Speaker
Description
Cosmological neutral hydrogen (HI) surveys offer a powerful tomographic
probe of the post-reionization universe to test the standard model of cosmology. Efficiently simulating this signal is essential for optimizing the outcome of upcoming surveys. We present a fast simulation method of the cosmological HI distribution employing the halo model framework. Our approach uses the approximate but fast PINOCCHIO code to generate the past light cone of dark matter halos, which are then populated with HI based on an empirical HI-halo mass relation. Given the large volumes and high mass resolution required for 21 cm intensity mapping, we simulate a past light cone covering declinations between -15° and -35° and frequencies from 700 to 800 MHz, matching HIRAX (the Hydrogen Intensity and Real-time Analysis eXperiment). To achieve this, we simulate a 1 h^{−3}Gpc^3 box with 6700^3 simulation particles, which is replicated to cover the entire lightcone, resolving halos masses down to 4.3 × 10^9 M_⊙ and capturing over 97 % of the expected HI density. From the past light cone, we construct HI maps with 5 MHz bandwidths. We compare key statistical properties, such as the mass function, mass density, power spectrum, and angular power spectrum against an analytical dark matter and HI halo model, which we implemented in PyCosmo, a Python package for theoretical cosmological predictions. Strong agreement is found between the analytical model and the simulation, with discrepancies within a few percent for the one-point statistics and under 10 % for the two-point statistics. Our approach enables efficient forecasting and forward-modeling for dedicated intensity mapping experiments like HIRAX, as well as future cosmological surveys with the SKAO, with strong potential for cross-correlation studies.
