EDGES reported the first evidence of star formation in the high-redshift Universe (z~18) in 2018 (Bowman et al, 2018; hereafter B18). The extraordinary depth and flattened profile of the reported absorption feature in the sky-averaged 21 cm spectrum compared to theoretical expectation has caused a great deal of both excitement and concern in the community. The EDGES team has continued to expand the set of verification tests presented in B18, which originally tested that the absorption feature was recovered under different instrument configurations (alternate ground planes, different identical antennas, different antenna orientations) and processing options (with/without beam chromaticity correction, different beam models, alternate smooth foreground models) and data cuts (different times of day and LST cuts).
This talk will present updates in three categories: (i) new data and alternative hardware configurations that aim to expose systematics, (ii) improved analyses and (iii) upgraded hardware -- but will focus on my work on (ii). This work is fundamentally concerned with the reproducibility of the 2018 results and improving the analysis to properly account for calibration systematics within a Bayesian framework. While this work is ongoing, I will show preliminary results that explore whether uncertainties in our gain calibration significantly bias our cosmological interpretation. Finally, I will give updates on EDGES-3, the next-generation experiment that will feature in-field real-time calibration, a less chromatic beam, and higher calibratability with fewer systematics in the band. This update is expected to yield order-of-magnitude improvements in the resulting model residuals, and expose extant contributions from instrumental systematics.
Meeting ID: 937 0923 8875