{"id":"fa1157c3-cbe1-4f8c-be22-2d07b94f92f8","arxiv_id":"1908.07216","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Polarimetry of upstream H-alpha emission should show a 90-degree flip in polarization direction when upstream protons are decelerated by cosmic rays, providing a new diagnostic of cosmic-ray modified shocks.","lead":"Supernova remnant shocks are thought to accelerate cosmic rays, but whether the cosmic rays push back on the incoming gas is hard to observe. This paper predicts that the direction of H-alpha polarization flips by 90 degrees when such a cosmic-ray modified shock exists, offering a new way to detect the slowdown directly.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The polarization flip relies on a cold, step-like decelerated proton population; if precursor proton heating or a smooth velocity gradient broadens the 200 km/s separation, the Ly-beta absorption asymmetry that drives the diagnostic is washed out.","rationale":"The reader's weakest_assumption identifies the cold shifted-Maxwellian treatment of the decelerated proton population as the load-bearing premise. I agree: the entire polarization-flip mechanism is the direction-dependent Ly-beta absorption arising from the clean 200 km/s bulk-velocity separation between pre-existing and charge-exchanged hydrogen atoms. If proton heating broadens the decelerated distribution, or if the acceleration process produces a smooth velocity gradient rather than a step function, the separation is smeared and the sign reversal may vanish. This is not an ad hominem or a consensus disagreement; it is a correctness risk because the model explicitly assumes no proton heating and a single decelerated velocity throughout the precursor. The concern is concrete and testable: running the model at finite precursor proton temperatures directly probes it. I also considered the position-dependent sign of the unmodified case (Figure 6 shows positive polarization near the shock and negative farther upstream) and the unpolarized-Ly-beta approximation, but these are secondary: the paper's proposal is to observe the upstream precursor, and the authors explicitly justify the unpolarized assumption by Ly-beta trapping. The proton-heating assumption, by contrast, is unexamined and is the mechanism that makes the diagnostic specific to CR modification. The existing code from Shimoda & Laming (2019) and the standard atomic data give the model independent support, so the paper merits conditional acceptance pending the sensitivity test, not rejection.","tokens_in":16083,"tokens_out":20175,"duration_ms":215068,"concrete_test":"Recompute the Stokes Q/I profile of Figure 6 for case (iii) with the same shock parameters but with precursor proton temperatures T_p = 10^5, 10^6, and 10^7 K (keeping the 5% bulk deceleration and all other inputs fixed). As a second check, replace the step-function deceleration with a linear ramp in the proton bulk velocity from Vsh to 0.95Vsh across the precursor length. If the sign flip relative to the no-modification case no longer occurs over the precursor, the proposed diagnostic is not robust to the assumed precursor structure.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 3 case (iii) specifies a CR precursor in which protons are decelerated by 5% of Vsh with no proton heating, so all velocity distributions are shifted Maxwellians (stated immediately before Figure 4 and used in Eqs. 32-34). The polarization flip in Figure 6 is produced by the 200 km/s shift between pre-existing and charge-exchanged hydrogen atoms: Ly-beta photons propagating nearly along the shock plane (μ≈0) are absorbed by both populations, while photons along the normal (μ≈±1) are absorbed only by the pre-existing population because the 200 km/s shift moves the decelerated atoms out of the line core. This direction-dependent optical-depth asymmetry is the entire physical origin of the claimed sign reversal. If the decelerated protons are heated so that their thermal width becomes comparable to 200 km/s, or if the deceleration occurs gradually over the precursor so that charge-exchanged atoms are born over a continuum of bulk velocities, the μ≈±1 absorption is partially filled in and the asymmetry, and hence the diagnostic, disappears. The paper neither justifies the no-heating assumption nor tests its sensitivity; the central claim may therefore be an artifact of the assumed cold, single-velocity precursor.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper proposes a new observational diagnostic for cosmic-ray-modified shocks: linear polarimetry of upstream Hα produced by the Lyβ-to-Hα conversion. The authors set up a plane-parallel slab model of an SNR shock, use a published radiative-transfer and atomic-population model, and consider three upstream states: no precursor, an electron-heating precursor without velocity modification, and a precursor in which protons are decelerated by 5% of the shock velocity without being heated. They compute the hydrogen ionization structure, Hα emissivity, and Stokes Q/I as functions of upstream position for Vsh = 4000 km s−1, T0 = 6000 K, and preshock ionization degree 0.5. Their main finding is that the polarization direction flips between the unmodified case (polarization parallel to the shock surface at large Lyβ optical depth) and the modified case (polarization perpendicular to the shock surface), with a few-percent degree, and that a velocity modification of 200 km s−1 is sufficient to produce the flip. They also show that the upstream Hα surface brightness is comparable to the downstream value and that the Hβ/Hα ratio is insensitive to the velocity modification. The paper concludes that Hα polarimetry can be a unique diagnostic of the CR back-reaction in SNR shocks.","tokens_in":16365,"tokens_out":14101,"duration_ms":155406,"significance":"If the predicted sign flip is robust, this is a genuinely new, falsifiable probe of CR precursor physics; it goes beyond SED fitting and connects a direct observable to the bulk deceleration of the upstream plasma. The model is a forward calculation with no fitted data, it builds on published atomic data and prior models, and it makes concrete predictions for the sign, degree, and spatial profile of the polarization, as well as for the Hβ/Hα ratio. The connection to the existing polarized Hα detection in SN 1006 (Sparks et al. 2015) gives the proposal some observational plausibility. The main weakness is that the central prediction has not yet been shown to survive plausible variations in precursor thermodynamics, so the significance of the diagnostic is currently conditional.","major_comments":[{"comment":"The sign flip is produced by the assumption that the decelerated protons remain a cold, shifted Maxwellian with no heating, so that the 200 km s−1 shift places the charge-exchanged atoms outside the line core for μ≃±1 while they still absorb at μ≃0. If precursor proton heating broadens the decelerated distribution to a thermal width comparable to 200 km s−1, or if the deceleration occurs gradually so that charge-exchanged atoms are born over a continuum of bulk velocities, this direction-dependent absorption asymmetry is reduced and the predicted polarization flip can be washed out. The paper states the no-heating assumption but does not justify it or test its sensitivity; this is load-bearing for the abstract's central claim.","section":"Section 3 (case iii), before Fig. 4; Eqs. (32)-(34)"},{"comment":"The statement that the unmodified shock has polarization parallel to the shock surface is not true at all positions in the model. In case (i), the polarization is positive, i.e. perpendicular to the shock surface, close to the shock front and becomes negative only at Lyβ optical depth of order unity. Since a real edge-on observation will integrate the Stokes parameters over a spatial resolution element along the shock-normal coordinate, the sign comparison in the abstract may not correspond to the quantity that is actually measured. The paper should specify the spatial region it refers to and give the polarization predicted for an observationally motivated integration.","section":"Abstract and Fig. 6"},{"comment":"The quantitative claims of a few-percent polarization and sensitivity to a 5% velocity modification are computed for a single parameter point: Vsh=4000 km s−1, T0=6000 K, χ0=0.5, and ΔVsh=200 km s−1. The charge-exchange rate, Lyβ optical depth, and the precursor length scale all depend strongly on these parameters, so without a parameter scan it is unclear whether the sign flip and the few-percent degree are generic features or specific to this one case.","section":"Section 3, parameter set"}],"minor_comments":[{"comment":"In the discussion of the Hα FWHM, '30-50 km−1' should read '30-50 km s−1'.","section":"Section 4"},{"comment":"The estimate that multiply scattered Lyβ has negligible polarization is based on a dust-scattering calculation; a short estimate for resonantly trapped Lyβ in hydrogen would be more directly relevant, since the polarization budget of the trapped radiation is one of the stated simplifications.","section":"Section 2, footnote 2"},{"comment":"The numerical factor in the diffusion-coefficient lower bound is stated without derivation; showing the prefactor explicitly would help readers reproduce the constraint on the CR diffusion coefficient.","section":"Eq. (35)"},{"comment":"The vertical axis labels in Figures 6-8 would be clearer if the units (percent for polarization, specific units for surface brightness) appeared directly in the axis labels rather than only in the caption.","section":"Figs. 6-8"}],"recommendation":"major_revision","confidential_remarks":"The paper leans heavily on the authors' own previous model (Shimoda & Laming 2019) for the atomic populations and radiative transfer. That is legitimate, but the referee report should make clear that a more explicit description of which parts are being reused and which are new would help the reader. The central idea is interesting and appropriate for MNRAS, but the robustness of the sign flip to precursor thermodynamics needs to be addressed before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Jiro and Laming have a genuinely new idea here. The claim is that Hα from upstream of an SNR shock is linearly polarized via Lyβ scattering, and the polarization direction flips from parallel to the shock plane to parallel to the shock normal when a CR precursor decelerates the upstream protons by 5% of Vsh. I checked the cited prior literature and did not find this specific flip prediction. It is a clever extension of known charge-exchange and Rayleigh-scattering physics, built on their earlier radiative-transfer model (Shimoda & Laming 2019). The forward model is internally consistent, the equations are laid out clearly, and the atomic data (oscillator strengths, charge-exchange cross-sections) are standard published values. The paper also acknowledges the main geometric simplification (edge-on viewing) and the assumption of unpolarized Lyβ. No code or data are shipped, but the equations are explicit enough to reproduce.\n\nThe soft spots match what the stress-test note flags. The whole diagnostic rests on the decelerated protons remaining a cold, shifted Maxwellian. The 200 km/s separation is what makes the Lyβ absorption direction-dependent; if those protons are heated, or if the deceleration is gradual so charge-exchanged atoms are born over a spread of bulk velocities, the asymmetry that produces the flip gets washed out. The paper states the no-proton-heating assumption in Section 3 case (iii) but does not justify it or test sensitivity to it. That is a real gap, though not a fatal one for a first proposal—the authors are explicit that this is a proof-of-concept calculation. I would also like to see more than one shock velocity and more than a single 5% modification; the claim that the polarization 'responds sensitively' is demonstrated for one point in parameter space.\n\nSome minor issues: the abstract simplifies the position-dependent polarization signal into a single direction, which could mislead a casual reader; and Figure 6 shows the polarization varying in sign along the precursor, so the observed direction may depend on where you look. The paper leans on two prior papers by the same authors for the atomic model, but those are real, published models with independent atomic data inputs, so this is not a circularity problem.\n\nBottom line: this is a serious, useful proposal for a diagnostic that SED fitting cannot provide. It deserves refereeing. I would ask for a sensitivity study on proton heating and a broader parameter scan before accepting, but the core idea is sound and worth engaging.","headline":"A genuinely new diagnostic proposal with a physically plausible polarization flip; the main caveat is the assumed cold, single-velocity decelerated proton population, which the paper does not stress-test.","tokens_in":16848,"tokens_out":1924,"would_cite":true,"duration_ms":20766,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Edge-on H-alpha polarimetry of supernova remnant shocks can reveal whether a cosmic-ray precursor has decelerated the upstream plasma, because the polarization direction flips by 90 degrees when it has.","keywords":["cosmic-ray modified shocks","supernova remnants","H-alpha polarimetry","Ly-beta scattering","resonant scattering polarization","charge exchange","radiative transfer","diffusive shock acceleration"],"falsifier":"Take spatially resolved H$\\alpha$ polarimetry across the limb of a supernova remnant shock with $V_{\\rm sh} \\simeq 4000\\ {\\rm km\\,s^{-1}}$ and preshock density below about $30\\ {\\rm cm^{-3}}$, choosing a remnant with independent gamma-ray evidence for efficient cosmic-ray acceleration. The model predicts that if the shock is modified, the upstream polarization will sit along the shock velocity with a degree near five per cent and will rotate by $90^\\circ$ relative to an unmodified shock; an observation that keeps the polarization parallel to the shock surface across the entire precursor, at a sensitivity below one per cent, would falsify the diagnostic.","tokens_in":15873,"feed_emoji":"🔭","tokens_out":14158,"duration_ms":128053,"temperature":0.7,"pith_summary":"This paper proposes that the linear polarization of the hydrogen Balmer-alpha line emitted just ahead of a supernova remnant shock can reveal whether the shock is modified by the back-reaction of accelerated cosmic rays. Upstream Balmer-alpha is produced mainly when hydrogen atoms scatter Lyman-beta photons coming from the hot downstream plasma, and scattering makes the re-emitted line polarized. Viewed edge-on, the model says the polarization direction is parallel to the shock surface for an ordinary shock, but flips to the direction of the shock velocity when a cosmic-ray precursor has decelerated the upstream protons by as little as five per cent of the shock speed. The predicted polarization is a few per cent, a level already reached in existing observations of one supernova remnant, so the diagnostic is within observational reach. If it works, it would give astronomers a direct measurement of the cosmic-ray pressure gradient that gamma-ray and X-ray spectral fitting cannot currently distinguish.","feed_headline":"Edge-on H-alpha polarization can reveal cosmic-ray-modified shocks","feed_subtitle":"A 5 percent slowdown of upstream protons flips the H-alpha polarization angle by 90 degrees at SNR shocks.","key_machinery":"The load-bearing mechanism is the Ly$\\beta$-to-H$\\alpha$ conversion under resonant scattering: a hydrogen atom absorbs a Ly$\\beta$ photon, is excited to the $3p$ state, and decays to $2s$, emitting H$\\alpha$ whose linear polarization is fixed by angular-momentum conservation and by the scattering phase matrices for the $3p_{1/2}$ (unpolarized channel) and $3p_{3/2}$ (polarizing channel) sublevels. The sign of the Stokes $Q$ parameter is set by the angular anisotropy of the incident Ly$\\beta$ radiation field after attenuation, $\\exp(-\\tau_\\nu/|\\mu|)$. In an unmodified shock the Ly$\\beta$ beam from downstream is elongated along the shock normal, giving polarization parallel to the shock surface. In a modified shock, charge-exchange reactions with protons decelerated to $0.95V_{\\rm sh}$ create a second hydrogen population whose Doppler-shifted Ly$\\beta$ absorption preferentially removes photons travelling at $\\mu \\approx 0$, reversing the anisotropy and flipping the polarization to the shock-velocity direction. The full calculation combines this scattering with a collisional-radiative model of hydrogen level populations, ionization balance, and emissivities.","core_discovery":"The central claim is that the polarization angle of upstream H$\\alpha$ acts as a sign of cosmic-ray-modified shock. In the upstream region of a fast shock, H$\\alpha$ is dominated by the Ly$\\beta$-to-H$\\alpha$ conversion: hydrogen atoms absorb Ly$\\beta$ photons radiated by the downstream gas and re-emit H$\\alpha$, and because this is a scattering process the H$\\alpha$ is linearly polarized with a degree set by the anisotropy of the incident Ly$\\beta$ beam. In an unmodified shock the beam is elongated along the shock normal, producing polarization parallel to the shock surface; in a modified shock, charge exchange with protons decelerated to $0.95V_{\\rm sh}$ creates a second hydrogen population whose Doppler-shifted absorption removes the Ly$\\beta$ photons travelling at small angles to the shock surface, reversing the anisotropy and producing polarization along the shock velocity. The paper computes the effect with a hydrogen level-population and radiative-transfer model, obtaining degrees of about two per cent for the unmodified shock, about one per cent for an electron-heating precursor with no deceleration, and about five per cent for the modified shock. It also shows that the upstream H$\\alpha$ surface brightness is comparable to the downstream value, making the polarized precursor emission detectable in principle.","pith_inferences":["A testable extension would be to map the polarization position angle as a function of distance from the shock front; the spatial profile of the expected angle flip would trace the deceleration profile of the precursor and, with a model, the cosmic-ray pressure gradient.","The same Ly$\\beta$-scattering mechanism should imprint a weaker polarization signal on H$\\beta$ through Ly$\\gamma$ conversion, so a combined H$\\alpha$/H$\\beta$ polarimetric measurement could separate resonant-scattering polarization from collisional-excitation emission.","Because the polarization flip relies on the decelerated protons retaining a narrow velocity distribution, kinetic or hybrid simulations that include neutral coupling could test whether precursor turbulence or proton heating washes out the signal.","Combining upstream H$\\alpha$ polarimetry with downstream H$\\alpha$ polarimetry, which is sensitive to the acceleration efficiency, could in principle measure both the cosmic-ray pressure and the injection efficiency in the same shock."],"forward_implications":["An edge-on observation of an SNR limb can separate a cosmic-ray-modified shock from an unmodified one by the position angle of the H$\\alpha$ polarization alone, even when the precursor itself is spatially unresolved.","The predicted few-per-cent polarization is observationally accessible; polarized H$\\alpha$ at $2.0 \\pm 0.4$ per cent has already been detected toward SN 1006.","Because the H$\\beta$/H$\\alpha$ ratio barely changes with the velocity modification, polarimetry supplies information that the Balmer decrement cannot.","The condition that the precursor length exceed the charge-exchange length translates to a lower bound on the cosmic-ray diffusion coefficient, about $10^{25}\\ {\\rm cm^2\\,s^{-1}}$, and to a lower bound near $33$ TeV on the energy of the cosmic rays driving the modification in a Bohm-like field.","The upstream H$\\alpha$ surface brightness is comparable to the downstream brightness, so the polarized precursor emission should be detectable rather than lost against the shock itself."],"supporting_citations":[{"why":"Supplies the collisional-radiative line transfer model and hydrogen level-population calculations on which the three shock models are built.","marker":"Shimoda & Laming 2019"},{"why":"Provides the Rayleigh scattering formalism and phase matrices used to compute the Stokes Q of the scattered H-alpha.","marker":"Chandrasekhar 1960"},{"why":"Derives the polarization phase matrices for the 3p1/2 and 3p3/2 hydrogen transitions that determine the H-alpha polarization.","marker":"Hamilton 1947"},{"why":"Establishes that charge exchange with decelerated protons generates a neutral hydrogen population moving at the proton bulk velocity.","marker":"Ohira & Takahara 2010"},{"why":"Provides the electron-heating precursor scenario used as the intermediate case with no velocity modification.","marker":"Laming et al. 2014"},{"why":"Reports the 2.0 ± 0.4 per cent polarized H-alpha detection toward SN 1006 that sets the observational precedent for the predicted signal level.","marker":"Sparks et al. 2015"},{"why":"Supplies the charge-exchange rate coefficients used to compute the production of decelerated hydrogen atoms at the assumed 200 km/s relative velocity.","marker":"Janev et al. 2003"}],"fun_headline_variants":["H-alpha polarization angle reveals cosmic-ray-modified shocks","Cosmic-ray-modified shocks flip H-alpha polarization by 90 degrees","Edge-on H-alpha polarization spots cosmic-ray-modified shocks","Polarization of H-alpha tells if a shock is cosmic-ray-modified","H-alpha polarimetry diagnoses cosmic-ray-modified shocks"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The diagnostic assumes that the cosmic-ray precursor decelerates the upstream protons by five per cent of the shock speed without heating them, so that every velocity distribution stays a shifted Maxwellian; if proton heating broadens the decelerated distribution, the direction-dependent Ly$\\beta$ absorption asymmetry that produces the polarization flip could be smeared out and the signal lost.","fun_headline_variants_meta":{"raw":{"variants":["H-alpha polarization angle reveals cosmic-ray-modified shocks","Cosmic-ray-modified shocks flip H-alpha polarization by 90 degrees","Edge-on H-alpha polarization spots cosmic-ray-modified shocks","Polarization of H-alpha tells if a shock is cosmic-ray-modified","H-alpha polarimetry diagnoses cosmic-ray-modified shocks"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000873,"raw_usage":{"total_tokens":3794,"prompt_tokens":975,"completion_tokens":2819,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":591,"completion_tokens_details":{"reasoning_tokens":2738}},"tokens_in":591,"tokens_out":2819,"duration_ms":23126,"temperature":1.0,"reasoning_tokens":2738,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:23:34.981057+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take spatially resolved H$\\alpha$ polarimetry across the limb of a supernova remnant shock with $V_{\\rm sh} \\simeq 4000\\ {\\rm km\\,s^{-1}}$ and preshock density below about $30\\ {\\rm cm^{-3}}$, choosing a remnant with independent gamma-ray evidence for efficient cosmic-ray acceleration. The model predicts that if the shock is modified, the upstream polarization will sit along the shock velocity with a degree near five per cent and will rotate by $90^\\circ$ relative to an unmodified shock; an observation that keeps the polarization parallel to the shock surface across the entire precursor, at a sensitivity below one per cent, would falsify the diagnostic.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that charge exchange with decelerated protons generates a neutral hydrogen population moving at the proton bulk velocity."},{"cited_title":"M., Hwang U., Ghavamian P., Rakowski C., 2014, @doi [ ] 10.1088/0004-637X/790/1/11 , https://ui.adsabs.harvard.edu/abs/2014ApJ...790...11L 790, 11","cited_arxiv_id":null,"evidence_quote":"Provides the electron-heating precursor scenario used as the intermediate case with no velocity modification."},{"cited_title":"B., Pringle J","cited_arxiv_id":null,"evidence_quote":"Reports the 2.0 ± 0.4 per cent polarized H-alpha detection toward SN 1006 that sets the observational precedent for the predicted signal level."},{"cited_title":"K., Reiter D., Samm U., 2003, C ollision processes in low-temperature hydrogen plasmas","cited_arxiv_id":null,"evidence_quote":"Supplies the charge-exchange rate coefficients used to compute the production of decelerated hydrogen atoms at the assumed 200 km/s relative velocity."}],"review_version":1}