[Paper Review] The galaxies missed by Hubble and ALMA: the contribution of extremely red galaxies to the cosmic census at 3

Christina C. Williams, Stacey Alberts+34 more|arXiv (Cornell University)|Nov 13, 2023

Galaxies: Formation, Evolution, Phenomena6 citations

TL;DR

Using deep JWST imaging from JADES, JEMS, and SMILES, this study identifies a population of extremely red, dusty, and post-starburst galaxies at 3 < z < 8 that were previously missed by Hubble and ALMA. Including MIRI photometry reduces inferred stellar masses and SFRs by up to 0.6 dex and 10× respectively, revealing that these faint, obscured sources could double the estimated obscured star formation rate density at 4 < z < 6, significantly revising the cosmic census of early galaxy formation.

ABSTRACT

Using deep JWST imaging from JADES, JEMS and SMILES, we characterize optically-faint and extremely red galaxies at $z&gt;3$ that were previously missing from galaxy census estimates. The data indicate the existence of abundant, dusty and post-starburst-like galaxies down to $10^8$M$_\odot$, below the sensitivity limit of Spitzer and ALMA. Modeling the NIRCam and HST photometry of these red sources can result in extreme, high values for both stellar mass and star formation rate (SFR); however, including 7 MIRI filters out to 21$μ$m results in decreased mass (median 0.6 dex for log$_{10}$M$^*$/M$_{\odot}&gt;$10), and SFR (median 10$ imes$ for SFR$&gt;$100 M$_{\odot}$/yr). At $z&gt;6$, our sample includes a high fraction of little red dots (LRDs; NIRCam-selected dust-reddened AGN candidates). We significantly measure older stellar populations in the LRDs out to rest-frame 3$μ$m (the stellar bump) and rule out a dominant contribution from hot dust emission, a signature of AGN contamination to stellar population measurements. This allows us to measure their contribution to the cosmic census at $z&gt;3$, below the typical detection limits of ALMA ($L_{ m IR}&lt;10^{12}L_\odot$). We find that these sources, which are overwhelmingly missed by HST and ALMA, could effectively double the obscured fraction of the star formation rate density at $4

Motivation & Objective

• To identify and characterize optically-faint, extremely red galaxies at 3 < z < 8 that were previously undetected by Hubble and ALMA.
• To assess the contribution of these missed galaxies to the cosmic star formation rate density, particularly in the obscured (dusty) component.
• To determine whether extreme photometric modeling biases in stellar mass and SFR estimates can be corrected using deep MIRI photometry out to 21 μm.
• To evaluate the nature of Little Red Dots (LRDs) at z > 6, especially whether they host AGN or are consistent with evolved stellar populations.
• To quantify the impact of these previously missed sources on the overall cosmic census of galaxy formation in the early universe.

Proposed method

• Utilized deep near- and mid-infrared imaging from JWST’s NIRCam and MIRI instruments in the JADES, JEMS, and SMILES surveys.
• Applied photometric redshift and stellar population synthesis modeling using Prospector and EAZY to fit SEDs from rest-frame UV to 21 μm.
• Incorporated ALMA 1.3 mm data to constrain dust emission and assess AGN contamination.
• Used the stellar bump at rest-frame 3 μm as a diagnostic to distinguish evolved stellar populations from hot dust emission.
• Applied MIRI filters out to 21 μm to break degeneracies in stellar mass and SFR estimation from NIRCam-only data.
• Classified sources as Little Red Dots (LRDs) based on NIRCam color selection and confirmed their high-redshift nature via photometric redshifts.

Figure 1: F150W–F444W color vs F444W magnitude illustrating our selection from the JADES catalog with S/N $>$ 20 in F444W (black points). We identify very red sources by their F150W–F444W $>2.2$ color (all colored points) color coded by their EAZY photometric redshift. We further identify the optica

Experimental results

Research questions

• RQ1 What is the true contribution of extremely red, dusty galaxies to the cosmic star formation rate density at 3 < z < 8, particularly below the detection limits of Hubble and ALMA?
• RQ2 To what extent do stellar population modeling biases in stellar mass and SFR estimates arise from missing mid-infrared photometry, and how can MIRI data correct them?
• RQ3 Are the Little Red Dots (LRDs) at z > 6 dominated by AGN activity or by evolved stellar populations, and how can this be distinguished?
• RQ4 How do the physical properties of these missed galaxies—such as stellar mass, SFR, and dust attenuation—compare to those in standard galaxy catalogs?
• RQ5 To what extent do these previously undetected sources alter the inferred obscured fraction of the star formation rate density at 4 < z < 6?

Key findings

• Including MIRI photometry out to 21 μm reduces median stellar mass estimates by 0.6 dex and SFR estimates by a factor of 10 for sources with log10(M*/M⊙) > 10 and SFR > 100 M⊙/yr.
• A significant fraction of high-redshift galaxies at z > 6 are Little Red Dots (LRDs), with strong evidence for older stellar populations up to rest-frame 3 μm, ruling out dominant hot dust emission from AGN.
• These extremely red, dusty galaxies are overwhelmingly missed by Hubble and ALMA, with IR luminosities below the ALMA sensitivity limit (L_IR < 10^12 L⊙).
• The inclusion of these previously undetected sources could effectively double the estimated obscured star formation rate density at 4 < z < 6 compared to previous estimates.
• Five sources at 5.5 < z < 7.7 show clear Balmer breaks and high stellar masses, suggesting evolved populations, though spectroscopy is needed to confirm their nature.
• The study demonstrates that prior cosmic census estimates significantly underestimated the contribution of faint, dusty, and post-starburst galaxies to early galaxy evolution.

Figure 2: For sources with MIRI and ALMA coverage, we compare inferred photometric redshift, stellar mass and SFR from our prospector modeling of HST+NIRCam+MIRI+ALMA vs just the HST+NIRCam photometry. Left panel shows the inferred photometric redshift (sources with spectroscopic redshifts are exclu