[Paper Review] GOODS-ALMA 2.0: Source catalog, number counts, and prevailing compact sizes in 1.1 mm galaxies

Carlos Gómez-Guijarro, D. Elbaz+37 more|arXiv (Cornell University)|Jun 24, 2021

Galaxies: Formation, Evolution, Phenomena179 references85 citations

TL;DR

This study presents GOODS-ALMA 2.0, a deep, contiguous 1.1 mm survey of 72.42 arcmin² in GOODS-South using combined high- and low-resolution ALMA data to achieve homogeneous sensitivity (σ = 68.4 µJy beam⁻¹). It detects 88 galaxies, revealing that dust continuum emission is predominantly compact (median Re = 0″.10 ± 0″.05, or 0.73 ± 0.29 kpc), with compactness prevailing even at sub-mJy fluxes, challenging assumptions about extended emission in faint sources.

ABSTRACT

Submillimeter/millimeter observations of dusty star-forming galaxies with the Atacama Large Millimeter/submillimeter Array (ALMA) have shown that dust continuum emission generally occurs in compact regions smaller than the stellar distribution. However, it remains to be understood how systematic these findings are. Studies often lack homogeneity in the sample selection, target discontinuous areas with inhomogeneous sensitivities, and suffer from modest $uv$ coverage coming from single array configurations. GOODS-ALMA is a 1.1mm galaxy survey over a continuous area of 72.42arcmin$^2$ at a homogeneous sensitivity. In this version 2.0, we present a new low resolution dataset and its combination with the previous high resolution dataset from the survey, improving the $uv$ coverage and sensitivity reaching an average of $\sigma = 68.4\mu$Jy beam$^{-1}$. A total of 88 galaxies are detected in a blind search (compared to 35 in the high resolution dataset alone), 50% at $S/N_{peak} \geq 5$ and 50% at $3.5 \leq S/N_{peak} \leq 5$ aided by priors. Among them, 13 out of the 88 are optically dark or faint sources ($H$- or $K$-band dropouts). The sample dust continuum sizes at 1.1mm are generally compact, with a median effective radius of $R_{e} = 0"10 \pm 0"05$ (a physical size of $R_{e} = 0.73 \pm 0.29$kpc at the redshift of each source). Dust continuum sizes evolve with redshift and stellar mass resembling the trends of the stellar sizes measured at optical wavelengths, albeit a lower normalization compared to those of late-type galaxies. We conclude that for sources with flux densities $S_{1.1mm} > 1$mJy, compact dust continuum emission at 1.1mm prevails, and sizes as extended as typical star-forming stellar disks are rare. The $S_{1.1mm} < 1$mJy sources appear slightly more extended at 1.1mm, although they are still generally compact below the sizes of typical star-forming stellar disks.

Motivation & Objective

• To overcome observational biases in ALMA studies of dusty star-forming galaxies (DSFGs) by using a contiguous, homogeneous survey area with dual array configurations.
• To address the systematic uncertainty in dust continuum size measurements by combining high-resolution and low-resolution data to improve uv coverage and sensitivity.
• To investigate whether compact dust emission is a universal property of DSFGs across a wide range of flux densities, especially below 1 mJy.
• To identify and characterize optically dark or faint sources (H- or K-band dropouts) missed in previous surveys.
• To derive robust number counts and size distributions to test the prevalence of compact vs. extended dust emission in DSFGs.

Proposed method

• Conduct a blind source detection in a 72.42 arcmin² continuous field using combined high-resolution (compact array) and low-resolution (extended array) ALMA data at 1.1 mm.
• Apply a prior-based source-finding methodology to recover faint sources with S/Npeak between 3.5 and 5, improving completeness at the faint end.
• Perform flux and size measurements using a multi-scale Gaussian decomposition technique on the combined dataset to model dust continuum emission.
• Use Monte Carlo simulations to assess source completeness and flux reliability, particularly in the sub-mJy regime.
• Derive differential and cumulative number counts by combining results from high- and low-resolution datasets, correcting for sensitivity and beam effects.
• Measure dust continuum effective radii (Re) and study their evolution with redshift and stellar mass, comparing to optical stellar sizes.

Experimental results

Research questions

• RQ1 Is compact dust continuum emission at 1.1 mm a systematic property of dusty star-forming galaxies across different flux densities?
• RQ2 How does the dust continuum size distribution of DSFGs compare to the sizes of their stellar components, and does this depend on redshift or stellar mass?
• RQ3 To what extent do low-resolution data recover fainter, potentially more extended sources missed by high-resolution-only surveys?
• RQ4 What fraction of ALMA-detected sources are optically dark or faint (H- or K-band dropouts), and how do their dust sizes compare to more luminous sources?
• RQ5 How do the number counts of 1.1 mm sources in this survey compare to existing literature, and what does this imply about the completeness of previous surveys?

Key findings

• A total of 88 galaxies were detected in a blind search, with 44 sources at S/Npeak ≥5 (100% purity) and 44 additional sources detected via prior-based methods at 3.5 ≤ S/Npeak ≤ 5.
• Thirteen of the detected sources are optically dark or faint (H- or K-band dropouts), seven of which are newly identified in this work.
• The median dust continuum effective radius at 1.1 mm is Re = 0″.10 ± 0″.05, corresponding to a physical size of 0.73 ± 0.29 kpc, indicating predominantly compact emission.
• Dust continuum sizes evolve with redshift and stellar mass in a manner similar to optical stellar sizes, though with a lower normalization than late-type galaxies.
• For sources with S₁.₁ₘₘ > 1 mJy, compact dust emission is the norm, and sizes as extended as typical star-forming stellar disks are rare.
• At S₁.₁ₘₘ < 1 mJy, sources appear slightly more extended on average, but remain generally compact (below typical stellar disk sizes), with a larger scatter and possible hints of extended emission requiring deeper data for confirmation.