Pith. sign in

REVIEW 3 major objections 3 minor 3 references

Mrk 335's X-ray obscuration is a recurring, three-phase, clumpy disk wind that reshapes itself on day timescales in response to the ionising continuum.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-02 20:23 UTC pith:67DYMMER

load-bearing objection Solid RGS study of Mrk 335's multi-phase obscurer; three-phase detection robust, but day-scale NH variability claims rest on fixed covering fractions. the 3 major comments →

arxiv 2602.23255 v2 pith:67DYMMER submitted 2026-02-26 astro-ph.HE astro-ph.GA

Variability of the X-ray obscuring wind in Mrk 335 with XMM-Newton/RGS

Daniele Rogantini , Erin Kara , Luigi Gallo , S Komossa , Peter Kosec , Christos Panagiotou , Dan Wilkins , Ehud Behar
show 5 more authors
Joheen Chakraborty Dirk Grupe Missagh Mehdipour Ciro Pinto Irina Zhuravleva
This is my paper
classification astro-ph.HE astro-ph.GA
keywords AGNX-ray spectroscopyphotoionised outflowsobscuring windsMrk 335broad-line regionX-ray variabilityradiative driving
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper argues that the dramatic X-ray obscuration of the Seyfert galaxy Mrk 335 is produced by a structured, three-phase disk wind rather than a single cloud. Using five consecutive high-resolution RGS spectra from June 2021, it shows that the three wind phases change column density and ionisation on day timescales, that the coldest phase thins sharply during a flare, and that all phases subsequently accelerate. This links continuum flares to radiative acceleration of the wind and places the wind at broad-line-region scales, where it may carry enough kinetic power to matter for AGN feedback. The same three-phase structure seen in 2009 indicates that such multi-phase obscuration recurs over decade timescales.

Core claim

The stacked 440-ks RGS spectrum of Mrk 335 requires three photoionised absorbers with time-averaged log ξ ≃ 3.63, 3.10, and 2.01, outflowing at about 5820, 3210, and 2140 km/s. The same three phases were present in the 2009 intermediate state, so the obscurer is not a one-off transient. Across five epochs spanning eight days, NH and log ξ vary in all phases; the low-ξ phase's column density drops from roughly 1.2 × 10^22 to 9 × 10^20 cm^-2 during the Epoch 4 flare, and in Epoch 5 all three components show increased outflow velocities. The high-ξ phase tracks the ionising luminosity as if near photoionisation equilibrium, while the low-ξ phase shows at most a delayed response of about a day.

What carries the argument

The central tool is a self-consistent photoionisation model (the pion model) driven by the measured broadband SED, applied first to the stacked RGS spectrum to fix a three-absorber baseline and then to each epoch with the covering fractions frozen at their stacked values. The model computes ionic column densities and transmission from the ionising continuum, letting the authors extract NH, log ξ, and outflow velocity for each phase. Supporting machinery includes a broadband SED decomposition into accretion-disc, warm-Comptonisation, power-law, and reflection components, thermal stability S-curves for each phase, and order-of-magnitude distance estimates from escape velocity, absorber thickne

Load-bearing premise

The day-to-day column-density changes stand or fall on the assumption that each wind phase's covering fraction stayed fixed during the campaign, because the single-epoch spectra cannot distinguish a thinner absorber from a more covering one.

What would settle it

Take a high-signal-to-noise spectrum of Mrk 335 during a similar flare and allow all three covering fractions to float freely; if a model with constant NH and variable covering fits equally well, the reported opacity drop is not a column change. Also, monitor the low-ξ phase on a dense day-scale cadence: a recombination lag much shorter or longer than about one day would falsify the inferred ~0.1 pc distance and the associated wind energetics.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • The match between the 2021 and 2009 three-phase winds implies the same multi-phase outflow structure can persist for more than a decade rather than appearing as a single transient event.
  • Day-scale variability in all phases, with the low-ionisation phase thinnest during the flare, supports a clumpy wind whose line-of-sight opacity is shaped by radiative driving.
  • The high-ionisation phase responds to ionising-luminosity changes on timescales shorter than a day, placing it close to the black hole and near photoionisation equilibrium; the low-ionisation phase's ~1-day delay, if real, puts it farther out.
  • Order-of-magnitude distances of 10^3–10^5 Rg place the wind at broad-line-region scales, and the inferred kinetic power can reach the percent level of L_bol under plausible geometry and filling factors, making the wind a candidate feedback channel.
  • Within the paper's parametrisation, intrinsic continuum changes dominate the broadband X-ray variability, but wind absorption adds a non-negligible modulation (a factor of about 1.5 in transmission).

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the covering fractions actually vary between epochs, the reported day-scale NH changes—especially the Epoch 4 drop—could be overestimated; the paper tests this only for the low-ξ phase, leaving the degeneracy unquantified for the mid- and high-ionisation phases.
  • The tentative ~1-day lag of the low-ξ phase is a testable distance probe: denser time sampling that measures the recombination response would confirm or rule out the ~0.1 pc distance and the associated energetics.
  • The structural similarity to the Mrk 817 obscurer suggests a general class of BLR-scale clumpy winds; if so, long-term multi-epoch monitoring of many Seyferts could map the wind's global covering fraction and duty cycle, turning the feedback estimate from an order-of-magnitude bound into a measurement.
  • A simultaneous UV–X-ray flare campaign would test the radiative-acceleration picture directly: if the same wind is seen in UV absorption, the UV troughs should slow or thin after an X-ray flare in the same way the X-ray phases do.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 3 minor

Summary. The paper analyzes a coordinated XMM-Newton/NuSTAR campaign on Mrk 335 in June 2021, with five RGS epochs spanning eight days. After modeling the broadband SED, the stacked RGS spectrum is fit with three photoionized absorbers (log ξ ≈ 3.63, 3.10, 2.01; |v_out| ≈ 5820, 3210, 2140 km/s), supported by large ΔC improvements. The authors then fit each epoch independently, allowing NH, ξ, and v_out to vary while fixing covering fractions and some turbulent velocities to the stacked-spectrum values. They report strong day-scale variability, a low-ξ NH drop during a flare, increased outflow velocities in the following epoch, and a qualitative recurrence of the three-phase structure seen in 2009. Order-of-magnitude arguments place the absorber at BLR scales (≈10^3–10^5 Rg) and yield a kinetic-power estimate that may reach the percent level of L_bol.

Significance. If the variability interpretation holds, the paper provides a rare day-timescale view of a multi-phase obscuring wind and a concrete, testable coupling between a flare and subsequent outflow acceleration. The statistical evidence for three absorbing phases in the stacked spectrum is strong (ΔC = 990, 400, 90 over 5, 5, and 4 free parameters), and the paper is commendably explicit about order-of-magnitude assumptions, degeneracies, and alternative interpretations (thermal stability, collisional ionization, partial-covering). The comparison with Mrk 817 and with the 2009 Mrk 335 state places the result in a broader context. The main weakness is that the epoch-to-epoch variability claims depend on fixing covering fractions, a degeneracy the authors themselves acknowledge; this needs either additional fitting or a more guarded presentation.

major comments (3)
  1. [§4.2, Table 3, §5.2] The central variability claim—that the wind transmission changes mainly through NH and that strong day-scale NH changes occur in all phases—rests on epoch fits where fcov is fixed to the stacked-spectrum values. The paper states that the NH–fcov degeneracy cannot be broken at single-epoch S/N, yet only the low-ξ component in Epoch 4 is tested with fcov free (§5.2). For the mid- and high-ξ components, no such test is shown. I request a systematic treatment: for each epoch and each pion component, either free fcov and present NH–fcov confidence contours, or quantitatively report how much the fitted NH can change when fcov is varied within the stacked-spectrum uncertainties. Without this, the quantitative NH changes (e.g., low-ξ NH from 1.2 to 0.09 × 10^22 cm^-2 in Epoch 4; Table 3) are not uniquely determined by the data.
  2. [§5.2, Fig. 7, Table 3] The claimed Epoch 5 increase in outflow velocity for all three phases should be shown to be robust to jointly freeing fcov and turbulent velocity. Partial-covering changes can alter line-profile shapes and bias centroid measurements. The reported velocity shifts are of order 5–15% and are individually significant at the few-σ level, but the coherence across three phases is part of the evidence for radiative acceleration. Please provide a joint fit or a likelihood-ratio test where fcov (and, where feasible, v_turb) are free, or explicitly label the velocity-increase claim as derived under the fixed-fcov assumption.
  3. [§5.2, Epoch 2] The Epoch 2 'exchange' of ξ between the mid- and high-ξ components is used to support 'strong variability in all phases'. This is also the lowest-flux epoch, with v_turb fixed and larger uncertainties. The text mentions a grid of initial conditions and freeing fcov without improvement, but no details or confidence contours are provided. Please show the relevant parameter-space exploration or soften the conclusion so that the Epoch 2 behavior is not presented as a well-established phase interchange.
minor comments (3)
  1. [Fig. 7 caption and §5.2] The figure caption states that the left column is versus ionising luminosity and the right column is versus time, but §5.2 refers to 'the right-hand panels of Fig. 7' for the log ξ–L_ion trend. Please correct the caption or the text so the reader knows which panel shows which variable.
  2. [§4.2] The sentence 'the degeneracy between column density and covering fraction cannot be broken' is important. Consider adding a one-sentence description in the abstract or conclusions of how this assumption affects the variability results, rather than only in the body.
  3. [Table 3] The table note says the high-ξ fcov is fixed to unity; this is an additional assumption that should be highlighted in the main text when discussing the high-ξ NH variability, since fcov < 1 would change the inferred NH.

Circularity Check

0 steps flagged

No significant circularity: the spectral fits, variability analysis, and order-of-magnitude distance/energetics estimates are self-contained and do not reduce to their inputs by construction.

full rationale

The paper's derivation chain is self-contained. The ionizing SED is modeled from broadband data with standard continuum/reflection components, and the RGS absorption is then modeled with self-consistent photoionization (pion) components; the three-phase structure and outflow velocities are best-fit parameters, not outputs of a model built on the conclusion. The variability analysis fits NH, xi, and v_out per epoch against the RGS data; the only fixed quantity, f_cov, is fixed to the stacked value, and the paper explicitly tests freeing it for the low-xi phase and acknowledges that the NH-f_cov degeneracy cannot be fully broken (Section 4.2). That is a stated limitation, not a construction that makes the NH changes equal to the input. Distance estimates (Eqs. 1-3) and energetics (Eqs. 4-6) are explicitly order-of-magnitude scalings with stated assumptions (e.g., v_out ~ v_esc), not circular derivations. Comparisons to Mrk 817 (Zaidouni et al. 2024) and earlier Mrk 335 work are external context and consistency checks; even where co-authors are involved, the cited results are not load-bearing for the present fits. No step assumes the conclusion or reduces a prediction to a fitted parameter by definition.

Axiom & Free-Parameter Ledger

12 free parameters · 7 axioms · 0 invented entities

The central claims rest on spectral fitting to public X-ray data. The three-phase wind parameters (NH, xi, v_out) are fitted free parameters, not derived from first principles. The distance and energetics additionally assume v_out ~ v_esc and unknown geometry/clumpiness, making those derived quantities order-of-magnitude.

free parameters (12)
  • NH of high-ionisation wind = 1.3e23 cm^-2 (average); varies 3.4e22-4.0e23 across epochs
    Column density of the high-xi pion component, fitted to RGS spectra (Table 3).
  • NH of mid-ionisation wind = 2.0e22 cm^-2 (average); varies 1.2e21-7.4e22
    Column density of the mid-xi pion component, fitted to RGS spectra (Table 3).
  • NH of low-ionisation wind = 1.8e21 cm^-2 (average); varies 9e20-1.2e22
    Column density of the low-xi pion component, fitted to RGS spectra (Table 3).
  • log xi of high-ionisation wind = 3.63 (average)
    Ionisation parameter of the high-xi component, fitted (Table 3).
  • log xi of mid-ionisation wind = 3.10 (average)
    Ionisation parameter of the mid-xi component, fitted (Table 3).
  • log xi of low-ionisation wind = 2.01 (average)
    Ionisation parameter of the low-xi component, fitted (Table 3).
  • Outflow velocity of high-ionisation wind = -5820 km/s (average)
    Velocity shift of the high-xi absorber, fitted to line positions (Table 3).
  • Outflow velocity of mid-ionisation wind = -3210 km/s (average)
    Velocity shift of the mid-xi absorber, fitted (Table 3).
  • Outflow velocity of low-ionisation wind = -2140 km/s (average)
    Velocity shift of the low-xi absorber, fitted (Table 3).
  • Covering fractions (mid and low phases) = 0.57 (mid), 0.67 (low); high fixed to 1
    Partial covering fractions determined from the stacked spectrum and held fixed in single-epoch fits (Section 4.2).
  • Turbulent velocities = 90-780 km/s depending on phase
    Line-broadening parameters fitted to the stacked spectrum; fixed to average values in Epochs 2 and 3 (Table 3).
  • SED continuum parameters (Gamma, comt norm, dbb norm, refl scale) = Gamma 1.94-2.08; other parameters listed in Table 2
    Parameters of the broadband SED model (dbb+comt+pow+refl) fitted per epoch; determine the ionising luminosity L_ion used in pion modelling.
axioms (7)
  • domain assumption Photoionisation equilibrium (PIE) is assumed for the wind gas (pion model)
    The entire spectral analysis uses the pion model, which assumes a steady-state photoionised plasma. If the gas is not in PIE on the timescales probed, the derived xi and NH could be biased. Section 4.1.
  • domain assumption Protosolar abundances (Lodders et al. 2009) for all absorbers
    Abundances are fixed to protosolar values; unusual metal abundances in the outflow would change line strengths and inferred columns. Mentioned in Section 3.
  • domain assumption The simplified SED decomposition (dbb+comt+pow+refl) adequately captures the ionising continuum
    The reflection model is a simple neutral slab, and the soft excess is modelled phenomenologically with comt. A different continuum shape would affect L_ion and thus the photoionisation parameter xi. Section 3.
  • ad hoc to paper Covering fractions measured from the stacked spectrum apply unchanged to each individual epoch
    The epoch fits fix f_cov to the stacked values to break degeneracies; if f_cov varies, the reported NH variability could be misattributed. Section 4.2.
  • ad hoc to paper Outflow velocity approximately equals the local escape velocity for distance estimates
    Equation (2) uses v_out ~ v_esc to derive radii. The paper acknowledges this is an order-of-magnitude reference and tests alternative estimates, but the derived distances and kinetic powers depend on this assumption. Sections 5.4.1-5.4.2.
  • domain assumption Black hole mass MBH = 2.7e7 Msun (Grier et al. 2012)
    Used for Rg scaling and Eddington ratios. Adopted from prior reverberation mapping measurements.
  • domain assumption The ionising luminosity L_ion from the SED is the correct driver of the photoionisation state
    The pion model uses the SED-derived L_ion; if the intrinsic SED is mis-estimated, the xi values and the correlation with L_ion would be affected. Section 3 and 4.

pith-pipeline@v1.3.0-alltime-deepseek · 33125 in / 11148 out tokens · 97147 ms · 2026-08-02T20:23:47.827639+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Variability of the X-ray obscuring wind in Mrk 335 with XMM-Newton/RGS." pith.science (2026). https://pith.science/paper/67DYMMER

@misc{pith2026260223255,
  author       = {Pith},
  title        = {Pith review of: Variability of the X-ray obscuring wind in Mrk 335 with XMM-Newton/RGS},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/67DYMMER}},
  note         = {Machine review of arXiv:2602.23255}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Transient X-ray obscuration in Seyfert 1 galaxies likely arises from clumpy accretion-disk winds near the broad-line region (BLR), but the wind structure and short-timescale variability are difficult to measure because high-resolution spectra are often suppressed during deep low states. We analyse a coordinated XMM-Newton/NuSTAR campaign on Mrk 335 in June 2021, with long-term Swift monitoring, capturing the source in an intermediate-flux state with strong RGS absorption features. We model the broadband SED to determine the ionising continuum for self-consistent photoionisation modelling of the RGS spectra. The stacked RGS spectrum requires three photoionised absorbers with log xi ~ 3.69, 2.97, and 1.91, outflowing at |v_out| ~ 5800, 3200, and 2100 km/s, respectively. Their properties are consistent with the three-phase obscurer reported in 2009, indicating that a similar multi-phase obscuring wind can persist over decade timescales. Using five consecutive RGS observations, we track the wind evolution on day timescales and find strong variability in column density and ionisation in all phases, together with smaller but coherent velocity changes. During a flare, the low-ionisation phase shows a significant drop in opacity, while in the subsequent epoch all phases show increased outflow velocities, suggesting a possible connection between continuum variability and changes in the line-of-sight absorber. The high-ionisation phase responds most directly to changes in ionising luminosity, while the lowest-ionisation phase shows at most a delayed response. Order-of-magnitude constraints place the obscurer at BLR scales, ~10^3-10^5 Rg, with kinetic power potentially reaching the percent level of L_bol for plausible assumptions on geometry and clumpiness.

Figures

Figures reproduced from arXiv: 2602.23255 by Christos Panagiotou, Ciro Pinto, Daniele Rogantini, Dan Wilkins, Dirk Grupe, Ehud Behar, Erin Kara, Irina Zhuravleva, Joheen Chakraborty, Luigi Gallo, Missagh Mehdipour, Peter Kosec, S Komossa.

Figure 1
Figure 1. Figure 1: EPIC-pn and NuSTAR spectra of Mrk 335 obtained during the 2021 campaign. Open symbols show the EPIC-pn data (0.3–10 keV), while filled symbols indicate the simultaneous NuSTAR observations (3–50 keV). The coloured datasets correspond to the five 2021 campaign epochs, using the same colour scheme adopted for the light curves in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Multiwavelength light curves of Mrk 335. Top panel: Long-term Swift/XRT monitoring up to May 2025. The vertical dashed line marks the epoch of the joint XMM-Newton and NuSTAR campaign. We indicate the approximate high-, mid- and low- flux states with the color palette yellow, red and brown. Second panel: XMM-Newton/EPIC-pn light curves in the 0.3–10 keV band for the five observed epochs. The Swift/XRT poin… view at source ↗
Figure 3
Figure 3. Figure 3: Spectral energy distribution (SED) modelling of Mrk 335. Top: Time-averaged SED modeling with the best-fit model components overplotted (see text for a description of each component), together with the multiwavelength data used in the modelling: XMM/EPIC-pn (orange), NuSTAR (gray), Swift/UVOT (magenta), XMM/OM (purple), and XMM/RGS (cyan). The intrinsic SED (all absorption removed) is shown as a solid blac… view at source ↗
Figure 4
Figure 4. Figure 4: Segments of the stacked RGS spectrum of Mrk 335 (net exposure of 440 ks), shown in the rest frame of the source. The magenta line represents the best-fit photoionisation model. Prominent absorption and emission features are marked with blue and red labels, respectively, while Galactic absorption lines at z = 0 are indicated in black. The absorption features are systematically blueshifted, tracing ionised o… view at source ↗
Figure 5
Figure 5. Figure 5: Left panel: Transmission spectrum of Mrk 335, obtained by dividing the observed RGS spectrum by the intrinsic continuum model. The plot zooms in on two key regions: the Fe UTA (15.5–16.5 ˚A) and the O VIII Lyα line (∼18.7 ˚A). The low-ξ wind component provides the dominant contribution to the Fe UTA, while all three ionisation components are required, in different proportions, to reproduce the O VIII absor… view at source ↗
Figure 6
Figure 6. Figure 6: Evolution of the model transmission spectra of the photoionised wind in Mrk 335. From top to bottom, the panels show the high-, mid-, and low-ionisation components, and the total transmission (product of all components), for the five XMM-Newton epochs (color-coded). Changes in both the depth and detailed line structure, most prominently in the Fe xvii–O VIII band, trace variations in the physical state of … view at source ↗
Figure 7
Figure 7. Figure 7: Evolution of the ionised outflow parameters of Mrk 335 across the 2021 campaign. Left column: wind parameters as a function of ionising luminosity Lion (13.6 eV–13.6 keV). From top to bottom, the panels show the column density (NH), the ionisation parameter (log ξ), and the absolute outflow velocity (|vout|). Right column: same parameters as a function of time since the start of the first observation. The … view at source ↗
Figure 8
Figure 8. Figure 8: Outflow parameter correlations for Mrk 335. Left: column density NH versus ionisation parameter log ξ. Right: absolute outflow velocity |vout| versus log ξ. Colours denote the three wind phases (high-, mid-, and low-ξ; see legend). Open hexagons show the best-fit values for the individual epochs, while filled hexagons show the best-fit values for the stacked deep RGS spectrum. et al. 2019). During the 2009… view at source ↗
Figure 9
Figure 9. Figure 9: Thermal stability curves (S-curves) for each ionised absorber, computed for the SED seen by each component, and the location of the absorbers assuming thermal equilibrium. Marker sizes reflect the 1σ uncertainty on log ξ. No statistically significant UFO signature is detected in either the RGS or EPIC-pn spectra, consistent with previous XMM-Newton observations of Mrk 335 (e.g. H. Liu et al. 2021). However… view at source ↗
Figure 10
Figure 10. Figure 10: provides a visual summary of the short-timescale spectral variability discussed in the main text. We display the time evolution of the RGS spectra in the Fe xvii–O VIII band (14.8–19.2 ˚A), where the variability is most apparent. Each panel shows one epoch fitted independently (RGS1 and RGS2 fitted simultaneously), with the best-fit model overplotted. The depth of the O VIII Lyα absorption and the strengt… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

3 extracted references · 1 linked inside Pith

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter doi edition editor eprint howpublished institution journal key month number organization pages publisher school series title misctitle type volume year version url label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts ...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION format.url url empty "" new.block "" url * "" * if FUNCTION format.eprint eprint empty "" archivePrefix empty "" archivePrefix "arXiv" = new.block " " eprint * " " * new.block " " eprint * " " * if if if FUNCTION format.doi doi empty "" " " doi * " " * if FUNCTION format.pid doi empty eprint empty ur...

  3. [3]

    o lzer , G. and Fritsch , M. and Deutsch , M. and H \

    thebibliography [1] 20pt to REFERENCES 6pt =0pt \@twocolumntrue 12pt -12pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key o...

    Pith/arXiv arXiv 2017