http://arxiv.org/abs/1403.0941
We perform a detailed comparison of three different simulations of the neutral hydrogen distribution during the epoch of reionization (EoR). Our benchmark is a radiative transfer simulation (C2RAY). Such simulations can produce realistic results, but are computationally expensive. We compare it with two different semi-numerical techniques: one using the same halos as C2RAY as its sources (Sem-Num), and one using a conditional Press-Schechter scheme (CPS+GS). These are more computationally efficient than C2RAY, but use more simplistic physics. We evaluate them in terms of how well they can reproduce the history and morphology of EoR. We find that Sem-Num can produce an ionization history and morphology that is very close to C2RAY. Considering the effects of redshift space distortions due to peculiar velocities, we also study a number of statistics such as: the variance, spherically averaged power spectrum and various angular multipole moments of the power spectrum of the 21-cm signal from EoR,that will be observable by upcoming radio telescopes. We find that both semi-numerical models perform quite well in predicting these observables at length scales relevant for the present and future experiments. We also observe that predictions from Sem-Num are slightly better (by $\sim 10\%$) than CPS+GS for most of the observables. However, the predictions for the reionization history and the morphology of the ionization maps are significantly better (by $\sim 25-50\%$) in Sem-Num than CPS+GS. As the predictions for the 21-cm signal together with the reionization history would be required for parametrization of EoR from the observations, it would be better to use a semi-numerical simulation which can predict both with an acceptable accuracy. Our analysis shows that among the two semi-numerical simulations Sem-Num satisfies this criteria very well as it uses a source model very similar to C2RAY.
S. Majumdar, G. Mellema, K. Datta, et. al.
Thu, 6 Mar 14
32/53