The foreground wedge and 21-cm BAO surveys

Redshifted H i 21 cm emission from unresolved low-redshift large-scale structure is a promising window for ground-based baryon acoustic oscillations (BAO) observations. A major challenge for this method is separating the cosmic signal from the foregrounds of Galactic and extra-Galactic origins that are stronger by many orders of magnitude than the former. The smooth frequency spectrum expected for the foregrounds would nominally contaminate only very small k modes; however, the chromatic response of the telescope antenna pattern at this wavelength to the foreground introduces non-smooth structure, pervasively contaminating the cosmic signal over the physical scales of our interest. Such contamination defines a wedged volume in Fourier space around the transverse modes that is inaccessible for the cosmic signal. In this paper, we test the effect of this contaminated wedge on the future 21-cm BAO surveys using Fisher information matrix calculation. We include the signal improvement due to the BAO reconstruction technique that has been used for galaxy surveys and test the effect of this wedge on the BAO reconstruction as a function of signal to noises and incorporate the results in the Fisher matrix calculation. We find that the wedge effect expected at z = 1–2 is very detrimental to the angular diameter distances: the errors on angular diameter distances increased by 3–4.4 times, while the errors on H(z) increased by a factor of 1.5–1.6. We conclude that calibration techniques that clean out the foreground ‘wedge’ would be extremely valuable for constraining angular diameter distances from intensity-mapping 21-cm surveys.