{"id":"799bb069-e918-46fc-9e8e-fca9b4045196","arxiv_id":"1909.00793","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Eta Carinae's gamma-ray emission is tied to hadronic acceleration in its colliding stellar winds, and CTA observations of its orbital variability could map the wind geometry and the injected proton and positron flux.","lead":"This conference paper describes how the colliding winds in the massive binary Eta Carinae can accelerate particles and emit gamma rays, and predicts that the future Cherenkov Telescope Array will see fast orbital variations. A generalist might read it to see how a single binary system can serve as a natural accelerator and what new telescope data may reveal about cosmic-ray protons and positrons leaving the system.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed hadronic explanation and reproduced variability rest on a single periastron; the 2014.5 periastron null is acknowledged but unexplained.","rationale":"The reader's weakest assumption focuses on the Parkin et al. 3D hydrodynamic simulation correctly representing the shocked gas geometry and density at all phases. That is a reasonable underlying assumption, since the whole calculation uses that grid, but it is indirect. The manuscript itself contains a more direct empirical challenge: the 2014.5 periastron did not show the hard component that the same model predicted. This is a falsifiable prediction that failed, and the text acknowledges it without explaining it. This concern lands more precisely on the central claim than the hydro-simulation dependence alone, because it does not rely on questioning the simulation's realism; it uses the model's own predictive record. The 2009-2012 match may still be physically correct, and the hadronic interpretation may still hold, but the claim that the variability pattern is 'reproduced' is weakened when the very next periastron is inconsistent. The reader's rationale did mention this issue, but not as the weakest assumption, hence partial agreement. The verdict remains CONDITIONAL: the model deserves attention, but the predictive claim needs a quantitative treatment of the 2014.5 null and, as the reader notes, the promised companion papers on gamma-gamma opacity and CTA simulations.","tokens_in":12986,"tokens_out":3635,"duration_ms":38164,"concrete_test":"Re-analyze Fermi-LAT 10-300 GeV data for the 2014.5 periastron using the same event selection, temporal binning, and stacked analysis as in Figure 2b for the 2009 periastron. Compare the resulting light curve with the model-predicted hard-component flux over the same phase range. If the predicted flux exceeds the observed 95% upper limits at several orbital phases (or the observed flux is inconsistent with the predicted modulation at >3 sigma), then the claimed reproducibility and the fixed 2.4% acceleration efficiency require revision; if the prediction falls within the limits, the concern is mitigated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claim is that the hard component above 10 GeV 'could be explained uniquely with a hadronic contribution' and that pion decay with 2.4% of shock mechanical energy into protons reproduces the variability from periastron 2009 through apastron 2012. Section 3 immediately notes the next periastron failed: 'in the following 2014.5 periastron passage the hard component did not show up again, remaining at a flux level compatible with the apastron.' This is a direct predictive failure. If the orbit-dependent shock geometry and acceleration efficiency are correct, the hard component should have reappeared at 2014.5; it did not, and no mechanism (changed wind parameters, clumpiness, efficiency, or geometry) is offered to reconcile the null. Thus the 'reproduces the variability pattern' claim is based on a single cycle, and the 'uniquely hadronic' inference is not robust. The concern is not that 3D hydro simulations are imperfect in general, but that the model made a falsifiable prediction that was not borne out while the text still treats the 2009-2012 match as convincing evidence.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper argues that the gamma-ray emission of Eta Carinae, the first non-compact gamma-ray binary, arises in the colliding winds of the LBV primary and its hot companion: a low-energy (0.3-10 GeV) inverse-Compton component from accelerated electrons and a hard (>10 GeV) pion-decay component from accelerated protons. The authors use the 3D hydrodynamic simulations of Parkin et al. to compute, in each adaptive cell, the shock velocity, mechanical power, and maximum particle energy, balancing acceleration with radiative cooling. They report that a 2.4% proton acceleration efficiency reproduces the observed hard-component variability from periastron 2009 to apastron 2012, and that gamma-gamma absorption in the anisotropic UV field, with head-on versus tail-in collisions, modulates the hard emission by a factor of about 40 in intensity and more than 20 in peak energy. They conclude that CTA should detect these variations on timescales of a few days, disentangling the intrinsic particle cutoff from opacity and constraining the system geometry, magnetic field, and cosmic-ray injection.","tokens_in":13226,"tokens_out":3496,"duration_ms":40861,"significance":"If correct, the hadronic interpretation would establish Eta Carinae as a PeV-scale particle accelerator in a colliding-wind binary, with implications for cosmic-ray injection from massive-star winds. The paper's strengths are its explicit coupling of the 3D hydrodynamical grid to particle acceleration and cooling, its separate calculation of anisotropic gamma-gamma opacity with orientation-dependent head-on/tail-in collisions, and its concrete, falsifiable CTA predictions. The CTA simulations and the factor-40 opacity variation are useful forward-modeling results. However, the central evidential claim is not yet robust: the hard component's non-reappearance at the 2014.5 periastron is acknowledged but not explained, and the match relies on fitted acceleration efficiencies and on one specific hydrodynamical grid.","major_comments":[{"comment":"The claim that the >10 GeV hard component 'could be explained uniquely with a hadronic contribution' and that a 2.4% proton efficiency reproduces the 2009-2012 variability is not supported by the full dataset, because the paper itself states that 'in the following 2014.5 periastron passage the hard component did not show up again, remaining at a flux level compatible with the apastron.' Since the model is orbit-dependent, this is a predictive failure that should have produced a reappearing hard component at 2014.5. No physical mechanism (e.g., changes in wind parameters, clumpiness, acceleration efficiency, or shock geometry) is offered to reconcile the null. The 'reproduces the variability pattern' claim is therefore based on a single cycle and should be explicitly reframed as a post-diction of one event, or accompanied by a quantitative model for why the 2014.5 periastron behaved differently.","section":"Section 3, Fig. 2a"},{"comment":"The agreement with the observed light curves is obtained by normalizing the electron and proton acceleration efficiencies (f_e and f_p, with f_p ~ 2.4%) and by choosing a surface magnetic field in the range 0.4-1 kG. These are free parameters of the model, and no degeneracy study, uncertainty estimate, or sensitivity analysis is presented. The text says the LE light curve is reproduced 'very well' and that the hard component 'could well reproduce' the variability, but without quantified uncertainties on f_p, f_e, and B, the reader cannot judge whether the 2014.5 null is a 1-sigma fluctuation, a 5-sigma contradiction, or something in between. Please provide error bars or at least a demonstration that the conclusions are stable across the allowed range of these parameters.","section":"Section 3, Fig. 2"},{"comment":"The LE light-curve match, including the post-periastron peak, depends crucially on the 'reverse bubble cavity' in the Parkin et al. hydrodynamical simulation, which roughly doubles the shock area for about a tenth of the orbit. The authors acknowledge that 'this secondary peak... could in reality collapse' because the details of magnetic field geometry, radiation transfer, and hydrodynamics are uncertain. Since the entire cell-by-cell particle acceleration calculation inherits the accuracy of this hydro grid, the central claim would be considerably strengthened by testing the sensitivity of the predicted light curves and spectra to plausible variations of the grid, such as including radiative cooling, clumping, or the difference between the Orbit-RD and Orbit-IA simulations. Without such tests, the agreement could be an artifact of the chosen simulation.","section":"Section 3, paragraph on reverse bubble cavity"}],"minor_comments":[{"comment":"The sentence 'These relativistic particles subsequently dissipates non-thermal radiation' has a subject-verb agreement error and is semantically imprecise; the particles radiate non-thermal emission rather than dissipating radiation.","section":"Abstract"},{"comment":"In the text accompanying Fig. 3, 'a an arbitrary spectrum' should read 'an arbitrary spectrum.' Please also state the assumed exposure time and the version of the CTA instrument response clearly in the caption.","section":"Section 3, caption of Fig. 3"},{"comment":"There are several typographical inconsistencies: '10∼ 40 M⊙' and '10 49.7 ergs' lack proper spacing and exponent formatting; 'Observations by Sir John Herschel [10], confirmed' contains an errant comma. These should be cleaned up.","section":"Section 2"},{"comment":"References [38] and [40] are cited as 'in preparation' and 'submitted,' respectively, yet they are used to support quantitative claims (the dashed variability curve in Fig. 4 and the detailed prospects for CTA observations). For the published record, either provide the actual evidence in this paper or cite the available data with a clear indication that the claims are preliminary.","section":"References [38] and [40]"}],"recommendation":"major_revision","confidential_remarks":"This is a conference proceedings rather than a full research paper, and the claims would normally require a longer companion publication with more quantitative detail. The most serious issue is the 2014.5 periastron null: the authors should be encouraged to address it head-on rather than treating the 2009-2012 match as the sole validation. The paper's forward-modeling machinery and CTA predictions are valuable and should be preserved; the revision should focus on honesty about predictive power and on parameter sensitivity."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a conference proceedings that mostly restates Balbo & Walter 2017 and points to companion papers for the actual gamma-gamma opacity calculation. If you know the 2017 paper, there is very little new here. If you don't, it is a clear, compact summary of the hadronic interpretation and the CTA prospects.\n\nThe one genuinely new idea—strong orbital modulation of gamma-gamma absorption around periastron, between head-on and tail-on collisions—is plausible but not demonstrated here. Figure 4 shows a dashed 'in preparation' curve, and the derivation is in references [38] and [40]. The physical argument is sensible, but the paper gives no numbers, no independent calculation, and no data. The CTA spectrum in Fig. 3 is from a 50-hour simulation of an 'arbitrary spectrum', so it is an illustration, not a prediction.\n\nCredit where it is due: the paper openly states that the hard component above 10 GeV did not recur at the 2014.5 periastron, remaining at apastron-like levels. That is a direct failure of the model's predictive loop, and the authors do not hide it. This honesty matters, and it is more than many would do.\n\nThe main soft spot is the claim that the hard component 'could be explained uniquely with a hadronic contribution'. That is too strong. The 2009–2012 light-curve match is obtained by tuning the proton acceleration efficiency to 2.4% and the electron efficiency to fit the low-energy component, so agreement is built in. The single-cycle match is then undermined by the 2014.5 null, and the paper offers no mechanism—no change in wind parameters, clumpiness, or acceleration efficiency—to reconcile it. The stress-test note is correct on this point.\n\nThe gamma-gamma absorption argument is the most independent piece, but it is a preview. The citation pattern is fine; self-citation to the 2017 paper and the in-preparation companion papers is appropriate when those papers contain the actual results.\n\nFor a reader wanting a fast overview of the hadronic model for Eta Carinae and a roadmap for CTA, this proceedings is useful. It is not a primary source. As a journal submission, I would desk-reject it because the new content is referenced, not presented, and the load-bearing claim is weakened by the unexplained null. As a proceedings contribution, it is fine.","headline":"A readable summary of the authors' hadronic model for Eta Carinae and a preview of a gamma-gamma opacity study, but the new results are only referenced and the model faces an unexplained null at the 2014.5 periastron.","tokens_in":13753,"tokens_out":2786,"would_cite":false,"duration_ms":27103,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Eta Carinae's hard gamma rays are made by proton collisions.","keywords":["Eta Carinae","colliding wind binaries","gamma-ray binaries","hadronic acceleration","neutral pion decay","gamma-gamma absorption","Cherenkov Telescope Array","cosmic-ray protons"],"falsifier":"A CTA campaign across a full orbit that sees no reappearance of the hard component at periastron, or no shift of the gamma-gamma absorption cutoff peak across the predicted factor of more than 20 in energy, would falsify the hadronic-plus-opacity model; likewise, a 1-100 MeV instrument detecting inverse-Compton emission far stronger than the secondary-lepton prediction would rule out purely hadronic acceleration.","tokens_in":12781,"feed_emoji":"🌠","tokens_out":7094,"duration_ms":66276,"temperature":0.7,"pith_summary":"Eta Carinae is the first known gamma-ray binary without a compact object: the dense wind of the massive primary star slams into the fast wind of a hidden companion, and the shock accelerates particles. This paper argues that the hard gamma-ray component above about $10\\,\\mathrm{GeV}$ cannot come from electrons and must be produced by proton-proton collisions through neutral-pion decay. It shows that channeling about $2.4\\%$ of the shock's mechanical energy into proton acceleration reproduces the Fermi-LAT light curve from periastron 2009 through apastron 2012, with protons reaching near $10^{15}\\,\\mathrm{eV}$ at closest approach. It also predicts that gamma-gamma absorption against the anisotropic ultraviolet photon field varies by a factor of about 40 in intensity and more than 20 in peak energy around periastron, and that the Cherenkov Telescope Array should see these swings on timescales of days, separating the intrinsic particle cutoff from opacity effects. If correct, this single system would show where and how protons reach the knee of the cosmic-ray spectrum and how many positrons the binary injects into the interstellar medium.","feed_headline":"Eta Carinae's hard gamma rays are made by proton collisions","feed_subtitle":"A 2.4 percent proton share reproduces its 2009-2012 light curve; CTA can test the prediction in days.","key_machinery":"The load-bearing machinery is a grid of adaptive cells from the 3D hydrodynamic simulation [20], which models the interacting winds of Eta Carinae at each orbital phase. In every cell the authors balance the characteristic diffusive-shock-acceleration time against synchrotron, inverse-Compton and bremsstrahlung cooling to obtain the local shock velocity, mechanical luminosity and maximum particle energy, then sum the cell emissivities to build light curves and spectra. The hadronic branch rides on neutral-pion decay, $pp\\to\\pi^0\\to 2\\gamma$, with a proton acceleration efficiency of about $2.4\\%$. The opacity mechanism is $\\gamma$-$\\gamma$ pair production in the anisotropic UV field, whose cross-section peaks at $\\xi=E_\\gamma k_B T/(m_e c^2)^2\\simeq 1.4$ for head-on and $\\xi\\simeq 30$ for tail-in collisions; the orbital phase controls which geometry the photons traverse along the line of sight, producing the predicted strong absorption modulation.","core_discovery":"The paper's central claim is that the orbital variability of Eta Carinae from X-rays to very-high-energy gamma rays is the summed emission of shock-accelerated particles computed cell by cell in a 3D hydrodynamic simulation of the colliding winds, and that the hard component above $\\sim 10\\,\\mathrm{GeV}$ is uniquely hadronic. In this picture, the low-energy ($0.3$-$10\\,\\mathrm{GeV}$) light curve is inverse-Compton emission from accelerated electrons tracking the orbital motion, while neutral-pion decay from protons carrying about $2.4\\%$ of the shock mechanical energy produces the hard component, with proton energies approaching $10^{15}\\,\\mathrm{eV}$ at periastron. Because the gamma rays must cross the anisotropic ultraviolet photon fields of both stars, the observed spectrum is a convolution of the intrinsic particle cutoff and $\\gamma$-$\\gamma$ absorption whose strength changes by a factor of about 40 and whose absorption peak shifts by more than a factor of 20 as the line of sight swings from tail-in to head-on collisions. The paper predicts that CTA, with its sensitivity for short exposures, will detect these modulations on few-day timescales above $30\\,\\mathrm{GeV}$, disentangling the intrinsic proton cutoff from opacity and fixing the geometry, magnetic-field configuration, and the flux of relativistic protons and positrons released into the interstellar medium.","pith_inferences":["The paper focuses on one binary; a natural extension is to ask whether the same cell-by-cell recipe and $2.4\\%$ efficiency, applied to the population of colliding-wind binaries, contributes a measurable fraction of the Galactic cosmic rays near the knee, since these systems stay active much longer than supernova remnants.","The 2014.5 periastron did not reproduce the hard component seen in 2009, so the 'reverse bubble cavity' that doubles the shock area may be stochastic; repeated CTA periastron campaigns would turn that shock instability into a measurable quantity rather than a nuisance parameter.","A sharper test than total flux would be the phase lag between periastron passage and the shift of the $\\gamma$-$\\gamma$ cutoff energy: that lag is set by the binary orientation and UV field geometry, so timing it would determine the line-of-sight geometry independently of the spectral normalization."],"forward_implications":["The hard component above $\\sim 10\\,\\mathrm{GeV}$ requires a hadronic (neutral-pion decay) contribution; leptonic inverse-Compton emission alone cannot explain the 2009 periastron very-high-energy spectrum.","With about $2.4\\%$ of the shock mechanical energy going into proton acceleration, the predicted pion-decay light curve matches Fermi-LAT observations from periastron 2009 through apastron 2012.","Protons can be accelerated to nearly $10^{15}\\,\\mathrm{eV}$ at periastron, close to the cosmic-ray knee, and colliding-wind binaries can sustain this acceleration over much of their lives, injecting up to $10^{48}$-$10^{49}$ erg of cosmic-ray energy.","The $\\gamma$-$\\gamma$ absorption varies by roughly a factor of 40 in intensity and more than a factor of 20 in peak energy around periastron; CTA should detect these modulations as flux variability on timescales of a few days above $30\\,\\mathrm{GeV}$.","An instrument sensitive in the 1-100 MeV band could discriminate the lepto-hadronic model from a purely hadronic one by measuring the strength of the inverse-Compton component."],"supporting_citations":[{"why":"Supplies the 3D hydrodynamic simulation of the colliding winds at each orbital phase; every shock velocity, mechanical power, and maximum particle energy in the model is computed on this grid.","marker":"[20]"},{"why":"Provides the Fermi-LAT low- and high-energy light curves and the two-component spectral decomposition that the cell-by-cell model is built to reproduce.","marker":"[26]"},{"why":"Gives the H.E.S.S. very-high-energy detection and the steep drop around 1 TeV that motivates the intrinsic-cutoff versus gamma-gamma-absorption discussion.","marker":"[27]"},{"why":"Supplies the alternative hadronic interpretation and the factor-of-about-40 contrast in gamma-gamma absorption between head-on and tail-in collision geometries.","marker":"[34]"},{"why":"Provides the angular dependence of electron-positron pair production by gamma rays in an anisotropic soft-photon flux, including the head-on and tail-in peak positions used in the opacity calculation.","marker":"[36]"},{"why":"Fixes the absolute orientation of Eta Carinae's binary orbit, which determines when the line of sight is head-on versus tail-in.","marker":"[37]"}],"fun_headline_variants":["Eta Carinae's hard gamma rays point to proton collisions","Hadronic origin for Eta Carinae's high-energy gamma rays","Proton collisions explain Eta Carinae's gamma-ray spectrum","CTA will probe Eta Carinae's proton acceleration","Eta Carinae's gamma rays: protons, not electrons, dominate"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire calculation assumes the 3D hydrodynamic simulation [20] correctly gives the location, shape, and density of the shocked gas at every orbital phase, including the temporary 'reverse bubble cavity' that doubles the shock area for about a tenth of the orbit; if that simulated wind-collision geometry is wrong, the matched light curves and CTA predictions built on it are not reliable.","fun_headline_variants_meta":{"raw":{"variants":["Eta Carinae's hard gamma rays point to proton collisions","Hadronic origin for Eta Carinae's high-energy gamma rays","Proton collisions explain Eta Carinae's gamma-ray spectrum","CTA will probe Eta Carinae's proton acceleration","Eta Carinae's gamma rays: protons, not electrons, dominate"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00023,"raw_usage":{"total_tokens":1585,"prompt_tokens":1150,"completion_tokens":435,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":766,"completion_tokens_details":{"reasoning_tokens":348}},"tokens_in":766,"tokens_out":435,"duration_ms":4820,"temperature":1.0,"reasoning_tokens":348,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:35:36.517172+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A CTA campaign across a full orbit that sees no reappearance of the hard component at periastron, or no shift of the gamma-gamma absorption cutoff peak across the predicted factor of more than 20 in energy, would falsify the hadronic-plus-opacity model; likewise, a 1-100 MeV instrument detecting inverse-Compton emission far stronger than the secondary-lepton prediction would rule out purely hadronic acceleration.","supporting_citations":[{"cited_title":"Spiralling out of control: 3D hydrodynamical modelling of the colliding winds in $\\eta\\thinspace$Carinae","cited_arxiv_id":"1011.0778","evidence_quote":"Supplies the 3D hydrodynamic simulation of the colliding winds at each orbital phase; every shock velocity, mechanical power, and maximum particle energy in the model is computed on this grid."},{"cited_title":"Fermi acceleration along the orbit of {\\eta} Carinae","cited_arxiv_id":"1705.02706","evidence_quote":"Provides the Fermi-LAT low- and high-energy light curves and the two-component spectral decomposition that the cell-by-cell model is built to reproduce."},{"cited_title":"First Results of Eta Car Observations with H.E.S.S.II","cited_arxiv_id":"1708.01033","evidence_quote":"Gives the H.E.S.S. very-high-energy detection and the steep drop around 1 TeV that motivates the intrinsic-cutoff versus gamma-gamma-absorption discussion."},{"cited_title":"On the origin of \\gamma-ray emission in \\eta\\ Carina","cited_arxiv_id":"1502.04056","evidence_quote":"Supplies the alternative hadronic interpretation and the factor-of-about-40 contrast in gamma-gamma absorption between head-on and tail-in collision geometries."},{"cited_title":"Electron-positron pair production by gamma rays in an anisotropic flux of soft photons, and application to pulsar polar caps","cited_arxiv_id":"1710.04021","evidence_quote":"Provides the angular dependence of electron-positron pair production by gamma rays in an anisotropic soft-photon flux, including the head-on and tail-in peak positions used in the opacity calculation."},{"cited_title":"Constraining the Absolute Orientation of Eta Carinae's Binary Orbit: A 3-D Dynamical Model for the Broad [Fe III] Emission","cited_arxiv_id":"1111.2226","evidence_quote":"Fixes the absolute orientation of Eta Carinae's binary orbit, which determines when the line of sight is head-on versus tail-in."}],"review_version":1}