pith. sign in

arxiv: 2604.23185 · v1 · submitted 2026-04-25 · 🌌 astro-ph.HE · astro-ph.SR

The line modulations of H-like Fe, Ca, Ar, and S observed with XRISM/Resolve in Cyg X-3

Tomohiro Hakamata , Hirokazu Odaka , Ryota Tomaru , Hironori Matsumoto , Taiki Kawamuro , Ralf Ballhausen , Tim Kallman , Daiki Miura
show 8 more authors
Hiroya Yamaguchi Teruaki Enoto Natalie Hell Shunji Kitamoto Hiroshi Nakajima Shin Watanabe Shinya Yamada Kazutaka Yamaoka
This is my paper

Pith reviewed 2026-05-08 07:33 UTC · model grok-4.3

classification 🌌 astro-ph.HE astro-ph.SR
keywords Cygnus X-3stellar windX-ray linesDoppler modulationmass-loss rateXRISMWolf-Rayet starH-like ions
0
0 comments X

The pith

Different Doppler phase offsets of iron, calcium, argon and sulfur lines reveal their distinct locations in the stellar wind of Cygnus X-3.

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

The paper reports measurements of the orbital modulations in the Doppler shifts of the Lyα lines from H-like Fe, Ca, Ar, and S in Cygnus X-3 using the XRISM Resolve instrument. The phase offsets of these modulations increase with decreasing atomic number, showing that iron ions are located closest to the compact object while sulfur ions extend farther across the system. Comparing the observed trends to a model of the ultraviolet radiation-driven wind from the companion Wolf-Rayet star allows an estimate of the wind's mass-loss rate. This work demonstrates how line modulations can probe the ionization and density structure of the stellar wind in X-ray binaries.

Core claim

The Doppler shifts of the H-like Lyα lines for Fe, Ca, Ar, and S show phase offsets of 0.04, 0.09, 0.11, and 0.17 respectively relative to the compact object's orbital motion. This trend indicates that Fe ions are concentrated near the compact object, S ions are distributed across the binary, and Ca and Ar are intermediate. A calculation using a UV-accelerated stellar wind model reproduces this qualitative trend for a mass-loss rate of approximately 5×10^{-6} to 1×10^{-5} solar masses per year.

What carries the argument

Orbital-phase dependent Doppler modulations of the Lyα complexes from H-like ions, which serve as tracers of the radial distribution of each ion species in the accelerating stellar wind.

If this is right

  • H-like Fe ions concentrate near the compact object while H-like S ions are distributed more broadly across the binary.
  • The stellar wind is primarily accelerated by ultraviolet radiation from the Wolf-Rayet star.
  • The mass-loss rate of the stellar wind is in the range of 5×10^{-6} to 1×10^{-5} solar masses per year.
  • Inhomogeneous structures such as an accretion wake or bow shock are suggested by the misalignment of absorption peaks from phase 0.5.

Where Pith is reading between the lines

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

  • This ion-tracing method using multiple species could be applied to map wind structures in other high-mass X-ray binaries.
  • The derived mass-loss rate constrains the amount of material available for accretion onto the compact object.
  • Future observations with higher spectral resolution could separate contributions from the main wind flow and bow shock structures.

Load-bearing premise

The ultraviolet-driven stellar wind model with its assumed acceleration law and ionization balance accurately captures the spatial distributions of the ions without major systematic effects from accretion-related structures.

What would settle it

An independent measurement of the Wolf-Rayet star's mass-loss rate, for example through radio free-free emission or optical line profiles, that falls significantly outside the range 5×10^{-6} to 1×10^{-5} solar masses per year.

Figures

Figures reproduced from arXiv: 2604.23185 by Daiki Miura, Hirokazu Odaka, Hironori Matsumoto, Hiroshi Nakajima, Hiroya Yamaguchi, Kazutaka Yamaoka, Natalie Hell, Ralf Ballhausen, Ryota Tomaru, Shin Watanabe, Shinya Yamada, Shunji Kitamoto, Taiki Kawamuro, Teruaki Enoto, Tim Kallman, Tomohiro Hakamata.

Figure 1
Figure 1. Figure 1: Light curve in the 2–12 keV band observed by XRISM/Resolve. The selected event grade is Hp. The lower horizontal axis shows the elapsed time in seconds from MJD 60393. The upper horizontal axis shows the orbital phase, which increases by +1 for each period. The green, orange, yellow, blue, magenta, skyblue, red, and brown crosses represent the observed count rates in phase 0.000–0.125, 0.125–0.250, 0.250–0… view at source ↗
Figure 2
Figure 2. Figure 2: Spectra of Fe Lyα lines. In each panel, the upper and lower parts show the observed spectrum and the residual, respectively. The crosses represent the observed spectra. The correspondence between each color and phase is the same as in view at source ↗
Figure 3
Figure 3. Figure 3: Spectra of Ca Lyα lines, similar to those in view at source ↗
Figure 4
Figure 4. Figure 4: Spectra of Ar Lyα lines, similar to those in view at source ↗
Figure 5
Figure 5. Figure 5: Spectra of S Lyα lines, similar to those in view at source ↗
Figure 6
Figure 6. Figure 6: Orbital modulations of Lyα1 lines. The orbital modulations of Fe, Ca, Ar, and S are shown from left to right, with those of the emission line velocities, widths, and normalizations (black), and the absorption line strengths (brown) represented from top to bottom. In each panel, the crosses and dashed lines represent the observed values. The purple and green lines in the top four panels represent the calcul… view at source ↗
Figure 7
Figure 7. Figure 7: Analysis results of the phase difference using the CCF. In the upper four panels, the φt for Fe, Ca, Ar, and S, along with their corresponding CCF values measured from the best-fit values of the emission line velocities, are presented. In the bottom four panels, the error distributions of φdiff obtained by the Monte Carlo method were shown by the black solid lines. The red dashed lines represent the φdiff … view at source ↗
Figure 8
Figure 8. Figure 8: φdiff of each element. The confidence interval of the error bar is 90%. a model including absorption lines for the S Lyα lines, but their analysis still yielded a velocity amplitude smaller than that derived from XRISM. This is probably because the orbital phase was di￾vided into only four segments due to the lack of photon statistics. Additionally, our analysis using a two-line component model of Lyα1 and… view at source ↗
Figure 9
Figure 9. Figure 9: Grid definition for the CLOUDY calculation for φ = 0. The red and blue arrows denote the X and Z axes, respectively. The compact object located at the origin. object are blocked by the WR star and do not reach directly, re￾sulting in the negligibly small fractions of H-like Fe, Ca, Ar, and S ions. We modeled the plasma density in each grid based on the CAK model. In the CAK model, the stellar wind velocity… view at source ↗
Figure 10
Figure 10. Figure 10: Calculated nion(rs) distribution on the orbital surface. From left to right, the density distributions of H-like Fe, Ca, Ar, and S are shown, and from top to bottom, the calculation results for the mass loss rates of 1, 2, 5, 10, 20×10−6 M⊙ yr−1 are presented. The green solid line represents the Roche lobe. The green and magenta arrows indicate the lines of sight at phases 0.0 and 0.5, respectively, when … view at source ↗
Figure 11
Figure 11. Figure 11: Observed image with NuST AR/FPMA in the energy band of 20– 79 keV. The extended structure observed immediately to the right of Cyg X-3 is the stray light. The green and yellow circles represent the region from which we extracted spectra of Cyg X-3 and background, respectively. to FPMB (constant) to account for a 5% systematic calibration difference in normalization between the two modules (Harrison et al.… view at source ↗
Figure 12
Figure 12. Figure 12: Observed spectra by NuST AR. The upper and lower parts show the observed spectrum and the residual, respectively. The black crosses represent the observed spectra by NuST AR/FPMA and FPMB. The red solid lines represent the best-fit model. The red dotted lines represent each component view at source ↗
read the original abstract

Cygnus X-3, hosting a Wolf-Rayet (WR) star whose dense wind produces various spectral lines due to photoionization by X-rays from a compact object, provides an ideal laboratory for studying wind dynamics and density structure. We measured the orbital modulations of the Fe, Ca, Ar, and S Ly$\alpha$ lines observed with the X-ray microcalorimeter (Resolve) onboard the $XRISM$, taking account of both emission and absorption lines of the Ly$\alpha$ complexes. The modulations of Doppler shifts of the Fe, Ca, Ar, and S Ly$\alpha$ lines showed amplitudes of 500 km s$^{-1}$ and phase offsets of 0.04, 0.09, 0.11, and 0.17, respectively, in units of an orbital period (4.8 hours) relative to the orbital motion of the compact object. This result indicated that H-like Fe most closely follows the compact object's motion. The line widths ranged from 400 to 1000 km s$^{-1}$. The intensities of both emission and absorption lines reached their minima around orbital phase 0.0 and their maxima around phase 0.5. The absorption peaks, however, did not align exactly with phase 0.5, suggesting the inhomogeneous structures such as an accretion wake and/or a bow shock. We compared the observed modulations with calculations based on a stellar wind model, accelerated by ultraviolet radiation from the WR star. One of the calculations qualitatively reproduced the observed trend that H-like Fe ions were concentrated near the compact object, whereas H-like S was distributed across the binary system, with H-like Ca and Ar showing intermediate spatial distributions. From this comparison, we estimated that a mass-loss rate of the stellar wind was approximately $5 \times 10^{-6}$-$1 \times 10^{-5}\ M_{\odot}\ {\rm yr}^{-1}$.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper reports XRISM/Resolve observations of orbital modulations in the Lyα lines of H-like Fe, Ca, Ar, and S in Cyg X-3. It measures Doppler amplitudes of ~500 km s^{-1} with phase offsets 0.04, 0.09, 0.11, and 0.17 (relative to the compact object) for Fe through S, line widths of 400–1000 km s^{-1}, and intensity minima near phase 0.0 for both emission and absorption components. Comparison to a UV-driven stellar wind model shows that one calculation qualitatively reproduces the trend of Fe ions concentrated near the compact object and S distributed across the system; from this the authors infer a Wolf-Rayet mass-loss rate of 5×10^{-6}–1×10^{-5} M_⊙ yr^{-1}.

Significance. The directly measured phase offsets and intensity variations provide new constraints on the spatial distribution of ionized species in the wind. If the mass-loss rate could be placed on a quantitative footing with reported uncertainties and degeneracy checks, the result would be useful for models of wind accretion and ionization in this high-mass X-ray binary. The current inference, however, rests on a single qualitative model match.

major comments (2)
  1. [model comparison and abstract] The mass-loss rate range is obtained by tuning the model normalization to match the observed modulation amplitudes and phase offsets (abstract and model-comparison paragraph). No χ², residual analysis, or parameter scan is presented, so it is unclear whether other rates or modest changes to the β-law or ionization balance yield equally plausible matches within the reported 400–1000 km s^{-1} line widths.
  2. [discussion of inhomogeneous structures] The UV-driven wind model used to convert the observed ordering of phase offsets into a density scale does not include the accretion wake or bow-shock structures that the data themselves suggest (intensity peaks not exactly at phase 0.5). This omission is load-bearing for the claim that the model spatial distributions are sufficiently accurate to yield a quantitative mass-loss rate.
minor comments (2)
  1. [abstract] The abstract states that both emission and absorption lines reach minima at phase 0.0, but the precise definition of the absorption-line intensity (equivalent width or peak depth) is not clarified in the provided text.
  2. [results on line widths] Line-width values are given as a range (400–1000 km s^{-1}) without specifying which ion corresponds to which end of the range or whether the widths are resolved or instrumental.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major comment point by point below and have revised the manuscript to improve clarity on the model comparison and limitations.

read point-by-point responses

  1. Referee: The mass-loss rate range is obtained by tuning the model normalization to match the observed modulation amplitudes and phase offsets (abstract and model-comparison paragraph). No χ², residual analysis, or parameter scan is presented, so it is unclear whether other rates or modest changes to the β-law or ionization balance yield equally plausible matches within the reported 400–1000 km s^{-1} line widths.

    Authors: We agree that the comparison is qualitative and that no formal χ² analysis or exhaustive parameter scan is presented. The mass-loss rate range is estimated from the normalization that reproduces the observed trend in phase offsets and amplitudes across ions. We have revised the abstract and model-comparison section to explicitly describe the estimate as qualitative, to note the absence of a full scan, and to discuss possible degeneracies with the β-law and ionization balance within the observed line widths. A quantitative fit would require an expanded model grid that was beyond the scope of this work. revision: partial

  2. Referee: The UV-driven wind model used to convert the observed ordering of phase offsets into a density scale does not include the accretion wake or bow-shock structures that the data themselves suggest (intensity peaks not exactly at phase 0.5). This omission is load-bearing for the claim that the model spatial distributions are sufficiently accurate to yield a quantitative mass-loss rate.

    Authors: The referee is correct that the adopted UV-driven wind model is smooth and omits explicit accretion wake or bow-shock structures, which are suggested by the intensity maxima not aligning exactly with phase 0.5. We already note these possible inhomogeneities in the manuscript. The phase offsets of the Doppler shifts, however, primarily constrain the radial distribution of ions through the ionization structure, which the model captures. Azimuthal structures mainly affect intensity rather than the velocity amplitudes and phase offsets used for the mass-loss estimate. We have added a dedicated paragraph in the discussion clarifying this distinction and the model's limitations while maintaining that the radial trend still provides a useful constraint. revision: yes

Circularity Check

0 steps flagged

No significant circularity; mass-loss rate inferred via standard model comparison

full rationale

The paper measures observed phase offsets (0.04–0.17) and Doppler amplitudes (~500 km s⁻¹) for the four H-like ions, then compares these to outputs of an external UV-driven stellar wind model that computes ionization balance and spatial distributions as a function of mass-loss rate. One model run is stated to qualitatively reproduce the observed ordering (Fe concentrated near the compact object, S distributed across the system). The mass-loss range is then read off from that match. This is parameter inference against an independent physical model; no equation reduces the estimated rate to an input by construction, no self-citation supplies the core model, and the result is not relabeled as a prediction. The derivation remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The mass-loss estimate depends on an external UV-driven wind model whose ionization and velocity laws are taken as given; no new entities are postulated.

free parameters (1)
  • mass-loss rate normalization
    The single scalar adjusted to match the observed modulation amplitudes of the four ions.
axioms (2)
  • domain assumption The wind is accelerated solely by UV line driving from the WR star with a standard beta-velocity law.
    Invoked when comparing observed ion locations to the model calculation.
  • domain assumption Photoionization equilibrium is reached instantaneously and is dominated by the compact-object X-rays.
    Underlying the assumption that line strengths trace local density and ionization parameter.

pith-pipeline@v0.9.0 · 5740 in / 1483 out tokens · 89864 ms · 2026-05-08T07:33:34.766267+00:00 · methodology

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

2 extracted references

  1. [1]

    2024, A&A, 692, A188 Publications of the Astronomical Society of Japan (2025), Vol

    Abalo, L., et al. 2024, A&A, 692, A188 Publications of the Astronomical Society of Japan (2025), Vol. 00, No. 0 13 Fig

    2024

  2. [2]

    The NuSTAR Data Analysis Software Gu ide V ersion 1.9.7

    01 ∗ The absorbed X-ray flux in 2-100 keV band. Note: The errors are 90 % confidence interval. C and degree of freedom are 2493.78 and 837, respectively. Antokhin, I. I., Cherepashchuk, A. M., Antokhina, E. A., & Ta tarnikov, A. M. 2022, ApJ, 926, 123 Antokhin, I. I., & Cherepashchuk, A. M., Antokhina, E. A. 201 9, ApJ, 871, 244 Blondin, J. M., Kallman, T. ...

    2022