[Paper Review] First results from the ATCA 18 GHz pilot survey

R. Ricci, E. M. Sadler+8 more|arXiv (Cornell University)|Jul 7, 2004

Radio Astronomy Observations and Technology1 references3 citations

TL;DR

This paper presents the first results from a pilot survey using the Australia Telescope Compact Array (ATCA) at 18 GHz, covering 1,216 deg² of the southern sky with a 60 mJy 4σ sensitivity limit. It confirms 221 sources, finds extragalactic source counts well described by a power-law index of −2.2, and reveals that over 70% have flat or inverted spectra, with high optical identification rates indicating a luminous, high-frequency-selected population distinct from lower-frequency surveys.

ABSTRACT

As a pilot study for the first all-sky radio survey at short wavelengths, we have observed 1216 sq. deg. of the southern sky at 18GHz (16mm) using a novel wide-band (3.4GHz bandwidth) analogue correlator on one baseline of the Australia Telescope Compact Array (ATCA). We scanned a region of sky between declination -71 and -59 degrees with an rms noise level of 15mJy. Follow-up radio imaging of candidate sources above a 4 sigma detection limit of 60mJy resulted in 221 confirmed detections, for which we have measured accurate positions and flux densities. For extragalactic sources, the survey is roughly 70% complete at a flux density of 126mJy and 95% complete above 300mJy. Almost half the detected sources lie within a few degrees of the Galactic plane, but there are 123 sources with |b|>5 degrees which can be assumed to be extragalactic. The differential source counts for extragalactic sources in the range 0.1Jy < S_{18GHz} < 3Jy are well fitted by a relation of the form n(S)=57(S/{Jy})^{-2.2\pm0.2} [Jy^-1 sr^-1], in good agreement with the 15GHz counts published by Taylor et al. (2001) and Waldram et al. (2003). Over 70% of the extragalactic sources have a flat radio spectrum (α_{0.843}^{18}> -0.5, S_ν\proptoν^α), and 29% have inverted radio spectra (α_{0.843}^{18}> 0). The optical identification rate is high: 51% of the extragalactic sources are identified with stellar objects (candidate QSOs), 22% with galaxies and only 27% with faint optical objects or blank fields.

Motivation & Objective

• To conduct a pilot all-sky survey at 18 GHz to probe the high-frequency radio source population.
• To assess the feasibility of wide-area, high-frequency surveys using a wide-band analog correlator on a compact array.
• To measure source counts and spectral properties for extragalactic sources to inform CMB foreground contamination and future surveys like Planck.
• To evaluate the optical identification rate and polarization properties of high-frequency radio sources.
• To determine the completeness and sensitivity of the survey for future all-sky surveys at 20 GHz.

Proposed method

• Utilized a wide-band (3.4 GHz) analog correlator on a single baseline of ATCA to achieve high bandwidth and fast scanning.
• Conducted rapid 0.2 s integration scans across a 1,216 deg² region between declinations −71° and −59°.
• Achieved a 4σ sensitivity limit of 60 mJy with an rms noise level of 15 mJy.
• Performed follow-up synthesis imaging on 4σ candidates to confirm detections and measure accurate flux densities and positions.
• Used a 3.4 GHz bandwidth with 16 lag channels, yielding 8 frequency channels of 425 MHz each for high signal-to-noise measurements.
• Applied source count modeling using a power-law fit: n(S) = 57 (S/Jy)−2.2±0.2 Jy⁻¹ sr⁻¹ for extragalactic sources.

Experimental results

Research questions

• RQ1 What is the source count distribution of extragalactic radio sources at 18 GHz, and how does it compare to lower-frequency surveys?
• RQ2 What fraction of high-frequency radio sources have flat or inverted spectra, and how does this affect CMB foreground contamination?
• RQ3 How does the optical identification rate of 18 GHz-selected sources compare to that of 1.4 GHz surveys?
• RQ4 What is the polarization distribution of high-frequency radio sources, and can it affect CMB B-mode measurements?
• RQ5 Can a wide-band, fast-scanning interferometric survey achieve sufficient sensitivity and completeness for future all-sky surveys at 20 GHz?

Key findings

• The survey confirmed 221 sources with a 4σ detection threshold of 60 mJy, achieving a 15 mJy rms noise level over 1,216 deg².
• Extragalactic source counts in the 0.1–3 Jy range are well described by a power-law with index −2.2±0.2, consistent with 15 GHz surveys.
• Over 70% of extragalactic sources have flat or inverted spectra (α₁₈ > −0.5), including 29% with inverted spectra (α₁₈ > 0).
• The optical identification rate is 73%, with 51% identified as QSOs and 22% as galaxies—significantly higher than at 1.4 GHz.
• The median polarization is 2.1%, consistent with known flat-spectrum populations and low enough to not hinder Planck’s E-mode measurements.
• The survey confirms that a full-sky 20 GHz survey with 40 mJy sensitivity is feasible and would yield ~10⁴ extragalactic sources, enabling a robust calibration network for ALMA.