We model the infrared to submillimeter spectral energy distribution of 11 nearby galaxies of the KINGFISH sample using Spitzer and Herschel data and compare model extrapolations at 870um (using different fitting techniques) with LABOCA 870um observat
ions. We investigate how the differences between predictions and observations vary with model assumptions or environment. At global scales, we find that modified blackbody models using realistic cold emissivity indices (beta_c=2 or 1.5) are able to reproduce the 870um observed emission within the uncertainties for most of the sample. Low values (beta_c<1.3) would be required in NGC0337, NGC1512 and NGC7793. At local scales, we observe a systematic 870um excess when using beta_=2.0. The beta_c=1.5 or the Draine and Li (2007) models can reconcile predictions with observations in part of the disks. Some of the remaining excesses occur towards the centres and can be partly or fully accounted for by non-dust contributions such as CO(3-2) or, to a lesser extent, free-free or synchrotron emission. In three non-barred galaxies, the remaining excesses rather occur in the disk outskirts. This could be a sign of a flattening of the submm slope (and decrease of the effective emissivity index) with radius in these objects.
Recent results have suggested that the well known mass-metallicity relation has a strong dependence on the star formation rate, to the extent that a three dimensional `fundamental metallicity relation exists which links the three parameters with mini
mal scatter. In this work, we use a sample of 4253 local galaxies observed in atomic hydrogen from the ALFALFA survey to demonstrate, for the first time, that a similar fundamental relation (the HI-FMR) also exists between stellar mass, gas-phase metallicity, and HI mass. This latter relation is likely more fundamental, driving the relation between metallicity, SFR and mass. At intermediate masses, the behaviour of the gas fundamental metallicity relation is very similar to that expressed via the star formation rate. However, we find that the dependence of metallicity on HI content persists to the highest stellar masses, in contrast to the `saturation of metallicity with SFR. It is interesting to note that the dispersion of the relation is very low at intermediate stellar masses (9< log(M*/Msun) <11), suggesting that in this range galaxies evolve smoothy, in an equilibrium between gas inflow, outflow and star formation. At high and low stellar masses, the scatter of the relation is significantly higher, suggesting that merging events and/or stochastic accretion and star formation may drive galaxies outside the relation. We also assemble a sample of galaxies observed in CO. However, due to a small sample size, strong selection bias, and the influence of a metallicity-dependent CO/H2 conversion factor, the data are insufficient to test any influence of molecular gas on metallicity.
