Position-dependent power spectra of the 21-cm signal from the epoch of reionization

Abstract

The 21-cm signal from the epoch of reionization is non-Gaussian. Current radio telescopes are focused on detecting the 21-cm power spectrum, but in the future the Square Kilometre Array is anticipated to provide a first measurement of the bispectrum. Previous studies have shown that the position-dependent power spectrum is a simple and efficient way to probe the squeezed-limit bispectrum. In this approach, the survey is divided into subvolumes and the correlation between the local power spectrum and the corresponding mean density of the subvolume is computed. This correlation is equivalent to an integral of the bispectrum in the squeezed limit, but is much simpler to implement than the usual bispectrum estimators. It also has a clear physical interpretation: it describes how the smallscale power spectrum of tracers such as galaxies and the 21-cm signal respond to a large-scale environment. Reionization naturally couples large and small scales as ionizing radiation produced by galactic sources can travel up to tens of Megaparsecs through the intergalactic medium during this process. Here we apply the position-dependent power spectrum approach to fluctuations in the 21-cm background from reionization. We show that this statistic has a distinctive evolution in time that can be understood with a simple analytic model. We also show that the statistic can easily distinguish between simple "inside-out" and "outside-in" models of reionization. The position-dependent power spectrum is thus a promising method to validate the reionization signal and to extract higher-order information on this process.

Using the sample variance of 21 cm maps as a tracer of the ionisation topology

“…We note that other sample shapes could have been considered instead of slices. For example, Giri et al (2019) consider the power spectrum measured in sub-cubes of a simulation. Since observational data cubes will consist of slices at different (but gradually increasing) frequencies, our findings can be easier translated to the results from observations.…”

“…We compute the local variance obtained from the 21 cm brightness temperature fields of all sub-cubes and compare their values in different redshifts. Instead, we would consider a wide field-ofview at a given frequency (or redshift), and divide it into subpatches -an approach already imagined in Giri et al (2019).…”

“…However, for smaller fields, it will provide us with information about the morphology imprinted in the field x(r), as it -contrary to usual one-point statistics, encompasses information on the correlations between pixels. It is interesting to note that previous works have also considered estimators computed in subvolumes of data: for example, Chiang et al (2015), Giri et al (2019) investigate the power spectrum of sub-volumes, also referred to as the position-dependent power spectrum. However, in contrast to these approaches, our estimator will benefit from its simplicity.…”

Investigating X-ray sources during the epoch of reionization with the 21 cm signal

“…These measurements can already rule out some, albeit extreme, EoR models (Ghara et al 2020;Greig et al 2020;Mondal et al 2020). Higher-order statistics, such as the bispectrum (Shimabukuro et al 2016(Shimabukuro et al , 2017Majumdar et al 2018;Hutter et al 2020), the three-point correlation function (Hoffmann et al 2019), position-dependent power spectra (Giri et al 2019), and multi-correlations (e.g. kSZ-kSZ-21 cm correlations as in Ma et al 2018b andLa Plante et al 2020, or [CII]-[CII]-21 cm correlations as in Beane & Lidz 2018), have also been proposed to study the EoR.…”

Tools21cm: A python package to analyse the large-scale 21-cm signal from the Epoch of Reionization and Cosmic Dawn

“…Ross, Dixon, Iliev, & Mellema, 2017;Ross, Dixon, Ghara, Iliev, & Mellema, 2019). It can also characterise the spatial fluctuations in the signal through spherically and cylindrically averaged power spectra Jensen et al, 2013;Ross et al, 2017) and position dependent power spectra (Giri, D'Aloisio, et al, 2019). It also has the capability to find interesting features, such as ionized regions, in (tomographic) image data (Giri et al, , 2018b and from these to derive statistical quantities, such as size distributions (Giri et al, 2018a) and topological quantities such as the Euler characteristic .…”