GMRT observation towards detecting the Post-reionization 21-cm signal

TL;DR

This study analyzes GMRT 610 MHz data to detect the redshifted 21-cm signal from the post-reionization epoch, employing foreground removal techniques to set upper limits on the neutral hydrogen fraction.

Contribution

First application of foreground removal and polynomial fitting to GMRT data for 21-cm signal detection at z=1.32, establishing new upper limits on neutral hydrogen.

Findings

Foregrounds dominate the measured power spectrum.

Residuals after foreground removal are consistent with noise.

Upper limit on x_HI b > 7.95 at 3-sigma.

Abstract

We have analyzed 610 MHz GMRT observations towards detecting the redshifted 21-cm signal from z=1.32. The multi-frequency angular power spectrum C_l(Delta nu) is used to characterize the statistical properties of the background radiation across angular scales ~20" to 10', and a frequency bandwidth of 7.5 MHz with resolution 125 kHz. The measured C_l(Delta nu) which ranges from 7 mK^2 to 18 mK^2 is dominated by foregrounds, the expected HI signal C_l^HI(Delta nu) ~10^{-6}- 10^{-7} mK^2 is several orders of magnitude smaller. The foregrounds, believed to originate from continuum sources, is expected to vary smoothly with Delta nu whereas the HI signal decorrelates within ~0.5 MHz and this holds the promise of separating the two. For each l, we use the interval 0.5 < Delta nu < 7.5 MHz to fit a fourth order polynomial which is subtracted from the measured C_l(Delta nu) to remove any smoothly varying component across the entire bandwidth Delta nu < 7.5 MHz. The residual C_l(Delta nu), we find, has an oscillatory pattern with amplitude and period respectively ~0.1 mK^2 and Delta nu = 3 MHz at the smallest l value of 1476, and the amplitude and period decreasing with increasing l. Applying a suitably chosen high pass filter, we are able to remove the residual oscillatory pattern for l=1476 where the residual C_l(Delta nu) is now consistent with zero at the 3-sigma noise level. We conclude that we have successfully removed the foregrounds at l=1476 and the residuals are consistent with noise. We use this to place an upper limit on the HI signal whose amplitude is determined by x_HI b where x_HI and b are the HI neutral fraction and the HI bias respectively. A value of x_HI b greater than 7.95 would have been detected in our observation, and is therefore ruled out at the 3-sigma level. (abridged)