{"id":"b6926543-8f77-4d3c-b71d-8f116354c711","arxiv_id":"2507.05085","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Relaxing the constant-temperature assumption in radiation-hydrodynamic disc wind simulations does not rescue line driving: the simulated winds remain overionized and roughly a hundred times weaker than older CAK-style models predicted.","lead":"The authors ran new computer simulations of winds from accreting white dwarf discs, this time letting the gas heat and cool instead of forcing a constant temperature. The simulated winds are still far too weak and too ionized compared with observed outflows, so earlier doubts about line-driven winds stand.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The weak-wind result may be a branch-selection artifact: all runs start from hydrostatic equilibrium (§2.6) and the paper never tests whether a self-shielded strong-wind branch persists, leaving the central overionization conclusion conditional on initial conditions.","rationale":"The paper's central claim has two parts: (1) relaxing the isothermal approximation does not change the wind outcome, and (2) the predicted outflows are too highly ionized to explain observed AWD winds. The first part is well supported by the four-model comparison and robustness tests, and is unaffected by branch selection since all models share the same initialization. The second part is the physically important conclusion and is where the branch-selection concern bites. The reader's weakest assumption focused on the adopted SED and instantaneous ionization equilibrium; these are parameter/microphysics uncertainties that the authors partly acknowledge. The branch-selection concern is more fundamental because it questions whether the simulations have found the physically realized solution at all: observations show UV resonance lines that indicate winds avoid over-ionization, which is exactly what a self-shielded strong branch would provide. The paper's own discussion (§4) implicitly raises this possibility but does not test it. A single initial-condition experiment would settle the question. For these reasons the verdict should remain CONDITIONAL, with the condition expanded to include a demonstration that the weak branch is the only stable attractor under the detailed MC-RHD treatment.","tokens_in":17500,"tokens_out":7692,"duration_ms":97543,"concrete_test":"Restart Model A (ideal gas, Hybrid macro-atom) with the velocity and density fields initialized to the steady CAK line-driven wind solution of Proga et al. (1998) for the same WD parameters; if such a snapshot is unavailable, impose a strong radial outflow (Mdot_wind ≈ 10^-4 Mdot_acc) for the first ~1000 s and then let it evolve freely. If the outflow decays to the weak branch (Mdot_wind ≈ 6×10^-14 Msun/yr) and remains there, the weak-wind conclusion is robust. If it settles into a self-shielded strong wind (Mdot_wind ≈ 10^-4 Mdot_acc), the reported efficiency problem in Model A is an initial-condition artifact, and the paper's conclusion must be restricted to the hydrostatic-start branch.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The most load-bearing assumption is not the SED or the ionization-equilibrium approximation but the implicit assumption that the weak-wind state is the only stable solution. All four simulations are initialized from a hydrostatic, zero-velocity configuration (§2.6) and converge to the same weak branch (Mdot_wind ≈ 6×10^-14 Msun/yr, §3). Line-driven disc winds are known to exhibit bistability: a dense, self-shielded strong wind can coexist with a weak/failed wind for the same external parameters, because the strong wind's higher density increases recombination, lowers ionization, and boosts the force multiplier. The paper shows the launch-region force multiplier is only M ≈ a few × 100 (§3.2, Fig. 6), just above the threshold for driving, and acknowledges that observed AWD outflows do manage to avoid over-ionization (§4). If a self-shielded strong branch exists, the reported overionization is a branch-selection artifact of the chosen initial conditions rather than a robust property of line-driven winds. The paper reports no test of alternate initial conditions, so this possibility is unaddressed. If the strong branch is stable under the detailed MC-RHD treatment, the central claim that line driving cannot produce the observed winds would be incorrect; at minimum it is unproven.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents four Monte-Carlo radiation hydrodynamic simulations of line-driven winds from accreting white dwarfs, crossing the equation of state (ideal gas vs isothermal) with two radiative transfer modes (Hybrid macro-atom and Classic). The main result is that all four models produce nearly identical wind geometries and mass-loss rates (within a factor of ~2), with Mdot_wind ≈ 6 × 10^-14 M_sun/yr (Mdot_wind/Mdot_acc < 10^-5), confirming the earlier finding that detailed multidimensional ionization and radiative transfer suppress line-driving relative to CAK-style simulations. The temperature near the sonic surface is ≈5 × 10^4 K, similar to the assumed isothermal value of 4 × 10^4 K, explaining the insensitivity. Synthetic UV spectra show weak resonance lines compared to observed AWD winds. The authors conclude that relaxing the isothermal approximation does not change the overionization problem.","tokens_in":17773,"tokens_out":7046,"duration_ms":83512,"significance":"This is a valuable technical contribution: it is the first line-driven wind RHD simulation with an ideal-gas equation of state and full energy equation, and the first to use the macro-atom formalism in this context. The 2×2 comparison cleanly isolates the effect of the thermodynamic treatment and the radiative transfer mode. The code is publicly archived, and the paper includes numerous parameter-sensitivity tests (ΔT_RAD, N_hat, rho_d, domain size, damping). If the results are robust, they strengthen the case that line-driven winds in AWDs are weak and overionized, with implications for AGN. The main caveat, detailed below, is that the simulations do not test whether a different dynamical branch (a self-shielded strong wind) exists.","major_comments":[{"comment":"All four simulations are initialized from a zero-velocity hydrostatic configuration (Eq. 21 with v_r = v_θ = 0) and evolve to a single quasi-steady state. The paper does not test alternate initial conditions. This matters because line-driven disc winds can exhibit bistability: a self-shielded, high-density strong wind can coexist with the weak/failed wind for the same external parameters, since higher density increases recombination, lowers ionization, and boosts the force multiplier. The launch-region force multiplier is only M ≈ a few × 100 (Fig. 6, §3.2), just above the threshold for driving, and the paper itself notes that observed AWD outflows do manage to avoid over-ionization (§4). If a strong branch exists and is stable under the MC-RHD treatment, the central conclusion that line-driving produces only overionized weak winds would be incomplete or incorrect. I recommend adding a simulation that initializes the flow with a wind-like density/velocity structure (e.g., taken from a CAK-style model, or by temporarily boosting the radiation force) to see whether the flow converges to the same branch. This is a load-bearing test for the claim in the abstract that the predicted outflows are too highly ionized.","section":"§2.6 and §3"}],"minor_comments":[{"comment":"The sentence 'Is also acts as a proof-of-concept' should read 'It also acts as a proof-of-concept.'","section":"§2.6"},{"comment":"Typo: 'bottem-rightpanel' should be 'bottom-right panel.'","section":"Fig. 2 caption"},{"comment":"Typo: 'Much 1 surface' should be 'Mach 1 surface.'","section":"Fig. 6 caption"},{"comment":"In the text after Eq. (12), 'represents the the Doppler width' has a duplicated 'the.'","section":"§2.4"},{"comment":"Equation (9) uses 'Thompson' but the correct spelling is 'Thomson' (as in Thomson scattering).","section":"§2.3"},{"comment":"The comparison of synthetic spectra with observed AWD spectra is qualitative ('weak or absent'); a quantitative measure, such as predicted equivalent widths or column densities for the resonance lines, would make the claim that the simulations fail to reproduce the observed UV wind signatures more precise.","section":"§3.2 and Fig. 5"}],"recommendation":"major_revision","confidential_remarks":"The paper is technically thorough and the 2×2 comparison is a clean design. The only substantive issue is the branch-selection concern. If the authors can demonstrate that the weak-wind branch is the robustly selected solution (for example by starting from a strong-wind initial condition and showing it decays to the weak branch, or by finding the strong branch is unstable), the paper would be acceptable. Without such a test, the headline conclusion remains conditional on the initial conditions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The real news here is a null result that holds up better than I expected. The paper runs four combinations of equation of state (ideal gas vs isothermal) and radiative transfer mode (Classic vs Hybrid macro-atom), and all four converge to the same weak-wind branch, with time-averaged mass-loss rates within a factor of two of each other. The temperature at the sonic surface lands near 5e4 K, close to the 4e4 K assumed in the isothermal runs, which gives a concrete physical reason why thermodynamics does not matter much in this regime. That is a genuine and useful finding, and it is new relative to Higginbottom et al. (2024). The comparison of synthetic spectra against RW Sex and IX Vel is honest: the predicted UV lines are weak or absent, which is a real failure mode, not a cosmetic one.\n\nThe soft spots are real but mostly acknowledged. The biggest one, which the stress-test note pushes hard, is branch selection. All runs start from hydrostatic equilibrium with zero poloidal velocity, and line-driven disc winds are known to have bistable solutions. A self-shielded strong branch, if it exists, would not be found from these initial conditions. The paper does not test alternate initial conditions, and it does not argue that the weak branch is unique. That said, I think the stress-test note overstates the case slightly: the bistability literature is mostly about CAK-style parameterized force multipliers, and here the force multiplier is computed from a detailed ionization calculation at M ~ a few hundred, just above threshold. It is entirely possible that a strong branch does not exist under this treatment. But the paper does not show that, so the central claim should be read as \"for these initial conditions and this SED, the weak branch is the outcome,\" not as a categorical proof that line driving cannot produce strong AWD winds.\n\nThe other caveats are the Shakura-Sunyaev blackbody-annuli SED with no central source and the instantaneous ionization equilibrium assumption. The paper flags both, including the likely breakdown of ionization equilibrium in fast, low-density regions, and argues the effect is minor there. I find that argument plausible but not quantitative. The code is archived on Zenodo, but the custom problem generators and input files are only available on request, which is a legitimate reason to hold the paper to a higher bar on reproducibility.\n\nOverall this is a solid, well-scoped paper that deserves a serious referee. The reader's conditional verdict is about right. My main ask would be a robustness test with a different initial condition, but I would not block publication on it if the authors clearly state the limitation. I would cite this work, and I would bring it to a reading group focused on radiation-hydrodynamics or disc winds.","headline":"A careful, credible negative result: relaxing the isothermal approximation does not rescue line-driven disc winds in AWDs, but the weak-wind branch is only probed from one set of initial conditions.","tokens_in":18321,"tokens_out":674,"would_cite":true,"duration_ms":10272,"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":"Replacing the isothermal approximation with a full ideal-gas energy equation does not change the picture: line-driven disc winds stay overionized, with mass-loss below $10^{-5}$ of the accretion rate and weak ultraviolet wind lines.","keywords":["radiation hydrodynamics","line-driven winds","accretion disc winds","accreting white dwarfs","overionization","force multiplier","Monte Carlo radiative transfer","ultraviolet spectroscopy"],"falsifier":"Re-run the fiducial simulation with time-dependent (non-equilibrium) ionization and recombination instead of instantaneous equilibrium; if the accelerating wind freezes into lower ionization stages and $\\dot{M}_{\\rm wind}/\\dot{M}_{\\rm acc}$ rises above $10^{-5}$ with strong C IV and N V absorption in the synthetic spectra, the overionization conclusion would be overturned.","tokens_in":17316,"feed_emoji":"💨","tokens_out":9099,"duration_ms":88404,"temperature":0.7,"pith_summary":"The paper asks whether the earlier isothermal simulations of line-driven accretion-disc winds reached the right conclusion despite neglecting the wind's thermal state. It replaces the fixed-temperature equation of state with an ideal gas law, solves the full energy equation including radiative heating and cooling, and runs four radiation-hydrodynamic simulations around an accreting white dwarf. The result is that the thermal treatment does not change the overall picture: the predicted winds remain overionized, the mass-loss rates stay below $10^{-5}$ of the accretion rate, and the synthetic ultraviolet spectra still lack the strong C IV and N V wind lines seen in real systems. The authors conclude that the line-driving efficiency problem is not removed by a more detailed treatment of thermodynamics.","feed_headline":"Full heating/cooling leaves line-driven disc winds 100x too weak","feed_subtitle":"Solving the wind's temperature structure confirms overionization that suppresses ultraviolet C IV and N V lines.","key_machinery":"The central mechanism is the coupled radiation-hydrodynamic loop: a Godunov-type hydrodynamics code advances the ideal-gas energy equation, while a Monte Carlo radiative transfer code supplies the radiation field, ionization state, heating and cooling rates, and direction-dependent ultraviolet fluxes. From these, a force multiplier $M(t)$ is computed with an independent line-list code containing more than 450,000 transitions, and the radiative acceleration is assembled as a sum over 36 directions using $g_i = [1+M(t_i)]\\sigma_e F_{UV,i}/c$, with the optical-depth parameter $t_i$ depending on the local velocity gradient and thermal velocity. The temperature structure that emerges near the Mach 1 surface, roughly $5\\times10^4$ K, is close to the $4\\times10^4$ K assumed in the isothermal runs, and that closeness is the direct reason the two treatments give similar winds.","core_discovery":"Relaxing the isothermal approximation in Monte Carlo radiation-hydrodynamic simulations of line-driven disc winds around accreting white dwarfs does not alter the conclusions drawn from isothermal models. With an ideal gas equation of state and the full energy equation solved alongside frequency-dependent radiative transfer and ionization equilibrium, the wind that develops is still overionized in its launching region: the force multiplier reaches only a few hundred rather than the near-optimal few thousand, the dominant driving species are strong oxygen lines rather than the many weak iron lines that drive hot-star winds, and the wind mass-loss rate remains about two orders of magnitude below the values from older CAK-style simulations. The temperature near the sonic surface turns out to be close to the $40\\,000$ K assumed in the isothermal runs, which is why the two approaches agree. Testing two different radiative-transfer modes produces the same outcome, indicating the result is not sensitive to the transfer mode either.","pith_inferences":["If overionization in the simulations is an artifact of assuming instantaneous ionization equilibrium, then including time-dependent recombination might restore lower ionization stages and stronger line acceleration in fast-moving regions; the paper itself notes that such freeze-in may occur.","The observed strong ultraviolet wind lines in high-state accreting white dwarfs imply that real outflows avoid overionization, so some combination of a softer-than-assumed disc SED, sub-grid wind clumping, or an additional driving mechanism such as magnetic fields is likely needed.","A direct numerical test of this would be to rerun the fiducial model with a reduced ionizing flux and check whether the mass-loss rate rises above $10^{-5}$ and the ultraviolet lines appear, which would show that line-driving can work despite the fiducial overionization.","The factor-of-two difference in mass-loss rate between the two radiative-transfer modes is smaller than the intrinsic time variability of the winds, suggesting that comparisons with observations will be limited by variability rather than transfer details."],"forward_implications":["The isothermal approximation was not the source of the weak winds; solving the full energy equation leaves the wind mass-loss rate within a factor of two of the isothermal models.","The line-driving efficiency problem is confirmed: the predicted $\\dot{M}_{\\rm wind}/\\dot{M}_{\\rm acc} < 10^{-5}$ is about one hundred times lower than earlier CAK-style simulations.","Synthetic ultraviolet spectra remain inconsistent with observations of high-state accreting white dwarfs such as RW Sex and IX Vel, with C IV, N V, and Si IV resonance lines weak or absent.","With the thermal treatment now included, the same simulation framework can be applied to line-driven disc winds in AGN, where the more complex SEDs make the isothermal approximation less reliable."],"supporting_citations":[{"why":"It supplies the isothermal baseline model and the coupled radiation-hydrodynamic framework that this paper extends to an ideal gas equation of state.","marker":"Higginbottom et al. 2024"},{"why":"It gives the earlier CAK-style AWD wind simulations with $\\dot{M}_{\\rm wind}/\\dot{M}_{\\rm acc}\\simeq10^{-4}$ that the new low mass-loss rates are compared against.","marker":"Proga et al. 1998"},{"why":"It provides the thin-disc temperature profile and blackbody-annuli spectral energy distribution used to generate the photon packets.","marker":"Shakura & Sunyaev 1973"},{"why":"It introduces the CAK line-driving formalism on which the force multiplier method and the comparison with hot-star winds rest.","marker":"Castor, Abbott, & Klein 1975"},{"why":"It supplies the method for generating force-multiplier lookup tables from a large atomic line list.","marker":"Parkin & Sim 2013"},{"why":"It describes the two radiative-transfer modes (Classic and Hybrid macro-atom) whose results are compared here.","marker":"Matthews et al. 2025"},{"why":"It provides the macro-atom formalism used to enforce co-moving-frame energy conservation in the hybrid mode.","marker":"Lucy 2002, 2003"}],"fun_headline_variants":["Full thermodynamics confirm disc winds 100x too weak","Overionization persists without isothermal assumption","Even realistic heating cannot strengthen disc winds","Temperature structure unchanged: winds stay too weak","Line-driven disc winds remain 100x weak with full energy"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The overionization conclusion rests on assuming that the disc radiates like a standard Shakura-Sunyaev blackbody-annuli spectrum with no central source and that ionization is always in instantaneous equilibrium; if the real spectrum is softer or ionization freezes into lower stages in fast-flowing gas, the driving could be much stronger.","fun_headline_variants_meta":{"raw":{"variants":["Full thermodynamics confirm disc winds 100x too weak","Overionization persists without isothermal assumption","Even realistic heating cannot strengthen disc winds","Temperature structure unchanged: winds stay too weak","Line-driven disc winds remain 100x weak with full energy"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000261,"raw_usage":{"total_tokens":1642,"prompt_tokens":1042,"completion_tokens":600,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":658,"completion_tokens_details":{"reasoning_tokens":529}},"tokens_in":658,"tokens_out":600,"duration_ms":7349,"temperature":1.0,"reasoning_tokens":529,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T20:20:13.255213+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the fiducial simulation with time-dependent (non-equilibrium) ionization and recombination instead of instantaneous equilibrium; if the accelerating wind freezes into lower ionization stages and $\\dot{M}_{\\rm wind}/\\dot{M}_{\\rm acc}$ rises above $10^{-5}$ with strong C IV and N V absorption in the synthetic spectra, the overionization conclusion would be overturned.","supporting_citations":[{"cited_title":"S., Matthews J","cited_arxiv_id":null,"evidence_quote":"It supplies the isothermal baseline model and the coupled radiation-hydrodynamic framework that this paper extends to an ideal gas equation of state."},{"cited_title":"R., Sim S","cited_arxiv_id":null,"evidence_quote":"It supplies the method for generating force-multiplier lookup tables from a large atomic line list."}],"review_version":1}