Euclid: Scaled-up little red dots and other sources with v-shaped spectral energy distributions at z>4

Euclid Collaboration

First listed 2026-04-17 | Last updated 2026-04-17

Abstract

Little Red Dots (LRDs) are some the most intriguing galaxy populations recently identified at z>~4 with JWST. They constitute the most extreme class of a more abundant population of sources with `V-shaped' spectral energy distributions (SEDs) and compact morphologies, which includes also Little Blue Dots (LBDs). Finding brighter analogues to these sources requires surveying sky areas which are significantly larger than those covered with JWST. Euclid deep images are ideally suited for this purpose. We make use of Euclid near-infrared images, complemented by Spitzer Infrared Array Camera (IRAC) data, over 0.75 sq. deg. of the COSMOS field to select a sample of 233 sources with `V-shaped' SEDs at z>4. Out of those, we identify 16 sources with compactness >1sigma above the median of all z>4 galaxies, which we consider robust LRD/LBD candidates in our sample. The stellar masses of these 16 sources are in the range 10^{8.5} - 10^{10.5} Msun, so they are significantly more massive than typical JWST-selected LRDs/LBDs. Interestingly, half of them are about as old as the Universe at their redshifts. In addition, we find that the median photometric properties of the Euclid LRDs/LBDs are similar to those of the so-called Blue Dust-Obscured Galaxies (Blue DOGs). Less than 10% of all our `V-shaped' SED sources, including only one of the Euclid LBDs, correspond to known AGN. The latter mostly constitute a population disjoint to the `V-shaped' SED sources. Spectroscopic follow up of the Euclid LRDs/LBD candidates remains necessary to probe whether they host BLAGN as fainter analogues do and whether constitute a transition phase from these fainter sources to standard AGN.

Short digest

This paper extends the LRD discussion from JWST to Euclid by identifying compact high-redshift sources with similarly V-shaped spectral energy distributions. The main result is that the Euclid sample contains a small set of robust LRD/LBD candidates whose photometric properties resemble the more extreme JWST little red dots while also connecting to a broader dusty population. The paper matters because it pushes the question from a handful of JWST discoveries toward a survey-scale demographic context.

Key figures to inspect

• Figure 1 is the sanity-check figure: it compares photometric and spectroscopic redshifts, so read this first if you want to know how robust the sample is before worrying about the LRD interpretation.
• Figure 2 is where to see what the selected sources actually look like in SED space, including the representative V-shaped continua and the photometric bands that drive the cuts.
• Figure 3 is the must-see selection figure: it maps the continuum-slope plane and shows exactly how the 'Cut 1' and 'Cut 2' populations are carved out from the parent sample.
• Figure 4 is the quickest demographic check, because it shows whether the selected Euclid candidates really sit in the intended high-redshift regime.