arxiv: 2604.16196 · v1 · submitted 2026-04-17 · 🌌 astro-ph.HE · astro-ph.GA

Recognition: unknown

A XRISM Study of Highly Ionized Iron Emission Lines from the Low-Eddington-ratio AGN in NGC 7213

Kaito Murakami , Taiki Kawamuro , Ryota Tomaru , Hirokazu Odaka , Elias Kammoun , Shoji Ogawa , Stefano Bianchi , Hirofumi Noda

show 5 more authors

Claudio Ricci Yuichi Terashima Yoshihiro Ueda Satoshi Yamada Hironori Matsumoto

Authors on Pith no claims yet

Pith reviewed 2026-05-10 07:17 UTC · model grok-4.3

classification 🌌 astro-ph.HE astro-ph.GA

keywords XRISMNGC 7213active galactic nucleiiron emission linesphotoionizationcollisional ionizationEddington ratioX-ray spectroscopy

0 comments

The pith

XRISM spectra of NGC 7213 show that highly ionized iron lines are explained by a single ionization zone without needing an extra component.

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

The paper uses XRISM Resolve and NuSTAR data to study the Fe XXV Heα and Fe XXVI Lyα lines in the low-Eddington-ratio AGN NGC 7213. Gaussian fits first suggest the lines may have different velocity widths, prompting tests of one-zone versus two-zone models under both photoionization and collisional ionization. The results indicate that an additional ionized component is not required by the data, and the observations cannot decide which ionization process dominates. Comparison with M 81* at even lower Eddington ratio further suggests the gas density producing these lines declines as the accretion rate drops.

Core claim

Gaussian fits indicate possible velocity widths of 790 km s^{-1} for Fe XXV and 2610 km s^{-1} for Fe XXVI, with the Heα resonance and forbidden lines each at roughly 0.5–0.6 × 10^{-5} ph s^{-1} cm^{-2} while the intercombination lines remain undetected. One- and two-zone photoionized and collisionally ionized models applied to the 2–60 keV continuum and iron-K band both show that a second zone is not significantly needed. The data leave open whether photoionization or collisional ionization produces the lines, and the line ratios are hard to match if the intercombination lines are truly suppressed. Direct comparison with M 81* reveals lower density in the iron-emitting gas of NGC 7213, hint

What carries the argument

One- and two-zone photoionization and collisional ionization models fitted to the broadband continuum and the Resolve iron-K spectrum, using Gaussian line profiles to test velocity widths and intensity ratios.

If this is right

A single ionization zone is sufficient to reproduce the observed Fe XXV and Fe XXVI lines.
Current data cannot distinguish whether photoionization or collisional ionization dominates.
The Fe XXV intensity pattern with suppressed intercombination lines is difficult to produce in either model.
The density of gas emitting highly ionized iron decreases as Eddington ratio drops, as shown by the NGC 7213 versus M 81* comparison.

Where Pith is reading between the lines

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

If the density trend holds, the iron-emitting gas in lower-Eddington systems may lie at larger radii or in more tenuous regions of the accretion flow.
Higher-resolution spectra could confirm whether the line-width difference is real and thereby tighten constraints on the geometry of the emitting plasma.
The difficulty reproducing the observed line ratios may point to missing atomic data or additional processes such as resonant scattering not included in the current models.

Load-bearing premise

The apparent difference in velocity widths between the Fe XXV and Fe XXVI lines is a real physical effect rather than an artifact of noise or fitting degeneracies.

What would settle it

A deeper XRISM or future microcalorimeter observation that measures identical velocity widths for Fe XXV and Fe XXVI or detects the intercombination lines at strengths predicted by the models would rule out the need to invoke separate physical zones or suppressed ratios.

Figures

Figures reproduced from arXiv: 2604.16196 by Claudio Ricci, Elias Kammoun, Hirofumi Noda, Hirokazu Odaka, Hironori Matsumoto, Kaito Murakami, Ryota Tomaru, Satoshi Yamada, Shoji Ogawa, Stefano Bianchi, Taiki Kawamuro, Yoshihiro Ueda, Yuichi Terashima.

**Figure 1.** Figure 1: Phenomenological fit results for the Resolve spectrum (§3.2). The left panel shows the fit result in which the velocity widths and redshifts of the Fe Heα and Fe Lyα lines are tied together, while the right panel shows the one where these parameters are allowed to vary independently. Black crosses indicate the observed Resolve data, and the sum of all model components is shown as a pink line. The lower pan… view at source ↗

**Figure 2.** Figure 2: Input SED model of nthcomp (pink) used for generating the pion table model. In the lower panel, residuals, χ = (data – model)/(1σ error), are shown. Black, gray, blue, and green crosses represent the Resolve, Xtend, FPMA, and FPMB data, respectively. Alt text: A panel showing the input spectral energy distribution model of nthcomp used to generate the pion table model. Γ = 1.8 in the subsequent photoioniza… view at source ↗

**Figure 3.** Figure 3: Top left: Resolve spectrum (black crosses) and the best-fitting model (pink solid line) adopting the single pion component. The pink dashed lines represent the cutoff power-law component, narrow and broad Fe Kα lines, and Fe Kβ line. The blue dashed line corresponds to the pion component. The gray dashed line indicates the NXB. The lower panel shows residuals of χ = (data − model)/(1σ error). Top right: En… view at source ↗

**Figure 4.** Figure 4: Ion fractions of iron ions from Fe XVI and higher calculated with our photoionization model. The horizontal axis shows the ionization parameter, and the vertical axis shows the ionic fraction. Alt text: A plot showing ion fractions of iron ions from Fe XVI and higher calculated with a photoionization model, as a function of the ionization parameter. served line-of-sight velocity (vobs), we adopt vrot = vo… view at source ↗

**Figure 5.** Figure 5: Top left: The best-fitting model obtained with the one component bapec model. The pink dashed lines represent the cutoff power-law continuum together with the narrow and broad Fe Kα lines and the Fe Kβ line. The blue and gray dashed lines correspond to the bapec and NXB components, respectively. The sum of all these components is shown as the pink solid line. Top right: Enlarged view of the iron emission l… view at source ↗

**Figure 6.** Figure 6: Spectra in the Fe Heα region obtained from our pion model for column densities ranging from 1021 to 1025 cm−2 , normalized to the intensity of the forbidden line at 1021 cm−2 . The left panel assumes a velocity broadening of 100 km s−1 for comparison, while the right panel assumes a velocity broadening of 800 km s−1 , which was obtained from the fit of the phenomenological model. In both panels, the ioniza… view at source ↗

**Figure 7.** Figure 7: Line ratios of Fe Heα as a function of column density ranging from 1021 to 1025 cm−2 from our pion model. The left, middle, and right panels show x/w, y/w, and z/w, respectively. In each panel, both the case with a velocity broadening of 800 km s−1 , obtained from the fit of the phenomenological model, and the case with 100 km s−1 are shown for comparison. The ionization parameters of 3.4 in log units, obt… view at source ↗

read the original abstract

We present an analysis of XRISM and NuSTAR data obtained for the nearby low-Eddington active galactic nucleus NGC 7213. Our goal is to examine whether its He-like and H-like iron emission lines can be reproduced by photoionization or collisional ionization processes. Using the broad-band energy coverage of our data (2-60 keV), we first constrained the continuum shape. Then, we focused on the iron-K band in the Resolve spectrum. Gaussian fits to Fe XXV He$\alpha$ and Fe XXVI Ly$\alpha$ lines suggest that they may have different velocity widths: $v_\sigma=790^{+370}_{-240}$ km s$^{-1}$ for Fe XXV and $v_\sigma=2610^{+1700}_{-1580}$ km s$^{-1}$ for Fe XXVI. In this case, the He$\alpha$ resonance line (w) and forbidden line (z) have similar intensities of $\approx0.5$-$0.6\times10^{-5}$ ph s$^{-1}$ cm$^{-2}$, while the intercombination lines (x+y) are not significantly detected with upper limits of $\lesssim 0.2\times10^{-5}$ ph s$^{-1}$ cm$^{-2}$. Motivated by the possible difference in the line widths, we tested one- and two-zone photoionized and collisionally ionized models. Our results show that the additional ionized component is not significantly required, and the current data cannot uniquely determine whether photoionization or collisional ionization dominates. Moreover, if the Fe XXV He$\alpha$ complex implies that the weak x+y lines are suppressed relative to the w and z lines, such a structure is difficult to reproduce with either ionization model adopted. Finally, by comparing NGC 7213 with M 81$^\ast$, accreting at a much lower Eddington ratio of $\lambda_{\rm Edd}\sim 10^{-5}$, we found a decrease in the density of the gas responsible for highly ionized iron emission, which may imply that the density decreases with decreasing $\lambda_{\rm Edd}$.

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.

Desk Editor's Note private letter to a colleague

XRISM spectra of NGC 7213 leave iron-line ionization ambiguous and show a tentative density drop versus M81, but the reported width difference is marginal.

read the letter

The main things to know are that the new XRISM Resolve data on NGC 7213 do not let us decide between photoionization and collisional ionization for the highly ionized iron lines, and a comparison to M81 suggests the emitting gas density may fall as Eddington ratio decreases. The paper applies the instrument's resolution to the Fe XXV He-alpha and Fe XXVI Ly-alpha lines after fixing the continuum with combined XRISM and NuSTAR coverage. Gaussian fits give widths of 790 km/s and 2610 km/s respectively, with the He-alpha resonance and forbidden lines roughly equal in strength and the intercombination lines weak. One- and two-zone models for both ionization mechanisms are tested, and the extra zone is not required at high significance. The line ratios are noted as difficult to match in either case. The density trend rests on the Fe XXV complex and the two-object comparison at very different lambda_Edd values. This is the clearest new piece: actual high-resolution spectra plus the explicit cross-check with M81. The modeling is described plainly and the non-uniqueness result follows directly from the fits shown. The soft spot is the velocity-width difference. The 1-sigma ranges overlap substantially, so a single common width remains allowed. That reduces the motivation for splitting into two zones and adds uncertainty to how the density is tied to the Fe XXV emission. The paper already states the second component is not needed, so the central claim is not overturned, but the width contrast should be treated as tentative rather than firm. The work is aimed at X-ray spectroscopists working on low-accretion AGN. It brings fresh data to a narrow question without overclaiming. A referee can check the exact fit statistics and atomic-model assumptions, but the paper is clear and honest enough to merit that step.

Referee Report

2 major / 2 minor

Summary. The manuscript analyzes XRISM Resolve and NuSTAR data of the low-Eddington-ratio AGN NGC 7213 to study the highly ionized iron lines Fe XXV Heα and Fe XXVI Lyα. Gaussian fits indicate possible different velocity widths for the lines. One- and two-zone photoionization and collisional ionization models are tested, leading to the conclusion that an additional ionized component is not significantly required and the data cannot uniquely determine the dominant ionization process. A comparison with M81* suggests that the gas density decreases with decreasing Eddington ratio.

Significance. If the modeling results hold after addressing statistical robustness, this work provides useful constraints on the origin of highly ionized iron emission in low-accretion-rate AGNs using XRISM's high-resolution spectroscopy. The comparison suggesting a density trend with Eddington ratio is potentially interesting for understanding gas properties across accretion states, though it inherits uncertainties from the line decomposition.

major comments (2)

[Abstract] Abstract: The reported velocity dispersions (Fe XXV: 790^{+370}_{-240} km s^{-1}; Fe XXVI: 2610^{+1700}_{-1580} km s^{-1}) have substantially overlapping 1σ ranges (roughly 550–1160 km s^{-1} vs. 1030–4310 km s^{-1}). This indicates that a single common width remains statistically allowed, weakening the motivation for two-zone models and requiring a quantitative test (e.g., F-test or posterior odds) of whether the width difference is significant before using it to guide model comparisons.
[Results and Discussion] Results and Discussion sections: The claim that 'the additional ionized component is not significantly required' and that 'the current data cannot uniquely determine whether photoionization or collisional ionization dominates' must be supported by explicit fit statistics (Δχ² values, degrees of freedom, and null probabilities) for the one-zone versus two-zone models, including any degeneracies with the weak x+y lines and continuum parameters. The density comparison with M81* should include error propagation from the line-profile decomposition uncertainties.

minor comments (2)

[Abstract] The abstract would benefit from stating the exact fit statistic (e.g., χ²/dof or C-stat) used to conclude that the second component is not required.
[Results] Notation for line intensities (e.g., 0.5–0.6×10^{-5} ph s^{-1} cm^{-2}) should be consistent with the model tables in the results section.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thoughtful and constructive comments, which help strengthen the statistical foundation of our analysis. We address each major comment below and will incorporate revisions to improve clarity and rigor.

read point-by-point responses

Referee: [Abstract] Abstract: The reported velocity dispersions (Fe XXV: 790^{+370}_{-240} km s^{-1}; Fe XXVI: 2610^{+1700}_{-1580} km s^{-1}) have substantially overlapping 1σ ranges (roughly 550–1160 km s^{-1} vs. 1030–4310 km s^{-1}). This indicates that a single common width remains statistically allowed, weakening the motivation for two-zone models and requiring a quantitative test (e.g., F-test or posterior odds) of whether the width difference is significant before using it to guide model comparisons.

Authors: We agree that the 1σ confidence intervals overlap substantially, so a common velocity width cannot be ruled out at high significance. The manuscript describes the widths as suggestive ('may have different velocity widths') rather than definitive, and the two-zone models were explored as a motivated but non-unique possibility. To address the concern rigorously, we will add an F-test (or equivalent Bayesian comparison) between single-width and two-width Gaussian models in the revised Results section, reporting the Δχ², degrees of freedom, and null-hypothesis probability. This will quantify whether the width difference justifies the additional model complexity. revision: yes
Referee: [Results and Discussion] Results and Discussion sections: The claim that 'the additional ionized component is not significantly required' and that 'the current data cannot uniquely determine whether photoionization or collisional ionization dominates' must be supported by explicit fit statistics (Δχ² values, degrees of freedom, and null probabilities) for the one-zone versus two-zone models, including any degeneracies with the weak x+y lines and continuum parameters. The density comparison with M81* should include error propagation from the line-profile decomposition uncertainties.

Authors: We acknowledge that the current text relies on qualitative statements about model preference without tabulating the quantitative fit statistics. In the revised manuscript we will explicitly report Δχ², degrees of freedom, and null probabilities for all one-zone versus two-zone comparisons (both photoionization and collisional ionization), and we will discuss parameter degeneracies involving the weak x+y lines and the underlying continuum. For the density comparison with M81*, we will propagate the uncertainties arising from the line-profile decomposition (including the velocity-width errors) into the derived densities and state the resulting range explicitly. revision: yes

Circularity Check

0 steps flagged

No circularity: standard observational fitting against external atomic models

full rationale

The paper's chain consists of continuum fitting on broadband data, Gaussian line-profile fits to Resolve spectra yielding velocity widths and fluxes, and subsequent tests of one- and two-zone photoionization/collisional-ionization models using standard external codes. None of these steps reduce a claimed result to a fitted input by construction, nor rely on self-citations whose content is itself unverified or defined by the present work. The density comparison with M81* is an external cross-check performed after independent model fits on each source. The velocity-width difference is reported directly from data with explicit error ranges; model selection follows from those fits rather than presupposing the conclusion.

Axiom & Free-Parameter Ledger

4 free parameters · 1 axioms · 0 invented entities

Analysis rests on standard X-ray spectral modeling assumptions and multiple fitted parameters for continuum shape, line components, and ionization state; no new entities postulated.

free parameters (4)

Fe XXV velocity dispersion
Fitted Gaussian width to He-alpha complex in Resolve spectrum.
Fe XXVI velocity dispersion
Fitted Gaussian width to Ly-alpha line.
ionization parameter and density in photoionized models
Adjusted to reproduce observed line intensities and ratios.
temperature and density in collisionally ionized models
Adjusted to match line strengths.

axioms (1)

domain assumption Standard assumptions in X-ray spectral modeling such as solar elemental abundances, power-law or cutoff power-law continuum, and atomic data from databases like AtomDB or SPEX.
Invoked when constraining the broadband continuum and fitting iron lines.

pith-pipeline@v0.9.0 · 5753 in / 1412 out tokens · 48238 ms · 2026-05-10T07:17:15.816246+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

51 extracted references

[1]

Arnaud, K. A. 1996, in Astronomical Society of the Pacific Conference

1996
[2]

R., McDuffie, J

Ballantyne, D. R., McDuffie, J. R., & Rusin, J. S. 2011, ApJ, 734, 112,

2011
[3]

C., McKee, C

Begelman, M. C., McKee, C. F., & Shields, G. A. 1983, ApJ, 271, 70,

1983
[4]

C., Denney, K

Bentz, M. C., Denney, K. D., Grier, C. J., et al. 2013, ApJ, 767, 149,

2013
[5]

2008, MNRAS, 389, L52,

Bianchi, S., La Franca, F., Matt, G., et al. 2008, MNRAS, 389, L52,

2008
[6]

2002, A&A, 387, 76,

Bianchi, S., & Matt, G. 2002, A&A, 387, 76,

2002
[7]

R., Drake, G

Blumenthal, G. R., Drake, G. W. F., & Tucker, W. H. 1972, ApJ, 172, 205,

1972
[8]

2018, MNRAS, 474, 1206,

Bu, D.-F., & Gan, Z.-M. 2018, MNRAS, 474, 1206,

2018
[9]

2020, A&A, 636, A73,

Duras, F., Bongiorno, A., Ricci, F., et al. 2020, A&A, 636, A73,

2020
[10]

Elitzur, M., & Ho, L. C. 2009, ApJL, 701, L91,

2009
[11]

E., Nicastro, F., & McHardy, I

Emmanoulopoulos, D., Papadakis, I. E., Nicastro, F., & McHardy, I. M. 2013, MNRAS, 429, 3439,

2013
[12]

2019, A&A, 630, A94,

Giustini, M., & Proga, D. 2019, A&A, 630, A94,

2019
[13]

N., McLaughlin, D

Gofford, J., Reeves, J. N., McLaughlin, D. E., et al. 2015, MNRAS, 451, 4169,

2015
[14]

2025, A&A, 704, A146,

Gu, L., Fukumura, K., Kaastra, J., et al. 2025, A&A, 704, A146,

2025
[15]

A., Craig, W

Harrison, F. A., Craig, W. W., Christensen, F. E., et al. 2013, ApJ, 770, 103, HI4PI Collaboration, Ben Bekhti, N., Flöer, L., et al. 2016, A&A, 594, A116,

2013
[16]

Ho, L. C. 2008, ARA&A, 46, 475,

2008
[17]

L., Awaki, H., et al

Ishisaki, Y ., Kelley, R. L., Awaki, H., et al. 2025, Journal of Astronomical

2025
[18]

S., Mewe, R., & Nieuwenhuijzen, H

Kaastra, J. S., Mewe, R., & Nieuwenhuijzen, H. 1996, in UV and X-ray Spectroscopy of Astrophysical and Laboratory Plasmas, ed. K. Yamashita & T. Watanabe, 411–414

1996
[19]

2001, ApJS, 133, 221,

Kallman, T., & Bautista, M. 2001, ApJS, 133, 221,

2001
[20]

2025, XRISM/Resolve reveals the complex iron structure of NGC 7213: Evidence for radial strat- ification between inner disk and broad-line region

Kammoun, E., Kawamuro, T., Murakami, K., et al. 2025, XRISM/Resolve reveals the complex iron structure of NGC 7213: Evidence for radial strat- ification between inner disk and broad-line region

2025
[21]

S., Netzer, H., et al

Kaspi, S., Smith, P. S., Netzer, H., et al. 2000, ApJ, 533, 631,

2000
[22]

R., & Pounds, K

King, A. R., & Pounds, K. A. 2003, MNRAS, 345, 657,

2003
[23]

P., Reeves, J

Lobban, A. P., Reeves, J. N., Porquet, D., et al. 2010, MNRAS, 408, 551,

2010
[24]

2009, Landolt Börnstein, 4B, 712,

Lodders, K., Palme, H., & Gail, H.-P. 2009, Landolt Börnstein, 4B, 712,

2009
[25]

S., & Kallman, T

Mehdipour, M., Kaastra, J. S., & Kallman, T. 2016, A&A, 596, A65,

2016
[26]

S., & Raassen, A

Mehdipour, M., Kaastra, J. S., & Raassen, A. J. J. 2015, A&A, 579, A87,

2015
[27]

S., Eckart, M

Mehdipour, M., Kaastra, J. S., Eckart, M. E., et al. 2025, A&A, 699, A228,

2025
[28]

M., Behar, E., Awaki, H., et al

Miller, J. M., Behar, E., Awaki, H., et al. 2025, ApJL, 985, L41,

2025
[29]

1994, ApJL, 428, L13,

Narayan, R., & Yi, I. 1994, ApJL, 428, L13,

1994
[30]

Nayakshin, S., & Kallman, T. R. 2001, ApJ, 546, 406,

2001
[31]

2015, ARA&A, 53, 365,

Netzer, H. 2015, ARA&A, 53, 365,

2015
[32]

2025, PASJ,

Noda, H., Mori, K., Tomida, H., et al. 2025, PASJ,

2025
[33]

2020, MNRAS, 494, 3616,

Nomura, M., Ohsuga, K., & Done, C. 2020, MNRAS, 494, 3616,

2020
[34]

Peterson, B. M. 1997, An Introduction to Active Galactic Nuclei

1997
[35]

Phillips, M. M. 1979, ApJL, 227, L121,

1979
[36]

2010, Space Science Reviews, 157, 103–134, Ramos Almeida, C., & Ricci, C

Porquet, D., Dubau, J., & Grosso, N. 2010, Space Science Reviews, 157, 103–134, Ramos Almeida, C., & Ricci, C. 2017, Nature Astronomy, 1, 679,

2010
[37]

R., Fabian, A

Ross, R. R., Fabian, A. C., & Young, A. J. 1999, MNRAS, 306, 461,

1999
[38]

2022, The Astrophysical Journal, 926, 209,

Shi, F., Zhu, B., Li, Z., & Yuan, F. 2022, The Astrophysical Journal, 926, 209,

2022
[39]

Starling, R. L. C., Page, M. J., Branduardi-Raymont, G., et al. 2005, MNRAS, 356, 727,

2005
[40]

2013, Publications of the Astronomical Society of Japan, 65,

Takeuchi, S., Ohsuga, K., & Mineshige, S. 2013, Publications of the Astronomical Society of Japan, 65,

2013
[41]

2025, PASJ, 77, S1,

Tashiro, M., Kelley, R., Watanabe, S., et al. 2025, PASJ, 77, S1,

2025
[42]

2023, MNRAS, 523, 3441,

Tomaru, R., Done, C., Odaka, H., & Tanimoto, A. 2023, MNRAS, 523, 3441,

2023
[43]

2019, MNRAS, 490, 3098,

Tomaru, R., Done, C., Ohsuga, K., Nomura, M., & Takahashi, T. 2019, MNRAS, 490, 3098,

2019
[44]

N., et al

Tombesi, F., Cappi, M., Reeves, J. N., et al. 2010, A&A, 521, A57,

2010
[45]

J., et al

Trakhtenbrot, B., Ricci, C., Koss, M. J., et al. 2017, MNRAS, 470, 800,

2017
[46]

2021, ApJ, 914, 62,

Waters, T., Proga, D., & Dannen, R. 2021, ApJ, 914, 62,

2021
[47]

Woo, J.-H., & Urry, C. M. 2002, ApJ, 579, 530,

2002
[48]

M., Behar, E., et al

Xiang, X., Miller, J. M., Behar, E., et al. 2025, ApJL, 988, L54, Xrism Collaboration, Audard, M., Awaki, H., et al. 2025, Nature, 641, 1132,

2025
[49]

2018, MNRAS, 475, 1190,

Yan, Z., & Xie, F.-G. 2018, MNRAS, 475, 1190,

2018
[50]

2007, Astrophysical Journal, 669, 830,

Young, A., Nowak, M., Markoff, S., Marshall, H., & Caneares, C. 2007, Astrophysical Journal, 669, 830,

2007
[51]

2004, ApJ, 612, 724, —

Yuan, F., & Narayan, R. 2004, ApJ, 612, 724, —. 2014, ARA&A, 52, 529,

2004