Speaker
Description
The reionization history of the intergalactic medium (IGM) is not perfectly constrained, although the midpoint is generally accepted to occur near $z\sim8$ and the whole process is likely concluded by $z\sim5.3$. The evolution of the neutral hydrogen fraction $x_\mathrm{HI}$ with redshift is being converged upon through a combination of many observational probes, but one that is particularly sensitive to the end stages at $z<6$ is the Lyman alpha forest (LAF). Metrics of the LAF, including the 1D flux power spectrum, autocorrelation function, and effective optical depth distributions have already been used to constrain the timeline, but with a more systematic approach to the investigation, the uncertainties may be reduced even further. To facilitate generation of improved late-stage constraints, we have developed a framework for simulating reionization using the Nyx cosmological hydrodynamics code paired with the AMBER (Trac et al. 2022) model for inhomogeneous reionization. AMBER parametrizes the reionization history in terms of a duration, midpoint, and asymmetry, which are paired along with Nyx-specific parameters such as the ionization front (i-front) heating and UVB details. To enable a more fine-grained evaluation of the parameters, we are creating neural network emulators of several LAF metrics based on a grid of these simulations, starting with the LAF power spectrum. Exploring the general trends, we find that the pressure smoothing scale within the IGM is strongly correlated with the adiabatic index of the temperature-density relation. We find that while models with 20,000 K i-front heating are better able to reproduce the shape of the observed $z=5$ 1D flux power spectrum than colder ones, they fail to match the highest wavenumbers. Further, we note that longer and more symmetric histories with greater i-front heating produce higher values of the autocorrelation functions and wider optical depth distributions, both of which are better matches to the observed LAF metrics than shorter, asymmetric, and cold i-front histories.
