ABSTRACT We explore how the characteristics of the cross-correlation functions between the 21-cm emission from the spin-flip transition of neutral hydrogen (H $\scriptstyle \rm I $) and early Lyman α (Ly α) radiation emitting galaxies (Ly α emitters, LAEs) depend on the reionization history and topology and the simulated volume. For this purpose, we develop an analytic expression for the 21-cm-LAE cross-correlation function and compare it to results derived from different astraeus and 21cmfast reionization simulations covering a physically plausible range of scenarios where either low-mass (≲ 109.5 M⊙) or massive (≳ 109.5 M⊙) galaxies drive reionization. Our key findings are: (i) the negative small-scale (≲ 2 cMpc) cross-correlation amplitude scales with the intergalactic medium’s (IGM) average H $\scriptstyle \rm I $ fraction (〈χH i〉) and spin-temperature weighted overdensity in neutral regions (〈1 + δ〉H i); (ii) the inversion point of the cross-correlation function traces the peak of the size distribution of ionized regions around LAEs; (iii) the cross-correlation amplitude at small scales is sensitive to the reionization topology, with its anticorrelation or correlation decreasing the stronger the ionizing emissivity of the underlying galaxy population is correlated to the cosmic web gas distribution (i.e. the more low-mass galaxies drive reionization); (iv) the required simulation volume to not underpredict the 21-cm-LAE anticorrelation amplitude when the cross-correlation is derived via the cross-power spectrum rises as the size of ionized regions and their variance increases. Our analytic expression can serve two purposes: to test whether simulation volumes are sufficiently large, and to act as a fitting function when cross-correlating future 21-cm signal Square Kilometre Array and LAE galaxy observations.
Read full abstract