# Deciphering an evolutionary sequence of merger stages in infrared-luminous starburst galaxies at z ~ 0.7

## TL;DR

This study uses multi-wavelength observations of starburst galaxies at z ~ 0.7 to propose an evolutionary sequence of merger stages, linking obscuration, galaxy size, shock activity, and AGN emergence.

## Contribution

It identifies correlations among attenuation, galaxy size, line widths, and shock indicators, establishing an evolutionary framework for merger stages in infrared-luminous starbursts.

## Key findings

- Total attenuation anti-correlates with radio size.
- More obscured starbursts show enhanced shock signatures.
- Emerging AGNs are more common in compact, late-stage mergers.

## Abstract

Based on optical/near-IR Magellan FIRE spectra of 25 starburst galaxies at 0.5 < z < 0.9, Calabr\`o et al.(2018) showed that their attenuation properties can be explained by a single-parameter sequence of total obscurations ranging from A(V)=2 to A(V)=30 towards the starburst core centers in a mixed stars and dust configuration. We investigate here the origin of this sequence for the same sample. We show that total attenuations anti-correlate with the starburst sizes in radio (3 GHz) with a significance larger than 5sigma and a scatter of 0.26 dex. More obscured and compact starbursts also show enhanced N2 (=[NII]/Halpha) ratios and larger line velocity widths that we attribute to an increasing shock contribution toward later merger phases, driven by deeper gravitational potential wells at the coalescence. Additionally, the attenuation is also linked to the equivalent width (EW) of hydrogen recombination lines, which is sensitive to the luminosity weighted age of the relatively unobscured stellar populations. Overall, the correlations among A(V), radio size, line width, N2 and EW of Balmer/Paschen lines converge towards suggesting an evolutionary sequence of merger stages: all of these quantities are likely to be good time-tracers of the merger phenomenon, and their large spanned range appears to be characteristic of the different merger phases. Half of our sample at higher obscurations have radio sizes approximately 3 times smaller than early type galaxies at the same redshift, suggesting that, in analogy with local Ultraluminous Infrared galaxies (ULIRGs), these cores cannot be directly forming elliptical galaxies. Finally, we detect mid-IR AGN torus for half of our sample and additional X-ray emission for 6 starbursts; intriguingly, the latter have systematically more compact sizes, suggestive of emerging AGNs towards later merger stages, possibly precursors of a later QSO phase.