{"id":"67b00339-7921-4ecb-80cc-bbe9bc12f115","arxiv_id":"2506.05089","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"In well-characterized multi-planet systems, adjacent planets show moderate radius uniformity but weak or absent mass uniformity, and similarly sized neighbours frequently have very different inferred interiors.","lead":"This study re-examines whether planets in the same star system resemble each other, using a large sample of multi-planet systems with measured masses and radii. It finds that neighbouring planets often share similar sizes but differ in mass, density, and likely interior composition, so similar size is not a reliable sign of similar planetary interiors.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The interior-dissimilarity claim rests on assigning planets to a two-layer Zeng et al. (2016) grid, and the 26/15 split in I is computed for only 41 of 149 similar-radius pairs, so the headline may not generalize beyond planets that fall near that grid.","rationale":"The reader's weakest assumption and my load-bearing concern are the same: the interior-dissimilarity conclusion is conditioned on the adequacy of the Zeng et al. (2016) two-layer composition grid and on the subset of HCR pairs for which the interior distance can be computed. This is the most vulnerable link in the central claim because the title and abstract assert something about interiors, whereas the directly measured quantities are masses and bulk densities. The mass and density conclusions (weak mass correlation, moderate density correlation in HCR, no overall mass correlation) are supported by multiple analyses including error accommodation and null-model comparisons, so the paper's descriptive results about masses and densities are likely robust. What is not robust is the interpretive step from bulk density to interior composition: with a grid that excludes H/He-rich interiors and other realistic structures, a planet assigned to a 'water-world' curve could actually be a rocky core with a modest H/He envelope, and the nearest-curve classification is degenerate. Additionally, the small number of pairs (41 of 149) used for the headline statistic means the phrase 'often' is not established for the full population of similar-sized neighbouring planets. These considerations do not overturn the paper's main empirical findings, nor do they change the conditional status that the reader assigned: the paper is an honest analysis with appropriate caveats in the discussion, but the strongest claim is stronger than the grid-based evidence can bear. A targeted recomputation with an extended composition grid would settle whether the interior claim survives contact with more realistic interior models.","tokens_in":30075,"tokens_out":14275,"duration_ms":166718,"concrete_test":"Recompute the interior distance I for all 149 HCR pairs using a grid that includes H/He-envelope models (e.g., extended Zeng et al. (2019) or Lopez & Fortney (2014) curves) in addition to the Fe/MgSiO3/H2O grid, applying the same nearest-curve assignment and the I>=0.1 threshold. If the fraction of HCR pairs classified as having dissimilar interiors drops below roughly 50%, or if the 26/15 split changes materially, the headline claim would not generalize beyond the originally adopted grid. As part of the same check, inspect the public GitHub code to confirm whether the exclusion condition in Section 3.2 implements I_pi > 0.1 (as the accompanying explanation indicates) or the literally written I_pi <= 0.1.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline 'Planets similar in size are often dissimilar in interior' depends on the interior distance I introduced in Section 3.2 (Eq. 4), which assigns each planet to the nearest Zeng et al. (2016) two-layer Fe-MgSiO3-H2O model curve and treats I>=0.1 as indicating substantially different interiors. This is load-bearing because the claim is about interiors, not just bulk densities. Two issues compound. First, the adopted grid omits H/He envelopes, mixed ice-rock compositions, and non-solar Fe/Si ratios, so planets with such interiors are either mapped onto a misleading water-world or rocky curve or are dropped from the analysis. Second, I is computed for only 41 of the 149 HCR pairs (Table 3), because pairs with any planet outside the grid's validity range (M_pl > 32 M_Earth) or with both planets far from all curves are excluded. The resulting 26/15 split therefore measures dissimilarity only for planets that happen to lie near the two-layer grid; it does not establish that small neighbouring planets with similar radii in general have disparate interiors. The paper acknowledges in Section 4.3 that the volatile-rich population's constituents are 'currently only theorized', but the abstract and title present the interior conclusion without this caveat. The mass and density dissimilarity in the HCR subsample is model-independent and robust, but the specific interior claim goes beyond what the grid can support.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript re-examines 'peas in a pod' correlations in multi-planet systems using a catalogue-based sample of systems with at least two directly measured masses and radii. It reports a moderate intrasystem radius correlation (R=0.516), a weak density correlation (R=0.265), and no overall mass correlation (R=-0.007) in the main sample, with system-wide dispersion consistently smaller in radius than in mass or density. The authors construct high-correlated-in-radius (HCR) subsamples and, using a new interior-distance metric I based on the Zeng et al. (2016) two-layer composition grid, find 26 of 41 HCR pairs with I>=0.1, concluding that similar-size neighbouring planets often have dissimilar interiors. The analysis is accompanied by multiple robustness checks, including Monte Carlo error accommodation, bootstrap tests, Kepler-only subsamples, Gaia DR3 stellar radii, and stability-constrained null populations from the non-parametric Ning et al. (2018) mass-radius relation.","tokens_in":30515,"tokens_out":7820,"duration_ms":88779,"significance":"If the result holds, the paper is an important counterweight to the common interpretation that intrasystem radius uniformity implies density and compositional uniformity: it would show that radius similarity within a system is not a reliable proxy for interior similarity, and it would pose a direct constraint on formation models that produce uniform compositions within a system. The paper is valuable for its large assembled sample, its transparent multiple robustness checks, and its public code and data availability. However, the headline interior claim rests on a selected subset of the HCR sample and on a grid-based metric whose definition and coverage need to be clarified, so the significance is conditional on resolving the issues described below.","major_comments":[{"comment":"The definition of the interior distance I is inconsistent with its verbal description. The text states that I is the mean of the two gap distances, but Eq. (4) gives I=|I_p1+I_p2|/2 with I_pi defined as signed density gaps relative to the reference curve. Opposite-sign gaps can cancel in this formula, yielding a small I for a genuinely dissimilar pair, and the formula is not the mean of the two distances unless the individual gaps are taken in absolute value. Please clarify the sign convention and, if the intended quantity is the mean absolute gap, use I=(|I_p1|+|I_p2|)/2 and re-check the 26/15 classification under the corrected definition.","section":"Section 3.2, Eq. (4)"},{"comment":"The headline interior claim is based on only 41 of the 149 HCR pairs. The sample is further restricted to planets with M_pl <= 32 M_Earth and to pairs in which at least one planet lies near the Zeng et al. (2016) Fe-MgSiO3-H2O two-layer grid; pairs whose planets are far from all grid curves are excluded by construction. Because inclusion therefore depends on the same composition model used to define I, the 26/15 split cannot be read as evidence about small neighbouring planets with similar radii in general. The paper should report the selection fractions explicitly, test how the split changes when the excluded HCR pairs are assigned conservative or model-independent I values, and qualify the abstract and Section 5 statements accordingly.","section":"Section 3.2, Table 3"},{"comment":"The abstract's claim of a 'weak correlation in densities' is not robust to the precision cut used in the same table: the subsample with sigma_P <= 0.5 P gives R=0.654 (n=44, p=1.4e-6), more than double the main-sample value R=0.265. The text mentions this result but the abstract and conclusions present the weak-density finding without this caveat. The authors should either present the precision-restricted result as the better estimate, or provide a quantitative selection-bias or sample-size argument for discounting it; as written, the density conclusion is sample-dependent.","section":"Section 3.1.1, Table 1"}],"minor_comments":[{"comment":"The sentence excluding 'pairs where both planets have I_pi <= 0.1' contradicts its explanatory clause about planets being 'farther than 0.1 from any MRR curve'; the inequality direction and the definition of I_pi (distance to the nearest curve versus signed distance to the reference curve) should be corrected.","section":"Section 3.2"},{"comment":"With the chosen indicator f(x,y)=(1/2(x^2+y^2))^{1/2}, the denominator in g should be ((x^2+y^2)/2)^{1/2}, not (x^2+y^2)^{1/2}; the printed formula is missing a factor of sqrt(2).","section":"Section 3.2, Eq. (3)"},{"comment":"The caption refers to 'dashed black lines' connecting HCR pairs, while the text in Section 3.2 describes 'dotted lines'; these should be made consistent.","section":"Figure 3"},{"comment":"The phrase 'see e.q. Wang (2017); Otegi et al. (2022)' should read 'e.g.'.","section":"Section 3.1.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for A&A and the data/code availability is a clear strength. The main concern is the coverage bias in the interior-distance analysis and the sensitivity of the density correlation to the precision cut; these are fixable with a revised analysis or a properly qualified presentation. I do not see a novelty-disclosure problem, and the self-citation to Hatalova et al. (2023) is not load-bearing."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Hi — I read the Mamonova et al. paper. The result worth remembering is that in a large sample of multi-planet systems with directly measured masses and radii, adjacent-planet masses show no correlation (R≈-0.007) while radii show a moderate correlation (R≈0.52), and the intrasystem dispersion story goes the same way. That is a real, useful update to the peas-in-a-pod literature, and it is likely robust: they cross-validated masses against two databases to exclude theoretical MRR values, ran error-accommodating Monte Carlo draws, checked Kepler-only and Gaia DR3 radii, and built stability-preserving nulls. I would cite the mass-null result.\n\nThe HCR exercise is the cleanest part of the paper. Taking the 149 pairs that are highly correlated in radius and showing that their masses and densities are only weakly correlated (R≈0.33 and 0.46) makes the point that size similarity does not imply mass/density similarity without relying on any composition model. That part deserves to be published.\n\nThe soft spots are in the interior claim. First, the density correlation is not stable: R=0.265 in the full sample jumps to R=0.654 under the σ≤0.5P cut (n=44). They report this honestly, but it means the headline 'weak density correlation' is sample-dependent. Second, the interior distance I is computed for only 41 of the 149 HCR pairs, and it depends on the Zeng two-layer grid, which omits H/He envelopes, mixed ice-rock interiors, and non-solar Fe/Si ratios. The 26/15 split is therefore a statement about planets that happen to fall near that grid, not about similar-radius neighbours in general. The paper's own Section 4.3 concedes that the volatile-rich population's constituents are only theorized, but the title and abstract do not carry that caveat. Third, Eq. 4 as written takes the absolute value of the sum of signed gaps, so a pair straddling the reference curve can look similar by cancellation; that needs a fix or at least a clarification. The moving-window stellar-property results are exploratory but not load-bearing.\n\nThis is a revise-and-resubmit, not a reject. The mass-null and HCR mass/density results are solid; the interior interpretation needs to be restructured around the 41-pair subsample, with the grid limitation and the metric definition fixed, and the title softened or justified. I would send it to a serious referee. Good reading-group material.","headline":"The mass-null result is solid and worth citing; the interior-dissimilarity headline is only as strong as a 41-pair subset mapped onto a two-layer composition grid.","tokens_in":30972,"tokens_out":5085,"would_cite":true,"duration_ms":62431,"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":"Neighbouring planets similar in size often have very different masses and interiors, so radius is a weak guide to composition.","keywords":["peas in a pod","exoplanet interiors","mass-radius relation","multi-planet systems","bulk density","intrasystem uniformity","interior distance","planet composition"],"falsifier":"Inspect the atmospheres or interiors of the 26 pairs the paper labels compositionally dissimilar ($I \\geq 0.1$) using independent diagnostics such as transmission spectroscopy, atmospheric escape, or asteroseismic density constraints; if most of those pairs turn out to share a similar bulk composition, the interior-distance interpretation is refuted.","tokens_in":29886,"feed_emoji":"🪐","tokens_out":15203,"duration_ms":151508,"temperature":0.7,"pith_summary":"The paper sets out to test whether the 'peas in a pod' pattern — the observed tendency for planets in the same multi-planet system to be similar in radius, mass, and spacing — extends to planet composition and interior structure. Using a large sample of systems in which at least two planets have directly measured masses and radii, it finds a moderate correlation in adjacent-planet radii but only a weak correlation in bulk density and no overall correlation in mass. Among the pairs that are most similar in radius, most sit on different theoretical composition curves, indicating distinct interior structures and volatile contents. The paper concludes that within a system, similarity in radius is not a reliable proxy for similarity in mass, density, composition, or physical nature. This is consequential because radius is the most easily measured exoplanet property and is routinely used to classify planets by type.","feed_headline":"Same-size planet neighbours often have different interiors","feed_subtitle":"Radius is a weak guide to what planets are made of, even for neighbours in the same system.","key_machinery":"The load-bearing object is the 'interior distance' metric $I$, defined in Eq. (4). For each planet in a pair, the planet is assigned the nearest curve in a fine grid of theoretical mass–density curves for two-layer interiors made of Fe, MgSiO$_3$, and H$_2$O; the closer of the two curves becomes the pair's reference curve, and $I$ is the average of the signed density gaps between each planet and that reference at the planet's own mass. A threshold of $I \\geq 0.1$ corresponds, on the grid, to compositions differing by at least about 30% in component content, so the metric converts the qualitative picture of straddling curves into a count of compositionally similar versus dissimilar pairs. Supporting machinery includes the scale-independent parameter gap $g$ of Eq. (3), the construction of a 'high-correlated in radius' subsample by trimming pairs with radius ratios far from unity, and a Monte Carlo 'error-accommodation' routine plus null populations drawn from a non-parametric mass-radius relation to assess whether the observed correlations and dispersions could arise from measurement noise.","core_discovery":"The paper's central discovery is that the intrasystem uniformity known as the 'peas in a pod' trend holds for planet radius but largely evaporates for mass, density, and interior structure. In the main sample of 184 adjacent planet pairs with directly measured masses, the Pearson correlation coefficient for mass is $R = -0.007$, statistically indistinguishable from zero, whereas for radius it is $R = 0.516$ and for bulk density $R = 0.265$. When the analysis is restricted to pairs that are highly similar in radius, the correlations in mass and density remain weak. For planets up to about 32 Earth masses, placing these similar-radius pairs on a grid of theoretical two-layer composition curves (iron, rock, water) shows that most pairs straddle widely separated curves. Quantifying this with an 'interior distance' metric $I$, the paper counts 26 pairs with $I \\geq 0.1$, interpreted as compositions differing by at least 30% in the grid, versus 15 pairs with $I < 0.1$. The conclusion is that neighbouring planets with similar radii can belong to different compositional families, such as rocky versus water-rich worlds, so radius similarity does not imply similarity in composition or interior structure.","pith_inferences":["If radius similarity does not track interior similarity, the 'peas in a pod' pattern measured on radius-dominated samples such as the Kepler sample may overstate the architectural uniformity of the underlying planet population, since the signal rides on the easiest-to-measure quantity.","Recomputing interior distances with interior models that include hydrogen-helium envelopes or mixed ice-rock mixtures would test how many of the paper's 'similar-interior' pairs survive; adding more composition families could push more pairs above the dissimilarity threshold.","The pairs the paper flags as compositionally diverse despite matching radii are natural targets for atmospheric characterisation, because they offer the most information about how different interiors can arise from a single protoplanetary disc."],"forward_implications":["Radius-based planet classification becomes unreliable: two planets of the same size in the same system can be a rocky world and a water world, so demographic cuts by radius alone mix different physical populations.","Part of the scatter in the observed mass-radius relation is intrasystem, not merely an inter-system effect of different formation environments, so population models must explain diversity produced within a single protoplanetary disc.","Formation models that predict strong mass uniformity within a system need to reconcile with the absence of mass correlation in the full sample, while the enhanced uniformity found around cool, old, low-metallicity stars marks the conditions where such models may apply.","Radius by itself cannot indicate a planet's physical nature: same-size neighbours in one system can have dissimilar densities, masses, and implied volatile content, so composition estimates must combine mass and radius."],"supporting_citations":[{"why":"Supplies the fine grid of two-layer Fe–MgSiO3–H2O mass-radius curves on which the interior distance metric is defined.","marker":"Zeng et al. (2016)"},{"why":"Provides the non-parametric mass-radius relation used to draw null-hypothesis populations of mass and density for the correlation and dispersion tests.","marker":"Ning et al. (2018)"},{"why":"The most directly comparable prior study of intrasystem similarity in radius, mass, and density, whose methods and error criteria the paper reproduces and extends.","marker":"Otegi et al. (2022)"},{"why":"Established the radius-correlation 'peas in a pod' result that this paper re-tests and confirms on a larger sample.","marker":"Weiss et al. (2018)"},{"why":"Reported strong intrasystem similarity in both mass and radius; the paper's contrasting mass result tests this earlier claim.","marker":"Millholland et al. (2017)"},{"why":"Independent study of core and water mass fractions around M dwarfs used to validate the interior-distance classifications.","marker":"Rodríguez Martínez et al. (2023)"},{"why":"Provides the dynamical stability criterion used to reject unstable pairs when constructing mock populations for the null hypothesis.","marker":"Deck et al. (2013)"},{"why":"Earlier mass-uniformity trend that the paper's larger sample re-evaluates and finds weaker than previously claimed.","marker":"Wang (2017)"}],"fun_headline_variants":["Radius lies: similar-sized exoplanets hide different insides","Peas in a pod? Not for planet interiors","Similar radii, dissimilar interiors: exoplanet neighbours fool us","Same size doesn't mean same stuff: planet interiors vary"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion rests on the assumption that the grid of two-layer iron–rock–water compositions used to define 'interior distance' adequately represents the real interiors of planets up to about 32 Earth masses, so that a large interior distance indicates a genuine compositional difference and not, for example, the presence of hydrogen-helium atmospheres or mixed ice-rock material that the grid simply does not include.","fun_headline_variants_meta":{"raw":{"variants":["Radius lies: similar-sized exoplanets hide different insides","Peas in a pod? Not for planet interiors","Similar radii, dissimilar interiors: exoplanet neighbours fool us","Same size doesn't mean same stuff: planet interiors vary"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000575,"raw_usage":{"total_tokens":2794,"prompt_tokens":1101,"completion_tokens":1693,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":717,"completion_tokens_details":{"reasoning_tokens":1637}},"tokens_in":717,"tokens_out":1693,"duration_ms":14387,"temperature":1.0,"reasoning_tokens":1637,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T10:24:56.997507+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Inspect the atmospheres or interiors of the 26 pairs the paper labels compositionally dissimilar ($I \\geq 0.1$) using independent diagnostics such as transmission spectroscopy, atmospheric escape, or asteroseismic density constraints; if most of those pairs turn out to share a similar bulk composition, the interior-distance interpretation is refuted.","supporting_citations":[{"cited_title":"D., & Jacobsen, S","cited_arxiv_id":null,"evidence_quote":"Supplies the fine grid of two-layer Fe–MgSiO3–H2O mass-radius curves on which the interior distance metric is defined."},{"cited_title":"2018, , 869, 5","cited_arxiv_id":null,"evidence_quote":"Provides the non-parametric mass-radius relation used to draw null-hypothesis populations of mass and density for the correlation and dispersion tests."},{"cited_title":"2022, Astronomy & Astrophysics, 658, A107","cited_arxiv_id":null,"evidence_quote":"The most directly comparable prior study of intrasystem similarity in radius, mass, and density, whose methods and error criteria the paper reproduces and extends."},{"cited_title":"M., Marcy, G","cited_arxiv_id":null,"evidence_quote":"Established the radius-correlation 'peas in a pod' result that this paper re-tests and confirms on a larger sample."},{"cited_title":"2017, The Astrophysical Journal Letters, 849, L33","cited_arxiv_id":null,"evidence_quote":"Reported strong intrasystem similarity in both mass and radius; the paper's contrasting mass result tests this earlier claim."},{"cited_title":"M., Payne, M., & Holman, M","cited_arxiv_id":null,"evidence_quote":"Provides the dynamical stability criterion used to reject unstable pairs when constructing mock populations for the null hypothesis."},{"cited_title":"2017, Research Notes of the AAS , 1, 26","cited_arxiv_id":null,"evidence_quote":"Earlier mass-uniformity trend that the paper's larger sample re-evaluates and finds weaker than previously claimed."}],"review_version":1}