Murriyang cryogenic phased array feed: spectral-line results and noise-reduction methods

TL;DR

This paper presents spectral-line results from a new cryogenic phased array feed on the Murriyang telescope, demonstrating improved survey capabilities and novel noise-reduction methods using higher-order SVD techniques.

Contribution

Introduction of a cryogenic phased array feed with enhanced field of view and survey speed, and development of advanced 3D SVD-based noise reduction methods for radio astronomy.

Findings

CryoPAF achieves a system temperature to efficiency ratio of 25 K within 0.3 deg of the optical axis.

HI observations align with previous results but suggest additional low-column-density gas.

3D SVD techniques significantly improve noise reduction and faint signal detection in RFI-rich environments.

Abstract

Spectral-line results from a new cryogenic phased array feed (cryoPAF) on the Murriyang telescope at Parkes are presented. This array offers a significant improvement in field of view, aperture efficiency, bandwidth, chromaticity and survey speed compared with conventional horn-fed receivers. We demonstrate this with measurements of sky calibrators and observations of 21-cm neutral hydrogen (HI) in the LMC and the nearby galaxy NGC 6744. Within 0.3 deg of the optical axis, the ratio of system temperature to dish aperture efficiency is 25 K and the ratio with beam efficiency is 21 K (at 1.4 GHz). For the previously measured $T_{sys} = 17$ K, respective efficiency values 0.7 and 0.8 are derived. Our HI observational results are in good agreement with previous results, although detailed comparison with multibeam observations of the LMC suggests that the earlier observations may have missed an extended component of low-column-density gas ($8\times 10^{18}$ cm$^{-2}$). We use the cryoPAF zoom-band and wideband data to make a preliminary investigation of whether the large number of simultaneous beams (72) permits the use of novel data reduction methods to reduce the effects of foreground/background continuum contamination and RFI. We also investigate if these methods can better protect against signal loss for the detection of faint, extended cosmological signals such as HI intensity maps. Using robust higher-order singular value decomposition (SVD) techniques, we find encouraging results for the detection of both compact and extended sources, including challenging conditions with high RFI occupancy and significant sky continuum structure. Examples are shown that demonstrate that 3D SVD techniques offer a significant improvement in noise reduction and signal capture compared with more traditional layered 2D techniques.