Unit Star Formation Rate Research Articles

Mid-IR emission of galaxies in the Virgo cluster and in the Coma supercluster

We study the relationship between the mid-IR (5–18 μm) emission of late-type galaxies and various other star formation tracers in order to investigate the nature of the dust heating sources in this spectral domain. The analysis is carried out using a sample of 123 normal, late-type, nearby galaxies with available data at several frequencies. The mid-IR luminosity (normalized to the H-band luminosity) correlates better with the far-IR luminosity than with more direct tracers of the young stellar population such as the Hα and the UV luminosity. The comparison of resolved images reveals a remarkable similarity in the Hα and mid-IR morphologies, with prominent HII regions at both frequencies. The mid-IR images, however, show in addition a diffuse emission not associated with HII regions nor with the diffuse Hα emission. This evidence indicates that the stellar population responsible for the heating of dust emitting in the mid-IR is similar to that heating big grains emitting in the far-IR, including relatively evolved stars responsible for the non-ionizing radiation. The scatter in the mid-IR vs. Hα, UV and far-IR luminosity relation is mostly due to metallicity effects, with metal-poor objects having a lower mid-IR emission per unit star formation rate than metal-rich galaxies. Our analysis indicates that the mid-IR luminosity is not an optimal star formation tracer in normal, late-type galaxies.

Nebular emission from star-forming galaxies

We present a new model for computing consistently the line and continuum emission from galaxies, based on a combination of recent population synthesis and photoionization codes. We use effective parameters to describe the H ii regions and the diffuse gas ionized by single stellar generations in a galaxy, among which the most important ones are the zero-age effective ionization parameter, the effective gas metallicity and the effective dust-to-heavy element ratio. We calibrate the nebular properties of our model using the observed [O iii]/Hβ, [O ii]/[O iii], [S ii]/Hα and [N ii]/[S ii] ratios of a representative sample of nearby spiral and irregular, starburst and H ii galaxies. To compute whole (line plus continuum) spectral energy distributions, we include the absorption by dust in the neutral interstellar medium (ISM) using a recent simple prescription, which is consistent with observations of nearby starburst galaxies. Our model enables us to interpret quantitatively the observed optical spectra of galaxies in terms of stars, gas and dust parameters. We find that the range of ionized-gas properties spanned by nearby galaxies implies factors of 3.5 and 14 variations in the Hα and [O ii] luminosities produced per unit star formation rate (SFR). When accounting for stellar Hα absorption and absorption by dust in the neutral ISM, the actual uncertainties in SFR estimates based on the emergent Hα and [O ii] luminosities are as high as several decades. We derive new estimators of the SFR, the gas-phase oxygen abundance and the effective absorption optical depth of the dust in galaxies. We show that, with the help of other lines such as [O ii], Hβ, [O iii], [N ii] or [S ii], the uncertainties in SFR estimates based on Hα can be reduced to a factor of only 2–3, even if the Hα line is blended with the adjacent [N ii] lines. Without Hα, however, the SFR is difficult to estimate from the [O ii], Hβ and [O iii] lines. The reason for this is that the absorption by dust in the neutral ISM and the ionized-gas parameters are then difficult to constrain independently. This suggests that, while insufficient by itself, the Hα line is essential for estimating the star formation rate from the optical emission of a galaxy.