An Investigation Into The Selection and Colors of Little Red Dots and Active Galactic Nuclei

TL;DR

This study investigates the nature and selection of Little Red Dots (LRDs) and their connection to AGN activity in the early universe using uniform JWST photometry and spectroscopy across multiple deep fields. It compiles LRDs from existing catalogs and combines them with broad-line (Type I) and obscured (Type II) AGN samples identified in JWST surveys, enabling a comprehensive comparison of rest-frame UV/optical colors and slopes. Key results show obscured AGNs at z<6 have colors dissimilar to LRDs, while unobscured AGNs at z<6 span a wide range with only a subset matching LRDs; at z>6, many unobscured AGNs resemble LRDs, potentially linked to higher bolometric luminosities. The work highlights that LRDs occupy a unique color-space that makes them easier to target, but also reveals biases in pre-selection that undercount many broad-line AGNs and that LRD selection includes sources affected by emission-line flux boosts, underscoring the need to consider selection effects when inferring high-z AGN demographics.

Abstract

Recently, a large number of compact sources at $z > 4$ with blue UV slopes and extremely red rest-frame optical slopes have been found in James Webb Space Telescope (JWST) extragalactic surveys. As a subsample of these sources, commonly called ``little red dots'' (LRDs), have been spectroscopically observed to host a broad-line active galactic nucleus (AGN), they have been the focus of multiple recent studies in an attempt to understand the origin of their UV and optical emission. Here, we assemble a sample of 123 LRDs from the literature along with spectroscopic and photometric JWST-identified samples of AGNs to compare their colors and spectral slopes. We find that while obscured AGNs at $z < 6$ have highly dissimilar colors to LRDs, unobscured AGNs at $z < 6$ span a wide range of colors, with only a subsample showing colors similar to LRDs. At $z > 6$, the majority of the unobscured AGNs that have been found in these samples are LRDs, but this may be related to the fact that these sources are at large bolometric luminosities. Because LRDs occupy a unique position in galaxy color space, they are more straightforward to target, and the large number of broad-line AGNs that do not have LRD colors and slopes are therefore underrepresented in many spectroscopic surveys because they are more difficult to pre-select. Current LRD selection techniques return a large and disparate population, including many sources having $2-5μ$m colors impacted by emission line flux boosting in individual filters.

Figures (1)

• Figure 1: Photometric and spectroscopic redshift stacked histogram for the assembled AGN and LRD samples. Each bar in the stacked histogram shows all of the sources that went into that bin, without any overlapping between the different samples. For the LRD sources, we plot the JADES+NGDEEP+CEERS LRD photometric and spectroscopic redshifts from kocevski2024 with purple and the JADES LRDs from perezgonzalez2024 with pink. For the broad-line sources, we plot the spectroscopic redshifts for Type I AGNs in JADES from maiolino2023 and Juodžbalis et al. (in prep) in red, the spectroscopic redshifts for the broad-line AGNs found using FRESCO in matthee2024 in maroon, the spectroscopic redshifts for the broad-line AGNs found in CEERS in harikane2023 in lime green, and the Type I AGNs from sun2024 in slate grey. For obscured sources, we plot the spectroscopic redshifts for Type II AGNs in JADES from scholtz2023 in dark blue, and the photometric and spectroscopic redshifts for SED-derived AGNs at $z > 2$ across JADES/SMILES in lyu2024 in light blue. In these plots, objects are only counted once, even if they are contained in more than one sample.