Interpreting the Spitzer/IRAC Colours of 7Prior to the launch of JWST, Spitzer/IRAC photometry offers the only means of
studying the rest-frame optical properties of z>7 galaxies. Many such high
redshift galaxies display a red [3.6] – [4.5] micron colour, often referred to
as the “IRAC excess”, which has conventionally been interpreted as arising from
intense [OIII]+Hbeta emission within the [4.5] micron bandpass. An appealing
aspect of this interpretation is similarly intense line emission seen in
star-forming galaxies at lower redshift as well as the redshift-dependent
behaviour of the IRAC colours beyond z~7 modelled as the various nebular lines
move through the two bandpasses. In this paper we demonstrate that, given the
photometric uncertainties, established stellar populations with Balmer (4000 A,
rest-frame) breaks, such as those inferred at z>9 where line emission does not
contaminate the IRAC bands, can equally well explain the redshift-dependent
behaviour of the IRAC colours in 7<z<9 galaxies. We discuss possible ways of
distinguishing between the two hypotheses using ALMA measures of [OIII] 88
micron and dust continuum fluxes. Prior to further studies with JWST, we show
that the distinction is important in determining the assembly history of
galaxies in the first 500 Myr.

Prior to the launch of JWST, Spitzer/IRAC photometry offers the only means of
studying the rest-frame optical properties of z>7 galaxies. Many such high
redshift galaxies display a red [3.6] – [4.5] micron colour, often referred to
as the “IRAC excess”, which has conventionally been interpreted as arising from
intense [OIII]+Hbeta emission within the [4.5] micron bandpass. An appealing
aspect of this interpretation is similarly intense line emission seen in
star-forming galaxies at lower redshift as well as the redshift-dependent
behaviour of the IRAC colours beyond z~7 modelled as the various nebular lines
move through the two bandpasses. In this paper we demonstrate that, given the
photometric uncertainties, established stellar populations with Balmer (4000 A,
rest-frame) breaks, such as those inferred at z>9 where line emission does not
contaminate the IRAC bands, can equally well explain the redshift-dependent
behaviour of the IRAC colours in 7<z<9 galaxies. We discuss possible ways of
distinguishing between the two hypotheses using ALMA measures of [OIII] 88
micron and dust continuum fluxes. Prior to further studies with JWST, we show
that the distinction is important in determining the assembly history of
galaxies in the first 500 Myr.

http://arxiv.org/icons/sfx.gif