Science

Evolution of galaxies

An important set of observables from the very early universe are the abundances of the primordial elements (e.g. He, D, Li). Key among these is helium. The effect to determine the primordial abundance of helium accurately would be aided by the existence of a larger sample of very metal poor galaxies. Ideal work for this would possess strong emission lines, to facilitate the abundance determination. Further, they would tend to have weak oxygen lines, due to their low metallicity. The focus of previous searches on the contents of the [OII]-selected surveys may then have created a huge bias against finding more low metallicity galaxies. Selecting by Hα as OTELO does, may well lead to the discovery of many previously unidentified low metallicty galaxies with which to investigate the primordial helium abundance.

OTELO survey will be able not only of separating Hα from [NII]658.4 nm, but to measure [NII]658.4 nm accurately enough down to Z=1/10 solar, for more than 1000 objects of a wide variety of morphological types. As it will be shown below, the Hα/[NII] ratio is a metallicity estimator that will allow studying the chemical evolution of galaxies, deriving accurate SFR and H₂/CO conversion factors and opening new windows for the study of the chemical evolution of the Universe.

Nitrogen as metallicity estimator

The [NII]658.4 nm is the most intense nitrogen line that can be observed at optical wavelengths. It is usually assumed to be too affected by excitation with respect to the [OII]+[OIII] to be used as metallicity indicator. However, it is more difficult to determine metallicity via the oxygen indicators in far away galaxies. For this reason, several authors ( Kewley et al. 2001; Denicoló et al. 2002; Melbourne and Salzer 2002) have suggested the use of [NII]658.4/Halpha as metallicity estimator, providing different empirical calibrations from 1/50th to twice the solar value ( Denicoló et al. 2002).

Chemical evolution of galaxies

Although some theoretical models exists ( Edmunds and Phillips 1997) the data available are still too scarce and biassed towards low extinction targets. They essentially consist in (i) Lyman Break Galaxies at high z ( Pettini et al 1994) (ii) 14 ELGs at 0.11 < z < 0.5 ( Kobulnicky and Zaritsky 1999) and (iii) UV selected galaxies at 0 < z < 0.4 ( Contini et al. 2002)

However, Contini et al. 2002 and Pettini et al. 2002 have detected more dispersion in [NII] with respect to oxygen. They claim that it might be due to a delay in the release of nytrogen, produced in intermediate stars as primary element, with respect to oxygen, produced in Type II supernovae. Nevertheless, this could be caused by the bias towards low extinction targets (of low dust content) of their samples, that might be undergoing their first metal production. This effect could be overcomed by OTELO survey, much less biased towards this type of objects. Also, it should not affect objects with continuous star formation such as spirals. To check the impact of this delayed nytrogen release in the metallicities, specially in the case of dwarfs, oxygen lines will be also observed in a subsample of OTELO targets selected within a range of morphological types and nytrogen content within each type. This will allow checking the nytrogen-metallicity calibration as well.

Finally, to study the chemical evolution of galaxies it is necessary to distinguish morphological types, since the evolution is very different for spiral-irregulars, dwarfs and early-types. For this reason, OTELO narrow band survey must be accompanied by a broad band survey of the same fields. This broad band auxiliary survey will allow not only identifying morphological types but providing an approximate redshift that will help to distinguish whether the emission line observed corresponds, for example, to Hα of a dwarf at z=0.4 or to [OIII] of a irregular at z=0.9.

Star Formation Rates and metallicity

Star Formation Rates are usually derived from Halpha or [OII] luminosities via a constant ratio ( Hunter and Gallagher 1986, Kennicutt 1998). It is generally assumed that SFR derived from [OII] can be heavily affected from different metal content, and that this line is not as good SFR indicator as Hα. However, even using Hα, the constant depends on different factors, including metallicity. From population synthesis models using a Salpeter IMF, Weilbacher and Fritze-v.Alvensleben (2001) found that SFR derived from Hα can suffer from large errors due to different metallicity content, including the IMF used. For example SFR of BCD, that have metallicities as low as ~1/10 Zsun, or of low metallicity systems, can be overestimated by more than a factor 3. OTELO will provide a first order correction of this effect.

Metallicity and CO conversion factor

The most used mm emission lines in extragalactic astronomy are those of CO in the different transitions between rotational states, such as the fundamental J=1—0 transition at 115 Ghz (2.6 mm). The observation of these lines is then heavily dependent on the metal content of the galaxy observed. This might be a problem when observing dwarf metal poor galaxies at any z or less evolved galaxies, especially since the molecular hydrogen content is evaluated via CO lines intensity using conversion factors that depend on the metal content. For example, for a galaxy of [12+log(O/H)]=7.6, the molecular hydrogen content derived from CO observations would be underestimated by almost an order of magnitude. It follows that the study of the metal content of the different Hubble types vs. redshift is a useful preparation for ALMA observations.

The faint end of the M-Z relation

The mass-metallicity (M-Z) relation has been widely studied for galaxies at different redshifts and in different environments (e.g., Lara-Lopez et al. 2009, Hughes et al. 2013). However, there are just a few studies targeting very low metallicity galaxies (e.g., Lee et al. 2006). The OTELO survey will allow us to detect and measure galaxies with very low metallicities. We will be able to prove the faint end of the M-Z relation for a statistically significant sample of galaxies at different redshifts. Furthermore, we will be able to confront observations with current models of metallicity evolution for galaxies with low abundances. Thanks to the tunable filters of OSIRIS, we will be able to deblend Hα from [NII]6563, allowing us to measure metallicities, and also SFRs using the Ha line.

Evolution of galaxies

The OTELO sample will be useful as low and intermediate redshift comparison group to the galaxy population at higher redshift. We will be able to measure more accurately than ever before the space densities of starburst galaxies locally, which may well have an impact on resolving the faint blue galaxy problem. In addition, the survey will also provide a more accurate estimate of the local star formation rate (SFR) density which, when combined with high-redshift studies, will lead to a more precise understanding of the evolution of the SFR as a function of cosmic epoch.

See also: Related Publications