Evolution of N/O abundance ratios and ionization parameters from z ∼ 0 to 2 investigated by the direct temperature method*†
Abstract
We present N/O abundance ratios, ionization parameters qion, and oxygen abundances O/H for a total of 41 galaxies (11 individual galaxies and a 30-galaxy stack) including Lyα emitters and Lyman break galaxies at z ∼ 2, and investigate galaxy evolution from z ∼ 0 to 2 in conjunction with 208529 local galaxies from the Sloan Digital Sky Survey (SDSS) and nine green pea galaxies (GPs). In contrast with most of the recent studies, we obtain the N/O ratio, qion, and O/H measurements by direct Te methods using [O III] λ4363 and O III] λ1665 lines. Based on these reliable measurements, we find that there are z ∼ 2 galaxies with excesses of N/O falling beyond the N/O-O/H relation of the local galaxies while the majority of the z ∼ 2 galaxies have N/O ratios that are nearly comparable with those of z ∼ 0 galaxies in the N/O-stellar mass relation. Our galaxies place the upper limit of the N/O ratio log(N/O) ≤ -1.26 on average, suggesting that the N/O ratio evolves, if at all, by <0.17 dex. Obtaining the reliable measurements free from the N/O-qion-O/H degeneracies, we demonstrate, for the first time, that z ∼ 2 galaxies with offsets in the Baldwin-Phillips-Terlevich (BPT) diagram show (1) only an N/O excess, (2) only a qion excess, or (3) both N/O and qion excesses. We argue that the BPT offsets at z ∼ 2 are not made by one of the (1)-(3) galaxy populations alone, but a composite of (1)-(3) populations. We confirm that these (1)-(3) populations also exist at z ∼ 0, such as GPs and SDSS low-mass and high-SFR galaxies (LMHSs).
- Publication:
-
Publications of the Astronomical Society of Japan
- Pub Date:
- June 2017
- DOI:
- arXiv:
- arXiv:1605.03436
- Bibcode:
- 2017PASJ...69...44K
- Keywords:
-
- galaxies: abundances;
- galaxies: evolution;
- galaxies: high-redshift;
- galaxies: ISM;
- Astrophysics - Astrophysics of Galaxies
- E-Print:
- 22 pages, 13 figures, 6 tables, the data and discussions in the v2/v3 are updated in response to the latest reports from other papers. Accepted for publication in PASJ. In press