MEOW: JWST/MIRI Unveils a Hidden Population of Ancient Black Holes
The MEOW survey on JWST with the MIRI camera covered 95 square arcminutes in the GOODS-N and GOODS-S fields in the F1000W and F2100W filters down to depths of 0.5 and 3.6 μJy (5σ). Modeling the spectral energy distribution using data from HST, NIRCam, and SCUBA-2 revealed 16 AGN at z=4.5–7.2 (12 spectroscopically confirmed) with bolometric luminosities of 10^44.6–10^46.4 erg/s. Twelve of them are newly discovered, including at least five narrow-line AGN — representatives of the dust-hidden population invisible in the optical range. Two broad-line AGN showed different mid-infrared properties: one is a 'little red dot' (LRD), the other is either a regular AGN or an LRD with bright hot dust. The bolometric luminosity function of obscured AGN at z=4.5–6 is comparable to those of unobscured broad-line AGN and LRDs, indicating that dusty black holes contribute significantly to overall growth. Narrow-line AGN reside in diverse galaxies, and absorption is detected both in the nucleus and the host system — a sign of multiple shielding mechanisms. Thus, MIRI/JWST proves to be a critical link in the multifaceted approach to a complete census of early supermassive black hole activity.
Context
Modern astrophysics faces a paradox: James Webb and other telescopes spot hundreds of quasars hosting black holes of a billion solar masses just 700 million years after the Big Bang. To grow so fast, these monsters must have gobbled up matter at breakneck speed, yet observations show their active phase is brief—only a few million years. This tension can be resolved if much of the growth is hidden from us by dust clouds. That's exactly the "shadow" population the MEOW survey sought, tapping MIRI's unique sensitivity on JWST to the thermal glow of dust. According to Edwin Hubble's law, distant galaxies race away from us, stretching light into the red, and a key sign of activity are broad hydrogen emission lines like Hα, discovered by Johann Balmer.
Methods
The researchers used 43 MIRI pointings for the MEOW survey, covering 95 square arcminutes in the GOODS-N and GOODS-S fields. The F1000W and F2100W filters (at 10 and 21 microns) capture the thermal glow of carbon-based dust heated to about 1200 K by a black hole's accretion disk. Cross-matching with archival Hubble and NIRCam data enabled the construction of spectral energy distributions (SEDs) for thousands of galaxies. Then modeling with the CIGALE code teased out the AGN contribution. By combining deep spectroscopy from NIRCam and NIRSpec, the team confirmed redshifts for most objects, including Hα and [OIII] lines.
Results
In the final sample of 16 AGN, 12 turned out to be new, missed by previous surveys. Five of the seven objects with available Hα line spectra showed no broad emission lines, meaning they are narrow-line type II AGN—classic "obscured" sources. The other two did show broad lines, but with different infrared emissions: one matches a "little red dot" (LRD), the other a typical quasar with a powerful hot dust component. Bolometric luminosities range from 10^{44.6} to 10^{46.4} erg/s. The luminosity function at z=4.5–6 shows that the space density of such hidden AGN is comparable to that of broad-line quasars and LRDs, and at high luminosities even surpasses it.
Implications
These results mean that the census of active nuclei in the young Universe was far from complete. A significant fraction of supermassive black hole growth occurred under "dark accretion," hidden from optical telescopes. This calls for a revision of models for the co-evolution of galaxies and their central monsters. Moreover, the variety of conditions in host galaxies—from dense circumnuclear tori to galaxy-wide obscuration—points to multiple obscuration mechanisms, complicating a unified picture.
Future development
In the future, as MIRI data accumulate over larger areas, we can push into fainter luminosities and higher redshifts. This will reveal whether torus or galactic-scale dust dominates at different epochs. Combining with X-ray surveys (e.g., the future Athena telescope) and more sensitive spectroscopy will provide a complete picture of accretion across cosmic history.
Impact
The discovery will impact early-universe cosmology, the theory of galaxy formation, and accretion physics. It also underscores the importance of infrared surveys for future missions like LUVOIR or HabEx.
Next steps
Deeper spectroscopic studies with NIRSpec will help refine black hole masses and dust properties. Expanding MIRI coverage to other deep fields is also needed to rule out chance fluctuations.
Key open problems
The work ties directly to the puzzle of supermassive black hole origins: how they managed to grow to billions of solar masses so quickly after the Big Bang. Hidden accretion likely partially solves this riddle, but questions about initial seeds and the efficiency of super-Eddington growth remain.
🎯 The survey's name MEOW is an acronym that plays on a cat's meow, which the authors say evokes curiosity and night hunting, as the survey "hunts" for hidden objects in the infrared.
Key numbers
- Surveyed area: 95 arcmin^2
- F2100W sensitivity: 3.6 µJy (5σ)
- Number of detected AGN: 16
- Of which new: 12
- Redshift range: 4.5–7.2