Pith. sign in

REVIEW 4 major objections 5 minor 1 references

Luminous Mid-IR Selected Obscured Quasars at Cosmic Noon in SDSS Stripe82 II: Spectroscopic Diversity and Broad H$\alpha$ Emissions

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read One-third of bright mid-infrared-selected quasars that appear to be Type-2 nonetheless show broad H$\alpha$ lines, identifying them as lower-redshift analogues of JWST's little red dots.

desk verdict Valuable spectroscopic sample of obscured quasars at cosmic noon, with a suggestive but currently imprecise broad-H-alpha fraction and a model-dependent [O III] comparison. read the letter →

arxiv 2508.02792 v1 pith:VZGARIPU submitted 2025-08-04 astro-ph.GA astro-ph.CO

Ben Wang , Yuzo Ishikawa , Joseph F. Hennawi , Zheng Cai , Gordon T. Richards , Nadia L. Zakamska , Daming Yang , Jan-Torge Schindler This is my paper · ORCID
classification astro-ph.GAastro-ph.CO
keywords obscuredquasarsType-2AGNmid-infraredselectionbroademissionlineslittlereddotscosmicnooncompositespectra
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper reports a spectroscopic survey of 23 luminous mid-infrared-selected Type-2 quasar candidates in SDSS Stripe 82 at $z=0.88$ to $3.49$, selected by bright WISE W4 detections and red $r-W4$ colors. It confirms 23 of 24 candidates as genuine obscured quasars and finds a wide range of emission-line widths: roughly one third show broad H$\alpha$ emission (FWHM $>2000$ km s$^{-1}$) despite the Type-2 classification, while the rest show only narrow lines. These broad-line obscured quasars have spectral energy distributions similar to those of JWST's "little red dots," so the paper argues they are lower-redshift analogues of the heavily obscured broad-line AGNs JWST reveals at higher redshift. It also finds that [O III] is relatively weak for their bolometric luminosities and constructs new composite spectra for the sample. If correct, the paper bridges the gap between well-studied low-redshift obscured quasars and the JWST-discovered obscured AGN population, bearing on whether the obscured fraction of supermassive black hole growth evolves through cosmic noon.

What carries the argument

The load-bearing machinery is the selection-and-confirmation chain. First, a single color cut in the SDSS Stripe 82 field, $r-W4>8.38$ AB with a bright WISE 22 $\mu$m detection ($f_{W4}>5$ mJy), isolates candidates whose optical light is extremely faint or red. Then deep rest-frame UV and optical spectroscopy confirms the objects as quasars and measures line profiles: Gaussian fits to Ly$\alpha$ and H$\alpha$ yield the FWHM that decides "broad" versus "narrow" at 2000 km s$^{-1}$. For the [O III] analysis, the paper converts W4/15 $\mu$m luminosities to bolometric luminosity with $L_{\rm bol}=6.03\,L_{15\,\mu\mathrm{m}}$, based on a composite SED with a torus template, and this conversion is what places the sample on the $L_{\rm [O\,III]}$--$L_{\rm bol}$ diagram. The combination of the color cut, the FWHM criterion, and the SED conversion carries the entire argument.

What would settle it

Measure each of the 13 [O III]-detected objects' bolometric luminosity independently, using X-ray luminosity or per-object torus SED fits, and re-plot $L_{\rm [O\,III]}$ against $L_{\rm bol}$; if the points fall onto the Lacy et al. relation, the reported [O III] weakness is an artifact of the shared SED scaling.

Watch

Extended reading notes

Core claim

The paper confirms 23 of 24 very red, mid-infrared-bright quasar candidates in SDSS Stripe 82 as genuine obscured (Type-2) quasars at $z=0.88$ to $3.49$, with 12 objects above $z=2$. Its central spectroscopic discovery is that the sample is not a single class: 8 of 23 objects show broad H$\alpha$ emission (FWHM $>2000$ km s$^{-1}$) even though they were selected to be optically faint and red, while most of the rest show only narrow lines; one object has broad lines in both Ly$\alpha$ and H$\alpha$. The broad-H$\alpha$ objects have rest-frame UV/optical SEDs similar to those of the JWST "little red dots," so the paper argues they are lower-redshift analogues of the heavily obscured broad-line AGNs uncovered at $z \simeq 4$--$9$. At the same time, the sample's [O III] $\lambda5007$ luminosities fall below the $L_{\rm [O\,III]}$--$L_{\rm bol}$ trend of lower-redshift Type-2 quasars even before extinction correction, and the coadded spectra show weak optical lines such as H$\beta$, [O II], and [S II] that are usable as future follow-up targets.

Load-bearing premise

The claim that [O III] is weak for the bolometric luminosity rests on a single SED conversion, $L_{\rm bol}=6.03 L_{15\,\mu\mathrm{m}}$, applied to every target and to the comparison sample; if individual mid-IR SED shapes differ, the deficiency could disappear.

Editorial extensions

If this is right

  • The high confirmation rate (23 of 24) means the $r-W4>8.38$ selection can produce large, clean samples of obscured quasars at $z\sim1.5$--$3.5$ from wide-area WISE data alone.
  • The presence of broad H$\alpha$ in about one third of a red-selected obscured sample shows that optical/UV classification as Type-2 does not rule out a broad-line region, so unification models must accommodate a population with broad Balmer lines seen through red SEDs.
  • The broad-H$\alpha$ objects provide a lower-redshift benchmark for the JWST little red dots, enabling comparisons of line widths, SED shapes, and host properties that are difficult for the high-redshift JWST objects.
  • The weak [O III] relative to bolometric luminosity, if real, implies that the narrow-line region in these luminous obscured quasars is either heavily extinguished or intrinsically different, and any model of obscured growth at cosmic noon will need to explain it.
  • The new composite spectra provide a template for identifying similar objects in future surveys and for planning JWST observations aimed at detecting weak rest-optical lines.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the broad-H$\alpha$ objects are true little-red-dot analogues, the Stripe 82 counts imply that such objects are not rare at $z\sim2$, and the same selection applied to full-sky WISE catalogs could yield many nearby analogues for detailed study.
  • A decisive way to separate a physically unobscured broad-line region from scattered or transmitted light would be spectropolarimetry: dust-scattered broad lines should be strongly polarized, and measuring that would settle whether the Type-2 classification is purely a line-of-sight effect or a partially obscured BLR.
  • The broad-to-narrow H$\alpha$ fraction could vary with luminosity or redshift; splitting a larger sample by $L_{\rm bol}$ and $z$ would show whether the broad-line fraction is a property of the most luminous obscured quasars or of the selection itself.
  • If the [O III] weakness is confirmed with per-object SEDs, it would strengthen the analogy with little red dots, whose emission lines are often weak relative to their red continua.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. This paper presents a Gemini/GNIRS and Keck/LRIS+KCWI spectroscopic survey of 23 luminous WISE W4-selected obscured quasar candidates in SDSS Stripe 82 at z = 0.88–3.49, confirming 23 of 24 candidates as Type-2 or reddened Type-1 quasars. The authors measure Ly-alpha and H-alpha line widths from Gaussian fits, report that roughly one-third of the sample shows broad H-alpha (FWHM > 2000 km/s) despite the Type-2 classification, compare their [O III] luminosities with bolometric luminosities derived from a mid-IR scaling, and construct composite spectra for the full sample and for broad/narrow H-alpha subgroups. They argue that the broad-line obscured objects resemble JWST-discovered little red dots and that the sample bridges the gap between low-redshift obscured quasars and high-redshift JWST AGNs.

Significance. If the central claims hold, this sample is valuable: it is one of the few spectroscopic samples of luminous obscured quasars at cosmic noon, it demonstrates spectral diversity in a color-selected obscured population, and it directly connects z~2 obscured AGNs to the JWST little-red-dot phenomenon. The public release of reduced spectra is a concrete asset. The broad-H-alpha detections themselves are supported by the displayed spectra, and the composite spectra are a useful reference. However, the headline 'one-third broad H-alpha' fraction is not currently robust because of inconsistent bookkeeping across the abstract, text, and tables, and because the FWHM measurements near the 2000 km/s threshold are vulnerable to the ignored [N II] doublet and instrumental resolution. The [O III] comparison also rests on a single assumed mid-IR SED scaling with no propagated uncertainties. These are load-bearing issues for the paper's main conclusions.

major comments (4)
  1. [Abstract; Section 3.3.3; Table 1; Section 7] The reported fraction of broad H-alpha emitters is inconsistent across the manuscript. The abstract states 8 of 23; Section 3.3.3 states '7/21 targets have broad H-alpha emission lines'; Table 1 lists 9 objects with FWHM > 2000 km/s (J0024-0012, J0130+0009, J0213+0024, J0215+0042, J0221+0050, J2243+0017, J2259-0009, J2329+0020, J2334+0031); and Section 7 conclusions (i) and (ii) give conflicting counts (four versus three objects with broad H-alpha and narrow Ly-alpha). Because the broad-line fraction is the paper's headline claim, the authors must reconcile these numbers, state which denominator (23, 21, or the 11 objects with both lines) is used, and correct the conclusion text.
  2. [Section 3.2; Section 2.2.1] The H-alpha FWHM measurements are derived from single-Gaussian fits at R~1100 that do not include the [N II] lambda-lambda 6548, 6584 doublet. The instrumental resolution is roughly 270 km/s at H-alpha, and the [N II] lines are separated from H-alpha by approximately -677 and +941 km/s, so they are unresolved at this resolution. A single Gaussian fit to H-alpha plus [N II] can be broadened by several hundred km/s depending on the [N II]/H-alpha ratio. Three objects lie within ~700 km/s of the 2000 km/s threshold (J2334+0031 at 2112 km/s, J0024-0012 at 2145 km/s, J0215+0042 at 2683 km/s), so their broad/narrow classification is not robust. The fits should include [N II] components with fixed wavelengths and atomic intensity ratios, and the sensitivity of the classification to the fitting model and to instrumental broadening should be quantified.
  3. [Section 6; Table 1; Section 4.2] The 'broad H-alpha' composite spectrum in Section 6 includes J0152-0024, whose Table 1 FWHM is 1382 km/s, based on the argument that its H-alpha line falls at the edge of the GNIRS coverage and its measured FWHM may be underestimated. This is a post-hoc reclassification that is not applied consistently to the tabulated line properties, and it changes the composite 'broad' group from a rule-based FWHM>2000 km/s sample to a mixed sample. Since the broad composite is used to support the LRD analogy, the grouping should be defined by a transparent criterion (e.g., Table 1 FWHM) or supported by a re-fit of J0152-0024 with proper treatment of the line profile.
  4. [Section 5; Eq. (2); Eq. (4); Figure 6] The [O III] deficiency and the '100x brighter' comparison rest on the scalings Lbol = 6.03 x L15um and Lbol = 6.01 x L13.5um, which are derived from the authors' own composite SED and a specific torus template (Stalevski et al. 2016) and then assumed to apply to every target and to the Lacy et al. (2013a) and Zakamska et al. (2016) comparison samples. No uncertainties are propagated into Lbol, and Figure 6 shows no error bars on the comparison points or on the new measurements. The claim that [O III] is weak at fixed Lbol is only as strong as this scaling; the authors should show how the result changes if the mid-IR SED shape varies, or at least provide a quantitative systematic uncertainty on the bolometric correction.
minor comments (5)
  1. [Section 3.3.1; Figure 5 caption] The number of objects with both narrow H-alpha and narrow Ly-alpha is given as five in Section 3.3.1 and in the Figure 5 caption, but the text in Section 3.3 says six; please make these counts consistent.
  2. [Section 7, conclusion (i)] The conclusion text refers to 'J2334+0041', but the corresponding table and figures list this target as J2334+0031.
  3. [References] The reference entry beginning 'S., Hamann F., et al. 2019' is missing the first author's name; this is likely a formatting error that should be corrected.
  4. [Section 5] The sentence 'Since our targets don't have H-beta line detection, we couldn't conduct the extinction correction using .' is incomplete and should be finished or removed.
  5. [Section 2.2.2] The instrumental setup description contains unclear notation, including '1.0.′′ long-slit' and '560 dichroic'; please clarify the slit width and dichroic wavelength.

Circularity Check

1 steps flagged · score 3.0 of 10

One self-referential bolometric calibration in Section 5; the broad H-alpha and composite-spectrum results are independent.

  1. self citation load bearing [Section 5, Eq. (2)]
    "We use the torus template from Stalevski et al. (2016) and our composite SED in Wang et al. (2025) to determine the relation between the luminosity at 15 μm (observed frame at z∼2) and the bolometric luminosity: Lbol= 6.03×L15μm. Using this relation, the luminosity at 15 μm in Lacy et al. (2013a) can be used to compute the bolometric luminosity, which assumes their targets have the same mid-IR SED shape as our targets."

    The bolometric luminosity used throughout Section 5 is defined from the authors' own composite SED for this same sample (Wang et al. 2025). The same self-derived conversion is then applied to the Lacy et al. comparison points, so the x-axis of Figure 6 is constructed from the paper's own template for both data sets. Consequently, the statement that 'our targets are about 100× brighter' and the claimed [O III] deficiency are not calibrated against an independent bolometric scale; they partly restate the W4 flux selection plus the adopted self-referential conversion. This affects only the Section 5 scaling-relation conclusion; the broad H-alpha detection, spectral diversity, and composite spectra are self-contained measurements.

full rationale

The central spectral results are independent measurements: the broad-versus-narrow H-alpha classification comes directly from Gaussian fits to the GNIRS spectra, and the composite spectra are built from the observed spectra with W4-based scaling. These do not reduce to any fitted parameter or to a self-citation chain. The main circularity-adjacent element is the bolometric correction in Section 5, where Lbol is taken from the authors' own composite SED (Wang et al. 2025) and then used both for their targets and for the literature comparison samples; the [O III]-deficiency conclusion therefore inherits that template assumption and is not fully externally anchored. I do not count the internal inconsistencies in the number of broad H-alpha objects (abstract 8/23, Section 3.3.3 7/21, Table 1 listing nine) as circularity; those are correctness and robustness concerns, not self-referential derivations. Overall the paper's headline diversity claim is not circular, but the Section 5 scaling conclusion is partly self-referential, warranting a score of 3.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The paper is observational; the main model-dependent inputs are the bolometric corrections from the authors' own SED fitting (Wang et al. 2025), which underpin the [O III] result. No new physical entities are introduced.

free parameters (2)
  • Bolometric correction Lbol/L15um = 6.03
    Derived from the authors' composite SED and Stalevski torus template in Wang et al. (2025); applied to all targets and to Lacy et al. 15um luminosities to compute Lbol (Section 5, Eq. 2). Assumes identical mid-IR SED shape for all sources.
  • Bolometric correction Lbol/L13.5um = 6.01
    Used to convert 13.5um luminosities from Zakamska et al. (2016) and LaMassa et al. (2010) to Lbol (Section 5, Eq. 4), again based on the authors' SED model.
assumptions (4)
  • domain assumption The composite SED and torus template from Wang et al. (2025) and Stalevski et al. (2016) are representative of all targets and of the comparison samples.
    Load-bearing for Lbol and therefore for the [O III] weakness result (Section 5, Eqs. 2 and 4).
  • domain assumption Emission lines follow simple Gaussian profiles; a single or double Gaussian adequately describes Ly-alpha, H-alpha, and [O III].
    Line fitting in Section 3.2; asymmetric or absorbed profiles (e.g., J0152-0024, J2334+0031) are noted but not modeled, so FWHM values may be biased.
  • domain assumption The redshift is determined by identified emission lines, and for sources with only one line the identification is unambiguous.
    Section 3.1-3.2; some redshifts rely on a single line (e.g., H-alpha for many GNIRS-only targets), and misidentification would alter Lbol and line luminosities.
  • domain assumption The selection criteria (Eq. 1) efficiently isolate obscured quasars without significant contamination.
    Section 2.1; 23 of 24 confirm as quasars, supporting this, but the one inconclusive source and possible completeness biases are acknowledged.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Luminous Mid-IR Selected Obscured Quasars at Cosmic Noon in SDSS Stripe82 II: Spectroscopic Diversity and Broad H$\alpha$ Emissions." pith.science (2026). https://pith.science/paper/VZGARIPU

@misc{pith2026250802792,
  author       = {Pith},
  title        = {Pith review of: Luminous Mid-IR Selected Obscured Quasars at Cosmic Noon in SDSS Stripe82 II: Spectroscopic Diversity and Broad H$\alpha$ Emissions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VZGARIPU}},
  note         = {Machine review of arXiv:2508.02792}
}
read the original abstract

We present a multiwavelength spectroscopic survey of 23 luminous mid-infrared-selected Type-2 quasars at redshifts z = 0.88 to 3.49. The targets were selected in the SDSS Stripe 82 field based on their bright WISE W4 detections (flux > 5 mJy) and extremely faint or red optical counterparts (e.g., r > 23 or r - W4 > 8.4), designed to identify heavily obscured quasars. Deep near-infrared (Gemini/GNIRS) and optical (Keck/LRIS and KCWI) spectroscopy confirm 23 out of 24 candidates as Type-2 quasars in this redshift range, including 12 objects at z > 2. The spectra exhibit strong rest-frame UV and optical emission lines (Ly-alpha, C IV, [O III], H-alpha) with a wide range of line widths, indicating significant spectral diversity. Approximately one-third of the sample (8 of 23) shows broad H-alpha emission (FWHM > 2000 km/s) despite their Type-2 classification, while the rest have only narrow lines (FWHM < 2000 km/s) characteristic of classical obscured quasars. Notably, these broad-line Type-2 quasars share similar spectral energy distributions with the JWST-discovered "little red dot" (LRD) AGNs, suggesting that our sample could be lower-redshift analogues of the heavily obscured broad-line AGNs uncovered by JWST. We also find that the [O III] 5007 angstrom emission is relatively weak for their high bolometric luminosities, deviating from trends seen in lower-redshift Type-2 QSOs. A new composite spectrum for Type-2 QSOs is built using our sample. Overall, our results demonstrate that mid-IR selection efficiently uncovers a diverse population of obscured quasars and that spectroscopic follow-up is crucial for revealing their true nature. This study provides new insights into heavily obscured SMBH growth at cosmic noon and bridges the gap to the obscured AGN populations now being revealed by JWST.

Figures

Figures reproduced from arXiv: 2508.02792 by the authors.

Figure 1
Figure 1. Example spectra from Keck and Gemini, and the fitted emission lines for 2 highest-𝑧 (𝑧 > 3) Type-2 QSO targets in our sample. The Left panel shows the spectra from Keck/LRIS(or Keck/KCWI) and Gemini/GNIRS, the green lines mark the detected emission line and the blue dashed lines are the undetected emission lines. The target name and redshift are shown in the title of the subplots. The right panel shows the Gaussian … view at source ↗
Figure 2
Figure 2. Example spectra and line fittings for 2 targets with narrow(< 2000km/s) H𝛼 emission line. The top target has a strong absorption feature at the wavelength of Ly𝛼 and NV emission. The H𝛼 is located at the edge of the GNIRS coverage, making it difficult to measure the line width. The bottom panel shows a target with both narrow Ly𝛼 and narrow H𝛼 emission lines. . MNRAS 000, 1–15 (2015) [PITH_FULL_IMAGE:figures/full_f… view at source ↗
Figure 3
Figure 3. Example spectra and line fitting for 2 targets with broad H𝛼 or Ly𝛼 emissions. The upper panel shows a target J0221+0050 with broad H𝛼 and narrow Ly𝛼 emission. The broad emission lines are not expected for typical obscured QSOs, but it is a common feature in some ERQs and JWST AGNs. The lower panel shows a target with broad Ly𝛼 but narrow H𝛼 emission line. . MNRAS 000, 1–15 (2015) [PITH_FULL_IMAGE:figures/full_fig_… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Example spectra and line fitting for 1 target with both broad Ly𝛼 and H𝛼 emission lines. This target J2334+0031 is the only target in our sample with both broad Ly𝛼 and H𝛼 emission lines. This target also has broad CIV, CIII, and MgII emission lines. . success rate (∼ …
Figure 5
Figure 5. Figure 5: The FWHM of H𝛼 and Ly𝛼 of 11 targets in our sample. The X-axis is the FWHM of the Ly𝛼 emission lines, and the y-axis is the H𝛼 FWHM. The 2000km/s is marked as black dashed lines. According to this definition, five targets have narrow H𝛼 and narrow Ly𝛼 emission lines, f…
Figure 6
Figure 6. Figure 6: The relation between the luminosity of the [O III] emission line and the bolometric luminosity. The relations from Lamastra et al. (2009), LaMassa et al. (2010), and Zakamska et al. (2016) are shown as blue, cyan, and red lines. The green dots are the Type-2 AGN measur…
Figure 7
Figure 7. Figure 7: The composite spectra for 23 targets in our sample. The top panel shows the composite spectra for all 23 targets, and the median panel shows the composite spectra for nine targets with broad H𝛼 emissions. The bottom panel shows the composite spectrum only for 12 target…

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

1 extracted references · 1 linked inside Pith

  1. [1]

    B., et al., 2024, arXiv e-prints, p

    S.,HamannF.,ZakamskaN.L.,AlexandroffR.M.,RupkeD.,WylezalekD., 2019, MNRAS, 488, 4126 Akins H. B., et al., 2024, arXiv e-prints, p. arXiv:2406.10341 Alexandroff R., et al., 2013, MNRAS, 435, 3306 Alexandroff R. M., et al., 2018, Monthly Notices of the Royal Astronomical Society, 479, 4936 Alonso-Herrero A., et al., 2006, ApJ, 640, 167 Antonucci R., 1993, A...

    arXiv 2015

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.