Abstract
We present chemical evolution models aimed at reproducing the observed (N/O) versus (O/H)
abundance pattern of star-forming galaxies in the Local Universe. We derive gas-phase abundances
from Sloan Digital Sky Survey (SDSS) spectroscopy and a complementary sample of
low-metallicity dwarf galaxies, making use of a consistent set of abundance calibrations. This
collection of data clearly confirms the existence of a plateau in the (N/O) ratio at very low
metallicity, followed by an increase of this ratio up to high values as the metallicity increases.
This trend can be interpreted as due to two main sources of nitrogen in galaxies: (i) massive
stars, which produce small amounts of pure primary nitrogen and are responsible for the (N/O)
ratio in the low-metallicity plateau; (ii) low- and intermediate-mass stars, which produce both
secondary and primary nitrogen and enrich the interstellar medium with a time delay relative
to massive stars, and cause the increase of the (N/O) ratio. We find that the length of the
low-metallicity plateau is almost solely determined by the star formation efficiency, which
regulates the rate of oxygen production by massive stars. We show that, to reproduce the high
observed (N/O) ratios at high (O/H), as well as the right slope of the (N/O) versus (O/H) curve,
a differential galactic wind – where oxygen is assumed to be lost more easily than nitrogen
– is necessary. No existing set of stellar yields can reproduce the observed trend without
assuming differential galactic winds. Finally, considering the current best set of stellar yields,
a bottom-heavy initial mass function is favoured to reproduce the data.
abundance pattern of star-forming galaxies in the Local Universe. We derive gas-phase abundances
from Sloan Digital Sky Survey (SDSS) spectroscopy and a complementary sample of
low-metallicity dwarf galaxies, making use of a consistent set of abundance calibrations. This
collection of data clearly confirms the existence of a plateau in the (N/O) ratio at very low
metallicity, followed by an increase of this ratio up to high values as the metallicity increases.
This trend can be interpreted as due to two main sources of nitrogen in galaxies: (i) massive
stars, which produce small amounts of pure primary nitrogen and are responsible for the (N/O)
ratio in the low-metallicity plateau; (ii) low- and intermediate-mass stars, which produce both
secondary and primary nitrogen and enrich the interstellar medium with a time delay relative
to massive stars, and cause the increase of the (N/O) ratio. We find that the length of the
low-metallicity plateau is almost solely determined by the star formation efficiency, which
regulates the rate of oxygen production by massive stars. We show that, to reproduce the high
observed (N/O) ratios at high (O/H), as well as the right slope of the (N/O) versus (O/H) curve,
a differential galactic wind – where oxygen is assumed to be lost more easily than nitrogen
– is necessary. No existing set of stellar yields can reproduce the observed trend without
assuming differential galactic winds. Finally, considering the current best set of stellar yields,
a bottom-heavy initial mass function is favoured to reproduce the data.
Original language | English |
---|---|
Article number | stw532 |
Pages (from-to) | 3466-3477 |
Number of pages | 12 |
Journal | Monthly Notices of the Royal Astronomical Society |
Volume | 458 |
Issue number | 4 |
Early online date | 10 Mar 2016 |
DOIs | |
Publication status | Published - 1 Jun 2016 |
Keywords
- stars: abundances
- ISM: abundances
- ISM: evolution
- galaxies: abundances
- galaxies: evolution
- galaxies: ISM