{"id":"048b19a2-0752-4b7f-9685-6fbc4db43632","arxiv_id":"2411.09194","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"Non-equal-mass planet systems are less dynamically stable than equal-mass systems at the same Hill-radius spacing, especially away from first-order resonances, which may help explain the observed peas-in-a-pod mass uniformity.","lead":"This paper uses computer simulations to show that planetary systems with more equal planet masses stay stable longer than systems with very different masses, as long as the planets are not trapped in a special orbital resonance. It suggests this 'survival bias' may be part of why the Kepler telescope saw many systems with similar-sized planets, the so-called peas-in-a-pod pattern.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed Gm–stability correlation may be an artifact of the strong anti-correlation between Gini index and minimum planet mass; the sampling does not control for this confound.","rationale":"I chose the m_min-Gm confound as the single most load-bearing concern. The reader's stated weakest assumption (close encounter threshold) is valid but seems conservative: high-Gm systems usually have a small innermost planet, so the threshold R_h,inner is smaller than in equal-mass runs, meaning instability would be detected later, not earlier; the observed shorter times thus likely persist under a more physical mutual-Hill-radius criterion. The m_min confound, by contrast, attacks the interpretation: if the anti-correlation is mediated by the minimum mass, then the Gini index is not a causal parameter, and the paper's title-level claim about 'similar masses' is not distinguished from 'no very small planet.' The proposed fixed-m_min test directly settles this. The absence of code and data is a reproducibility issue but not a logical flaw in the argument, so it does not change the conditional verdict.","tokens_in":8546,"tokens_out":17737,"duration_ms":205862,"concrete_test":"Run a new suite of REBOUND/Mercurius simulations identical to Section 2 but with the minimum planet mass fixed to a constant (e.g., 0.5 m⊕ or 1.0 m⊕) for every system, while still matching the total mass of 30 m⊕ and N=8; vary Gm by redistributing the remaining mass among the other seven planets (e.g., by concentrating mass in one or two planets). For each K > 4 and away from first-order MMRs, recompute the Spearman correlation between log(t/t0) and Gm. If the correlation becomes consistent with zero (within the same p-value thresholds used in Section 3.2.2), the reported anti-correlation is a consequence of the minimum planet mass, not of the Gini index, and the central claim should be rephrased.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's mass sampling (Section 2) fixes total mass at 30 m⊕ and N=8, imposes only a lower bound m_i ≥ 0.1 m⊕, and draws systems with Gm 'uniformly spread' between 0 and 0.97. For fixed total mass and N, the Gini index is almost monotonically tied to the minimum planet mass: a system with Gm≈0.9 must contain at least one planet near the 0.1 m⊕ floor, while a Gm≈0 system has all eight planets near 3.75 m⊕. The paper's own mechanism (Section 3.2.1, Eq. 4) predicts eccentricity growth e ∝ m^{-1/2}, so the smallest planet is the expected trigger of instability. Thus the observed Spearman anti-correlation between log(t/t0) and Gm (Section 3.2.2) could be entirely mediated by m_min rather than by 'mass uniformity' as an independent variable. Equivalently, the general statement 'non-equal mass systems are less stable than equal-mass systems' may reduce to the known and much weaker statement that systems containing a very low-mass planet are less stable. The paper does not report any run that holds m_min fixed while varying Gm, so this confound is unresolved. The close-encounter threshold issue raised by the reader is real but appears to bias against the reported effect when the innermost planet is typically small (which is the case for high-Gm systems), so it is not the primary threat.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript reports N-body simulations of eight-planet systems with equal mutual Hill spacing parameter K and fixed total planetary mass, varying the mass distribution as quantified by the adjusted Gini index Gm. The authors find that for K > 4 and away from first-order mean-motion resonances, the dynamical instability timescale t/t0 is anti-correlated with Gm, with high-Gm systems becoming unstable about two to four orders of magnitude sooner than equal-mass systems. They interpret this via energy equipartition (higher eccentricities in the smallest planets) and argue that survival bias may contribute to the 'peas-in-a-pod' mass uniformity observed in Kepler multi-planet systems.","tokens_in":8887,"tokens_out":6360,"duration_ms":60948,"significance":"If the central result is robust, it provides a simple dynamical-selection mechanism for intra-system mass uniformity that complements formation-based explanations. The paper's strengths include a well-defined integration protocol using REBOUND/Mercurius, exploration of three total masses (10, 30, and 100 Earth masses), an additional well-ordered mass configuration that addresses some observational constraints, and a clear statement of limitations in the comparison with observed near-resonant systems. The simulations are described in sufficient detail that the results are in principle reproducible, and the qualitative anti-correlation appears consistent across several K values and total masses. However, the central conclusion is currently under-supported because the analysis does not separate the effect of mass uniformity from the effect of the presence of a very low-mass planet, and because censored simulation times are treated as measured instability times in the correlation analysis.","major_comments":[{"comment":"The Spearman rank correlation in Figure 2 is computed on values of log(t/t0) where integrations that reach the maximum time of 10^8 yr are included as if they were actual instability times. For K > 8, a substantial fraction of low-Gm systems presumably survive to the maximum time, so these data points are right-censored. Treating censored values as exact times compresses the apparent distribution and can produce or exaggerate a spurious anti-correlation. The paper should either restrict the correlation analysis to K ranges with negligible censoring, use survival-analysis methods (e.g., log-rank tests or proportional hazards), or explicitly report the survival fraction as a function of Gm and K and show that the anti-correlation persists under a conservative imputation of the censored values.","section":"Section 3.2.2, Figure 2"},{"comment":"The sampling in Section 2 fixes total mass and N with only a lower bound m_i >= 0.1 Earth masses, so Gm is strongly anti-correlated with the minimum planet mass for fixed total mass and planet number. The mechanism proposed in Section 3.2.1 (Eq. 4) predicts e proportional to m^{-1/2}, so the smallest planet is expected to be the first to reach high eccentricity and trigger close encounters. Consequently, the observed anti-correlation between log(t/t0) and Gm may be fully mediated by the minimum planet mass rather than by the degree of mass uniformity. To support the general conclusion in Section 5 that 'non-equal mass systems are less stable than equal mass systems' as a statement about mass uniformity, the authors need to control for m_min, for example by generating systems with a fixed minimum mass while varying Gm, or by stratifying the correlation by m_min. Without such a test, the findings are consistent with the weaker statement that systems containing a very low-mass planet are dynamically fragile.","section":"Section 3.2.2 and Section 3.2.1"},{"comment":"The close encounter threshold is defined as the Hill radius of the innermost planet, Rh = a1(mu1/3)^(1/3). Because planetary masses are randomly assigned, the innermost mass varies substantially and is likely correlated with Gm: high-Gm systems often contain a very low-mass planet, and if that planet is innermost, the threshold is smaller, so a closer approach is required to register an instability. This makes the definition of t non-uniform across systems and can affect the quantitative claim of two to four orders of magnitude difference even if the qualitative anti-correlation is robust. I recommend using a pair-dependent mutual Hill radius for the encountering pair, or a fixed physical distance, and testing that the main result is insensitive to the choice of threshold.","section":"Section 2"}],"minor_comments":[{"comment":"The procedure for generating mass distributions with Gm uniformly spread between 0 and 0.97 is not described; please provide the algorithm (e.g., Dirichlet-like sampling with acceptance-rejection) so that readers can reproduce the initial conditions.","section":"Section 2"},{"comment":"The three Gm bins have different numbers of systems per K value, but the figure does not report the sample sizes in each bin or clearly define the error bars for the non-equal mass groups; please add this information in the caption or text.","section":"Figure 1"},{"comment":"The equipartition argument leading to e proportional to m^{-1/2} assumes that each planet acquires a similar random energy, but the systems start on circular, coplanar orbits; the source of the initial random energy (e.g., early resonant overlap or mutual encounters) should be stated more explicitly.","section":"Section 3.2.1"},{"comment":"The statement that instability timescales are 'two to four orders of magnitude' longer for equal-mass systems should be accompanied by a quantitative comparison based on medians at specific K values rather than a visual estimate from Figure 1.","section":"Section 3.2.2 and Section 5"},{"comment":"The comparison with observed near-resonant planet pairs, which states that they generally have Gm < 0.38, would be strengthened by specifying the sample selection and the definition of 'near-resonant' used in the cited Goyal et al. (2023) work.","section":"Section 4"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely question and the simulations are clearly defined, but the two main concerns—censoring in the correlation analysis and the confound between Gm and the minimum planet mass—are load-bearing for the central claim. Both are addressable with additional analysis or a more cautious interpretation. I would not recommend rejection, but the revision needs to either demonstrate robustness of the anti-correlation to these issues or substantially soften the general conclusions in the abstract and Section 5."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know this paper for one thing: it runs a clean, systematic N-body sweep varying the Gini index of planetary masses and shows a robust anti-correlation between stability time and mass unevenness. That is new relative to Rice & Steffen (2023), which capped mass deviations at 30%. They also show the effect vanishes near first-order MMRs and check both ordered and random mass arrangements. The writing is clear, the MMR discussion is careful, and the authors are honest that the mechanism does not explain the enhanced uniformity of near-resonant pairs.\n\nThe soft spot is a real confound, not a manufactured one. Their sampling fixes total mass (30 Earth masses, N=8) and only imposes m_i >= 0.1 Earth masses. For fixed total mass and N, the Gini index is almost monotonically tied to minimum planet mass: high-Gm systems must contain at least one planet near the 0.1 Earth mass floor, while Gm about 0 systems have all planets near 3.75 Earth masses. Their own mechanism (e proportional to m^-1/2) targets the smallest planet as the instability trigger, so the observed anti-correlation may be entirely mediated by the minimum planet mass. No run holds the minimum mass fixed while varying Gm, so the paper does not distinguish mass uniformity from absence of a very small planet. This is the main threat, and it undercuts the strong conclusion that non-equal mass systems are intrinsically less stable.\n\nThe close-encounter issue the reader raised is secondary. Using the innermost planet's Hill radius as the threshold biases against the reported effect in high-Gm systems, because the innermost planet is often small in those systems; so it is conservative, not fatal. Still, it is a sloppy definition and should be fixed. There is also no code or data release, which is a limitation for a simulation paper.\n\nThe central numerical trend is real and worth reporting, but the interpretation as a survival-bias mechanism for peas-in-a-pod is not yet convincing. If the effect is actually about minimum mass, then the relevant statement is that systems with a very low-mass planet are unstable, which is not the same as saying similar-mass systems are preferred, and it may not transfer to Kepler systems where planets are not as extreme as 0.1 Earth masses.\n\nI would send this to peer review, because the question is important and the sweep is useful, but I would insist the authors add controlled simulations that fix the minimum mass while varying Gm, and release their initial conditions. Without that, the headline claim should be softened. For a reading group, it is a good case study in how sampling choices can create a confound in N-body experiments.","headline":"Useful Gini-index stability sweep, but the headline anti-correlation is likely confounded by minimum planet mass; worth publishing after that is addressed.","tokens_in":754,"tokens_out":868,"would_cite":true,"duration_ms":59146,"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":"Planetary systems with equal-mass planets remain dynamically stable two to four orders of magnitude longer than unequal-mass systems at the same spacing, according to N-body simulations.","keywords":["exoplanets","planetary systems","dynamical instability","Gini index","mass uniformity","mean motion resonances","N-body simulations","peas-in-a-pod"],"falsifier":"Re-run the simulation suite with the instability threshold set to a fixed physical separation (for example, one Solar radius, or a fixed fraction of the mutual Hill radius of the encountering pair) instead of the innermost planet's Hill radius; if the two-to-four-order-of-magnitude gap between equal-mass and high-Gini systems disappears, the reported anti-correlation is an artifact of the mass-dependent definition.","tokens_in":8374,"feed_emoji":"🪐","tokens_out":8445,"duration_ms":84921,"temperature":0.7,"pith_summary":"This paper uses N-body simulations to ask whether planetary systems whose planets have similar masses are dynamically more stable than systems with the same total mass and spacing but unequal masses. It finds that, away from first-order mean-motion resonances and for spacing parameter $K>4$, instability time decreases as the Gini index of planetary masses increases: equal-mass systems survive two to four orders of magnitude longer than high-Gini systems. The proposed mechanism is equipartition of random energy, which drives the smallest planets to higher eccentricities and earlier close encounters. The authors argue that this survival bias should make mass-uniform systems more observable, contributing to the 'peas-in-a-pod' pattern seen in Kepler multi-planet systems.","feed_headline":"Equal-mass planet systems stay stable 10,000x longer","feed_subtitle":"Simulations tie the 'peas-in-a-pod' pattern to survival bias: unequal-mass systems go unstable sooner.","key_machinery":"The machinery is a controlled simulation grid: eight planets on circular, coplanar orbits around a solar-mass star, fixed total mass of 30 Earth masses, and constant spacing parameter $K=(a_{i+1}-a_i)/R_{\\mathrm{H}}$ in units of the mutual Hill radius, with $K$ scanned from 3 to 10. Mass heterogeneity is quantified by the adjusted Gini index $G_m$ (0 for equal masses, approaching 1 for a dominant single planet). The dynamical explanation is carried by equipartition of random energy: since random energy is approximately $\\tfrac{1}{2} m e^2/a$, equal sharing gives $e \\propto m^{-1/2}$, so the smallest planets become the most eccentric and trigger the first close encounter.","core_discovery":"The central claim is that mass uniformity is itself a stabilising property: for non-resonant planetary systems with fixed total mass and equal spacing in units of mutual Hill radius, non-equal-mass systems are less stable than equal-mass systems. In the simulations, the scaled instability time $t/t_0$ is anti-correlated with the adjusted Gini index $G_m$ for $K>4$ away from first-order mean-motion resonances, with Spearman rank coefficients near $-0.7$ for non-resonant $K$ and $p<10^{-4}$; equal-mass systems remain stable two to four orders of magnitude longer than those with high $G_m$. Near first-order resonances the relation is non-monotonic because small mass differences shift planet pairs off exact resonance and can stabilise them. The authors attribute the destabilisation to equipartition of random energy, which gives $e \\propto m^{-1/2}$, so lower-mass planets reach high eccentricities sooner. They conclude that survival bias likely contributes to the observed mass uniformity of Kepler multi-planet systems, while noting it cannot by itself explain the enhanced uniformity of near-resonant planet pairs.","pith_inferences":["Beyond the paper, one can test the survival-bias explanation directly by comparing the Gini index of compact multi-planet systems across stellar ages: if older systems are more mass-uniform, the bias is at work, whereas a flat age dependence would point to formation.","Beyond the paper, the equipartition mechanism predicts that within a system about to go unstable, the smallest planet's eccentricity should grow fastest; reconstructions of eccentricity evolution in simulated or observed systems could confirm or refute this ordering.","Beyond the paper, since the simulations terminate at first close encounter, extending them through mergers and collisions would let the final Gini index of survivors be compared with observed systems, directly quantifying how much of the 'peas-in-a-pod' pattern comes from post-instability processing.","Beyond the paper, the dependence of the stability gap on planet number and total mass is unexplored; scaling the same constant-total-mass setup to 4, 6, or 10 planets would show whether the two-to-four-order-of-magnitude gap is generic or specific to eight-planet systems."],"forward_implications":["For observed non-resonant multi-planet systems with $K>4$, the simulations predict that the surviving population should be biased toward low $G_m$, so mass uniformity should be more common among older or more tightly packed systems.","Near first-order mean motion resonances the trend is non-monotonic: slight mass inequality can move planet pairs off exact resonance and lengthen stability, so the anti-correlation between $t/t_0$ and $G_m$ is weaker or absent there.","Systems with $K>8.5$ and nearly equal masses can remain stable beyond 100 Myr in these simulations, implying that such configurations are the most likely to be detected.","The simulations stop at the first close encounter, so the reported instability times are times to first strong interaction, not to full collisional relaxation; the observed mass uniformity of near-resonant pairs ($G_m < 0.38$) is not explained by survival bias alone."],"supporting_citations":[{"why":"Supplies the equal-mass baseline: log($t/t_0$) grows linearly with $K$, with dips at MMRs, against which non-equal-mass runs are compared.","marker":"Chambers et al. 1996"},{"why":"Prior study of equal-separation systems with non-equal masses that this paper extends by varying the degree of mass non-uniformity.","marker":"Rice & Steffen 2023"},{"why":"Source of the adjusted Gini index $G_m$ used to quantify mass heterogeneity.","marker":"Goyal & Wang 2022"},{"why":"Provides the equipartition of random energy relation $e \\propto m^{-1/2}$, the mechanism invoked for smaller planets destabilising the system.","marker":"Kokubo & Ida 2012"},{"why":"Shows that equipartition of angular momentum deficit excites eccentricities of lower-mass planets more, supporting the proposed destabilisation channel.","marker":"Wu & Lithwick 2011"},{"why":"Documents equal-mass instability timescales and MMR overlap deviations used as a comparison point.","marker":"Obertas et al. 2017"},{"why":"Provides the REBOUND N-body code used for the integrations.","marker":"Rein & Liu 2012"},{"why":"Provides the Mercurius integrator used for the dynamical simulations.","marker":"Rein et al. 2019"},{"why":"Reports that observed near-resonant planet pairs have $G_m < 0.38$, the observational constraint the paper compares against.","marker":"Goyal et al. 2023"}],"fun_headline_variants":["Equal-mass planets are 10,000x more stable than uneven ones","Mass uniformity in systems increases stability away from resonances","Survival bias may explain 'peas-in-a-pod' exoplanet patterns","Similar-mass planets resist dynamical instability longer"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that a close encounter closer than the Hill radius of the innermost planet marks the same 'instability' event across systems with very different mass distributions, even though that radius depends on the innermost planet's mass.","fun_headline_variants_meta":{"raw":{"variants":["Equal-mass planets are 10,000x more stable than uneven ones","Mass uniformity in systems increases stability away from resonances","Survival bias may explain 'peas-in-a-pod' exoplanet patterns","Similar-mass planets resist dynamical instability longer"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000232,"raw_usage":{"total_tokens":1481,"prompt_tokens":931,"completion_tokens":550,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":547,"completion_tokens_details":{"reasoning_tokens":479}},"tokens_in":547,"tokens_out":550,"duration_ms":39785,"temperature":1.0,"reasoning_tokens":479,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T20:55:25.709055+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the simulation suite with the instability threshold set to a fixed physical separation (for example, one Solar radius, or a fixed fraction of the mutual Hill radius of the encountering pair) instead of the innermost planet's Hill radius; if the two-to-four-order-of-magnitude gap between equal-mass and high-Gini systems disappears, the reported anti-correlation is an artifact of the mass-dependent definition.","supporting_citations":[{"cited_title":"2017, , 293, 52","cited_arxiv_id":null,"evidence_quote":"Documents equal-mass instability timescales and MMR overlap deviations used as a comparison point."},{"cited_title":"E., Wetherill , G","cited_arxiv_id":null,"evidence_quote":"Supplies the equal-mass baseline: log($t/t_0$) grows linearly with $K$, with dips at MMRs, against which non-equal-mass runs are compared."},{"cited_title":"R., & Steffen , J","cited_arxiv_id":null,"evidence_quote":"Prior study of equal-separation systems with non-equal masses that this paper extends by varying the degree of mass non-uniformity."},{"cited_title":"2012, Progress of Theoretical and Experimental Physics, 2012, 01A308","cited_arxiv_id":null,"evidence_quote":"Provides the equipartition of random energy relation $e \\propto m^{-1/2}$, the mechanism invoked for smaller planets destabilising the system."},{"cited_title":"V., Dai , F., & Wang , S","cited_arxiv_id":null,"evidence_quote":"Reports that observed near-resonant planet pairs have $G_m < 0.38$, the observational constraint the paper compares against."}],"review_version":1}