Recovering the 21-cm signal from simulated FAST intensity maps

Abstract

ABSTRACT The 21-cm intensity mapping (IM) of neutral hydrogen (H i) is a promising tool to probe the large-scale structures. Sky maps of 21-cm intensities can be highly contaminated by various foregrounds, such as Galactic synchrotron radiation, free–free emission, extragalactic point sources, and atmospheric noise. Here we present a model of foreground components and a method of removal, in particular to quantify the potential of the Five-hundred-meter Aperture Spherical radio Telescope (FAST) for measuring H i IM. We consider a 1-yr observational time with a survey area of $20\,000\, {\rm deg}^{2}$ to capture significant variations of the foregrounds across both the sky position and angular scales relative to the H i signal. We first simulate the observational sky and then employ the principal component analysis (PCA) foreground separation technique. We show that by including different foregrounds, and thermal and 1/f noises, the value of the standard deviation between the reconstructed 21-cm IM map and the input pure 21-cm signal is $\Delta T = 0.034\, {\rm mK}$, which is well under control. The eigenmode-based analysis shows that the underlying H i eigenmode is just less than the 1 per cent level of the total sky components. By subtracting the PCA-cleaned foreground + noise map from the total map, we show that PCA method can recover H i power spectra for FAST with high accuracy.