{"id":"49f97ee8-9fca-4ede-bd14-6a19b5fb4b62","arxiv_id":"1908.02527","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Two-dimensional multi-fluid simulations indicate that WASP-12b's atmosphere escapes in a stellar-captured double stream at about 10^12 g/s, making a bow-shock explanation for the observed early ingress implausible.","lead":"Simulations of the hot Jupiter WASP-12b show that, when tidal forces are included, the planet's escaping atmosphere forms two streams flowing toward and away from the star, losing mass at about 10^12 grams per second. This study tests whether a stellar wind bow shock explains the planet's observed early transit ingress and ultraviolet absorption, and finds the bow shock scenario unrealistic for this planet.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 2D axisymmetric model excludes Coriolis force and forces the SW flow along the planet-star line, so the predicted no-early-ingress geometry is not established for WASP-12b.","rationale":"The paper's main new result is the captured-by-the-star regime, and its observational relevance rests on the claim that no early ingress is produced in that regime. For that claim to hold, the simulated MgII distribution must faithfully represent the real 3D flow. The model is 2D axisymmetric with the symmetry axis along the planet-star line; this geometry forces the outflow into symmetric streams and cannot represent the leading-side versus trailing-side asymmetry set by the 226 km/s orbital motion. The authors explicitly acknowledge the omission of Coriolis force and the co-directionality assumption in Section 2 and the Conclusions, but the assertion that MgII within 3 Rp is correctly reproduced is supported only by a qualitative statement, not by a quantitative error estimate. This is a correctness risk of the central argument rather than a mere disagreement with external consensus. I therefore keep the CONDITIONAL judgment rather than accepting the abstract's conclusion, but I do not reject the paper: the multi-fluid model, the inclusion of hydrogen/helium/magnesium chemistry, and the explicit comparison of two scenarios are real strengths, and the missing Coriolis and oblique-flow check is a well-defined, fixable gap.","tokens_in":19647,"tokens_out":7064,"duration_ms":79386,"concrete_test":"Run scenario 1 again in a non-axisymmetric (3D) geometry that includes the Coriolis term, keeping the same inner boundary, XUV spectrum, and SW parameters from Table 1, and recompute the MgII h-line transit light curve with the observed viewing geometry. If the pre-ingress (phase < -0.01) absorption increases by more than about 2% relative to the axisymmetric run, or if the stellarward stream is deflected by more than about 10 degrees within 5 Rp, the conclusion that the captured regime produces no early ingress is falsified. An inexpensive analytical first check is to evaluate Ωr/v along the published streamlines and verify that it stays below about 0.1 out to 5 Rp.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2 states that the Coriolis force is not included, restricting validity to about ±5 Rp, and Appendix A assumes co-directionality of the SW velocity and the XUV flux. For WASP-12b this is not a small effect: the orbital velocity V_K = 226 km/s is comparable to the adopted fast-wind speed V_SW = 417 km/s in scenario 1 and is perpendicular to the planet-star line, so the true ram-pressure geometry is oblique rather than axisymmetric. The conclusion that the captured regime forms only two symmetric streams and produces no early ingress follows from a symmetry the real system does not have; any leading-side MgII column that could produce early ingress is excluded by construction. The manuscript's assertion (Section 4) that Coriolis spiraling is small near 3 Rp is unquantified. Using the stated expansion speeds of 5-10 km/s, the Coriolis deflection angle Ωr/v at r=3Rp is of order 0.1-1 rad, and the associated transverse Doppler shifts are tens of km/s, comparable to the widths of the simulated MgII absorption profiles. Thus the central claim that the realistic scenario cannot produce an early ingress is not robust to the leading omitted physical effect.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript presents two-dimensional, multi-fluid hydrodynamic simulations of the upper atmosphere of the hot Jupiter WASP-12b, including hydrogen photo-chemistry and minor species Mg and He, under different stellar XUV fluxes and stellar wind conditions. Two interaction regimes are considered: a 'captured by the star' scenario with tidal force and a fast stellar wind, and a 'blown by the wind' scenario without tidal force and a slow wind. The authors compute MgII h-line absorption and compare it with HST/COS near-ultraviolet observations, reporting that the captured-by-the-star scenario produces two streams and a mass-loss rate of about 10^12 g/s but no early ingress, whereas the blown-by-the-wind scenario forms a bow shock but is judged unrealistic. They conclude that the observed early ingress is not reproduced by the realistic scenario and that the bow-shock interpretation is disfavoured.","tokens_in":19923,"tokens_out":4967,"duration_ms":54153,"significance":"If the modelling assumptions are accepted, the paper contributes a quantitative, multi-fluid description of atmospheric escape and MgII absorption for WASP-12b, and it argues against the bow-shock explanation of the observed early ingress in favour of a stellar-capture regime. The model is more self-consistent than earlier prescribed-boundary 3D simulations in that it couples XUV-driven outflow, charge exchange, and minor-species ionization, and it produces specific predictions for absorption-line profiles and mass-loss rates that can be compared with future observations. However, the significance is substantially tempered by the axisymmetric geometry, which excludes the Coriolis force and aligns the stellar wind with the planet-star line, and by the fact that the closest match to the observed MgII absorption is obtained by varying the poorly constrained stellar wind density. These issues directly affect the central claim that the realistic scenario cannot produce early ingress.","major_comments":[{"comment":"The axisymmetric 2D model omits the Coriolis force and assumes co-directionality of the stellar wind velocity and the XUV flux. For WASP-12b the Keplerian velocity V_K=226 km/s (stated in Section 3.2) is comparable to the adopted fast-wind speed V_SW=417 km/s in scenario 1, so the true interaction is oblique rather than axisymmetric. The two-stream structure and the resulting conclusion that scenario 1 produces no early ingress (Section 4) follow from this imposed symmetry. The manuscript's assertion that Coriolis spiraling is 'still small' at about 3 R_P (Section 4) is unquantified; using the stated expansion speeds of 5-10 km/s, the deflection angle Omega*r/v at r=3 R_P is of order 0.1-1 rad, and the associated transverse Doppler shifts are tens of km/s, comparable to the widths of the simulated MgII absorption profiles. The central claim that the realistic scenario cannot explain the early ingress is therefore not robust to this omitted leading physical effect.","section":"Section 2 and Appendix A"},{"comment":"The claimed reproduction of the observed ~4% MgII absorption is obtained by selecting stellar wind densities N_sw=3e4 and 3e5 cm^-3 from a small grid (Table 1) rather than from an independent constraint on WASP-12's wind. The abstract's statement that the simulations 'enable ... to reproduce the times of ingress and egress' is not supported by the presented light curves: in the realistic scenario 1 no early ingress is formed, while in scenario 2 the computed ingress coincides with the optical transit (Fig. 7b). The comparison with observations is therefore a fit, not a prediction, and it does not validate the model's geometry for the early-ingress phenomenon.","section":"Section 4, Figs. 6-7 and Table 1"},{"comment":"The MgII abundance and absorption are sensitive to several unconstrained inputs, including the low-energy Mg + He+ double charge-exchange cross-section, which the manuscript states is 'currently unknown', and the heating efficiency eta_h=0.5. These parameters directly control the MgII column density and hence the simulated absorption depth. The paper does not provide a sensitivity analysis over plausible ranges of these values, so the reported match to the observed 4% absorption cannot be distinguished from a parameterised fit rather than a robust model outcome.","section":"Section 2 and Appendix A"}],"minor_comments":[{"comment":"The reference list contains a duplicated entry for Shaikhislamov et al. 2016 (two identical entries with the same journal and page numbers).","section":"References"},{"comment":"The bibliography entry 'Cubtz M., Saar S. H., Shkolnik E., 2010' appears to be a typo for 'Cuntz et al.'; the in-text citation 'Cuntz et al. 2000' does not have a matching reference-list entry.","section":"References"},{"comment":"The notation in Eq. (1) is unclear: the integral limit 'StR' and the subscripts 'v,out' and 'v,transit' are not defined, and the intensity I is used without an explicit definition.","section":"Section 4, Eq. (1)-(2)"},{"comment":"The concluding sentence states that absorption is dominated by 'resonant thermal line broadening', but the text in Section 4 and Eq. (11) indicate that the line wings are controlled by natural broadening; this wording should be checked for consistency.","section":"Conclusions"}],"recommendation":"major_revision","confidential_remarks":"The paper is built on the authors' own previous modelling framework, and the key comparison with observations relies on an unconstrained stellar wind density. The axisymmetry limitation is acknowledged but its quantitative impact on the early-ingress conclusion is not assessed. I recommend major revision, with emphasis on either a 3D treatment or a quantitative estimate of the Coriolis deflection and a sensitivity study of the uncertain Mg and heating parameters."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the paper is worth reading for the captured-by-the-star result and the mass-loss numbers, but the MgII absorption story is weaker than the abstract suggests, and the no-early-ingress conclusion is not established outside the 2D setup.\n\nWhat's new: the authors apply their multi-fluid framework to WASP-12b, with tidal force, MgII including He double charge-exchange, and compute synthetic MgII absorption. The result that even a fast, dense stellar wind cannot stop the tidally driven outflow, and that mass loss is ~1-2e12 g/s controlled by stellar gravity rather than XUV, is a concrete number for evolution models. The argument that the bow-shock scenario requires ignoring tidal force and using a slow, dense wind is reasonable and useful.\n\nSoft spots. The match to the observed ~4% MgII absorption is obtained by choosing the stellar wind density (3e4 or 3e5 cm^-3), which is poorly constrained. That makes the comparison a fit, not a prediction. The abstract says the simulations enable reproduction of the times of ingress and egress; in the realistic scenario 1 there is no early ingress, and in scenario 2 the ingress/egress align with optical times. The observed early ingress is not reproduced in either scenario. So the abstract overstates this.\n\nThe larger issue is the 2D geometry. With no Coriolis force and the stellar wind and XUV forced co-directional, WASP-12b's orbital velocity of 226 km/s is comparable to the adopted wind speed. At the 3 Rp where the MgII column is concentrated, the Coriolis deflection is tens of degrees, not negligible. The symmetric double-stream shape and the absence of a leading-side absorbing column are, at least in part, built into the assumed axisymmetry. The manuscript acknowledges this, but then asserts the near-planet MgII distribution is still reproduced correctly without a quantitative justification. That assertion does not hold up under the simple deflection estimate.\n\nNet: the regime classification and mass-loss rates are useful, but the observational comparison needs reframing, and the Coriolis omission should be addressed or quantified before the early-ingress claim is trusted. This paper deserves a serious referee; it should not be desk rejected. I'd send it to review, with a request to quantify the Coriolis effect and bring the abstract in line with what the model actually shows.","headline":"Despite a solid multi-fluid model application and a physically plausible captured-by-the-star regime, the paper's headline claims about reproducing the MgII ingress are undercut by the model's axisymmetry and a fitted stellar wind density.","tokens_in":20544,"tokens_out":7607,"would_cite":true,"duration_ms":77271,"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":"Tidal gravity, not a bow shock, drives WASP-12b's escape","keywords":["hot Jupiter","WASP-12b","atmospheric escape","stellar wind","MgII absorption","hydrodynamic modelling","Roche lobe overflow","early ingress"],"falsifier":"Run a 3D simulation of WASP-12b that includes the Coriolis force and lets the stellar wind arrive at an angle to the XUV flux: if the resulting MgII column ahead of the planet shifts the computed transit ingress earlier than the optical contact, the paper's central conclusion of no early ingress in the realistic regime is falsified. The observed variable early ingress in the NUV transit light curves is the corresponding existing test.","tokens_in":19463,"feed_emoji":"🪐","tokens_out":7333,"duration_ms":70612,"temperature":0.7,"pith_summary":"Using a 2D multi-fluid hydrodynamic model of WASP-12b's hydrogen-dominated upper atmosphere, this paper argues that the planet's escape is controlled by the host star's tidal pull rather than by the stellar wind. When the tidal force is included, even a fast and dense stellar wind cannot stop the planetary outflow; the material instead splits into two streams moving toward and away from the star, with a saturated mass-loss rate near $10^{12}\\ \\mathrm{g\\,s^{-1}}$. That 'captured by the star' regime is presented as the realistic one for WASP-12b, whose Roche lobe is only $1.37\\,R_P$. The paper also computes MgII h-line absorption from the simulated outflow and finds 4--10% absorption in the realistic case, with no early ingress, while the previously proposed bow-shock scenario only appears when tidal effects are artificially switched off. If correct, the simulations rule out the bow-shock explanation for the observed variable early ingress and reframe WASP-12b as a planet whose atmosphere is being accreted by its star.","feed_headline":"Tidal gravity, not a bow shock, drives WASP-12b's escape","feed_subtitle":"Simulations put the hot Jupiter in the captured-by-the-star regime with ~10^12 g/s mass loss, no early MgII ingress.","key_machinery":"The load-bearing object is the 2D axisymmetric multi-fluid hydrodynamic model of Shaikhislamov et al. (2014, 2016) and Khodachenko et al. (2015, 2017), applied to WASP-12b with hydrogen ionization/recombination chemistry and minor species He and Mg treated as separate fluids. The crucial knob is the averaged tidal gravitational potential of equation (A.5), which shortens the Roche radius to $R_{L1}=1.37\\,R_P$; switching it on or off defines the two scenarios ('captured by the star' versus 'blown by the wind'). Absorption is computed by integrating Voigt-profile optical depth through the simulated MgII density, velocity, and temperature fields, with a resonant double-charge-exchange reaction between MgI and HeII included because it removes MgII near the planet.","core_discovery":"The central claim is that WASP-12b's upper atmosphere is in a 'captured by the star' escape regime: the stellar gravitational pull extends the planetary wind beyond the Roche lobe and splits it into two streams along the planet-star axis, so the mass-loss rate (saturated at roughly $1{-}2\\times10^{12}\\ \\mathrm{g\\,s^{-1}}$) is set by gravity, not by the XUV heating rate or the stellar wind ram pressure. In this regime the stellar wind only compresses the streams laterally and picks up ions near the boundary; it does not stop the outflow. A bow shock ahead of the planet, the configuration previously invoked to explain early NUV ingress, forms only in a test case with no tidal force and a slow, dense wind, which the authors call unrealistic for WASP-12b. Simulated MgII h-line absorption reaches 4--10% in the realistic scenario (larger for rarer winds), and the simulated transit curves in the shock scenario begin at the optical transit, meaning no early ingress is produced; the paper concludes that a complete account of the observed early ingress would require three-dimensional geometry.","pith_inferences":["If the captured-by-the-star regime is real, WASP-12b is a natural laboratory for star-feeding accretion streams; a 3D simulation with the Coriolis force would predict whether the two streams bend prograde and create the observed variable early ingress by placing MgII ahead of the planet along the orbit.","Because the simulated mass-loss rate in the realistic scenario depends only weakly on XUV flux, reconstructing the host star's XUV history from WASP-12b's current escape rate would be unreliable; the escape rate is mostly a gravity diagnostic instead.","The same two-scenario test could be applied to other ultra-hot Jupiters near their Roche limit: for planets with larger $R_{L1}/R_P$, the bow-shock regime may become realistic, making the presence of early ingress a diagnostic of orbital distance and stellar wind conditions."],"forward_implications":["WASP-12b's outflow is gravitationally captured by the star, so the planet is likely feeding material into a circumplanetary or circumstellar torus rather than shedding it as a comet-like tail.","The mass-loss rate in the realistic regime is 10--15 times higher than in the no-tide case and saturates near $10^{12}\\ \\mathrm{g\\,s^{-1}}$, so evolutionary estimates based on XUV-driven energy-limited escape alone would underestimate the planet's mass loss.","Bow-shock interpretations of WASP-12b's variable early NUV ingress are not supported by a self-consistent wind solution; the observed early absorption would need explanation from 3D structure, such as Coriolis bending or a torus of previously escaped material.","MgII absorption is more sensitive to stellar wind density than to XUV flux in the captured regime, so NUV MgII transit depth can serve as a probe of the local stellar wind environment around the planet."],"supporting_citations":[{"why":"Supplies the NUV transit observations of MgII and FeII and the early-ingress detection that the simulations aim to reproduce.","marker":"Fossati et al. (2010a)"},{"why":"Provides the second NUV dataset and the evidence for a torus of escaped material around the system.","marker":"Haswell et al. (2012)"},{"why":"The 2D multi-fluid hydrodynamic model and the 'captured by the star' versus 'blown by the wind' regime classification are taken from these works.","marker":"Shaikhislamov et al. (2014, 2016)"},{"why":"Supplies the model equations, reaction list, and the averaged tidal gravitational potential used in scenario 1.","marker":"Khodachenko et al. (2015, 2017)"},{"why":"Proposes the bow-shock formation hypothesis for the early ingress that scenario 2 tests and the paper argues is unrealistic.","marker":"Vidotto et al. (2010, 2011)"},{"why":"Gives the specific bow-shock scenario whose shock structure and early-ingress prediction are compared against the simulations.","marker":"Llama et al. (2011)"},{"why":"Provides the resonant double charge-exchange cross-section of Mg with He+ that controls the MgII abundance near the planet.","marker":"DuBois (1986)"},{"why":"Cited for the analogous regime of planetary material captured by stellar gravity.","marker":"Tremblin & Chiang (2013)"},{"why":"Source of the analytical absorption cross-section used for computing the MgII line profiles.","marker":"Bourrier & Lecavelier des Etangs (2013)"},{"why":"Supplies the optical transit light curve to which the simulated MgII transit curves are compared.","marker":"Hebb et al. (2009)"}],"fun_headline_variants":["WASP-12b's escape is gravity-driven, not a shock","Tidal capture, not bow shock, sets WASP-12b's mass loss","Stellar gravity, not shock, forms WASP-12b's twin streams","WASP-12b: gravity dominates escape, bow shock unrealistic","Tidal force, not a bow shock, governs WASP-12b's evaporation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The model's 2D axisymmetric geometry ignores the Coriolis force and assumes the stellar wind and XUV flux arrive along the same direction, so the computed MgII distribution is trusted only within about five planetary radii; if Coriolis bending redirects the absorbing stream, the conclusion of no early ingress could change.","fun_headline_variants_meta":{"raw":{"variants":["WASP-12b's escape is gravity-driven, not a shock","Tidal capture, not bow shock, sets WASP-12b's mass loss","Stellar gravity, not shock, forms WASP-12b's twin streams","WASP-12b: gravity dominates escape, bow shock unrealistic","Tidal force, not a bow shock, governs WASP-12b's evaporation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000695,"raw_usage":{"total_tokens":3198,"prompt_tokens":1055,"completion_tokens":2143,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":671,"completion_tokens_details":{"reasoning_tokens":2040}},"tokens_in":671,"tokens_out":2143,"duration_ms":16224,"temperature":1.0,"reasoning_tokens":2040,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:41:01.391594+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run a 3D simulation of WASP-12b that includes the Coriolis force and lets the stellar wind arrive at an angle to the XUV flux: if the resulting MgII column ahead of the planet shifts the computed transit ingress earlier than the optical contact, the paper's central conclusion of no early ingress in the realistic regime is falsified. The observed variable early ingress in the NUV transit light curves is the corresponding existing test.","supporting_citations":[{"cited_title":"A., Fossati L., Ayres T., France K., Froning C","cited_arxiv_id":null,"evidence_quote":"Provides the second NUV dataset and the evidence for a torus of escaped material around the system."},{"cited_title":"L., Shaykhislamov I., Lammer H., Prokopov P","cited_arxiv_id":null,"evidence_quote":"Supplies the model equations, reaction list, and the averaged tidal gravitational potential used in scenario 1."},{"cited_title":"D., Physical Review A, 1986, 34(4),","cited_arxiv_id":null,"evidence_quote":"Provides the resonant double charge-exchange cross-section of Mg with He+ that controls the MgII abundance near the planet."},{"cited_title":"A., Bouchy F., Stempels H","cited_arxiv_id":null,"evidence_quote":"Supplies the optical transit light curve to which the simulated MgII transit curves are compared."}],"review_version":1}