{"id":"f000c157-1e4e-4b6c-a216-c0b784089def","arxiv_id":"1908.06108","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Unstable periodic orbits and dynamical stability maps can predict long-term instability and white dwarf pollution in one-planet plus one-asteroid systems, and match 10 Gyr N-body simulations.","lead":"This paper uses the mathematics of periodic orbits and chaos maps to predict when an asteroid will be flung into a white dwarf star, polluting its atmosphere with metals. It offers a quick, analytic way to survey planetary system architectures that would take billions of years to simulate directly.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Predictive maps depend on an unverified 250 kyr DFLI cutoff, and the reported N-body validation samples only chaotic regions, so the false-negative rate over 10 Gyr is not established.","rationale":"The reader's conditional verdict identifies the same soft spot: the DFLI tmax = 250 kyr adequacy is asserted rather than demonstrated, and the N-body comparison is partly a selection effect because initial conditions are drawn from chaotic regions. I agree that this is the most load-bearing assumption for the central claim that the DS maps are reliable diagnostics. The paper has genuine strengths: the periodic orbit families are computed with standard continuation and monodromy methods, the DFLI maps are a recognized chaos indicator, and the N-body runs use a different integrator (mercury Bulirsch-Stoer) and physical units, so the purple-diamond concentration in chaotic regions is real supporting evidence. But that evidence does not bound the false-negative rate in the dark regions, nor does it test whether 250 kyr is long enough to reveal all chaos that matters over 10 Gyr. The proposed check — recomputing the DFLI maps at longer tmax — would directly settle whether the cutoff is adequate. If the classification is stable, the central claim survives; if not, the maps need recalibration before use as quick diagnostics. No change to the reader's conditional verdict is needed.","tokens_in":13374,"tokens_out":8836,"duration_ms":89535,"concrete_test":"Recompute the DFLI maps of Fig. 1 on the identical (eA,eP) grid with tmax = 250 kyr, 2.5 Myr, and 25 Myr, keeping all other numerical settings fixed, and record the fraction of grid cells whose classification changes from dark (regular) to pale (chaotic) as tmax grows. If any non-negligible fraction (e.g., >5%) converts, the 250 kyr cutoff is not adequate and the maps would under-predict white dwarf pollution; if the classification is invariant, the time-scale extrapolation is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2.2 states that DFLI integration to tmax = 250 kyr 'has been proved to be adequate for revealing chaos in the RTBP with the masses we use', but no derivation, citation, or convergence test is given. Every DS map in Section 3 inherits this cutoff. The central claim is that these maps can estimate pollution prospects and timescales for one-planet architectures over ~10 Gyr. A finite-time chaos indicator at 250 kyr cannot by itself certify either direction of that extrapolation: an orbit with weak chaos or slow Arnold diffusion can appear regular within 250 kyr yet become collisionally unstable on Gyr timescales, and an orbit that is chaotic on 250 kyr need not collide with the white dwarf within 10 Gyr. The scaling from normalized RTBP units to the physical white-dwarf system is also only sketched in Section 4.2 (Eq. 2 plus 'the time remained the same'), so it is not demonstrated that a fixed 250 kyr normalized integration corresponds to a physically meaningful, architecture-independent time window. The N-body check in Section 5 is not a clean external test: 'we chose the initial conditions for which the DFLI traced chaoticity (DFLI>15)', so agreement is tested only inside chaotic regions. Dark/stable regions, which are the ones a diagnostic needs to flag as safe, are not sampled. Thus the maps' false-negative rate over Gyr timescales is the load-bearing unproven quantity.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes unstable periodic orbits and DFLI-based dynamical stability maps in the planar and spatial circular/elliptic restricted three-body problems as fast diagnostic tools for white dwarf metal pollution in one-planet architectures. It generalizes Antoniadou & Veras (2016) to eccentric planets (eP = 0.048) and inclined asteroids, concentrating on the 2:1 interior mean-motion resonance. The authors compute DS maps on several planes of initial conditions (Figs. 1–6), then run Bulirsch-Stoer N-body integrations for up to 14 Gyr in a limited phase-space region, and compare outcomes (stable, ejected, planet-collision, WD-collision) with the DFLI>15 chaotic regions (Figs. 7–10). They conclude that the maps agree well with N-body outcomes and can be used to estimate pollution prospects and timescales for one-planet white dwarf systems.","tokens_in":13632,"tokens_out":5284,"duration_ms":55772,"significance":"If the claimed agreement holds, the work provides a computationally cheap, non-circular prior for N-body explorations and for interpreting the growing sample of polluted white dwarfs. The central validation is not circular: the periodic orbits and DS maps are computed first, and the N-body simulations are used as an independent comparison; this is a genuine strength. The extension to the ERTBP and to inclined orbits via vertical stability is also a useful step beyond the previous circular coplanar treatment. However, the predictive claim is currently stronger than the evidence: the paper itself limits the N-body comparison to a small region of phase space, the finite-time DFLI cutoff is asserted without convergence support, and only chaotic (DFLI>15) initial conditions are tested, leaving the false-negative rate in stable regions unquantified. The maps may well be useful diagnostics, but the quantitative basis for their use as a global tool needs to be established.","major_comments":[{"comment":"The assertion that a DFLI integration time of tmax = 250 kyr is 'adequate for revealing chaos in the RTBP with the masses we use' is load-bearing for every DS map in the paper, yet no derivation, citation, or convergence test is provided. Since the maps are subsequently used to make statements about evolution over ~10 Gyr, the authors should either supply a quantitative justification for this time window (e.g., DFLI maps computed for several tmax values and shown to be stable) or restrict the conclusions to the chaotic-tracing capability on that particular timescale.","section":"§2.2"},{"comment":"The scaling between the normalized RTBP units and the physical white dwarf system is not adequately specified. The sentence 'The rest of the orbital elements and the time remained the same' is not sufficient: rescaling the semimajor axis by ζ^(1/3) changes the orbital period unless a compensating time-unit transformation is applied, and the physical masses used in the N-body runs are not stated. The authors need to demonstrate that a fixed 250 kyr normalized integration corresponds to a well-defined, architecture-independent physical time window for aP = 10 au and mWD = 0.6 M⊙, or explain the convention under which the time coordinate is unchanged.","section":"§4.2, Eq. (2)"},{"comment":"The N-body validation is not an unbiased external test of the maps because the initial conditions are selected from the chaotic regions: Section 5 states that 'we chose the initial conditions for which the DFLI traced chaoticity (DFLI>15)'. Stable or regular regions are therefore never sampled, so the false-negative rate over Gyr timescales is unknown. The abstract's claim that the maps 'can be used as tools' to estimate pollution prospects requires both chaotic and stable regions to be tested; at minimum, the authors should run N-body simulations for a sample of DFLI<10 initial conditions and report the resulting classification rates.","section":"§5"},{"comment":"The agreement between the DS maps and N-body outcomes is assessed only qualitatively, by visual inspection of where purple diamonds concentrate. With only 10–50 asteroids per simulation set and no reported run counts per panel, the statistical weight behind the claim is unclear. The authors should quantify the comparison, for example by giving the fraction of WD-collision outcomes that fall inside the DFLI>15 region and the number of initial conditions in each class, to support the predictive-power statement.","section":"§5.1, Fig. 7"}],"minor_comments":[{"comment":"The text refers to the bottom panel of Fig. 7 as the (θ1,θ2) = (0,0) configuration, but the figure caption lists the bottom panel as (θ1,θ2) = (π,π); this inconsistency should be fixed.","section":"§5.1"},{"comment":"The phrase 'aP = 10 au (or scaled as aP(N) = 8.439009789 au)' is confusing: if the physical planet semimajor axis is 10 au, it is unclear why the simulation uses 8.439 au, and the physical masses mWD and mP used in the integrator are not explicitly given. The authors should state the exact unit conventions and masses used in Mercury.","section":"§4.2"},{"comment":"The text says the integration stops when the DFLI reaches the threshold 10^30, while Section 5 uses DFLI>15 as the chaos criterion; the relationship between these two thresholds (linear versus logarithmic DFLI) should be clarified.","section":"§2.2"},{"comment":"The sign convention for eA and eP in Figs. 1 and 6 is explained in words, but the plots would be easier to read if the positive/negative branches were labeled directly on the eccentricity axes.","section":"§3.1"},{"comment":"In Fig. 10, the white crosses mark the four planar periodic orbits, but the caption does not state which of the four are horizontally stable or unstable; adding this information would help connect the 3D-ERTBP results to the stability discussion in Fig. 1.","section":"§5.2"}],"recommendation":"major_revision","confidential_remarks":"The central idea is sound and the non-circular comparison is a strength, but the paper currently overstates the predictive power of the maps. The key missing pieces are a justification or convergence test for the 250 kyr DFLI window, a clear demonstration of the unit scaling from the RTBP to physical white dwarf systems, and N-body tests that include stable/regular initial conditions. These are fixable within the scope of the manuscript, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a solid, well-scoped extension of the authors' earlier CRTBP-based maps to eccentric and inclined architectures, and the central idea is sound. The new results are the DS maps for the 2D-ERTBP, 3D-CRTBP, and 3D-ERTBP for a Jupiter-mass planet with eccentricity 0.048 in the 2:1 MMR, plus a 10 Gyr N-body suite that is independent of the map construction. The paper clearly explains the periodic-orbit families, the four symmetric configurations, and the distinction between horizontal and vertical stability. The figures showing purple diamonds (WD collisions) concentrated near unstable periodic orbits are genuinely supportive, and Section 5 gives an honest account of the initial-condition selection.\n\nThe main soft spot is that the N-body comparison only samples chaotic initial conditions (DFLI > 15), so the false-negative rate of the maps over 10 Gyr is never tested. The maps are meant to be quick diagnostics; if regular regions hide slow instabilities that the 250 kyr DFLI misses, the tool would misclassify some systems. This is the strongest criticism and is directly addressable with a small set of N-body simulations drawn from regular regions, or at least a quantitative statement about why the cutoff is adequate.\n\nSecond, the tmax = 250 kyr claim in Section 2.2 is asserted without citation or convergence test. A finite-time chaos indicator can miss weak chaos, so the leap to 10 Gyr predictions needs more support. A simple convergence check would fix this.\n\nThird, the unit scaling in Section 4.2 is compressed. The sentence \"the rest of the orbital elements and the time remained the same\" is ambiguous; the authors likely mean the mapping from their 2016 paper, but an explicit derivation or worked example would help. This is minor but relevant to the timescale claims.\n\nThe paper is honest about its limitations and does not overclaim beyond the abstract's \"limited region of phase space.\" It deserves a serious referee and likely publication after minor-to-moderate revision. The missing stable-region N-body sample and the DFLI convergence test should be requested, but they do not invalidate the core approach. I'd cite this if working on WD pollution diagnostics, and I'd bring it to a reading group focused on dynamical astronomy or exoplanetary evolution.","headline":"Useful extension of periodic-orbit stability maps to eccentric and inclined WD-planet systems, but the validation samples only chaotic ICs and the 250 kyr DFLI cutoff is unproven.","tokens_in":14216,"tokens_out":7526,"would_cite":true,"duration_ms":74784,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["70F15","70F16","70H12"],"pacs":[],"model":"deepseek-v4-flash","headline":"Using unstable periodic orbits in the restricted three-body problem, this paper shows that chaotic regions identified by dynamical stability maps predict asteroid collisions with white dwarfs in one-planet systems, offering a fast…","keywords":["white dwarf pollution","planetary debris","periodic orbits","restricted three-body problem","chaos indicators","DFLI","2:1 mean motion resonance","N-body simulations"],"falsifier":"Take an initial condition that the DS maps mark chaotic and that was not included in the paper's Section 5 comparison, for example an asteroid with $e_A$ near 0.9 and $e_P$ near 0.2 in the $(\\theta_1,\\theta_2)=(\\pi,\\pi)$ configuration, integrate it with the same Bulirsch-Stoer integrator for 10 Gyr, and check whether the fraction of white-dwarf collisions matches the map's chaotic fraction; a systematic mismatch would falsify the map's predictive power. A second check: rerun the DFLI computations with $t_{\\max}$ increased from 250 kyr to, say, 1 Myr and see if orbits currently classified regular become chaotic.","tokens_in":13107,"feed_emoji":"🪐","tokens_out":9369,"duration_ms":79814,"temperature":0.7,"pith_summary":"White dwarf atmospheres are polluted by rocky debris, but finding which planetary architectures deliver asteroids to the star normally requires $N$-body simulations running up to $10^{10}$ years. This paper argues that unstable eccentric periodic orbits in the restricted three-body problem, together with chaos maps based on the DFLI indicator, can do the job much faster and with no need to rescale time. The authors compute such maps for a white dwarf with one Jupiter-like planet and an inner asteroid near the 2:1 mean-motion resonance, in planar, elliptic, and inclined versions of the problem. They then show that the chaotic regions of the maps coincide with the outcomes of 10-Gyr $N$-body simulations, in particular with asteroid-white-dwarf collisions. If the correspondence holds, the maps become quick diagnostics for pollution prospects in one-planet white dwarf systems.","feed_headline":"Chaos maps flag asteroids that hit white dwarfs","feed_subtitle":"Unstable periodic orbits around one Jupiter-like planet predict pollution without 10-Gyr simulations.","key_machinery":"The central objects are the periodic orbits of the restricted three-body problem in the 2:1 interior mean-motion resonance, continued in eccentricity and inclination, together with their linear (horizontal) and vertical stability. Stable periodic orbits organise regular regions; unstable ones seed chaos. The DFLI (detrended fast Lyapunov indicator, the finite-time Lyapunov indicator divided by integration time) is computed on grids of initial conditions to build dynamical stability maps, with pale regions marking chaos and white marking close-encounter integration failure. The map is computed in dimensionless RTBP units over $t_{\\max}=250$ kyr and then rescaled to physical units through the scaling $a^{(N)}_P = a_P \\zeta^{1/3}$, keeping time and angles fixed, so that the same chaotic structures can be overlaid on $N$-body outcomes.","core_discovery":"The central claim is that the chaotic domains traced by unstable periodic orbits in the 2D-ERTBP, 3D-CRTBP, and 3D-ERTBP accurately predict long-term instability in white dwarf systems with one major and one minor planet, and classify that instability as ejection, planet collision, or white dwarf collision. In the planar elliptic case with $e_P = 0.048$ (Jupiter-like), the 2:1 resonant periodic orbits organise the $(e_A,e_P)$ plane into stable islands (1:1 secondary resonance) and chaotic seas; the authors find that white dwarf collisions concentrate in the chaotic regions, especially for the configurations $(\\theta_1,\\theta_2)=(\\pi,0)$ and $(\\pi,\\pi)$, and that for $(\\pi,\\pi)$ pollution can occur for any asteroid eccentricity once $e_P > 0.1$. In the 3D-CRTBP, the unstable spatial family is surrounded by only a small chaotic region for prograde inclinations, so low-inclination circular-planet systems should not produce pollution; in the 3D-ERTBP, vertically unstable planar periodic orbits yield instability events even at low inclination. The paper concludes that the DFLI-based dynamical stability maps, computed for only 250 kyr in normalized units, agree well with the outcomes of expensive 10-Gyr $N$-body simulations in the tested region of phase space.","pith_inferences":["The same periodic-orbit-plus-DFLI strategy could be carried to other interior mean-motion resonances (for example the 3:1 or 4:1) to see whether the map-to-collision correspondence holds there, extending the diagnostic beyond the 2:1 case studied here.","Because the maps are scale-free and cheap, an ensemble-level use suggests itself: match the observed incidence of metal-polluted white dwarfs against the chaotic fractions of one-planet architectures with different planet eccentricities.","The 250-kyr DFLI threshold is a practical calibration choice; increasing it might expose weakly chaotic orbits currently classified regular, so the maps' reliability at Gyr timescales could be tested by recomputing a few maps at longer $t_{\\max}$."],"forward_implications":["The DFLI-based dynamical stability maps can be used to estimate pollution prospects and timescales for one-planet white dwarf architectures without running 10-Gyr $N$-body simulations.","For the 2:1 resonance, the configuration $(\\theta_1,\\theta_2)=(\\pi,0)$ with $\\Delta\\varpi=\\pi$ produces white dwarf collisions near the resonance, while $(\\pi,\\pi)$ with $\\Delta\\varpi=0$ yields collisions for any asteroid eccentricity once $e_P > 0.1$.","In the 3D-CRTBP, small prograde asteroid inclinations have only a tiny surrounding chaotic region, so circular-planet, low-inclination systems are unlikely to pollute the white dwarf.","For the 3D-ERTBP, vertically unstable planar periodic orbits generate instability events even at low inclination, so vertical stability is a useful additional diagnostic.","Asymmetric unstable periodic orbits, when used as $N$-body initial conditions, lead predominantly to white dwarf collisions or ejections rather than stable evolution."],"supporting_citations":[{"why":"Introduced the periodic-orbit and DS-map approach to white dwarf pollution in the planar CRTBP; this paper's scaling and N-body comparison build directly on it.","marker":"Antoniadou & Veras (2016)"},{"why":"Supplies the 2:1 resonant symmetric periodic orbit families and the asymmetric Family 1 orbits used as initial conditions.","marker":"Antoniadou & Libert (2018a)"},{"why":"Provides the 3D-CRTBP unstable periodic orbit family and the vertical-stability bifurcation framework for the 3D-ERTBP.","marker":"Antoniadou & Libert (2019)"},{"why":"Earlier single-planet scattering study whose 2:1 resonance boundary behaviour motivates the comparison of map predictions with N-body outcomes.","marker":"Debes et al. (2012)"},{"why":"Describes the Bulirsch-Stoer integrator used for all N-body simulations.","marker":"Chambers (1999)"},{"why":"Defines the fast Lyapunov indicator that the DFLI chaos indicator is derived from.","marker":"Froeschlé et al. (1997)"},{"why":"Defines the DFLI and its use in dynamical stability maps for the RTBP.","marker":"Voyatzis (2008)"},{"why":"Full-lifetime N-body simulations that motivate the need for a fast alternative by limiting exploration to a narrow band of phase space.","marker":"Veras et al. (2016)"},{"why":"Shows old polluted white dwarfs exist with cooling ages of 5-8 Gyr, justifying the need for Gyr-scale predictions from short chaos integrations.","marker":"Hollands et al. (2018)"}],"fun_headline_variants":["Periodic-orbit chaos maps spot white dwarf polluters","Eliminate 10-Gyr runs: chaos maps predict white dwarf hits","Eccentric orbits flag asteroids doomed to white dwarf impacts","Fast stability maps forecast white dwarf metal pollution"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The prediction rests on the assumption that chaotic behavior detected by the DFLI indicator over 250 kyr in normalized units, then rescaled to a real white dwarf system, correctly identifies orbits that will actually collide, be ejected, or survive over billions of years.","fun_headline_variants_meta":{"raw":{"variants":["Periodic-orbit chaos maps spot white dwarf polluters","Eliminate 10-Gyr runs: chaos maps predict white dwarf hits","Eccentric orbits flag asteroids doomed to white dwarf impacts","Fast stability maps forecast white dwarf metal pollution"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000799,"raw_usage":{"total_tokens":3607,"prompt_tokens":1134,"completion_tokens":2473,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":750,"completion_tokens_details":{"reasoning_tokens":2405}},"tokens_in":750,"tokens_out":2473,"duration_ms":17570,"temperature":1.0,"reasoning_tokens":2405,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:54:41.172191+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take an initial condition that the DS maps mark chaotic and that was not included in the paper's Section 5 comparison, for example an asteroid with $e_A$ near 0.9 and $e_P$ near 0.2 in the $(\\theta_1,\\theta_2)=(\\pi,\\pi)$ configuration, integrate it with the same Bulirsch-Stoer integrator for 10 Gyr, and check whether the fraction of white-dwarf collisions matches the map's chaotic fraction; a systematic mismatch would falsify the map's predictive power. A second check: rerun the DFLI computations with $t_{\\max}$ increased from 250 kyr to, say, 1 Myr and see if orbits currently classified regular become chaotic.","supporting_citations":[{"cited_title":"I., Veras D., 2016, MNRAS, 463, 4108","cited_arxiv_id":null,"evidence_quote":"Introduced the periodic-orbit and DS-map approach to white dwarf pollution in the planar CRTBP; this paper's scaling and N-body comparison build directly on it."},{"cited_title":"I., & Libert, A.-S.\\ 2019, MNRAS, 483, 2923","cited_arxiv_id":null,"evidence_quote":"Provides the 3D-CRTBP unstable periodic orbit family and the vertical-stability bifurcation framework for the 3D-ERTBP."},{"cited_title":"H., Walsh, K","cited_arxiv_id":null,"evidence_quote":"Earlier single-planet scattering study whose 2:1 resonance boundary behaviour motivates the comparison of map predictions with N-body outcomes."},{"cited_title":"E.\\ 1999, MNRAS, 304, 793","cited_arxiv_id":null,"evidence_quote":"Describes the Bulirsch-Stoer integrator used for all N-body simulations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the DFLI and its use in dynamical stability maps for the RTBP."},{"cited_title":"J., G \\\"a nsicke, B","cited_arxiv_id":null,"evidence_quote":"Full-lifetime N-body simulations that motivate the need for a fast alternative by limiting exploration to a narrow band of phase space."},{"cited_title":"A., G \\\"a nsicke, B","cited_arxiv_id":null,"evidence_quote":"Shows old polluted white dwarfs exist with cooling ages of 5-8 Gyr, justifying the need for Gyr-scale predictions from short chaos integrations."}],"review_version":1}