Characterisation of the stellar content of SDSS EELGs through self-consistent spectral modelling

TL;DR

This study demonstrates that accurately characterizing the stellar content of extreme emission line galaxies requires self-consistent spectral modelling that accounts for nebular emission, revealing biases in traditional methods.

Contribution

It introduces the use of the spectral synthesis code FADO for EELGs, highlighting the importance of including nebular emission in spectral analysis for reliable stellar property estimates.

Findings

Stellar mass and age estimates are biased when nebular emission is ignored.

Including nebular emission yields more accurate stellar properties.

Differences between spectral synthesis codes relate to star-formation activity.

Abstract

Extreme emission line galaxies (EELGs) are a notable galaxy genus, ultimately being regarded as local prototypes of early galaxies at the cosmic noon. Robust characterisation of their stellar content, however, is hindered by the exceptionally high nebular emission present in their optical spectroscopic data. This study is dedicated into recovering the stellar properties of a sample of 414 EELGs as observed by the SDSS Survey. Such is achieved by means of the spectral synthesis code FADO, which self-consistently considers the stellar and nebular emission in an optical spectrum. Additionally, a comparative analysis was carried on, by further processing the EELGs sample with the purely stellar spectral synthesis code Starlight, and by extending the analysis to a sample of 697 normal star-forming galaxies, expected to be less affected by nebular contribution. We find that, for both galaxy samples, stellar mass and mean age estimates by Starlight are systematically biased towards higher values, and that an adequate determination of the physical and evolutionary properties of EELGs via spectral synthesis is only possible when nebular continuum emission is taken into account. Moreover, the differences between the two population synthesis codes can be ascribed to the degree of star-formation activity through the specific star-formation rate and the sum of the flux of the most prominent emission lines. As expected, on the basis of the theoretical framework, our results emphasise the importance of considering the nebular emission while performing spectral synthesis, even for galaxies hosting typical levels of star-formation activity.