{"id":"00f4544f-3491-41ec-aa7b-4903ce624726","arxiv_id":"2506.05273","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Spectra from XRISM suggest the five fast X-ray outflows in PDS 456 are stratified by ionization, with a weak and not statistically significant trend v_out proportional to the ionization parameter to the power 0.14, plus a thermally unstable soft X-ray component.","lead":"Using new XRISM X-ray spectra of the quasar PDS 456, the authors fit six fast outflow components with a self-consistent photoionization model and tested all possible orderings of the absorbing layers. They find a tentative pattern in which slower outflows sit closer to the black hole, and argue the soft X-ray absorber is thermally unstable.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The headline vout–xi correlation is not statistically significant (p≈0.1, n=5) and is model-dependent, yet the abstract presents it as an established stratification result.","rationale":"The reader's weakest assumption (CF=1 and order-to-radius mapping) is important for the ordering conclusion, but the central quantitative claim—the vout–ξ scaling relation—is computed from the PION parameters and is stated to be stable across the tested sequences. The more direct threat is that the correlation is marginal: with five points, p~0.1, and scatter visible in Figure 2, the 'stratified ionization structure' is not established at conventional significance. The relation also appears only with PION; XABS and PHASE yield no correlation, so the result is model-dependent. A leave-one-out or permutation test would show whether the trend is robust. If it is not, the abstract's wording should be softened. I therefore treat the statistical fragility as the load-bearing concern, while agreeing that the covering-factor/ordering issue is a serious secondary concern. The appropriate verdict remains conditional: the analysis is valuable but the headline claim needs tempering and robustness checks.","tokens_in":28128,"tokens_out":8963,"duration_ms":106305,"concrete_test":"Recompute the vout–ξ correlation for the five hard X-ray UFOs after (a) removing UFO5, (b) removing UFO1, and (c) applying a permutation test on the five PION points to obtain an exact two-sided p-value. If the slope flips sign, drops below 1σ, or the permutation p-value exceeds 0.05, the claimed stratification is driven by one or two components, and the abstract's 'reveal' should be downgraded to 'tentatively suggest'.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim of a stratified ionization structure rests on the correlation vout∝ξ^(0.14±0.04) derived from only five hard X-ray UFO components. Table 2 reports Spearman ρ=0.70 (p=0.13) and Pearson ρ=0.81 (p=0.09) for the PION results, and the text explicitly states that none of the fits meet p<0.05. The abstract nevertheless states 'Our results reveal a stratified ionization structure, characterized by a relation...', overstating the statistical support. The relation is also model-dependent: XABS and PHASE fits show no correlation (|ρ|<0.5, p>0.6). The ODR slope uncertainty (0.04) is a regression standard error that can be heavily leveraged by the extreme points (UFO1 and UFO5); with n=5, removing a single component may destroy the trend. Moreover, the five points are not independent measurements but outputs of a joint spectral fit with shared continuum parameters, so the effective evidence is weaker than nominal p-values suggest. If the correlation does not survive simple robustness checks, the central physical interpretation of a single stratified outflow is unsupported.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes coordinated XRISM, XMM-Newton, and NuSTAR observations of the luminous quasar PDS 456, modeling six ultrafast outflow (UFO) components with the self-consistent photoionization code PION in SPEX. By permuting the order of the six PION absorbers in the multiplicative spectral model, the authors investigate whether screening effects constrain the relative line-of-sight positions of the components. They report a power-law correlation between outflow velocity and ionization parameter for the five hard X-ray UFOs, vout ∝ ξ^(0.14±0.04), and use this, together with C-statistic histograms and a limited Bayesian analysis, to argue for a stratified outflow in which slower components lie closer to the supermassive black hole. The soft X-ray UFO is found to be thermally unstable in all orderings, though its radial location remains ambiguous. The paper also presents simulations of future XRISM (gate-valve open) and NewAthena observations to show how the order of absorbers could be better constrained.","tokens_in":28413,"tokens_out":4406,"duration_ms":46599,"significance":"If the stratified structure and the vout–ξ correlation are robust, they would constitute a rare, direct observational probe of UFO stratification and acceleration in a prototypical AGN, complementing the line-profile diagnostics of Gallo et al. (2023) and providing constraints on driving mechanisms. The systematic exploration of all 720 order permutations with PION, the use of Bayesian evidence for three representative orders, and the forward-modeling of future instruments are methodologically valuable and go beyond the earlier XRISM collaboration paper. The paper is honest in many places—notably in Sections 4.2 and 4.3 and in the conclusions—about the tentative nature of the sequence and the weakness of the correlation. However, the abstract and several interpretive passages state the stratification as an established result, which is not supported by the paper's own statistics. The work is a useful contribution, but its headline claims need to be brought into line with the evidence.","major_comments":[{"comment":"The abstract's claim that the results \"reveal a stratified ionization structure, characterized by a relation vout ∝ ξ^(0.14±0.04)\" overstates the statistical support. Table 2 reports Spearman ρ=0.70 (p=0.13) and Pearson ρ=0.81 (p=0.09) for n=5, and the text in Section 4.3 explicitly states that \"none of the fits meet the conventional threshold for statistical significance (p<0.05)\". The conclusion section properly uses \"possibly stratified\" and \"tentative trend\", so the abstract is inconsistent with the body. The central physical interpretation of a single stratified outflow rests on this correlation, so the abstract should be reworded to match the caveated language used in the conclusions.","section":"Abstract; Section 4.3, Table 2"},{"comment":"The statistical preference for UFO6's layer is not as strong as claimed. The text states that for UFO6 \"the best-fit statistics occur simultaneously when it is placed at the second and third layers, with a ΔC−stat = 2 improvement over the second-best position\" and that \"no specific sequential combination emerges as statistically superior to the others\". The subsequent choice of the \"layer with the highest occurrence within the best C-stat bin\" is a histogram heuristic, not a model comparison, and a ΔC=2 difference is weak evidence. The abstract's phrase \"statistically favored — based on the evidence from both the C-statistic and Bayesian analysis\" is therefore an overinterpretation, especially because the Bayesian analysis in Section 4.1.5 computed evidence for only three of the 720 permutations (PION1, PION2, PION3) and cannot validate per-layer preferences across the full permutation space.","section":"Section 3.2.3; Section 4.1.5"},{"comment":"The assumption CF=1 (full covering) is load-bearing for the physical interpretation of screening and radial ordering, but it is only weakly tested. When covering factors were freed, only UFO6 was well constrained (CF=0.88±0.02); for UFO1–5 only lower limits (CF>0.14 and CF>0.5) were obtained, with a total improvement of ΔC−stat/ν=15/6. As the text acknowledges, this \"prevent[s] us from definitively distinguishing between scenarios of consecutive shells and clumpy outflows\". Since the screening-based ordering analysis assumes that the multiplicative order corresponds one-to-one to radial order, this assumption needs to be stated as a caveat in the abstract and in Section 3.2, or the ordering conclusions should be made conditional on the full-covering scenario.","section":"Section 4.1.4"},{"comment":"The vout–ξ correlation is model-dependent and based on a very small, non-independent sample. The XABS and PHASE fits show no correlation (|ρ|<0.5, p>0.6), and the PION relation is a fit to model output rather than a model-independent measurement. Moreover, the five PION points come from a single joint spectral fit with shared continuum parameters, so the effective number of independent measurements is smaller than n=5. The paper notes the model dependence but still uses the relation as evidence for a single stratified outflow. A robustness test—for example, omitting each of the five points in turn and recomputing the ODR slope, or showing the influence of the assumed uncertainties on the slope error—should be included before presenting the relation as a kinematic signature.","section":"Section 4.3; Figure 2"}],"minor_comments":[{"comment":"\"examing\" should be \"examining\".","section":"Section 5, first bullet"},{"comment":"The heading \"Figure B1-B1\" should read \"B1–B3\".","section":"Appendix 2"},{"comment":"The statement that \"any sequential combination satisfying these three conditions results in at most a ΔC−stat = 2 difference\" would be more informative if the number of such combinations were reported, since this directly indicates the degree of degeneracy among the 720 permutations.","section":"Section 3.2.3"},{"comment":"The p-values are quoted to two decimals; it would be useful to also state the effective number of independent measurements, given that the five points are derived from a joint fit with shared continuum parameters.","section":"Table 2"},{"comment":"The Bayesian evidence values are given as logZ ~ −516, −513, −508, and the text then reports log B3,2 = 5 and log B2,1 = 3. It would be clearer to define explicitly that logB refers to the difference in log-evidence, to avoid confusion between evidence and posterior odds.","section":"Section 4.1.5"}],"recommendation":"major_revision","confidential_remarks":"The paper is from a large, well-established collaboration and presents an unusually detailed permutation analysis of multi-component PION absorption in PDS 456. The main weaknesses are rhetorical rather than computational: the abstract and several interpretive statements present tentative results as established, while the body contains the appropriate caveats. A revision that aligns the abstract and the claims with the reported p-values and ΔC values, and that explicitly conditions the ordering conclusions on the full-covering assumption, would make the paper a valuable contribution. I see no evidence of a fundamental error that would require rejection, but the load-bearing statistical overreach should be fixed before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: this is a careful, genuinely useful analysis of new XRISM data that resolves five Fe-K UFOs in PDS 456, and the permutation-based ordering method is a real step forward. But the headline claim—a stratified outflow with vout∝ξ^0.14—is weaker than the abstract implies. The text itself admits none of the correlations reach p<0.05 (Spearman p=0.13, Pearson p=0.09, n=5), and the trend disappears when the same data are modeled with XABS or PHASE. So the abstract oversells a tentative hint. That mismatch is the main soft spot.\n\nWhat is genuinely new: the XRISM Resolve spectrum decomposed into six PION components, with a 720-permutation C-stat ordering analysis and Bayesian evidence for three representative sequences. That is a credible methodological contribution to multi-component absorber modeling. The paper also does good housekeeping: it tests variable abundances, checks whether residuals drive the sequence, tries free covering factors, and simulates future XRISM/NewAthena data. The finding that highly ionized absorbers are position-insensitive is well supported and useful.\n\nSoft spots, in proportion. First, the correlation is not statistically significant and is model-dependent; the five points are outputs of a joint spectral fit, not independent measurements, so the effective evidence is weaker than the nominal p-values. Second, the preferred position of UFO6 has a ΔC=2 ambiguity between layers 2 and 3, and the paper resolves it by counting occurrences in the best C-stat bin—reasonable but fragile. Third, Bayesian evidence covers only three of 720 sequences, so it does not validate the global ordering. Fourth, the full-covering (CF=1) assumption is load-bearing; the free-covering test constrains only UFO6, leaving clumpy-outflow alternatives viable. The paper acknowledges these limits in the body, which is to its credit, but the abstract reads as if they are settled.\n\nWho this is for: anyone working on AGN outflows, PION spectral modeling, or XRISM science. It deserves a serious referee; the weaknesses are addressable by softening the abstract, adding jackknife or leave-one-out robustness checks on the correlation, and being explicit that Bayesian evidence covers only selected sequences. I would send this to review.","headline":"Solid XRISM analysis of PDS 456 outflows with a clever ordering analysis, but the headline stratification relation is not statistically significant and the abstract oversells it.","tokens_in":29018,"tokens_out":1704,"would_cite":true,"duration_ms":22645,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"PDS 456's five ultra-fast outflows form a stratified wind","keywords":["PDS 456","ultra-fast outflows","AGN feedback","XRISM","photoionization modeling","PION","quasar winds","X-ray absorption spectroscopy"],"falsifier":"Observe PDS 456 with a high-resolution soft X-ray spectrometer capable of resolving the 0.7-1 keV band, such as XRISM with the gate valve open or NewAthena/X-IFU, for about 100 ks and repeat the six-component PION permutation analysis. The paper's own simulations predict that the correct sequence will be recoverable with $\\Delta C \\sim 900$ between orderings and that the soft X-ray outflow will be pinned to the third layer with $\\Delta C \\sim 100$; if the data instead leave the relative positions unconstrained or favor a different layer, the claimed stratification and the screening-based ordering method would be falsified.","tokens_in":27956,"feed_emoji":"🌪️","tokens_out":6842,"duration_ms":73466,"temperature":0.7,"pith_summary":"This paper asks whether the six outflow components recently resolved in the X-ray spectrum of the quasar PDS 456 are arranged in an ordered structure rather than being independent clumps. Using a photoionization model that recomputes ionization as each absorbing layer modifies the light reaching the next layer, the authors find that the five hard X-ray outflows follow a velocity-ionization relation $v_{\\rm out} \\propto \\xi^{(0.14\\pm 0.04)}$, and that trial orderings of the absorbing layers statistically favor the slower outflows sitting closer to the black hole. If true, this means the ultra-fast outflows in this quasar form a single stratified wind that accelerates outward, which would constrain how such winds are launched and how they feed energy back into the galaxy. The paper is careful to call the ordering tentative, since the correlation is only marginally significant and the preferred positions rest on small improvements in fit quality.","feed_headline":"Five quasar winds line up by speed and ionization","feed_subtitle":"XRISM data tie wind velocity to ionization, hinting slower outflows sit nearer the black hole.","key_machinery":"The load-bearing tool is PION, a self-consistent photoionization model that dynamically computes the ionization balance of each absorbing layer from the spectral energy distribution that actually reaches it, so an inner absorber that removes photons changes the ionization of the layers behind it. In spectral fitting, the six PION components are arranged in a multiplicative sequence, and the paper runs all 720 possible order permutations to see which radial ordering best matches the data. The order preference is read from C-statistic differences (up to $\\Delta C = 22$ across sequences) and confirmed by Bayesian evidence comparisons among three representative orderings, with the preferred PION3 ordering (UFO 5-4-6-3-2-1) favored over the others by $\\log B \\geq 3$. Stability curves computed by PION identify which ionization states sit on thermally unstable branches of the heating-cooling equilibrium.","core_discovery":"The central claim is that the five hard X-ray ultra-fast outflows in PDS 456, once modeled with the self-consistent photoionization code PION, show a stratified ionization structure in which outflow velocity increases with ionization parameter as $v_{\\rm out} \\propto \\xi^{(0.14\\pm 0.04)}$. The paper further claims that by permuting the order of the six PION absorption components along the line of sight, the data favor the slowest hard X-ray outflow ($\\log\\xi \\sim 4.1$, $v \\sim 0.23c$) at the innermost layer and the soft X-ray outflow ($\\log\\xi \\sim 3$, $v \\sim 0.27c$) in the middle layers, suggesting that slower outflows are launched closer to the supermassive black hole and accelerate as they propagate outward. The soft X-ray outflow is found to be thermally unstable in every ordering, while the hard X-ray outflows are thermally stable; its physical location remains uncertain between the broad-line region at sub-parsec scales and co-spatial with the hard X-ray outflows. The authors present this stratification as a tentative but physically plausible interpretation, supported by C-statistics and Bayesian model comparison.","pith_inferences":["An implication the authors leave implicit is that the measured slope 0.14 could serve as a probe of the radial density and ionization profile of the wind, and comparing this slope across quasars may reveal whether the stratification is universal or tied to Eddington ratio.","The order-permutation method could be exported to other AGN with multiple absorption systems once XRISM- or NewAthena-quality spectra are available, since the method only needs soft X-ray features sensitive to screening.","If UFO6 is truly co-spatial with the hard X-ray UFOs, its high density and tiny clump size imply a fine-spray geometry, which would change how covering factors and column densities are interpreted in all UFO studies.","A direct test of the stratification would be to check whether the velocity-ionization ordering persists during flaring and quiescent states separately, since the 0.8-0.9 keV residuals that drive the order preference appeared only during the flare."],"forward_implications":["The five hard X-ray UFOs should be treated as one stratified wind: velocity and ionization increase together, so measurements of one component can predict the others.","Slower-outflow-closer-to-the-black-hole ordering implies outward acceleration, which matches radiatively driven wind models where terminal velocity grows with radius.","Ignoring absorber order in photoionization fits, as pre-calculated codes do, can misassign ionization parameters in multi-absorber AGN, so future fits must treat screening explicitly.","The thermally unstable soft X-ray outflow explains its observed rapid variability and suggests the wind fragments into multiple phases, connecting the UFO to warm absorber and BAL phenomena.","If the same ordering analysis is repeated on other AGN, wind structure, not just wind presence, becomes measurable, allowing direct tests of feedback energetics."],"supporting_citations":[{"why":"Resolves the multiple UFO components in the Fe-K band and supplies the six-component model and XRISM data used here.","marker":"Paper I"},{"why":"Introduces the PION code that underlies the self-consistent photoionization modeling in this work.","marker":"Miller et al. 2015"},{"why":"Documents PION calculations and the differences from pre-calculated photoionization models.","marker":"Mehdipour et al. 2016"},{"why":"Provides the earlier detection of UFOs in PDS 456 with velocities of 0.25-0.34c.","marker":"Nardini et al. 2015"},{"why":"Provides the earlier detection of the soft X-ray UFO in PDS 456.","marker":"Reeves et al. 2016"},{"why":"Gives the MHD wind model predicting the velocity-ionization scaling used for comparison.","marker":"Fukumura et al. 2010"},{"why":"Supplies the momentum-conserving scaling $v \\propto \\xi^{0.5}$ and the UFO-warm absorber connection.","marker":"Tombesi et al. 2013"},{"why":"Supplies the energy-conserving outflow scaling $v \\propto \\xi^{1/3}$ used to interpret the observed slope.","marker":"King & Pounds 2015"},{"why":"Defines the S-curve stability analysis used to judge which UFO solutions are thermally stable.","marker":"Krolik et al. 1981"}],"fun_headline_variants":["Quasar winds stratified by speed and ionization","XRISM finds velocity-ionization ladder in PDS 456","Slowest ultra-fast outflows lie closest to black hole","Wind speed scales with ionization in quasar PDS 456","PDS 456 outflows: slower winds nearer the black hole"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The stratification and ordering conclusions assume that each absorbing component fully covers the X-ray source and that the sequence of components in the spectral model corresponds one-to-one to their physical radial order along the line of sight; if the outflows are clumpy and only partially cover the source, the screening between layers that carries the ordering signal is not guaranteed.","fun_headline_variants_meta":{"raw":{"variants":["Quasar winds stratified by speed and ionization","XRISM finds velocity-ionization ladder in PDS 456","Slowest ultra-fast outflows lie closest to black hole","Wind speed scales with ionization in quasar PDS 456","PDS 456 outflows: slower winds nearer the black hole"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000234,"raw_usage":{"total_tokens":1621,"prompt_tokens":1196,"completion_tokens":425,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":812,"completion_tokens_details":{"reasoning_tokens":343}},"tokens_in":812,"tokens_out":425,"duration_ms":5608,"temperature":1.0,"reasoning_tokens":343,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T10:21:45.202632+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Observe PDS 456 with a high-resolution soft X-ray spectrometer capable of resolving the 0.7-1 keV band, such as XRISM with the gate valve open or NewAthena/X-IFU, for about 100 ks and repeat the six-component PION permutation analysis. The paper's own simulations predict that the correct sequence will be recoverable with $\\Delta C \\sim 900$ between orderings and that the soft X-ray outflow will be pinned to the third layer with $\\Delta C \\sim 100$; if the data instead leave the relative positions unconstrained or favor a different layer, the claimed stratification and the screening-based ordering method would be falsified.","supporting_citations":[{"cited_title":"M., Kaastra , J","cited_arxiv_id":null,"evidence_quote":"Introduces the PION code that underlies the self-consistent photoionization modeling in this work."},{"cited_title":"S., & Kallman , T","cited_arxiv_id":null,"evidence_quote":"Documents PION calculations and the differences from pre-calculated photoionization models."},{"cited_title":"2010, , 715, 636","cited_arxiv_id":null,"evidence_quote":"Gives the MHD wind model predicting the velocity-ionization scaling used for comparison."}],"review_version":1}