Unlocking Sensitivity for Visibility-based Estimators of the 21 cm Reionization Power Spectrum

• Published in: The Astrophysical journal Vol. 852; no. 2; pp. 110 - 122
• Main Authors: Zhang, Yunfan Gerry, Liu, Adrian, Parsons, Aaron R.
• Format: Journal Article
• Language: English
• Published: Philadelphia The American Astronomical Society 10.01.2018; IOP Publishing
• Subjects: Antennas; Arrays; Astrophysics; Beams (radiation); Coherence; dark ages, reionization, first stars; Drift; Hydrogen; instrumentation: interferometers; Ionization; methods: data analysis; Pipelines; Radio interferometers; Redundancy; Sensitivity; Systematics; techniques: interferometric; Time lag; Visibility
• ISSN: 0004-637X; 1538-4357
• DOI: 10.3847/1538-4357/aaa029

Radio interferometers designed to measure the cosmological 21 cm power spectrum require high sensitivity. Several modern low-frequency interferometers feature drift-scan antennas placed on a regular grid to maximize the number of instantaneously coherent (redundant) measurements. However, even for such maximum-redundancy arrays, significant sensitivity comes through partial coherence between baselines. Current visibility-based power-spectrum pipelines, though shown to ease control of systematics, lack the ability to make use of this partial redundancy. We introduce a method to leverage partial redundancy in such power-spectrum pipelines for drift-scan arrays. Our method cross-multiplies baseline pairs at a time lag and quantifies the sensitivity contributions of each pair of baselines. Using the configurations and beams of the 128-element Donald C. Backer Precision Array for Probing the Epoch of Reionization (PAPER-128) and staged deployments of the Hydrogen Epoch of Reionization Array, we illustrate how our method applies to different arrays and predict the sensitivity improvements associated with pairing partially coherent baselines. As the number of antennas increases, we find partial redundancy to be of increasing importance in unlocking the full sensitivity of upcoming arrays.