{"id":"5900d894-5002-49bb-b538-82aa76fedea8","arxiv_id":"2411.18686","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Giant planets are unlikely to retain dilute cores through their evolution unless they form with low initial luminosity and steep internal compositional steps.","lead":"This paper simulates the long-term evolution of giant planets to test whether Jupiter-like dilute cores survive. It finds that dilute cores are destroyed when the planet starts with high luminosity, while bloating in hot Jupiters has only a small effect.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Mesh-resolution dependence of the dilute-core retention threshold is unresolved; the 'unlikely to be retained' conclusion in Sect. 5 is not shown to be converged.","rationale":"The reader's weakest_assumption pinpoints the mesh resolution, and I agree that this is the most load-bearing issue for the headline 'unlikely to retain a dilute core' conclusion. The mechanism by which dilute cores are retained in this model is the formation of sharp compositional steps (Sect. 3.1); Eq. 8 shows that the compositional gradient dX_j/d ln p is discretized on the mass mesh. When a step is present, dX_j is fixed by the step amplitude while d ln p depends on cell size, so the mesh spacing sets the maximum gradient that can inhibit convection (Sect. 3.6). The resolution study (Fig. 10) shows that at 2e4 points the extended structure fully mixes and the compact core shrinks to ~6% of the mass, whereas at 1e5 points the compact core grows to ~17% and the number of steps roughly doubles. This unconverged behavior directly controls the persistence threshold. The central claim's quantitative anchor is the initial-luminosity threshold of ~3e3 L_J (Sect. 3.7), but this threshold was only computed at the standard 5e4 resolution. If the threshold increases with resolution, the comparison to formation luminosities (Mordasini 2013: 2e4 to 6e4 L_J) changes qualitatively. The paper even concedes in Sect. 3.6 that it is unclear what mesh size is realistic, and in Sect. 4.1 that step locations are mesh dependent. Therefore the verdict should remain CONDITIONAL: the authors need to demonstrate convergence of the persistence threshold with mesh size (and ideally also quantify the delayed-convection prescription) before the population-level conclusion can be accepted. No additional objection is needed; the paper is otherwise clear and internally consistent.","tokens_in":19391,"tokens_out":6692,"duration_ms":59255,"concrete_test":"Re-run the Sect. 3.7 luminosity series (L_init = 1e3, 3e3, and 1e4 L_J) for at least the compact and extended initial structures with nmesh = 1e5 and 2e5, keeping the same delayed-convection prescription. Determine the maximum initial luminosity at which any dilute core (a surviving Z step) remains. If this threshold moves above ~6e3 to 2e4 L_J, the 'unlikely to be retained' conclusion loses its quantitative basis; if the threshold remains around 3e3 L_J, the conclusion is robust to resolution.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is that dilute cores cannot persist for initial luminosities much above ~3e3 L_J, making their retention in most giant planets unlikely (Sect. 5). The mechanism that allows persistence is the formation of sharp compositional steps that inhibit convection. Sect. 3.6 (Fig. 10) shows that this step formation is controlled by the mass-mesh resolution: with 2e4 points only the compact initial structure survives and only to ~6% of the mass; with 5e4 points all four structures retain cores out to ~35%; with 1e5 points the compact core extends to ~17% and the number of steps roughly doubles. However, the resolution study was performed only at the standard initial luminosity L_init = 1e3 L_J. The paper never checks whether the luminosity threshold of ~3e3 L_J (Sect. 3.7) is stable when the mesh is refined. Since higher resolution consistently makes dilute cores more persistent, the threshold could shift upward at nmesh = 1e5 or 2e5, potentially above the cold-start formation luminosities of ~2e4 to 6e4 L_J (Mordasini 2013) that the paper uses to conclude dilute cores are unlikely. The paper itself concedes in Sect. 3.6 that 'It is unclear what number of mesh points is most realistic' and in Sect. 4.1 that step locations and sizes are mesh dependent. Without a convergence test of the retention threshold, the headline conclusion is not numerically supported. The delayed-convection prescription (Sect. 4.2) is an additional ad hoc element, but the mesh-resolution issue is the more direct and quantifiable threat to the central claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript models the 4.5 Gyr evolution of giant planets with the 1D code completo21, adding Ledoux-criterion convection with composition gradients and a separate mass-mesh treatment of convective mixing. Starting from four initial heavy-element profiles (compact, extended, metal-rich, and Jupiter-like), the authors vary orbital distance and bloating, mixing length, semi-convection efficiency, opacity, mesh resolution, and initial luminosity. They find that dilute cores can be retained under some conditions, that semi-convection can shrink them, that bloating has a modest effect, and that mesh resolution strongly controls step formation and core extent. The headline claim is that dilute cores cannot persist at initial luminosities much above ~3e3 L_J for a Jupiter-mass planet, and the paper concludes that retention of dilute cores in a large fraction of giant planets is unlikely.","tokens_in":19758,"tokens_out":4764,"duration_ms":46500,"significance":"If the headline result were robust, it would be an important constraint for post-formation giant-planet evolution and for interpreting atmospheric versus bulk metallicities. The paper also contributes a systematic parameter study of hot Jupiters and explicitly tests the effect of an alternative luminosity distribution. The central result is not a fit: the main parameters are selected a priori, not tuned to produce dilute-core survival. However, the load-bearing claim is not numerically converged. The paper itself concedes that the appropriate number of mesh points is unclear and that step locations and sizes are mesh dependent. Because finer meshes consistently make dilute cores more persistent, the luminosity threshold in Sect. 5 could shift materially. The conclusion about the rarity of retained dilute cores is therefore conditional on an unvalidated numerical parameter; the significance is real but provisional.","major_comments":[{"comment":"The central claim is not shown to be converged with respect to the mass mesh. At L_init = 1e3 L_J, going from 2e4 to 5e4 to 1e5 mesh points changes whether four initial profiles retain a dilute core, how many steps form, and how far the compact core extends (from ~6% to ~17% of the mass). The resolution study is only performed at the standard L_init = 1e3 L_J; the key threshold of ~3e3 L_J (Sect. 3.7) is never tested at higher resolution. Since finer meshes consistently make dilute cores more persistent, the threshold could shift upward at 1e5 or 2e5 points, potentially above the cold-start luminosities of ~2e4 to 6e4 L_J invoked in Sect. 3.7. The paper's own statement in Sect. 3.6 that 'It is unclear what number of mesh points is most realistic' and the Sect. 5 caveat 'Assuming that the number of mesh points of 5e4 ... provides a good approximation' make this a load-bearing unresolved issue, not a presentation issue.","section":"§3.6, Fig. 10 and §5"},{"comment":"The luminosity distribution dL/dm = L/M is an ad hoc simplifying assumption, and the paper shows that the alternative dL/dm = -T dS/dt changes the early luminosity and radiative-conductive gradient by up to a factor of 3 in the 0.1-0.2 m/M region, which is precisely the usual extent of the dilute core. This alternative also reduces the number of compositional steps from 7 to 2 for the compact structure. Since the mixing criterion and the ability of a step to inhibit convection depend on the radiative-conductive gradient (Eqs. 5, 6, and 8), the simplified luminosity profile is a possible source of the reported retention threshold. At minimum, the paper should quantify how the step-size and luminosity thresholds change under a more physical entropy-based luminosity profile, or present a physical justification for why dL/dm = L/M is adequate for the central claim.","section":"§4.4, Eq. (3)"},{"comment":"The artificial 1 Myr delay in applying the radiative-conductive gradient for mixing purposes is a parameter that conditions the outcome. The manuscript states that without this delay, some initial compositions mix completely and lose their dilute core, whereas with the delay they retain it. The delay is justified qualitatively by hot-start accretion scenarios, but no sensitivity study is presented for its duration or the shape of the ramp. Because the central conclusion that dilute cores can persist at L_init = 1e3 L_J depends on this prescription for the compact and Jupiter-like cases, the paper needs to show that the retention does not hinge on the particular choice of 1 Myr or on the functional form of the delayed onset.","section":"§4.2 and §2.1"},{"comment":"The luminosity threshold itself is inferred from runs at only three initial luminosities (1e3, 3e3, and 1e4 L_J), and the paper uses the maximum luminosity reached during mixing rather than the literal initial value. The increase in luminosity during mixing is sizable (e.g., from 3e3 to up to 5.7e3 L_J for the metal-rich profile), but it is not shown how this maximum depends on mesh resolution or on the delayed-convection prescription. A combined resolution study at and above the stated threshold is needed before the conclusion 'it is unlikely that a large number of giant planets retain a dilute core' can be considered robust.","section":"§3.7 and §5"}],"minor_comments":[{"comment":"There is a typo: 'the size of the the thermodynamic evolution timestep' should read 'the size of the thermodynamic evolution timestep'.","section":"§2.1"},{"comment":"The phrase 'for which be do not observe a consistent relation' should read 'for which we do not observe a consistent relation'.","section":"§4.1"},{"comment":"The notation '3 x 1e3 LJ' is awkward; in the published version this should be typeset as 3 × 10^3 L_J consistently throughout the abstract and text.","section":"Abstract and §3.7"}],"recommendation":"major_revision","confidential_remarks":"The paper is candid about its limitations, and the stress-test concern about mesh resolution genuinely lands: the central claim is not numerically converged. This is fixable within the manuscript's scope by running the luminosity-threshold cases at finer resolution and by adding a sensitivity test for the delayed-convection prescription. If those tests show the threshold is stable, the paper would be acceptable after revision; if not, the conclusion must be softened accordingly. I see no novelty or attribution problems beyond the authors' own acknowledged dependence on previous work."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know first: this is a serious, carefully run evolution-model study of convective mixing and dilute-core retention in giant planets. The genuinely new pieces are the hot-Jupiter bloating treatment—showing that bloating slightly inhibits mixing by lowering the intrinsic luminosity—and the systematic luminosity threshold near 3e3 L_J above which dilute cores fail to survive. Both go beyond Vazan et al. (2015, 2018) and Müller et al. (2020), which had the qualitative trend but not this quantitative threshold. The parameter study itself is broad: mixing length, semi-convection, opacity, orbital distance, and mesh size. The paper is also admirably transparent, stating in Sect. 3.6 that it is unclear what number of mesh points is most realistic and in Sect. 4.1 that step locations and sizes are mesh dependent. That honesty earns real credit.\n\nThe soft spot is load-bearing. The Sect. 5 claim that dilute cores are unlikely to be retained by many giant planets depends on the standard mesh of 5e4 equal-mass points. Fig. 10 shows that with 2e4 points only the compact initial structure keeps a tiny dilute core at ~6% of the mass, while with 1e5 points the compact core extends to ~17% and the number of steps roughly doubles. More resolution consistently makes dilute cores more persistent. Yet the resolution test was done only at L_init = 1e3 L_J, not at the threshold 3e3 L_J. Since higher resolution shifts the outcome in the direction of persistence, the threshold could move upward at 1e5 or 2e5 points—potentially above the 2e4–6e4 L_J cold-start formation luminosities that the paper uses to conclude dilute cores are unlikely. So the headline conclusion is not numerically converged as stated.\n\nTwo secondary issues: the delayed-convection prescription (Sect. 4.2) is ad hoc, with no quantified sensitivity to the 1 Myr timescale, and the luminosity distribution approximation dl/dm = L/M versus T dS/dt changes results by up to a factor of ~3 in the dilute-core region (Sect. 4.4). Both are discussed honestly, but they add further uncertainty to the quantitative threshold.\n\nWho this is for: planetary interior and formation modelers, especially those working on atmospheric-to-bulk metallicity relations and Jupiter-like dilute cores. It deserves a serious referee, but the referee should require either a convergence test of the luminosity threshold across mesh sizes or a physical argument for why 5e4 points captures the real layering scale. Without that, the Sect. 5 conclusion should be softened to a tentative trend rather than a statement about the population.","headline":"Careful and unusually honest parameter study, but the headline dilute-core conclusion rests on an unresolved mesh-resolution dependence that the paper itself concedes.","tokens_in":781,"tokens_out":777,"would_cite":false,"duration_ms":24039,"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":"Evolution models of giant planets indicate that dilute cores—extended interiors enriched in heavy elements—rarely survive when the planet starts out brighter than a few thousand Jupiter luminosities, so most giants should end up fully…","keywords":["dilute cores","giant planet evolution","convective mixing","semi-convection","hot Jupiters","initial luminosity","compositional gradients","Jupiter interior"],"falsifier":"Repeat the paper's Jupiter-like simulation at an initial luminosity of $10^4\\,L_J$ with about $3\\times10^5$ equally spaced mass points, the resolution implied by the overshooting-length estimate quoted in Sect. 3.6; if a dilute core then survives, the luminosity ceiling is an artifact of the coarser mesh rather than a physical limit.","tokens_in":19174,"feed_emoji":"🪐","tokens_out":8164,"duration_ms":73009,"temperature":0.7,"pith_summary":"Giant planets may form with an extended region of enriched heavy elements in their deep interiors, but this paper argues that such dilute cores are usually erased. In a one-dimensional evolution model that tracks convective mixing, a Jupiter-mass planet retains its dilute core only if its post-formation luminosity starts below roughly $3\\times10^3$ Jupiter luminosities and its initial composition contains sharp enough steps. Standard formation models generally deliver higher starting luminosities, so the authors conclude that only a minority of giant planets should keep a dilute core through 4.5 billion years of evolution. The result matters for interpreting exoplanet atmospheres: if dilute cores are rare, atmospheric heavy-element abundances are a more direct probe of bulk composition than they would be if layered interiors were common.","feed_headline":"Most giant planets lose their dilute cores early on","feed_subtitle":"Models set the survival ceiling near 3,000 Jupiter luminosities; hotter starts fully mix the interior.","key_machinery":"The mechanism that carries the argument is compositional convection governed by the Ledoux criterion in a one-dimensional planetary evolution model. Convection begins where the radiative-conductive gradient $\\nabla_{\\mathrm{rad}}$ exceeds the sum of the adiabatic gradient and the compositional gradient $\\nabla_X$; mixing is then modelled as a diffusion process on a separate equal-mass mesh of $5\\times10^4$ points using mixing-length theory for the diffusion coefficient. Steep compositional steps—the 'stairs' in the heavy-element profile—are what hold a dilute core in place, because they make $\\nabla_X$ large enough to suppress mixing. The paper's key numerical control is the mesh resolution, since it sets the maximum compositional gradient that can be represented, and the key physical control is the initial luminosity, since $\\nabla_{\\mathrm{rad}}$ grows roughly linearly with it.","core_discovery":"The central claim is that dilute cores—deep interior regions where heavy elements are present at moderate enrichment rather than in a pure compact core—are difficult to preserve. The paper finds that for a Jupiter-like planet with the heavy-element profiles taken from formation models, the limiting factor is the initial luminosity, which sets the radiative-conductive gradient that drives convection in the deep envelope. Above about $3\\times10^3\\,L_J$, mixing destroys the compositional staircase that stabilises the dilute core; at $10^4\\,L_J$ the envelope mixes completely. Because hot-start and even many cold-start formation scenarios produce luminosities above this threshold, the paper concludes that retaining a dilute core through the whole evolution is an unlikely outcome for most giant planets, at least under the assumptions and resolution of the model.","pith_inferences":["If confirmed, the strong mesh dependence raises the possibility that the qualitative conclusion—that dilute cores are rare—is an artifact of unresolved layering; a physically motivated resolution would require resolving the estimated overshoot length, roughly $3\\times10^5$ mesh points, which the paper did not run at the decisive luminosity.","A corollary beyond the paper is that the atmospheric-versus-bulk metallicity mismatch should be systematically absent in young, bright giant planets and possibly present in old, dim ones.","The bloating results imply a testable orbital-distance trend: if hot Jupiters form in situ, their interior mixing should be minimal near 0.04–0.05 AU and stronger both farther out and closer in; a survey of atmospheric metallicities versus orbital distance could look for that non-monotonic signature.","The paper uses water for all heavy elements; if real interiors are richer in heavier molecules, the compositional gradient is larger and mixing is weaker, so the luminosity ceiling would move upward and the claim that most planets lose their dilute cores would be too strong."],"forward_implications":["If dilute cores are as fragile as this model says, JWST measurements of atmospheric metallicity in giant planets can be interpreted with relatively simple core-plus-homogeneous-envelope structures for most planets.","The first billion years matter almost exclusively: nearly all mixing happens early, so the conditions set by formation, not later evolution, decide whether a dilute core survives.","Close-in hot Jupiters should not simply be assumed convective: bloating lowers the intrinsic luminosity and can suppress mixing, leaving slightly more dilute cores than at wide orbits, provided they formed in place.","Strong semi-convection can shrink a dilute core and enrich the outer envelope, but it cannot fully erase a large initial core in the paper's models.","The threshold of roughly $3\\times10^3\\,L_J$ provides a formation-property test: planets formed by hot accretion are expected to be fully mixed, and any observed dilute core would point to cold accretion or early radiative zones."],"supporting_citations":[{"why":"Supplies the convective-mixing implementation—Ledoux criterion, mixing velocity, and diffusion coefficient—that the paper adapts to its mass mesh.","marker":"Vazan et al. (2015)"},{"why":"Provides the Jupiter-like initial composition used for one of the four starting structures and the low-luminosity benchmark that motivates the study.","marker":"Vazan et al. (2018)"},{"why":"Provides the compact, extended, and metal-rich post-formation heavy-element profiles that serve as the other three initial structures.","marker":"Müller et al. (2020)"},{"why":"Supplies the semi-convection diffusion-coefficient approximation used to test how semi-convection affects dilute-core survival.","marker":"Langer et al. (1983)"},{"why":"Establishes the post-formation luminosities of a Jupiter-mass planet, which the paper compares against its survival threshold of about $3\\times10^3\\,L_J$.","marker":"Mordasini (2013)"},{"why":"Provides the conductive opacities that, together with radiative opacities, set the radiative-conductive gradient in the deep envelope.","marker":"Cassisi et al. (2007)"}],"fun_headline_variants":["Dilute cores vanish above 3,000 Jupiter luminosities","Hot-start giant planets can't keep dilute cores","Most giants mix away their dilute cores early","Luminosity threshold dooms dilute cores in giants"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central conclusion depends on assuming that 50,000 equal-mass layers capture the real compositional layering, even though the paper shows the survival of the dilute core changes drastically with mesh resolution and offers no physical argument that this count matches the true layering scale.","fun_headline_variants_meta":{"raw":{"variants":["Dilute cores vanish above 3,000 Jupiter luminosities","Hot-start giant planets can't keep dilute cores","Most giants mix away their dilute cores early","Luminosity threshold dooms dilute cores in giants"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000723,"raw_usage":{"total_tokens":3267,"prompt_tokens":992,"completion_tokens":2275,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":608,"completion_tokens_details":{"reasoning_tokens":2211}},"tokens_in":608,"tokens_out":2275,"duration_ms":17334,"temperature":1.0,"reasoning_tokens":2211,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T10:58:27.813861+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the paper's Jupiter-like simulation at an initial luminosity of $10^4\\,L_J$ with about $3\\times10^5$ equally spaced mass points, the resolution implied by the overshooting-length estimate quoted in Sect. 3.6; if a dilute core then survives, the luminosity ceiling is an artifact of the coarser mesh rather than a physical limit.","supporting_citations":[{"cited_title":"2015, ApJ, 803, 32","cited_arxiv_id":null,"evidence_quote":"Supplies the convective-mixing implementation—Ledoux criterion, mixing velocity, and diffusion coefficient—that the paper adapts to its mass mesh."},{"cited_title":"2018, A&A, 610, L14","cited_arxiv_id":null,"evidence_quote":"Provides the Jupiter-like initial composition used for one of the four starting structures and the low-luminosity benchmark that motivates the study."},{"cited_title":"2013, A&A, 558, A113","cited_arxiv_id":null,"evidence_quote":"Establishes the post-formation luminosities of a Jupiter-mass planet, which the paper compares against its survival threshold of about $3\\times10^3\\,L_J$."},{"cited_title":"Y., Pietrinferni, A., Catelan, M., & Salaris, M","cited_arxiv_id":null,"evidence_quote":"Provides the conductive opacities that, together with radiative opacities, set the radiative-conductive gradient in the deep envelope."}],"review_version":1}