[Paper Review] The VLA-VIRMOS Deep Field I. Radio observations probing the microJy source population

M. Bondi. P. Ciliegi, G. Zamorani+33 more|ArXiv.org|Mar 17, 2003

Galaxies: Formation, Evolution, PhenomenaPhysics and Astronomy25 references108 citations

TL;DR

This paper presents deep 1.4 GHz VLA radio observations of a 1 deg² field in the VIRMOS VLT Deep Survey, achieving a 17 μJy rms noise and detecting 1,054 radio sources down to 80 μJy. The study corrects for resolution bias and clean bias using simulated source populations, yielding radio counts that show a significant change in slope below 1 mJy and are ~50% higher than those in the Hubble Deep Field, indicating field-to-field variance and improved completeness in this survey.

ABSTRACT

We have conducted a deep survey (r.m.s noise 17 microJy) with the Very Large Array (VLA) at 1.4 GHz, with a resolution of 6 arcsec, of a 1 square degree region included in the VIRMOS VLT Deep Survey. In the same field we already have multiband photometry down to I(AB)=25, and spectroscopic observations will be obtained during the VIRMOS VLT survey. The homogeneous sensitivity over the whole field has allowed to derive a complete sample of 1054 radio sources (5 sigma limit). We give a detailed description of the data reduction and of the analysis of the radio observations, with particular care to the effects of clean bias and bandwidth smearing, and of the methods used to obtain the catalogue of radio sources. To estimate the effect of the resolution bias on our observations we have modelled the effective angular-size distribution of the sources in our sample and we have used this distribution to simulate a sample of radio sources. Finally we present the radio count distribution down to 0.08 mJy derived from the catalogue. Our counts are in good agreement with the best fit derived from earlier surveys, and are about 50 % higher than the counts in the HDF. The radio count distribution clearly shows, with extremely good statistics, the change in the slope for the sub-mJy radio sources.

Motivation & Objective

• To conduct a deep, wide-area radio survey at 1.4 GHz with uniform sensitivity and high angular resolution (6 arcsec) to probe the sub-mJy and μJy source population.
• To address systematic effects such as clean bias and bandwidth smearing that can distort source detection and flux measurement in deep radio surveys.
• To derive a complete, flux-limited catalogue of radio sources in a 1 deg² field to enable statistical analysis of the radio source count distribution.
• To correct for resolution bias using simulated source populations based on the effective angular-size distribution of detected sources.
• To compare the derived radio counts with previous surveys and assess field-to-field variance, particularly in relation to the Hubble Deep Field.

Proposed method

• Conducted 56 hours of VLA observations in B-configuration at 1.4 GHz over 9 days, achieving a 17 μJy rms noise and 6 arcsec resolution.
• Applied rigorous data reduction techniques to minimize clean bias and bandwidth smearing, with detailed analysis confirming their negligible impact on source detection.
• Generated a complete catalogue of 1,054 radio sources using a 5σ detection threshold, with flux densities measured from the cleaned mosaic.
• Modelled the effective angular-size distribution of sources in the 0.4–1.0 mJy range to simulate a population of radio sources with realistic size and flux distributions.
• Used the simulated source population to statistically correct the observed counts for resolution bias, particularly in the faintest flux bins (S < 0.08 mJy).
• Fitted the corrected differential and integral counts with broken power laws to quantify the change in slope at sub-mJy levels.

Experimental results

Research questions

• RQ1 What is the true shape of the 1.4 GHz radio source count distribution down to 80 μJy, and does it show a significant change in slope below 1 mJy?
• RQ2 To what extent do resolution bias and clean bias affect the completeness and flux measurements in deep radio surveys with high dynamic range?
• RQ3 How do the radio counts in this survey compare with those from the Hubble Deep Field and other deep surveys, and what does this imply about field-to-field variance?
• RQ4 What is the contribution of different source populations (e.g., starbursts, early-type galaxies) to the sub-mJy and μJy radio source population, as inferred from multi-wavelength follow-up?
• RQ5 Can the observed source count distribution be accurately modeled using a broken power law, and what are the best-fit parameters for the faint and bright flux regimes?

Key findings

• The survey achieved a 17 μJy rms noise level and detected 1,054 radio sources down to a 5σ limit of approximately 80 μJy in a 1 deg² field.
• The corrected radio source counts show a significant change in slope below 1 mJy, with a differential slope of -2.28 ± 0.04 in the 0.08–0.6 mJy range, close to the Euclidean value.
• The counts in the 0.08–0.6 mJy range are approximately 50% higher than those derived by Richards (2000) in the Hubble Deep Field, indicating systematic undercounting in the HDF.
• The observed counts are in good agreement with the best-fit model from earlier surveys (Katgert et al. 1988), validating the completeness and accuracy of the corrected catalogue.
• Simulations based on the effective angular-size distribution of sources in the 0.4–1.0 mJy range confirmed that resolution bias significantly affects the faintest flux bins and that correction is essential for accurate counts.
• The study demonstrates that field-to-field variance in sub-mJy counts is real and not solely due to instrumental or reduction effects, as confirmed by comparison with lower-resolution surveys of the HDF region.