Modeling Dust and Starlight in Galaxies Observed by Spitzer and Herschel: The KINGFISH Sample
Dust and starlight are modeled for the KINGFISH project galaxies. With data from 3.6 micron to 500 micron, models are strongly constrained. For each pixel in each galaxy we estimate (1) dust surface density; (2) q_PAH, the dust mass fraction in PAHs; (3) distribution of starlight intensities heating the dust; (4) luminosity emitted by the dust; and (5) dust luminosity from regions with high starlight intensity. The models successfully reproduce both global and resolved spectral energy distributions. We provide well-resolved maps for the dust properties. As in previous studies, we find q_PAH to be an increasing function of metallicity, above a threshold Z/Z_sol approx 0.15. Dust masses are obtained by summing the dust mass over the map pixels; these "resolved" dust masses are consistent with the masses inferred from model fits to the global photometry. The global dust-to-gas ratios obtained from this study correlate with galaxy metallicities. Systems with Z/Z_sol > 0.5 have most of their refractory elements locked up in dust, whereas when Z/Z_sol < 0.3 most of these elements tend to remain in the gas phase. Within galaxies, we find that q_PAH is suppressed in regions with unusually warm dust with nu L_nu(70 um) > 0.4L_dust. With knowledge of one long-wavelength flux density ratio (e.g., f_{160}/f_{500}), the minimum starlight intensity heating the dust (U_min) can be estimated to within ~50%. For the adopted dust model, dust masses can be estimated to within ~0.07 dex accuracy using the 500 micron luminosity nu L_nu(500) alone. There are additional systematic errors arising from the choice of dust model, but these are hard to estimate. These calibrated prescriptions may be useful for studies of high-redshift galaxies.