Fade to grey: Systematic variation of galaxy attenuation curves with galaxy properties in the EAGLE simulations

James W. Trayford*, Claudia del P. Lagos, Aaron S.G. Robotham, Danail Obreschkow

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Downloads (Pure)

Abstract

We present a simple model for galaxy attenuation by fitting SKIRT radiative transfer calculations for ∼10 000 EAGLE galaxies at redshifts z = 2 − 0. Our model adapts the two-component screen model of Charlot & Fall, parametrizing the optical depth and slope of the interstellar medium screen using the average dust surface density, ∑dust. We recover relatively tight relations between these parameters for the EAGLE sample, but also provide the scatter in these parameters owing to the morphological variation and orientation of galaxies. We also find that these relations are nearly independent of redshift in the EAGLE model. By pairing our model with an empirical prescription for birth clouds below the resolution scale of the simulation, we reproduce the observed relation between attenuation slope and optical depth for the first time in a cosmological simulation. We demonstrate that this result is remarkably independent of the attenuation properties assumed for the birth cloud screen, merely requiring a boosted attenuation for infant stars. We present this model with a view to interpreting observations, as well as processing semi-analytic models and other hydrodynamic simulations.

Original languageEnglish
Pages (from-to)3937-3951
Number of pages15
JournalMonthly Notices of the Royal Astronomical Society
Volume491
Issue number3
Early online date19 Nov 2019
DOIs
Publication statusPublished - 1 Jan 2020

Keywords

  • Dust
  • Extinction
  • Galaxies: Formation
  • Galaxies: ISM
  • Galaxies: Structure
  • UKRI
  • STFC

Fingerprint

Dive into the research topics of 'Fade to grey: Systematic variation of galaxy attenuation curves with galaxy properties in the EAGLE simulations'. Together they form a unique fingerprint.

Cite this