{"id":"f33a1790-d792-4a71-8d48-d1c518bc22c5","arxiv_id":"1908.10395","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Pairs of circumbinary planets can be captured into stable, eccentric 1:1 coorbital (horseshoe) resonances during disc-driven migration.","lead":"Hydrodynamic simulations show that pairs of planets migrating inward through a disc around a binary star can become trapped in a shared 1:1 horseshoe resonance near the disc's inner edge. The result suggests a new class of stable circumbinary orbits that could exist in systems like Kepler-47 and Kepler-413.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'natural outcome' claim rests on a single viscosity/initial-condition slice; Kepler-413 also requires an unseen equal-mass companion.","rationale":"I read the paper as a proof-of-concept numerical study: with a viscous circumbinary disc and two low-mass planets, a 1:1 horseshoe resonance can form at the common parking radius near the cavity edge and survive after disc removal. The internal logic is coherent, and the n-body follow-up gives independent support that the reported configurations are not transient artifacts. No formal verification or released code exists, but the plots and numbers are consistent with the narrative.\n\nThe concern that carries the most weight is not internal inconsistency but external robustness. The phenomenon appears only in a narrow slice of the parameter space that the authors themselves explore: alpha=0.001 works, alpha=1e-4 fails, and no intermediate values are given. The mechanism depends on both planets avoiding gap opening and sharing a single parking orbit, which is precisely the quantity controlled by viscosity and disc thickness. The absence of any variation in h, initial separation, or mass ratio means we cannot tell whether the two successes are representative or accidental. The Kepler-413 case is additionally weakened by the insertion of an unseen second planet with exactly the observed mass. The authors are suitably cautious at the end, saying that such orbits 'remain to be seen' in reality, and they admit the simulated parking radii are larger than observed. That honesty is a point in their favor, but it reinforces that the conclusion is a model outcome, not a prediction.\n\nGiven these limitations, the reader's CONDITIONAL verdict is appropriate. My stress-test does not change that verdict: the proof-of-existence is not overturned, but the stronger claim of a 'natural outcome' is not yet established.","tokens_in":6416,"tokens_out":11004,"duration_ms":126291,"concrete_test":"Run a focused Kepler-47 parameter grid around the fiducial case: alpha in {3e-4, 1e-3}, h in {0.04, 0.05}, initial outer-planet separations of 2, 5, and 10 mutual Hill radii, and planet mass ratios 1:1 and 1.2:1. If a stable 1:1 horseshoe pair forms in fewer than half of these runs, or only in the exact fiducial box, the result is a special-case outcome rather than a natural one, and the conclusion should be reworded accordingly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's key inference is that stable 1:1 coorbital circumbinary pairs are a natural outcome of convergent disc migration. The evidence is two successful hydro runs plus short n-body follow-ups. The most load-bearing weakness is that success is demonstrated in a narrow parameter regime: alpha=0.001, h=0.04, equal-mass planets, and ad hoc initial separations (7.5/14 abin for Kepler-47; 9/11.5 abin for Kepler-413). Section 4 shows that lowering alpha to 1e-4 makes the inner planet ejected, so a factor-of-ten viscosity reduction destroys the phenomenon; no intermediate alpha, varying h, or varying initial separation is reported. This matters because the capture mechanism rests on both planets not opening gaps and parking at the same cavity edge, and that coincidence is exactly what the viscosity controls. For Kepler-413 the simulation additionally inserts a second 0.21 MJup planet that is not observed, so the system-specific statement in the abstract goes beyond the data. The authors frankly note the simulated planets park farther out than observed, further loosening the connection to real systems. These issues do not falsify the proof-of-existence, but they do prevent the stronger 'natural outcome' conclusion.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports 2D hydrodynamic simulations of circumbinary discs around Kepler-47 and Kepler-413 with two embedded, equal-mass planets that migrate inward under disc torques. In both systems the inner planet parks at the eccentric cavity edge and the outer planet subsequently converges onto the same orbit, producing a 1:1 coorbital configuration with horseshoe-type motion, orbital eccentricities of about 0.25 to 0.30, and aligned orbits. The authors extract the final states and show in pure n-body integrations without the disc that the configurations remain stable for at least 22,000 binary periods. They conclude that stable coorbital circumbinary planets can be a natural outcome of convergent disc migration and suggest that such systems should be searched for in the Kepler data.","tokens_in":6626,"tokens_out":4272,"duration_ms":47300,"significance":"If the result holds, it identifies a plausible formation channel for a previously unrecognized class of circumbinary resonant systems, and the paper gives a clear dynamical mechanism (parking at the same cavity edge) that is physically reasonable. The study has genuine strengths: the 1:1 resonance is an emergent outcome of the migration calculation rather than being inserted by hand; the follow-up n-body runs test long-term stability without the disc; and the authors explicitly acknowledge that their simulated planets lie farther from the binary than the observed planets. However, the significance is limited by the narrow parameter coverage and, for Kepler-413, by the insertion of an unseen equal-mass companion. The result is best read as a proof of existence for a specific disc model, not as a demonstrated general outcome for the observed systems.","major_comments":[{"comment":"The claim that coorbital capture is a natural outcome of convergent migration rests on a single viscosity and initial-condition slice. The authors themselves report that lowering alpha from 0.001 to 1e-4 in Kepler-47 causes gap opening and ejection of the inner planet, so a factor-of-ten change in viscosity destroys the phenomenon. Because alpha controls whether both planets remain gapless and park at the same cavity edge, the paper should either present a small parameter study (intermediate alpha, varied h, varied initial separations) or explicitly soften the wording from 'natural outcome' to 'possible outcome for the chosen disc parameters'.","section":"Section 4, viscosity dependence"},{"comment":"In the Kepler-413 model, a second planet of exactly the observed planet's mass is embedded at 11.5 a_bin, but only one planet is detected in that system. The abstract's statement that 'in Kepler-47 and -413 the planets are captured in a 1:1 resonance' therefore presumes an unseen equal-mass companion. The paper should state this assumption prominently in the abstract and frame the Kepler-413 result as conditional on a hypothetical second planet rather than as a statement about the observed system.","section":"Section 2, Kepler-413 model setup"},{"comment":"No convergence tests or quantitative error estimates are reported. Section 2 states that the 684x584 grid has been shown sufficient in earlier work, but the capture times, libration periods, and eccentricity values quoted in Section 3 could depend on resolution, on the 20,000 T_bin disc convergence criterion, and on the reported time-step. A convergence check or at least an explicit estimate of numerical uncertainty is needed before the quantitative values can be taken at face value.","section":"Section 2, numerical convergence"},{"comment":"The paper acknowledges that the simulated planets park farther from the binary than the observed planets and states that coorbitals are not expected to form very close to the stars, but no quantitative support for that expectation is given. Since this radius mismatch directly limits the applicability of the result to Kepler-47 and Kepler-413, the paper should either provide a physical argument based on existing simulations or explicitly identify this as an open question for future work.","section":"Section 4, radial mismatch with observations"}],"minor_comments":[{"comment":"There is a typo in 'we discuss out results'; it should read 'our results'.","section":"Section 2, first paragraph"},{"comment":"In 'stable coorbital configurations can exits', 'exits' should be 'exist'.","section":"Section 4, first paragraph"},{"comment":"The footnote 'These results will be published in a future study' is not a citable reference; the relevant results should be cited instead, or the statement should be marked as unpublished work.","section":"Section 4, footnote 1"},{"comment":"The caption says the top panels show semi-major axis difference against difference in mean longitude, but the reader must infer which color corresponds to which time; adding a color bar or stating the time interval more explicitly would improve clarity.","section":"Figure 3 caption"}],"recommendation":"major_revision","confidential_remarks":"The paper is a short Letter whose core simulations are internally consistent and whose resonance emerges from the migration calculation rather than being inserted by hand. The main risk is overstatement: the 'natural outcome' phrasing in the abstract and Section 4 goes beyond what a single-viscosity, single-initial-condition study can support, and the Kepler-413 application requires an unseen companion. I would encourage the editor to ask for a modestly revised version that either adds a small parameter robustness test or substantially weakens the generality claims. I do not see a circularity problem with the self-citations; they appropriately supply the code and disc-equilibrium framework."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a solid numerical proof-of-existence paper. For two low-eccentricity circumbinary systems (Kepler-47, Kepler-413) the authors show that two low-mass planets migrating in a viscous disc can be captured into a stable 1:1 horseshoe resonance near the cavity edge. That is genuinely new—previous coorbital work was on single stars or circular equal-mass binaries. They also include useful negative controls: for Kepler-35 and Kepler-34 (higher binary eccentricity) the planets get ejected, so low binary eccentricity looks important.\n\nThe paper does several things well. The resonance is not put in by hand; it emerges from the disc migration. The follow-up n-body integrations (without the disc) show the horseshoe configuration survives for tens of thousands of binary periods. And the authors are candid about the main discrepancy: their simulated planets park farther from the binary than the observed ones. They also state in the abstract that 'it remains to be seen if such orbits exist in reality,' which is the right level of caution.\n\nThe soft spots are real but not fatal. The parameter coverage is thin: one viscosity (alpha=0.001), one aspect ratio (h=0.04), equal planet masses, and a couple of chosen initial separations. The authors themselves show that dropping alpha to 1e-4 makes the inner planet get ejected, so the capture mechanism lives in a narrow regime. There are no convergence tests on resolution or timestep, and no error bars on the orbital elements. The Kepler-413 specific claim is weaker because only one planet is observed; the simulation inserts a second planet of exactly the observed mass. So the abstract's statement that Kepler-413's planets are captured in 1:1 resonance goes beyond the data—it is a model with an unseen companion. The 'natural outcome' phrase in the discussion is also a bit strong; what is demonstrated is a possible outcome for a particular disc model, not a general one.\n\nCitation pattern looks fine. The self-citations to Kley et al. (2019) and Thun & Kley (2018) provide the code and equilibrium disc; the target result is not already in those papers. No circularity problem.\n\nBottom line: worth a serious referee. The result is novel and the negative controls are informative. I would ask the authors to either widen the parameter study a little or soften the 'natural outcome' language to something like 'a possible outcome.' I would not block publication over the narrow regime—the proof-of-existence is the contribution.","headline":"A clean proof-of-existence that circumbinary disc migration can park two planets in a stable 1:1 horseshoe resonance, though the parameter regime is narrow and the Kepler-413 claim relies on an unseen planet.","tokens_in":7178,"tokens_out":2963,"would_cite":true,"duration_ms":28164,"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":"The paper claims that two migrating circumbinary planets can be captured into a stable 1:1 horseshoe resonance around low-eccentricity binaries, with aligned eccentric orbits, and that the configuration survives after the gas disc…","keywords":["circumbinary planets","1:1 mean-motion resonance","coorbital planets","horseshoe orbits","planet migration","protoplanetary discs","Kepler-47","Kepler-413"],"falsifier":"A simulation with $\\alpha = 10^{-4}$ for Kepler-47, which the paper itself reports, breaks the mechanism: the planets open a gap, the inner planet is ejected, and no coorbital resonance appears. Observationally, a re-analysis of the Kepler photometric archive for Kepler-413 that rules out a second planet of mass $\\sim 0.21\\,M_{\\rm Jup}$ near the predicted coorbital separation would falsify the claim that this specific system hosts a 1:1 resonance.","tokens_in":6186,"feed_emoji":"🪐","tokens_out":9714,"duration_ms":93171,"temperature":0.7,"pith_summary":"The paper sets out to explain how several planets can end up orbiting a binary star and, in particular, whether migration through the circumbinary disc can leave two planets sharing the same orbit. Through 2D viscous hydrodynamic simulations of Kepler-47 and Kepler-413, it argues that two low-mass planets of similar mass are captured into a $1{:}1$ mean-motion resonance at the disc's inner cavity, with fully aligned eccentric orbits and horseshoe motion. If this is right, it would be a new class of stable coorbital planet configuration around binary stars, different from the nearly circular coorbitals known in the Solar System and not yet observed among exoplanets.","feed_headline":"Two planets can share a single orbit around a binary star","feed_subtitle":"Simulations park equal-mass pairs in stable horseshoe resonance in Kepler-47 and Kepler-413.","key_machinery":"The central mechanism is the parking position: the inner edge of the precessing, eccentric circumbinary cavity acts as a trap that halts inward migration. Because the planets are small enough not to open a gap in the disc at $\\alpha = 10^{-3}$, both settle at the same orbit near the cavity edge and their orbits align with the precessing cavity; the incoming outer planet catches up and the pair enters horseshoe libration around $\\mathrm{L}_4$ and $\\mathrm{L}_5$. The paper then removes the disc and uses n-body integrations to show that the resulting horseshoe pair is dynamically stable on its own.","core_discovery":"The central claim is that in the low-eccentricity binary systems Kepler-47 and Kepler-413, convergent migration in a viscous circumbinary disc naturally parks two equal-mass planets on the same orbit. The planets stop at the precessing, eccentric inner cavity edge, the parking position, without opening a gap, and the second planet catches up to the first; the pair then librates in a $1{:}1$ resonance, executing horseshoe motion about $\\mathrm{L}_4$ and $\\mathrm{L}_5$. Their orbits remain aligned with each other and with the cavity, with mean eccentricities of about 0.25 to 0.30, and the minimum mean-longitude separation stays above about $20^\\circ$, so the planets never collide. After the disc is removed, n-body integrations over $22\\,000$ to $78\\,500$ binary periods show the horseshoe configuration is stable. The authors therefore conclude that stable coorbital circumbinary planets are a viable outcome of disc-driven migration, probably requiring low binary eccentricity.","pith_inferences":["An implicit extension is to map the boundary of the capture region in binary eccentricity and disc viscosity; the paper's failure cases suggest the mechanism occupies a low-eccentricity, high-viscosity corner of parameter space.","The long libration periods, roughly $10^3$ binary orbits, imply that detecting such pairs photometrically would require monitoring over many years, a signature the paper does not quantify.","If real, this configuration offers a clean test of disc-driven migration theory: the coorbital pair's presence would pin down the disc's viscosity at the time of planet migration, since low-viscosity discs destroy the resonance."],"forward_implications":["Coorbital circumbinary planets should be added to the list of possible outcomes of planet formation around binaries, not treated as dynamically forbidden.","Observational searches for a second planet in Kepler-47 and Kepler-413 are the most direct test; the simulations give a specific expected configuration: aligned, eccentric, horseshoe motion with a minimum longitude separation near $20^\\circ$.","The mechanism predicts a preference for low binary eccentricity: Kepler-35 ($e_{\\rm bin} = 0.14$) and Kepler-34 ($e_{\\rm bin} = 0.52$) did not produce coorbital capture, while $e_{\\rm bin} = 0.02$ and $0.04$ did.","The resonance survives disc dispersal, so the observed population of circumbinary planets could contain such pairs even though the gas that formed them is gone.","Moderate planet mass ratios still allow capture: a 20% mass difference in Kepler-47 gave a tadpole-type coorbital rather than a horseshoe, showing the phenomenon is not limited to exactly equal masses."],"supporting_citations":[{"why":"It supplies the locally isothermal disc model with aspect ratio $h = 0.04$ and the result that low-mass planet orbits align with the eccentric cavity.","marker":"Kley et al. 2019"},{"why":"It provides the 2D hydrodynamic method and grid setup, and shows the chosen domain and resolution suffice for circumbinary disc simulations.","marker":"Thun & Kley 2018"},{"why":"It reports the detection of the third planet in Kepler-47 and gives the planet masses used in the initial conditions.","marker":"Orosz et al. 2019"},{"why":"It provides the detected planet mass in Kepler-413, which the two embedded planets are set to.","marker":"Kostov et al. 2014"},{"why":"It demonstrates that convergent migration can capture planets into a 1:1 coorbital resonance, starting with horseshoe motion as seen here.","marker":"Cresswell & Nelson 2006"},{"why":"It shows that continued migration in a gas disc usually destabilises 1:1 pairs, which the parking position avoids by halting migration.","marker":"Cresswell & Nelson 2009"},{"why":"It establishes the n-body stability domain for eccentric coplanar coorbital planets that the new configurations are compared with.","marker":"Leleu et al. 2018"},{"why":"It documents the Janus-Epimetheus horseshoe coorbital pair, the known Solar System analogue the paper contrasts with its eccentric binary case.","marker":"Yoder et al. 1983"}],"fun_headline_variants":["Binary star systems may host co-orbiting planet pairs","Equal-mass planets can share a circumbinary orbit","Stable horseshoe resonance parks two planets on same orbit","Two planets around a binary star may lock in 1:1 resonance","New class of stable orbits predicted for planets around binary stars"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The main load-bearing premise is that the disc is viscous enough ($\\alpha = 10^{-3}$) to prevent gap opening and that a second planet is present; if the real discs are less viscous, or Kepler-413 has only the one observed planet, the capture mechanism fails.","fun_headline_variants_meta":{"raw":{"variants":["Binary star systems may host co-orbiting planet pairs","Equal-mass planets can share a circumbinary orbit","Stable horseshoe resonance parks two planets on same orbit","Two planets around a binary star may lock in 1:1 resonance","New class of stable orbits predicted for planets around binary stars"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000602,"raw_usage":{"total_tokens":2786,"prompt_tokens":893,"completion_tokens":1893,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":509,"completion_tokens_details":{"reasoning_tokens":1824}},"tokens_in":509,"tokens_out":1893,"duration_ms":15650,"temperature":1.0,"reasoning_tokens":1824,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:44:43.868165+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A simulation with $\\alpha = 10^{-4}$ for Kepler-47, which the paper itself reports, breaks the mechanism: the planets open a gap, the inner planet is ejected, and no coorbital resonance appears. Observationally, a re-analysis of the Kepler photometric archive for Kepler-413 that rules out a second planet of mass $\\sim 0.21\\,M_{\\rm Jup}$ near the predicted coorbital separation would falsify the claim that this specific system hosts a 1:1 resonance.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It supplies the locally isothermal disc model with aspect ratio $h = 0.04$ and the result that low-mass planet orbits align with the eccentric cavity."},{"cited_title":"& Nelson, R","cited_arxiv_id":null,"evidence_quote":"It demonstrates that convergent migration can capture planets into a 1:1 coorbital resonance, starting with horseshoe motion as seen here."},{"cited_title":"& Nelson, R","cited_arxiv_id":null,"evidence_quote":"It shows that continued migration in a gas disc usually destabilises 1:1 pairs, which the parking position avoids by halting migration."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It establishes the n-body stability domain for eccentric coplanar coorbital planets that the new configurations are compared with."},{"cited_title":"F., Colombo, G., Synnott, S","cited_arxiv_id":null,"evidence_quote":"It documents the Janus-Epimetheus horseshoe coorbital pair, the known Solar System analogue the paper contrasts with its eccentric binary case."}],"review_version":1}