A Comprehensive Photometric Selection of `Little Red Dots' in MIRI Fields: An IR-Bright LRD at $z=3.1386$ with Warm Dust Emission

TL;DR

This work presents a broad, photometrically driven census of Little Red Dots (LRDs) using JWST NIRCam and MIRI data across PRIMER, JADES, and UDS fields, extending color selection to $z \sim 3$–$9$ and reducing contamination from red galaxies and brown dwarfs. By integrating MIRI-based rest-frame UV-to-NIR colors and a compactness criterion, the authors identify 248 LRDs down to $F444W<27$ mag, revealing a higher number density (~0.93 arcmin$^{-2}$) than prior studies and a substantial fraction of new, bluer LRDs. The study analyzes UV-to-NIR and near-to-mid-IR colors to infer diverse dust emission properties, comparing to galaxy- and AGN-dominated SED models, and demonstrates substantial degeneracies in the IR peak origin between ISM dust and AGN torus scenarios. Two low-redshift LRDs with rich MIRI coverage are modeled in detail, illustrating how warm-to-hot dust can shape the IR SED and underscoring the need for deeper MIRI/future spectroscopy to disentangle the dust origins and constrain dust masses and temperatures in LRDs.

Abstract

JWST has revealed a population of compact `Little Red Dots' (LRDs) at $z\gtrsim4$, with red rest-frame optical and blue UV colors. These objects are likely compact dusty starbursts or heavily reddened AGNs, playing a pivotal role in early black hole growth, dust production, and stellar assembly. We introduce a new photometric selection to identify LRDs over a broad range in redshifts and rest-frame UV-to-NIR colors enabling a more complete census of the population. This method identifies 248 LRDs with F444W$<27$ mag over 263 arcmin$^2$ in the JADES, PRIMER-COSMOS, and UDS fields with MIRI coverage, increasing the number density by $\times$1.7 compared to previous samples, suggesting that previous census were underestimated. Most LRDs are detected in MIRI/F770W but only 7% (17) are detected in F1800W. We use MIRI-based rest-frame [1$-$3 $μ$m] colors to trace dust emission. F1800W-detected LRDs have a median [1$-$3 $μ$m]$=1.5$ mag, with a broad scatter indicative of diverse dust emission properties. About 20% exhibit [1$-$3 $μ$m]$<1$ mag colors consistent with negligible dust emission, but the majority show significant dust emission at 3 $μ$m (f$^{\rm dust}_{3μm}\lesssim0.8$) from the galaxy ISM or a hot-dust-deficient AGN torus. A correlation between bluer UV-to-NIR colors and stronger IR emission suggests that the bluest LRDs may resemble unobscured QSOs. We report a LRD at $z_{\rm spec}=3.1386$, detected in MIRI, Spitzer/MIPS, and Herschel/PACS. Its IR SED rises steeply at $λ_{\rm rest}>6~μ$m and peaks near $\sim40~μ$m, providing the first direct evidence of warm dust emission (T$=50-100$ K) in a LRD.

Figures (13)

• Figure 1: NIRCam color-color diagrams illustrating photometric selection methods for LRDs and their typical SEDs. Left: Best-fit SED models for 5 LRDs (solid lines) drawn from pg24a and wang24_lrd, covering a representative range of UV-to-NIR colors indicated in the text. The black and green dashed lines show typical SEDs for a young, low-extinction star-forming galaxy with extreme emission lines (EW$\sim$1000Å) and a type I QSO, respectively. The next panels display the color-redshift tracks of these models ranging from $z=3$ to $z=9$, as indicated by squares of increasing size. Center: Color-color selection diagram. The dark and light grey regions indicate the selection thresholds for LRDs used in barro24 and pg24a. The density plot shows the bulk of the galaxy sample color-coded by stellar mass. The color-redshift tracks suggest that this method tends to miss LRDs at low redshift ($z\lesssim5$) and those with bluer UV-to-NIR colors ([0.25$-$1 $\mu$m] $<3$ mag). Right: New color-color selection for LRDs. This method provides a more complete sample by selecting candidates within the region that encloses the color tracks of the representative LRD models from $z=3$ to $z=9$. The grey shaded area indicates the threshold used to reject brown dwarf contaminants.
• Figure 2: The flux ratio within apertures of radius r = 0.5" and 0.2" in F444W for the full photometric sample in PRIMER and JADES is shown by black dots. The unresolved stellar locus appears as a straight line at F$_{\rm F444W}$(0.5")/F$_{\rm F444W}$(0.2") $\sim1.2$. The black line marks the compactness threshold used to identify LRDs (red circles).
• Figure 3: Left: New color-color selection diagram (as in the right panel of Figure \ref{['fig:selection_criteria']}) comparing to other LRD samples from the literature in the same fields. The grey circles indicate our sample. The purple stars, blue circles, and green squares show the LRDs identified only in KOV24, KOI24, and those in common between the two. The histograms on the top and side indicate the overall distribution of the different subsets, as well as their median and percentiles. Right: F277W$-$F444W color vs. F444W magnitude for the same LRDs in the left panel. The black line indicates the LRD selection threshold of pg24a. Despite having a similar number of LRDs, the overlap between KOV24 and KOI24 is only $\sim$50%. The agreement is better for red and bright LRDs, while KOV24 tends to miss some of the red but faint, and KOI24 misses preferentially blue LRDs with F277W$-$F444W$<1$. The new method identifies nearly all the LRDs in the previous samples and an additional 40%, with the new LRDs being preferentially blue and faint.
• Figure 4: Same color-color and color-magnitude diagrams as in Figure \ref{['fig:selection_othersamples']}, but color-coded based on detection in different MIRI bands. Orange, blue, and grey circles represent LRDs detected in F1800W, F770W, and those not detected in MIRI, respectively. The majority of the LRD sample is detected in F770W, with the detection fraction primarily dependent on the F444W magnitude, ranging from 77% to 94% for F444W$<27$ mag and F444W$<26$ mag, respectively. In contrast, the detection fraction in F1800W is only $\sim7$% (17), predominantly concentrated among the brightest LRDs (F444W$\lesssim24$ mag). However, there is a weak trend of F1800W detections extending to fainter LRDs (F444W$>25$ mag) with bluer F277W$-$F444W colors.
• Figure 5: F1800W$-$F770W color vs. F444W magnitude for MIRI-detected LRDs. Orange circles represent F1800W detections, and blue triangles indicate upper limits for LRDs detected only in F770W. The blue lines show the running median and percentiles for the upper limits. Histograms on the right show the cumulative distribution of the color and upper limits. Shaded grey areas depict the typical color of a Type I QSO with high obscuration at different redshifts. The median color of F1800W detections, F770W$-$F1800W$=1.4^{+0.4}_{-0.6}$ mag, is much bluer than a dust-obscured AGN, with only two LRDs showing such IR red colors. Similarly, about 75% of F1800W non-detections have F770W$-$F1800W$<2.7$ mag, bluer than an obscured QSO at $z\gtrsim4$.