Weekly issue

Week 33, 2026

Aug 10–16, 2026

Week 33, 2026 includes 5 curated papers, centered on spectroscopy, high-z, QSO.

2608.12487v1

Ripples in the OCEANS: Broad Line Variability of Little Red Dots

Madisyn Brooks, Kelcey Davis, Jonathan R. Trump, Raymond C. Simons, Erini Lambrides, Pablo Arrabal Haro, Bren E. Backhaus, Nikko J. Cleri, Steven L. Finkelstein, Mauro Giavalisco, Norman A. Grogin, Michaela Hirschmann, Dale D. Kocevski, Anton M. Koekemoer, Rebecca L. Larson, Ray A. Lucas, Stephen M. Wilkins, Stijn Wuyts, Jorge A. Zavala

Theme match 5/5

Digest

Brooks et al. test the AGN interpretation of little red dots through multi-epoch JWST/NIRSpec spectroscopy of six OCEANS-selected LRDs, combining new high-resolution OCEANS data with archival CEERS and RUBIES epochs. After calibrating the spectra with narrow [O III] or [Ne III] emission, they find marginal broad Hα variability in OCEANS-100424/RUBIES-42232 (27% at 2.1σ) and OCEANS-35829/RUBIES-49140, the GlimmIr (50% at 1.5σ), while the other four sources yield only upper limits and none shows continuum variability. Comparison with SDSS-RM quasars indicates that the mixture of two variable and four non-variable objects departs from typical quasar broad-line variability, but the two tentative detections nevertheless favor an accreting AGN contribution over a pure scattering origin for the broad lines. The result emphasizes that repeat spectroscopy can separate genuinely variable LRD subtypes from a population whose broad Balmer profiles may not all trace a standard exposed BLR.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Figure 1. Shows where the six targets sit within the full OCEANS color-selected LRD population and immediately identifies OCEANS-100424 and the GlimmIr as the two marginally variable sources.
Figure 4. Presents the calibrated multi-epoch Hα profiles and fitted broad components directly, making the line-flux changes and the non-detections across the six-object sample inspectable source by source.
Figure 5. Synthesizes the central measurement by placing normalized broad Hα flux against rest-frame epoch separation for all six LRDs, separating the two tentative variables from the four upper-limit cases.
Figure 6. Provides the conclusion-driving comparison to SDSS-RM quasars in matched rest-frame time bins, framing the observed combination of LRD variability measurements and limits relative to ordinary quasar behavior.

Tags

2608.10832v1

ATLAS. III. Dust Around Little Red Dots: Hydrogen Line Ratios beyond Dust-free Non-Case B Models

Tomokazu Kiyota, Masami Ouchi, Hiroto Yanagisawa, Makoto Ando, Yuichi Harikane, Yuta Kageura, Minami Nakane, Yurina Nakazato, Yoshiaki Ono, Yui Takeda

Theme match 4/5

Digest

This ATLAS installment measures broad Balmer decrements in 20 LRDs, comprising five JWST/NIRSpec high-redshift sources and 15 local analogues, with careful fits that account for absorption and blended neighboring lines. Broad Hα/Hβ spans 6–30, far above Case B; joint Balmer–Paschen Cloudy modeling shows that high density can explain one local LRD, but the other two systems with Paschen constraints require substantial additional extinction even after non-Case-B radiative-transfer effects are included. Because the narrow-line decrements do not imply comparably strong reddening, the paper places the obscuring dust near the broad-line region and argues that dust obscuration may occur in at least roughly half of LRDs, potentially as a lower-column counterpart to an AGN dusty torus.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Figure 3. This is the population-level measurement that establishes the central anomaly: LRD broad Hα/Hβ ratios reach 6–30 and several exceed the maximum attainable in the dust-free Cloudy grid. It also places the high-redshift JWST sample, local LRDs, lower limits, and comparison SDSS AGNs on the same diagnostic.
Figure 5. The Balmer–Paschen diagnostic directly breaks the dust-versus-density degeneracy. Comparing the three Paschen-measured LRDs with the dust-free Cloudy grid and reddening vector shows why non-Case-B gas alone succeeds for only one source while others require extinction.
Figure 9. This figure connects the line-ratio inference to the available infrared evidence in the three local LRDs. Its comparison of inferred minimum broad-line attenuation, narrow-line attenuation, dust-template temperatures, and line widths supports dust associated with the compact broad-line-region environment rather than galaxy-scale reddening.
Figure 10. The concluding schematic distills the paper's physical interpretation: an LRD can retain visible broad lines while line-of-sight dusty material attenuates the broad-line and continuum emission. It provides the clearest visual synthesis of the proposed lower-column-density analogue of a classical AGN torus.

Tags

2608.08369v1

GATOS: Distinct Feedback Modes in AGN Central Regions Revealed by Spatially Resolved JWST Spectroscopy

Lulu Zhang, Chris Packham, Erin K. S. Hicks, Ismael García-Bernete, Almudena Alonso-Herrero, Ric I. Davies, Martin J. Ward, Daniel E. Delaney, Takuma Izumi, Dimitra Rigopoulou, Cristina Ramos Almeida, Omaira González-Martín, Rogemar A. Riffel, Claudio Ricci, Montserrat Villar-Martín, Francoise Combes, Miguel Pereira-Santaella, Sebastian F. Hoenig, Andrew J. Bunker, Peter G. Boorman, Enrica Bellocchi, Nancy A. Levenson, Santiago García-Burillo, Fergus R. Donnan, Anelise Audibert, Lindsay Fuller, Tanio Díaz-Santos, David J. Rosario

Theme match 3/5

Digest

Using JWST/MIRI MRS spectroscopy at 4–24 pc resolution, GATOS maps PAH bands alongside H₂ and ionized-gas diagnostics across the nuclei of five AGN spanning a broad luminosity range. The study finds that nuclear PAH suppression reflects two distinct feedback channels: AGN irradiation, which heats and ionizes PAHs before preferentially destroying smaller grains, and shock processing from jets or outflows, which preferentially removes ionized PAHs. Their coexistence explains the observed bimodality in PAH band ratios and helps reconcile conflicting PAH-based AGN results, while cautioning that PAH star-formation calibrations in AGN require feedback-aware treatment; the conclusions are based on a limited sample.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Figure 1. This figure establishes the observational problem by comparing resolved 6.2, 7.7, and 11.3 μm PAH surface brightness with neon-based SFR surface density in the five AGN nuclei. It shows where the usual star-forming PAH–SFR calibration fails on parsec scales, motivating the feedback-based interpretation.
Figure 4. The central diagnostic figure links PAH 6.2/7.7 and 11.3/7.7 ratios to independent indicators of shock processing and AGN irradiation, while placing the measurements against star-forming spaxels and PAH models. It directly visualizes the two PAH-ratio regimes that underpin the paper’s claim of distinct radiative and kinetic feedback modes.
Figure 5. This alternate-diagnostic version of the PAH-ratio analysis tests the conclusion using [Fe II]/[Ar II] for shocks and [Ne V]/[Ne III] for AGN irradiation. It matters because it demonstrates that the feedback interpretation is not tied to a single choice of line-ratio proxy.
Figure 6. The schematic distills the paper’s bottom-line physical picture: irradiation leaves a population weighted toward larger, more ionized PAHs, whereas shocks preferentially destroy ionized PAHs and favor large, neutral survivors. It is the clearest synthesis of why mixed feedback modes can produce bimodal PAH ratios and complicate PAH-based SFR estimates near AGN.

Tags

2608.10070v1

A cosmic collision in the making: JWST/NIRISS reveals a merging and maturing protocluster core at z~3 around a red quasar

C. Bertemes, D. Wylezalek, G. Noirot, R. Hviding, D. S. N. Rupke, N. L. Zakamska, S. Veilleux, L. Gatto, W. Liu, A. Vayner, Y. Ishikawa, Y. Chen, S. Sankar

Theme match 2/5

Digest

Using JWST/NIRISS wide-field slitless spectroscopy around the luminous red quasar SDSSJ165202.64+172852.3, Bertemes et al. identify “The Step,” a z~3 protocluster with at least 18 spectroscopic members. Its exceptionally dense core, >100-fold galaxy overdensity, 700–850 km/s line-of-sight span, and step-like position–velocity structure point to a merger between two massive haloes, potentially assembling a Coma-like descendant. The system is already environmentally mature: alongside mildly elevated star formation and two new optical-line-ratio AGN candidates, it hosts two line-less, D4000-break quiescent galaxies, making it a compelling snapshot of the transition toward passive cluster cores.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Figure 3. Use this field overview to show the confirmed membership, the concentration around J1652, and the relationship between the NIRISS-wide protocluster map and the much smaller NIRSpec quasar-core footprint.
Figure 5. This is the central structural diagnostic: the 3D position–velocity rendering makes the proposed northern and southern substructures and the distinctive step-like signature of a two-halo merger immediately visible.
Figure 1. Highlight the representative spectra, especially the two passive members with strong continuum breaks, because their line-less nature provides the direct evidence that this z~3 overdensity already contains quenched galaxies.
Figure 6. This SFR–stellar-mass comparison separates the putative infalling northern subgroup from the main halo and shows that member star formation is only slightly elevated relative to the field main sequence.
Figure 7. Use this literature-scale synthesis to place the Step’s halo-normalized total star formation on the lower envelope of comparable protoclusters, supporting the paper’s conclusion that its activity is typical to mildly suppressed for its halo mass.

Tags

2608.10280v1

Tabular foundation models for the estimation of probabilistic quasar photometric redshifts in S-PLUS

Raquel R. Valença, Lilianne Nakazono, Rafael Izbicki, Marco Henrique de Almeida Inácio, Bruno Marcondes e Resende, Maycon Jorge Deláqua da Silva, Kiana Coimbra Buin Lins, Natanael M. Cardoso, Claudia Mendes de Oliveira

Theme match 2/5

Digest

This paper tests whether frozen tabular foundation models can deliver full probabilistic quasar photo-z posteriors for the 12-band S-PLUS DR6 survey, where colour-redshift degeneracies and spectroscopic-to-photometric covariate shift make single-value estimates inadequate. Across training sets spanning 500 to 121,626 quasars and comparisons with eight specialised baselines, TabPFN 2.5 is best or statistically tied for best on nearly every density and point-estimation metric, with especially strong gains for small training samples, bright or faint sources, and high-redshift quasars. Its posteriors remain close to nominally calibrated after importance-weighted evaluation intended to emulate the photometric target catalogue, making it a strong default for catalogue-scale quasar redshift inference; the practical limitation is the accelerator-memory cost of inference on large support and target sets. SHAP results further identify WISE W1/W2 as the most informative individual inputs, with UV and optical bands providing useful refinements.

Open paper page arXiv abstract arXiv PDF Highlighted PDF

Key figures to inspect

Figure 1. This is the paper's central benchmark synthesis, showing how TabPFN 2.5 ranks against specialised density estimators and point-prediction methods across both ordinary and importance-weighted evaluation regimes. It makes the broad, metric-level basis for the claimed default-model status immediately visible, while flagging that the raw weighting regime has a very small effective sample size.
Figure 2. The PIT P-P curves provide the most direct visual test of whether predictive redshift densities remain calibrated when evaluation is shifted from the spectroscopic sample toward the photometric target population. This figure is essential for the paper's deployment-oriented claim that TabPFN 2.5 retains near-nominal calibration under covariate shift.
Figure 3. These targeted catastrophic-outlier examples show what aggregate scores can conceal: cases where TabPFN 2.5 resolves a high-redshift ambiguity, cases that defeat all methods, and cases where it fails while competing estimators succeed. The figure grounds the probabilistic photo-z comparison in concrete multimodal posteriors and preserves the paper's caveat that no method eliminates catastrophic failures.