PI: James Bolton

In Asthana et al. (2024, MNRAS, 533, 2843) we presented a new suite of late-end reionisation simulations performed with ATON-HE, a GPU-based radiative transfer code.  The simulations were performed as part of DiRAC project dp145 using the Cambridge Data Intensive and Edinburgh Extreme Scaling systems.   The simulations are designed to reproduce the observed properties of the intergalactic medium at z<6.   Our fiducial reionisation history has a mid-point at redshift z = 6.5, but we also explore an ‘Early’ reionisation history with a mid-point at z = 7.5 and an ‘Extremely Early’ reionisation history with a mid-point at z = 9.5 (see the three panels in the Figure).  We then considered the implications of these different reionisation models for the visibility of the Lyman-alpha emitting galaxies recently detected by JWST.  In our simulations, the fraction of haloes embedded in ionised bubbles that are large enough to allow JWST-detected Lyman-alpha emitters to be visible becomes close to unity much before the volume filling factor of ionised regions. For our fiducial reionisation history this happens at z = 8, but this is probably still too late to be consistent with the detection by JWSTof abundant Lyman-alpha emission out to z = 11.  We conclude a reionisation history similar to our ‘Early’ model or perhaps even our ‘Extremely Early’ model may be required to explain both the JWST observations and properties of the intergalactic medium at z<6.  

The figure shows maps of the neutral hydrogen fraction in the intergalactic medium and its evolution over time for three different models of reionisation.  Reionisation starts latest in the top panel, and earliest in the bottom panel.  These radiative transfer simulations were performed with DiRAC GPU allocations on the Cambridge Data Intensive and Edinburgh Extreme Scaling systems.  From Asthana et al, 2024, MNRAS, 533, 2843.