Weekly issue

Week 9, 2026

Feb 23 – Mar 1, 2026

Week 9, 2026 includes 5 curated papers, centered on LRD, high-z, broad Balmer.

2602.23310v1

Extreme Emission Line Galaxies in CEERS Are Powered by Star Formation, not AGN

Kelcey Davis, Madisyn Brooks, Jonathan R. Trump, Vital Fernández, Taylor A. Hutchison, Rebecca L. Larson, Anthony J. Taylor, Elizabeth J. McGrath, Guillermo Barro, Anton M. Koekemoer, Pablo Arrabal Haro, Mark Dickinson, Bren E. Backhaus, Nikko J. Cleri, Steven L. Finkelstein, Ananya Ganapathy, Raymond C. Simons, Ricardo O. Amorín, Alexander de la Vega, Norman A. Grogin, Michaela Hirschmann, Weida Hu, Jarrett L. Johnson, Jeyhan S. Kartaltepe, Dale Kocevski, Mario Llerena, Ray A. Lucas, Madeline A. Marshall, Fabio Pacucci, Laura Pentericci, Phoebe R. Upton Sanderbeck

Theme match 5/5

Digest

JWST/NIRSpec spectra from CEERS, CAPERS, RUBIES, and deep THRILS observations vet a photometric sample of extreme emission-line galaxies (EWobs >5000 Å in Hα or Hβ+[O III]) in CEERS and test a Little Red Dot (LRD) color–color cut. Six new broad-line AGN are found at 3.5<z<7 in ~8 hr G395M THRILS data, yet EELGs with and without broad lines have similar optical line ratios and [O III] emission becomes more compact with increasing EW. Photometric and spectroscopic EWs generally agree within a factor of ~3, but AGN often have inflated photometric EWs from blue UV/red optical continua, with only ~10% (rising to ~35% with deeper spectra) showing broad Balmer lines. When AGN are present the narrow Hα dominates—especially at the largest EWs—implying the extreme emission is powered primarily by young star formation, not the AGN.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Figure 1: Inspect how the EW>5000 Å selection maps to spectroscopic EWs; note which photometric EELGs fall below threshold after spectroscopy and that AGN are over-represented among the discrepant high-EW photometric outliers.
Figure 2: Check zphot–zspec systematics and the specific confusion tracks (Hα↔[O III], Lyman vs Balmer breaks); then compare EWphot vs EWspect to see the ~×3 agreement and identify BL AGN as EW-inflated outliers.
Figure 3: Contrast shallow RUBIES with deep THRILS for the same source; the deep exposure reveals broad Balmer plus He I λ5876, λ7065, illustrating why the BL-AGN fraction rises to ~35% with deeper/grating data.
Figure 4: Read the broad-component fraction vs EW; at the highest Hα or Hβ+[O III] EWs the narrow component dominates, directly supporting a star-formation origin for the extreme lines even in AGN hosts.

Tags

2602.22386v1

Little Red Dots as Obscured Little Blue Dots: A Super-Eddington Unification Model

Piero Madau, Roberto Maiolino

Theme match 5/5

Digest

Tests an orientation-based unification where Little Red Dots are dust-reddened, high-inclination counterparts of Little Blue Dots powered by super-Eddington accretion, using a thick, radiation-pressure supported flow with an anisotropic blue SED, an equatorial BLR, and a modest-covering dusty screen. Inclination-dependent SEDs fed to Cloudy reproduce the extreme broad Hα EWs with only ~15% global BLR covering factor and, with a gray AV=2.8 along LRD sightlines, match the V-shaped UV–optical continua, large Balmer decrements, and strong Balmer breaks only in the most obscured views. Weak high-ionization lines arise from orientation-dependent suppression of the XUV/soft X-ray continuum toward equatorial directions, removing the need for a fully enclosing cocoon; a compact equatorial dust component reradiates only a small fraction of Lbol (modest hot-dust bump; far-IR/sub-mm limits imply small dust masses). The model unifies LRD and LBD observables via orientation and predicts linked trends among Hα EW, Balmer decrement/break, high-ionization line strength, and IR output—offering concrete, testable diagnostics for early black-hole growth.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Figure 1: Check how ionizing photon rates above H I/He I/He II/N IV/Ne IV thresholds drop by 2–3 dex from face-on to edge-on, quantifying the self-shadowing that weakens high-ionization lines for LRD-like inclinations.
Figure 2: Inspect the 3D thick-torus and funnel geometry to see how an equatorial BLR plus a circumnuclear dusty screen intercepts polar-beamed radiation, linking inclination to continuum dimming and BLR illumination.
Figure 3: Read the Hα EW PDFs versus ionization parameter to see that matching the observed median requires only CF≈0.15–0.19 and that the high-EW tail maps to i≳70°, consistent with stacked LRD measurements.
Figure 4: Compare Hα and Hβ EW versus inclination to verify that large Balmer EWs favor i≈60°–80°, aligning the LRD selection with geometry and setting expectations for Balmer decrement/break trends.

Tags

2602.21002v1

Constraints on dynamically-formed massive black holes in Little Red Dots from X-ray non-detections

M. Liempi, D. R. G. Schleicher, M. A. Latif, R. Schneider, F. Flammini Dotti, A. Escala, M. C. Vergara

Theme match 4/5

Digest

Using a stack of 55 X-ray–non-detected Little Red Dots from JADES and NGDEEP, the authors model collision-driven black-hole seeding and 0.3–7 keV emission against the Chandra upper limits. They find that compact mass–radius scalings with exponents >0.55 (e.g., R_gal ∝ M_gal^0.6) naturally yield massive seeds and are compatible with the stacked non-detections under specific accretion/obscuration combinations. High Eddington ratios remain viable only with large columns and/or enhanced metallicity in the shielding gas, while moderate sub-Eddington growth also reconciles seed and final masses with the observed X-ray weakness. The upshot is that LRDs are prime birthplaces for massive BHs and, even if initially starbursts, should evolve into AGN.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Sample definition and redshift–mass panel: compare the 55 stacked sources to the 341-object parent LRD catalog; check how M* was inferred from M_UV and how the JADES/NGDEEP subsample spans z and luminosity.
Galaxy mass–radius scaling constraints: plot of the assumed R_gal–M_gal relation; inspect which exponents (>0.55) keep collision-driven seeds consistent with X-ray limits and how R ∝ M^0.6 changes predicted seed masses.
Predicted L_X versus stacked upper limits: curves for different accretion rates and duty cycles against the soft/hard-band non-detections; identify regions excluded by the stack.
Allowed parameter space in (λ_Edd, N_H, Z): map showing that high Eddington ratios require larger columns and/or higher metallicity, while sub-Eddington tracks satisfy limits with milder obscuration.
Seed-to-final BH mass evolution: trajectories demonstrating how collision-born seeds grow under constant or rising SFR scenarios and remain X-ray weak while reaching observed masses.

Tags

2602.20247v1

Little Red Dots: One Photometric Tag Concealing Diverse Spectroscopic Flavors of Massive Star Formation and Black Hole Activity

Pablo G. Pérez-González, Guillermo Barro, Stefano Carniani, Francesco D'Eugenio, George H. Rieke, Roberta Tripodi, Andrew J. Bunker, Xihan Ji, Rui Marques-Chaves, Daniel Schaerer, Giacomo Venturi, Flor Arévalo-González, Santiago Arribas, Pierluigi Rinaldi, Bruno Rodríguez Del Pino, Joris Witstok, Rachana Bhatawdekar, Leindert A. Boogaard, Stephane Charlot, Jacopo Chevallard, Luca Costantin, Mirko Curti, Emma Curtis-Lake, Emanuele Daddi, Kelcey Davis, Mark Dickinson, Callum T. Donnan, Fergus R. Donnan, James S. Dunlop, Daniel J. Eisenstein, Henry C. Ferguson, Román Fernández Aranda, Steven L. Finkelstein, Seiji Fujimoto, Giovanni Gandolfi, Mauro Giavalisco, Norman A. Grogin, Mahmoud Hamed, Michaela Hirschmann, Jeyhan S. Kartaltepe, Dale D. Kocevski, Anton M. Koekemoer, Gene C. K. Leung, Cristina M. Lofaro, Ray A. Lucas, Derek J. McLeod, Jens Melinder, Goran Östlin, Casey Papovich, Laura Pentericci, Borja Pérez-Díaz, Marcia Rieke, Jan Scholtz, Rachel S. Somerville, Thomas M. Stanton, Struan D. Stevenson, Irene Shivaei, Sandro Tacchella, Jonathan R. Trump, Hannah Übler, Xin Wang, Christina C. Williams, Christopher N. A. Willmer, L. Y. Aaron Yung, Yongda Zhu

Theme match 4/5

Digest

JWST/NIRSpec+MIRI spectroscopy for 249 Little Red Dots (2.3<z<9.3) yields a median 0.09–1.2 μm stack (SED to 4 μm) and subtype stacks showing wide diversity in continua and lines. The stacks point to compact systems hosting 10^6.0–10^6.5 M⊙ black holes embedded in dense, warm gas, with ubiquitous UV/optical Fe II indicating BLR sightlines and high, sub-Eddington accretion (λ_Edd≈0.6). Far-UV light can be ≈80% stellar in the bluest LRDs, with possible WR features (He II 4687, N lines) implying a 3–7 Myr nuclear starburst alongside strong Balmer breaks and atypical Balmer/Paschen/[O III]/He I ratios plus a ~4550 Å absorption. The authors interpret LRD “flavors” as the luminosity-weighted mix of a short (≲20 Myr; extreme 3–7 Myr) near-Eddington BH phase and a coeval compact starburst, giving BH-to-stellar mass ratios of 1–2% and a framework for early BH growth.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Figure 1: Inspect where the spectroscopic LRDs fall in the Barro et al. color–color and color–magnitude cuts and compare histograms (F444W, F115W–F200W, F200W–F444W, F277W–F444W) to judge how representative the prism subsample is relative to the full NIRCam-selected LRD set.
Figure 2: Compare redshift distributions of the spectroscopic versus photometric LRD samples and across optical-to-UV subtypes; check median z shifts with subtype to see whether the “bluest” versus “reddest” LRDs occupy different epochs.
Figure 3: Use the full-stack spectrum (with MIRI photometry) and zoom-ins to verify ubiquitous Fe II complexes, Balmer break strength, and the unusual Balmer/Paschen/[O III]/He I ratios; check the ~4550 Å absorption and how LRD stacks compare to QSO and LAE templates across Mg II, [O III]+Hβ, Hα, and NIR windows.
Figure 4: Read the placement of the median and subtype stacks in the C III] EW versus C III]/He II plane against AGN- and SF-dominated model loci to see which LRD flavors sit in the AGN, hybrid, or SF regimes and how optical-to-UV ratio tracks this shift.

Tags

2602.22305v1

You can't see me: super-Eddington growth hindering X-ray detection in high-z broad-line AGNs

Alessandro Trinca, Alessandro Lupi, Francesco Haardt, Piero Madau

Theme match 3/5

Digest

Jointly modeling broad-line widths with deep Chandra non-detections, this work reinterprets JWST-selected high-z broad-line AGNs using a self-shadowed, super-Eddington funnel plus slim-disc Comptonization framework in a full MCMC. Applied to 14 sources from Lupi (2024b) and Maiolino (2025), the X-ray weakness—extreme bolometric corrections, suppressed 2–10 keV, and 0.5–5 keV non-detections—emerges naturally when the corona is confined and radiatively over-cooled inside a narrow funnel. The posteriors are strongly bimodal but favor low-mass (~10^6–10^7 Msun), highly super-Eddington solutions over very massive, low-Eddington ones, yielding masses consistent with or below local MBH–Mstar. Predicted SEDs are redder with reduced hard-X-ray output and larger bolometric corrections, arguing that many high-z broad-line AGNs are rapidly growing super-Eddington black holes rather than overmassive outliers.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Fig. 1 (kTe and Γ vs funnel half-opening angle): shows how narrowing the super-Eddington funnel steepens the X-ray slope and cools the corona—the key lever producing X-ray weakness.
Predicted L2–10 keV and observed-frame 0.5–5 keV versus Chandra limits for the 14 objects: verify that model luminosities/fluxes fall below the upper limits and how this varies with redshift and funnel angle.
Corner/posterior plots for representative sources: inspect the bimodality (high-M/low-fEdd vs low-M/high-fEdd) and how adding X-ray non-detections to broad-line constraints drives the preference for super-Eddington, low-mass BHs.
MBH versus Mstar comparison: check that re-inferred masses align with or fall below the local MBH–Mstar relation, mitigating extreme BH/host ratios from single-epoch virials.
Grid of intrinsic SEDs from the agnslim library with funnel-dependent Comptonization: see the redder continua, reduced hard-X-ray output, and the implied large bolometric corrections relative to local AGN templates.