MEOW: The increase in the obscured AGN fraction in mid-infrared from 0 < z < 6 with JWST MIRI

abstract · abstract

Obscured active galactic nuclei (AGN) are often invoked to explain the rapid emergence of young quasars at high redshift and are crucial for building a complete census of AGN activity and black hole growth. The advent of the James Webb Space Telescope (JWST) extends the discovery space for obscured AGN into the mid-infrared (mid-IR) with unprecedented precision through reprocessed dust emission. In this work, we use deep JWST Mid-Infrared Instrument (MIRI) imaging from the MIRI Early Obscured AGN Wide Survey (MEOW), together with existing JWST Near Infrared Camera (NIRCam), spectroscopic, and Hubble Space Telescope imaging data, to identify a previously unrecognized population of obscured AGN out to z ~ 6. Using spectral energy distribution (SED) modeling of the MIRI-detected sources, we identify 883 AGN over an area of ~ 131 arcmin2 and construct the AGN bolometric luminosity function, including both obscured and unobscured sources, across five redshift bins. We find an excess in AGN abundance relative to UV-selected AGN luminosity functions, indicating a substantial obscured population missed by optical/UV surveys, with the inferred obscured fraction increasing with redshift and reaching ~ 98-99% in our highest-redshift bin, 4.5 < z < 6. We also find higher AGN abundances and obscured fractions than X-ray-based studies, consistent with a previously unrecognized population of heavily obscured, Compton-thick AGN revealed by mid-IR selection. These results suggest that a large fraction of supermassive black hole growth at early times occurs during heavily obscured phases largely inaccessible at other wavelengths.

S3.SS2: III.2 SED fitting / block 1 · html

Based on the CIGALE SED fitting results, we further identify and characterize the parent AGN sample in the following subsection. We additionally visually inspected all sources and removed spurious objects (about ≈ 1 % \approx 1\% of the total objects), such as saturated stars, also confirmed by the high χ 2 \chi^{2} reported in CIGALE .

S6.SS3: VI.3 Implications for SMBH growth / block 1 · html

The results of our study point towards high obscured fractions, as high as f obsc ≈ 98 − 99 % f_{\mathrm{obsc}}\approx 98-99\% in the highest redshift bin. Taken at face value, this could alleviate some of the tension raised by the short UV-luminous duty cycles of order ∼ 0.1 − 1 % \sim 0.1-1\% reported by recent studies at z ≳ 6 z\gtrsim 6 [ 36 , 94 , 103 , 53 ] , as it would imply that black hole accretion timescales are nearly two orders of magnitude larger.

S6.SS3: VI.3 Implications for SMBH growth / block 1 · html

Thus, our results support a picture in which a large fraction of luminous high-redshift AGN exist in highly dust enshrouded environments. This finding predicts that the short UV-bright activity inferred independently from quasar ionizing signatures, and population duty cycles can be reconciled by the fact that most of the growth ( ≈ 98 − 99 % \approx 98-99\% ) occurs in highly obscured phases that suppress the UV radiation from the quasars.

S7: VII Conclusions / block 1 · html

Our results reveal a substantial population of dust-enshrouded AGN at early cosmic times. In particular, at high redshift ( z ≳ 4.5 z\gtrsim 4.5 ) and high luminosity ( log ⁡ L bol ≳ 46 ​ erg ​ s − 1 \log L_{\mathrm{bol}}\gtrsim 46\,\mathrm{erg\,s^{-1}} ), the estimated obscured fraction approaches ≈ 98 − 99 % \approx 98-99\% . While the detailed luminosity dependence of the obscured fraction remains sensitive to the choice of type-1 QLF at the faint end, the high-luminosity regime is comparatively well constrained, suggesting that the black hole accretion timescales could be up to two orders of magnitude larger than previous estimates based on proximity zones and clustering of UV-luminous quasars. Future measurements of quasar lifetimes from damping-wing signatures in faint quasars (JWST#9180, PI: Hennawi) and from clustering analyses (JWST#7519, PI: Arita) will be key to further constraining this interpretation.

A3: Appendix C Comparison with the AGN identified in the JWST SMILES program / block 1 · html

We show the results of the cross-match in Fig. 11 . We recover 84 counterparts out of the 217 SMILES AGN. As expected, the matched sources are those with bolometric luminosities above the F1000W and/or F2100W detection limits, while the majority of the unmatched objects fall below our sensitivity threshold. After visual inspection of all the cut-outs, we identify six exceptions in which the lack of a match is not driven by sensitivity: two sources fall outside the F444W coverage, and four are near a bright source and are not deblended by our source detection procedure. We note however that our source injection simulations account for such incompleteness.

S2.SS5: II.5 Redshift determination / block 1 · html

We adopt the CIGALE redshifts for sources without spectroscopic measurements. This choice is motivated by the code’s ability to derive redshifts consistently with the full SED fitting, naturally accounting for spectral features within the model. We emphasize, however, that our results are not sensitive to the choice between eazy and CIGALE , as the redshift discrepancy for most sources is less than the width of the redshift bins used subsequently in the luminosity function calculations (§ IV ), and the small number of outliers has no statistically significant effect on the overall population (a detailed comparison is presented in Appendix B ).

S4.SS2: IV.2 Parametric approach / block 1 · html

To ensure continuous bolometric luminosity coverage in our luminosity function, necessary for meaningful comparison with existing literature (§ V ), we also fit our results with a double power law, following standard practice:

S4.SS2: IV.2 Parametric approach / block 1 · html

Lastly, we report our results from both the parametric and the non-parametric approach in Table 2 .

S5: V Results / block 1 · html

In this section, we use the luminosity functions described in § IV to quantify the obscured AGN population missed by optical/UV studies. We first qualitatively show how our luminosity function, including both obscured and unobscured sources, compares with the type-1 UV-bright quasar luminosity function (QLF) reported in [ 63 ] . To estimate the obscured AGN fraction in each luminosity and redshift bin (§ V.2 ), we integrate our MEOW/MIR luminosity function to determine the total AGN abundance (obscured + unobscured), and independently integrate the QLF for the unobscured population (see also Fig. 1 ).

S2.SS4: II.4 Black hole feedback / block 1 · html

We implement BH feedback through thermal energy injection, resulting from BH accretion. The energy injected to the nearby cold gas at each timestep δ ​ t \delta t is determined as

S3: III Results / block 1 · html

We compare our model outputs with various LRD observations, specifically considering two classes of BH populations: heavy DCBH seeds, and light stellar-remnant seeds that are accreting at Eddington/super-Eddington rates. For the heavy seed models, we consider 4 cases - the Eddington-limited case, super-Eddington limited growth (up to 1.5 times Eddington), forced super-Eddington growth (where all DCBHs accrete at super-Eddington rates when gas is available with f edd = 1.5 f_{\rm edd}=1.5 ), and the case imposing a higher LW flux threshold for DCBH formation (up to J crit = 3000 − 10000 J_{\rm crit}=3000-10000 ). The default heavy DCBH case used for our analysis is the Eddington-limited model, unless stated otherwise. For the light seeds, we consider 2 cases - super-Eddington limited (up to 1.5 times) and forced super-Eddington growth ( f edd = 1.5 f_{\rm edd}=1.5 ). The default model for light seeds is the super-Eddington limited case unless stated otherwise.

S6: VI Summary & Conclusions / block 1 · html

We investigate the effects of cosmic variance by randomly placing MIRI-like pointings (FOV ∼ 2.3 ​ arcmin 2 \sim 2.3\,\rm arcmin^{2} ) within the 2 ​ deg 2 2\,\rm deg^{2} lightcones. We are able to reproduce the observed roughly factor of 2 field-to-field variation seen in L. Sajkov et al. [85] among individual MIRI pointings. We also quantify the fractional error σ AGN / N ¯ ​ ( > S ) AGN \sigma_{\rm AGN}/\bar{N}(>S)_{\rm AGN} for bright, rare AGN as a function of survey area. This suggests that surveys of order 30-40 arcmin 2 such as SMILES have a roughly ∼ 30 % \sim 30\% cosmic variance error in the AGN counts.

S4.SS5: IV.5 Effect of Cosmic Variance / block 1 · html

We repeat this process 5000 times, computing the median and 1 σ \sigma confidence interval of the redshift distribution across the samples. Fig 13 shows the resultant median and 1 σ \sigma spread relative to the measured redshift distribution from L. Sajkov et al. [85] . For each monte-carlo realization, we perform a two-sample Kolmogorov-Smirnov (KS) test against the observed distribution. The two distributions are qualitatively similar, and the median and 1 σ \sigma spread show good agreement. There is a spread in the KS p p values due to cosmic variance, but they suggest the simulated and observed redshift distributions are statistically distinct. Some of the differences can be attributed by 15% of the MIRI F560W sources in L. Sajkov et al. [85] not having matches in COSMOS2020. These sources would most likely boost the z ∼ 1 − 5 z\sim 1-5 portion of the observed redshift distribution, leading to closer agreement with the model. We are also affected by small number statistics where each count per bin in the observed redshift distribution has a non-negligible Poisson error.

S4.SS1: IV.1 Identified BLAGNs / block 1 · html

Based on these selection criteria, we identify 252 BLAGNs. These objects are listed in Table 1 and their Gaussian fitting results are shown in the Appendix (Figure 7 ). Among the identified BLAGNs, ∼ \sim 100 sources have been previously reported in the literature. Within our sample, 15 AGNs are identified from the G395H grating, 14 from G235H, 88 from G235M, 121 from the G395M grating, and 14 from the G140M grating. All identified BLAGNs exhibit prominent broad-line emission in the H α \alpha line or, where available, the H β \beta line. There are 5 sources that do not pass our criteria though previous studies identified them as BLAGNs. Our measurements yield their FWHM broad \text{FWHM}_{\text{broad}} ∼ \sim 800 – 900 km s -1 , which is just below our 1000 km s -1 criteria. To be consistent with the literature, we still keep them in the Table 1 .

S1: I Introduction / block 1 · html

In addition, recent findings from JWST have revealed that compared to local AGNs, high- z z AGNs exhibit higher bolometric luminosities and Eddington ratios as a function of black hole mass, suggesting higher accretion efficiencies and rapid black hole growth [ 26 , 68 , 39 , 65 , e.g.,] . These results suggest that faint AGNs may represent an early evolutionary phase in black hole growth, and that SMBHs in the early universe grow faster than their host galaxies, enriching the universe with higher metallicity [ 68 , e.g.,] . However, the majority of these conclusions are based on single or small sample high-redshift AGNs [ 65 , e.g.,] . Therefore, their conclusions could be influenced by selection bias, as AGNs with higher accretion rates are intrinsically brighter and more easily detected, potentially leading to an overgeneralization of the accretion rates to all high-redshift AGNs.

abstract / block 1 · abstract

We investigate the capabilities of the mid-infrared instrument (MIRI) of James Webb Space Telescope (JWST) to advance our knowledge of AGN dust using the spectral fitting technique on an AGN collection of 21 nearby (z<0.05) AGN (7 type-1 and 14 type-2) observations obtained with the medium resolution spectroscopy (MRS) mode. This collection includes publicly available AGN and data from the collaboration of Galactic Activity, Torus, and Outflow Survey (GATOS). We developed a tool named MRSPSFisol that decomposes MRS cubes into point-like and extended contributions. We found statistically good fits for 12 targets with current AGN dust models. The model that provides good fits (chi2/dof<2) for {these 12 targets} assumes a combination of clumpy and smooth distribution of dust in a flare-disk geometry where the dust grain size is a free parameter. Still, two and one AGN statistically prefer the disk+wind and the classical clumpy torus model, respectively. However, the currently available models fail to reproduce 40% of the targets, likely due to the extreme silicate features not well reproduced by the models and signatures of water-ice and aliphatic hydrocarbon absorption features in most targets. New models exploring, for instance, new chemistry, are needed to explain the complexity of AGN dust continuum emission observed by JWST.

S5.SS3: V.3 Distribution functions / block 1 · html

Figure 9 shows the fitted distributions. For comparison, we also calculated the 1 / V max 1/V_{\mathrm{max}} distributions in bins of 0.5 0.5 dex. Overall, our results exhibit broad agreement in comparison to previous measurements in the literature. We wish to highlight that these astrophysical quantities have been inferred simultaneously with ProSpect , and that we have endeavored to obtain a representative census of M ⋆ \mathrm{M_{\star}} , SFR and L AGN \mathrm{L_{AGN}} . Subtle departures are most likely attributed to differences in methodologies as elucidated below.

S5.SS3: V.3 Distribution functions / block 1 · html

Shark , however, employs the P-Millenium dark-matter only simulation [ 11 ] as the basis for the semi-analytic model to achieve better resolution of low mass galaxies and smaller time increments between snapshots compared to the first version presented in Lagos et al. [75] . Our results for the SMF and SFRF at all redshifts are generally in agreement with Shark for the domain of M ⋆ \mathrm{M_{\star}} and SFR that we probe. The AGNLF in Shark is in general ≈ 1 \approx 1 dex higher in normalization with a steeper bright-end slope compared to our results in all redshift bins.

abstract · abstract

Characterizing the growth of supermassive Black Holes (SMBHs) is critical to the evolution of galaxies, however the majority of this activity is obscured, rendering traditional tracers of active SMBHs, such as in the restframe optical/UV, ineffective. The mid-infrared has been particularly successful in revealing obscured AGN activity however much of this work is confined to the local universe due to the lack of a far-IR telescope with the required sensitivity and wavelength coverage. In this work we demonstrate the effectiveness of PRIMA (PRobe far-Infrared Mission for Astrophysics), a concept 1.8m far-IR observatory, to detect and characterize deeply obscured galaxy nuclei over cosmic time. With the PRIMAger instrument covering 25 - 235 $μ$m, we find that we can accurately detect obscured nuclei via the deep silicate absorption at restframe $9.8 μ$m between $z=2-7$. Additionally, the FIRESS spectrograph can produce R$\sim$100 spectra of obscured nuclei out to $z\sim7$, detecting Polycyclic Aromatic Hydrocarbons (PAHs), ices, ionized and molecular gas. With the large number of deeply obscured nuclei PRIMA can detect and characterize, such a mission is critical to understanding the growth of SMBHs.

S1: 1 Introduction / block 1 · html

López-Corredoira et al. (2024) (hereafter Paper I) estimated the average age of the oldest stellar population in these 13 galaxies to be on average larger than 900 Myr within 95% CL, a result which is in tension with the age of the Universe in the standard model; similar tension was also obtained by Martínez-García (2025) with similar data and technique. For this estimation in Paper I, a fit of Spectral energy distribution (SED) obtained from photometry in 10 filters at visible and near infrared was carried out assuming a combination of two Single stellar populations (SSPs) with the same extinction; a lower extinction for the younger population would lack physical justification, and considering the presence of the Ly- α \alpha break this extinction should be low, avoiding degeneracy age-dust at longer wavelengths ( López-Corredoira et al. 2017 ; López-Corredoira et al. 2024 ; Gao, López-Corredoira & Wei 2024 ) . Extra stellar populations did not significantly improve the fits. Also, the inclusion of an active galactic nucleus (AGN) component was investigated by Paper I(Sect. 3.4), assuming an independent AGN extinction with A V ≤ 3 A_{V}\leq 3 magnitudes and it was concluded that, although a minor component due to AGN is possible on the average SED, its inclusion in the fit does not significantly change the range of the age of the oldest stellar population necessary to produce the fit. However, some possibilities of interstellar extinction within A V ≤ 3 A_{V}\leq 3 and the effect of the intergalactic extinction at reionization epoch was not explored.

S5.SS1: 5.1 SED fits with two single stellar populations and one AGN / block 1 · html

Table 4: Best-fitting results with two SSPs. Errors represent the limits within 95% CL ( ≡ 2 ​ σ \equiv 2\sigma ) (within the resolution of the templates), with maximum likelihood method (Paper I, Appendix A) for a ​ g ​ e old age_{\rm old} and Avni method (Paper I, Sect. 3.1) for the rest of the parameters. F277W-F444W is the color in AB-magnitudes. The ages are expressed in Gyr. Redshifts without error bars are spectroscopic redshifts, obtained from: ( Kocevski et al. 2023 ) : [Koc23]; ( Fujimoto et al. 2023 ) : [Fuj23]; W24; https://dawn-cph.github.io/dja/spectroscopy/nirspec/ : [S]; https://archive.stsci.edu/hlsp/jades : [J]. Stacked SED: [SA]: stacked of 31 galaxies (4th column of Table 5 ), 16 bins; Stacked SED: [SAB]: S13 except the points at λ = 1000 \lambda=1000 Å λ = 7499 \lambda=7499 Å 14 bins; [S13]: stacked of 13 galaxies with spectroscopic redshift (5th column of Table 5 ), 16 bins; [S13B]: 14 bins.

abstract · abstract

The advent of the James Webb Space Telescope has revolutionised our understanding of the high-redshift Universe through its detection of bright, massive galaxies up to $z\gtrsim 10$ and its identification of peculiar sources called `little red dots' (LRDs). The origin of both classes of objects remains uncertain but is likely linked to the formation and early growth of the first massive black holes (MBHs), which may be more easily explained by invoking phases of super-Eddington accretion. In this study, we used a state-of-the-art zoom-in cosmological simulation of a quasar host to investigate whether these objects could resemble any of the peculiar sources observed with JWST during their assembly. We find that the impact of MBH feedback on star formation is typically moderate, with outflows preferentially escaping perpendicular to the galactic disc. However, for approximately ten percent of the galaxy's lifetime, the system enters a distinct quenched phase following rapid MBH growth driven by super-Eddington accretion. This phase culminates in a powerful feedback event, during which the MBH jet and disc-driven winds interact directly with the galactic disc and carve out a central cavity. We also find that, during the history of the quasar host progenitor, the spectral properties of the system can resemble both LRDs and quenched galaxies, depending on the specific evolutionary stage considered. These findings suggest that both conditions may represent transient phases in the life cycle of high-redshift galaxies.