Escaping photons affect galaxy formation but supernovae dominant
6 Aug 2015
The new generation of low-frequency radio telescopes, including MWA (Murchison Widefield Array), LOFAR (LOw Frequency Array), PAPER (Precision Array for Probing the Epoch of Reionisation) and SKA (Square Kilometre array), will enable us to observe the evolution of neutral hydrogen during the reionisation of the Universe. The resulting measurements of the timing and structure of reionisation promise to probe the properties of the first galaxies.
In recent years though – while gearing up to using these new instruments – a great deal of attention has focused on theoretical modelling the effect of galaxies on the reionisation of the intergalactic medium (IGM). Large modern simulations commonly begin with an N-body simulation to generate a distribution of dark matter haloes, followed by relating dark matter halo mass to ionising luminosity. Radiative transfer methods (usually ray-tracing algorithms) then model the generation of ionised structure on large scales which is often run with lower resolution than the N-body code for computational efficiency.
One of main limitations in modelling of reionisation is the physics of the ionising sources. In their 2013 publication, however, CAASTRO Affiliate Dr Hansik Kim and his colleagues at the Universities of Melbourne and Durham overcame this limitation. They combined the semi-analytic galaxy formation model, as implemented in the Millennium-II dark matter simulation, with a semi-numerical scheme to describe the resulting ionisation structure. The researchers further had to account for the most important unknowns for the reionisation history: the fraction of ionising photons that escape from their host galaxies.
Observational estimates show a broad range of escape fraction values from a few per cent in the local Universe to possibly a few tens per cent at redshift z=1~3. However, there are no observational constraints on the escape fraction during the Epoch of Reionisation.
Dr Kim and his team decided to incorporate a variable escape fraction for ionising photons and were able to predict the redshifted 21-cm power spectrum for the resulting reionisation histories. They not only modelled dependencies of the escape fraction with halo mass and redshift, according to the default semi-analytic galaxy formation model that includes supernova feedback, but also the halo mass dependency of escape fraction without supernova feedback. The importance of SN feedback during reionisation is not constrained by current observations.
Their results show that an escape fraction which varies with host dark matter halo mass and redshift does in fact influence the structure of reionisation. However, the effect is smaller that the dominant astrophysical influence of SN feedback.