Measuring the redshift of reionization with a modest array of low-frequency dipoles

Abstract

The designs of the first generation of cosmological 21-cm observatories are split between single dipole experiments which integrate over a large patch of sky in order to find the global (spectral) signature of reionization, and interferometers with arcminute-scale angular resolution whose goal is to measure the 3D power spectrum of ionized regions during reionization. We examine whether intermediate scale instruments with complete Fourier (uv) coverage are capable of placing new constraints on reionization. We find that even without using a full power spectrum analysis, the global redshift of reionization, zreion, can in principle be measured from the variance in the 21-cm signal among multiple beams as a function of frequency at a roughly 1 degree angular scale. At this scale, the beam-to-beam variance in the differential brightness temperature peaks when the average neutral fraction was ∼ 50%, providing a convenient flag of zreion. We choose a low angular resolution of order 1• to exploit the physical size of the ionized regions and maximize the signal-to-noise ratio. Thermal noise, foregrounds, and instrumental effects should also be manageable at this angular scale, as long as the uv coverage is complete within the compact core required for low-resolution imaging. For example, we find that zreion can potentially be detected to within a redshift uncertainty of ∆zreion < ∼ 1 in > ∼ 500 hours of integration on the existing MWA prototype (with only 32 × 16 dipoles), operating at an angular resolution of ∼ 1• and a spectral resolution of 2.4 MHz.

Overcoming real-world obstacles in 21 cm power spectrum estimation: A method demonstration and results from early Murchison Widefield Array data

“…Understanding the behavior of the EoR window over a large redshift range is important, since there is still considerable uncertainty about the timing and duration of the EoR. Moreover, it is often argued that a tentative first detection of the cosmological signal will only be convincingly distinguishable from residual foregrounds if one can show that the 21 cm brightness temperature fluctuations peak at some redshift, since theory predicts that the midpoint of reionization should be marked by such a peak [7,62]. It is therefore essential to characterize the EoR window (and by extension, residual foregrounds) over a broad frequency range.…”

Global 21 cm signal experiments: A designer’s guide

“…Although it is strictly speaking non-zero, we can safely ignore cosmological anisotropy noise because it is negligibly small compared to the foreground noise. Through a combination of analytic theory [48] and simulation analysis [38], the cosmological anisotropies have been shown to be negligible on scales larger than ∼ 1 • to 2 • , which is a regime that we remain in for this paper.…”

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

“…From this PDF, analytic expression for statistical moments can be derived. Comparing how these statistical moments, derived numerically from more sophisticated models and simulations, deviate from these analytical expressions provide us with hints on the nature of reionisation (Gluscevic & Barkana 2010), such as its reionisation topology (e.g., inside-out or outside-in, see Watkinson & Pritchard 2014) and its global reionisation history (Bittner & Loeb 2011;Patil et al 2014), even when derived from dirty 21 cm signal images or after foreground removal (Harker et al 2009). Kittiwisit et al (2018) show that HERA will be able to detect the variance of the 21 cm brightness temperature field from reionisation with high sensitivity, by averaging over measurements from multiple fields.…”