest members of 190 kpc. They have comparable star formation efficiency (the ratio of star formation rate to Hi mass) to the remaining SINGG fields. The Choir member galaxies also match the wider SINGG sample in their radii, Ha equivalent width and surface brightness. I define a new, more robust calibration for the metallicity diagnostic for identifying tidal dwarf galaxy candidates in the absence of tidal tails, based on the luminosity-metallicity relation with a consistent metallicity definition. Using that calibration, SDSS dwarfs fainter than M_R = -16 have a mean metallicity of 12 + log(O/H) = 8.28 p 0.10, regardless of their luminosity. Tidal dwarf galaxy candidates in the literature are elevated above this at 12 + log(O/H) = 8.70 p 0.05 on average. Our hydrodynamical simulations also predict that tidal dwarf galaxies should have metallicities elevated above the normal luminosity-metallicity relation. I compare 53 star-forming galaxies in 9 of the HI gas-rich Choir groups and find those brighter than M_R ~ -16 to be consistent with the normal relation defined by the SDSS sample. At fainter magnitudes my sample has a wide range in metallicity, suggestive of varying Hi content and environment. Three (16%) of the dwarfs are strong tidal dwarf galaxy candidates (12 + log(O/H) g 8.6), and four (21%) are very metal poor dwarfs (12 + log(O/H) l 8.0); these are probably gas-rich dwarfs with recently ignited star formation. I fit model mass-follows-light rotation curves to optical slit spectroscopic observations of 22 dwarf galaxies in the sample. Due to observational limitations, ten of these are of sufficient quality to measure mass-to-light (M/L) ratios. These are low (M/L = 0.73 p 0.39 M⊙/L⊙), consistent with the star-forming nature of these galaxies, though in most cases I do not measure out to radii where normal galaxies are dark-matter-dominated. I find a suggestion of a trend towards higher M/L ratio with increasing luminosity and metallicity, albeit with large scatter. I find a relatively large pressure support component (s_D = 13.1p1.9 km/s), indicating that the galaxies are experiencing tidal fields. One galaxy has a strongly-falling rotation curve, which could be explained by 1) a rotating disk that becomes pressure-supported at large radii; 2) a tilted bar surrounded by a face-on disk; or, 3) a kinematic twist. I consider whether or not falling rotation curves can be measured, based on a tidal stripping model. The tidal radius r_tidal must be larger than twice the turnover radius r_turn where the rotation curve falls significantly, otherwise the baryons there do not remain bound and the measurement cannot be made. As much as half of this sample is affected by tidal stripping. Further to this, the Ha light at this radius must be sufficiently bright to be detectable; this is only the case for three (14%) galaxies in my sample. It seems that the falling rotation curves predicted for tidal dwarf galaxies are rarely, if ever, observable.