{"id":"eeccd73a-173c-4599-94ec-7c55f527bdce","arxiv_id":"2507.04954","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In dense solid-particle disks, the mass of the largest formed satellite scales roughly linearly with disk mass, with a stochastic spread large enough that duplicated initial conditions produce very different moons.","lead":"This paper runs many computer simulations of disks of rocky debris around rocky planets to see how moons form. It finds that moon formation is very random, that heavier disks make heavier moons, and estimates which moons could be seen by JWST.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The scaling law in Eq. (4) is fit through low-mass disks for which the authors themselves state planetary tides are not modeled and tidal migration timescales are comparable to formation timescales, so the low-mass anchor of the fit is not tested under the physics that dominates there.","rationale":"The paper's central claim, as summarized by the reader, is a quantitative scaling law from disk mass to largest-satellite mass plus a threshold for one- versus two-satellite systems. The most load-bearing condition for that claim is that the simulated satellite masses faithfully represent the physics in the disk-mass regime used in the fit. Section 4.1 concedes that for Mdisk <= 0.005 M_C, the formation process is driven by planetary tides not present in the model; the collision model also lacks fragmentation and tidal stripping. Those low-mass points are not outliers to be ignored; they are part of the dataset that determines the fitted coefficient and exponent in Eq. (4). If unmodeled tides suppress accretion or accelerate outward migration in this regime, the fitted normalization could drop and the threshold between single- and double-satellite outcomes could shift. The concern is acknowledged by the authors, which is a point in their favor; the missing piece is a quantitative estimate of its effect on the fitted parameters. The proposed exclusion test is cheap and decisive: if the fit is unchanged after removing the tidally-affected points, the central scaling survives; if not, the paper should either restrict the fit to Mdisk >= 0.008 M_C or add tides. I do not see an internal contradiction in Eq. (4), and the stochasticity is handled transparently with 30-run ensembles. The JWST detectability issue raised by the reader is real but secondary; the scaling law is the central claim.","tokens_in":19849,"tokens_out":7376,"duration_ms":85369,"concrete_test":"Refit Eq. (4) excluding Mdisk <= 0.005 M_C (and, as a second check, excluding Mdisk <= 0.008 M_C), comparing the fitted normalization and exponent with the full-fit values. If either parameter shifts by more than its quoted uncertainty, the low-mass points are controlling the scaling. As a complementary physical test, run one low-mass case (e.g., Mdisk = 0.005 M_C) with an added tidal migration torque or compare against the 1D ring-satellite models of Crida and Charnoz (2012) to see whether the mean largest-satellite mass is reduced by more than the error bars. Report the fit residuals for both versions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central result is the power-law fit Eq. (4), Ml,sat = (0.20 +/- 0.16) Mdisk^(0.98 +/- 0.16), obtained from the mass series in Section 4 with beta=3. That series includes Mdisk = 0.003 and 0.005 M_C. Section 4.1 explicitly says that for Mdisk <= 0.005 M_C, ttide (Eq. 5) becomes comparable to the timespan of second-satellite formation (10^6-10^7 T_C) and that the formation process would be driven by planetary tides, which are not included in the N-body model. The collision treatment (Section 2.2) also has no fragmentation and no tidal stripping, and the authors warn that this 'may overestimate satellite masses.' Because those two low-mass points are included in the fit, and because the same regime is where the two-satellite threshold (0.003-0.03 M_C) is defined, the headline mapping from disk mass to satellite mass is least secure at the low-mass end of the fit. This is not an internal inconsistency; it is an acknowledged physical omission whose quantitative effect on Eq. (4) is not estimated.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper uses three-dimensional N-body simulations of dense, self-gravitating solid-particle disks (50,000 equal-mass particles) to study satellite formation around terrestrial planets and super-Earths. For each model, 30 simulations are run with identical initial angular momentum but randomized angular elements. The authors explore the effects of the disk surface-density exponent beta at fixed disk mass and then vary the initial disk mass Mdisk from 0.003 to 0.1 MC at beta = 3. The main results are: (i) large run-to-run stochasticity in satellite masses, with approximately Gaussian distributions; (ii) disks above about 0.03 MC predominantly form one dominant satellite, whereas disks between about 0.003 and 0.03 MC tend to form two-satellite systems; (iii) a fitted power law for the mean largest-satellite mass, Ml,sat = (0.20 +/- 0.16) Mdisk^(0.98 +/- 0.16), approximately linear in disk mass and consistent with the analytical prediction of Ida et al. (1997); and (iv) applications to the Moon, Charon, Phobos, and the JWST detectability of exomoons. The central quantitative claim is Eq. (4).","tokens_in":20082,"tokens_out":5466,"duration_ms":63559,"significance":"The paper's main contribution is statistical: by running 30 simulations per parameter set, it quantifies the stochasticity that earlier single-run or few-run studies could not characterize, and it provides a simple mapping from initial disk mass to typical largest-satellite mass. If the scaling law holds, it is immediately useful for connecting giant-impact-generated disk masses to observable exomoon masses, and the JWST detectability estimate gives a concrete observational target. The recovery of the Hyodo et al. (2015) single/two-satellite threshold at about 0.03 MC and the agreement with the Ida et al. (1997) analytical scaling are genuine positive checks. However, the central result is an empirical fit to simulations whose collision model lacks fragmentation, tidal stripping, and planetary tides; the authors acknowledge that these omissions may overestimate satellite masses. The quantitative coefficient and the low-mass threshold therefore remain provisional until the sensitivity to these neglected physical effects is quantified.","major_comments":[{"comment":"The fitted power law in Eq. (4) is obtained from simulations that include Mdisk = 0.003 and 0.005 MC, yet Section 4.1 states that for Mdisk <= 0.005 MC the tidal migration timescale ttide becomes comparable to the second-satellite formation timescale (10^6-10^7 TC) and that formation would be driven by planetary tides, which are not included in the N-body model. These low-mass points anchor the low-mass end of the fit and define the 0.003-0.03 MC two-satellite regime, so the scaling law and the threshold are not tested under the physics claimed to dominate there. Please refit Eq. (4) with these points excluded (and possibly also Mdisk = 0.008) to show how the coefficient A and exponent b change, or otherwise provide a quantitative bound on the systematic error from missing tidal effects.","section":"Section 4, Eq. (4), and Section 4.1"},{"comment":"The collision treatment assumes that every gravitationally bound collision leads to perfect merging, with no fragmentation, no tidal stripping, and no mass loss, and planetary tides are not included in the N-body integration. As the authors note, this 'may overestimate satellite masses.' Because the normalization A = 0.20 in Eq. (4) and the inferred disk-mass requirements for Charon and Phobos in Section 4 depend directly on the absolute satellite masses, the paper should provide a quantitative estimate of this bias, for example by testing a simple fragmentation prescription or by comparing against SPH-calibrated collision outcomes, or explicitly reframe the fitted relation as an upper-limit estimate.","section":"Section 2.2 and Eq. (4)"},{"comment":"Equation (4) is used to infer that Charon formation would require Mdisk >= 0.3 MC and that direct Phobos formation requires Mdisk ~ 10^-7 - 10^-6 MC. These values lie outside the fitted range (0.003 - 0.1 MC) by factors of 3 and by several orders of magnitude, respectively. The authors do note that Eq. (4) is only 'certainly suitable' within the fitted range, but the Charon and Phobos passages are still framed as quantitative conclusions. Please either soften these inferences to extrapolation-level statements or, if the extrapolation is to be retained, justify the power law beyond the simulated range with an independent argument.","section":"Section 4 and Section 5"}],"minor_comments":[{"comment":"The reference list contains an extraneous passage in Portuguese between Goldreich and Soter (1966) and Goldreich and Tremaine (1980) beginning 'Obrigado por este comentario...' that appears to be leftover author-response text. This should be removed before publication.","section":"References"},{"comment":"The phrase 'how the initial surface density profile influences on the system' should read 'influences the system' or 'has an influence on the system.'","section":"Section 3, first paragraph"},{"comment":"The dashed horizontal line in Figure 5a is described in the body text as the Moon-to-Earth mass ratio, but this is not stated in the caption; please add it for self-containedness.","section":"Figure 5 caption"},{"comment":"The stopping criterion Mring = 0.05 Mdisk is said to be 'determined through test simulations,' but no details of those tests are given; a sentence describing the number of test runs and the criterion used to judge that satellite mass growth is negligible would aid reproducibility.","section":"Section 4"},{"comment":"In Eq. (5), the symbols n and a are not explicitly defined at first use; please define n as the satellite's mean motion and a as its semi-major axis, and clarify that all masses are in units of MC.","section":"Equation (5)"}],"recommendation":"major_revision","confidential_remarks":"The paper is suitable for the journal's scope. The statistical ensemble approach is a clear strength. However, the headline scaling law is fit through low-mass points for which the authors themselves state the dominant physics (planetary tides) is absent, and the perfect-merging assumption directly affects the fitted normalization. These are fixable by sensitivity tests and by restricting the claimed range of validity, so I recommend major revision rather than rejection. The accidental Portuguese insertion in the reference list should be caught in revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely new thing here is the 30-run ensemble characterization of satellite formation in dense solid-particle disks. The linear scaling and the 0.03 M_C single-satellite threshold were already known (Ida et al. 1997; Hyodo et al. 2015), and the authors say so. What you get beyond that is a quantitative handle on the scatter: mass-ratio distributions, beta dependence, and the fact that the largest satellite mass varies by an order of magnitude under identical initial conditions. That is useful, and the authors are transparent about their approximations.\n\nThe N-body setup is plausible: 50k equal-mass particles, surface density exponent beta, 30 runs per parameter set, stopping criterion based on ring mass. They compare against prior analytical work and single-run simulations and find consistency. The two-satellite pathway for low-mass disks and the transition at 0.03 M_C match Hyodo et al. (2015). The scaling law Ml,sat ≈ 0.2 Mdisk is a nice compact summary, though the uncertainties are large (0.20 ± 0.16, exponent 0.98 ± 0.16).\n\nNow the soft spots. The stress-test concern is real: the fit includes Mdisk = 0.003 and 0.005 M_C points, where the authors themselves say planetary tides are not modeled and tidal migration timescales are comparable to formation timescales. Those are the least reliable points, and they anchor the low-mass end of the scaling law. The authors flag the caveat but don't estimate how much it shifts the fit. That is a moderate, not fatal, issue, since the high-mass points and the prior analytical prediction agree.\n\nBigger problem: the JWST detectability estimate is wrong. They claim a 0.006 M_C satellite is detectable at 10 ppm around a Sun-like star. That's about 0.003 R_star for Earth-density, which gives a transit depth of ~3 ppm, not 10. The actual threshold is roughly four times more massive (or more, depending on satellite density). This error only affects the observational claim in the abstract, not the simulation results, but it's the kind of thing a referee should catch.\n\nAlso, there is a leftover Portuguese reviewer note in the reference list (between Goldreich & Soter and Goldreich & Tremaine). That has to be cleaned up.\n\nWho is this for? People working on exomoon formation and detectability, and to a lesser extent the Moon/Pluto/Mars satellite origin community. It deserves a serious referee because the core statistics are new and useful, but it needs revision: fix the detectability arithmetic, estimate the tidal/fragmentation bias on the fit, and clean the manuscript.","headline":"Useful ensemble statistics for satellite formation from dense solid-particle disks, with a clean scaling law, but the JWST detectability estimate is off by a factor of several in mass and the manuscript is not clean.","tokens_in":20680,"tokens_out":4532,"would_cite":true,"duration_ms":44931,"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":"In dense solid-particle disks left by giant impacts, the largest satellite's mass scales nearly linearly with the disk's mass, at about 20%, while disk mass decides whether one or two moons form.","keywords":["satellite formation","exomoons","circumplanetary disks","solid-particle disks","giant impacts","N-body simulations","surface density profile","Moon formation"],"falsifier":"Run the same 30-run ensembles at $M_{\\rm disk}=0.005\\,M_{\\rm C}$ and $\\beta=3$ with planetary tides switched on ($k_2/Q \\sim 10^{-3}$-$10^{-2}$): if the fraction of runs that still produce a second satellite with at least 10% of the largest satellite's mass falls below roughly 50%, the claimed two-satellite regime for $0.003$-$0.03\\,M_{\\rm C}$ does not survive the physics the paper itself identifies as missing. A parallel check: include collisional fragmentation at $M_{\\rm disk}=0.03\\,M_{\\rm C}$; if the mean $M_{\\rm l,sat}$ drops by more than about a factor of two, the fitted coefficient of about 0.2 is an artifact of perfect merging.","tokens_in":19628,"feed_emoji":"🌙","tokens_out":11569,"duration_ms":115251,"temperature":0.7,"pith_summary":"Giant impacts on terrestrial planets and super-Earths can leave behind dense disks of solid debris, and this paper asks what kind of moons such disks produce. Using thirty N-body simulations per disk model, it claims that the mass of the largest resulting satellite scales almost linearly with the initial disk mass, typically about a fifth of it, and that the disk mass sets the multiplicity: disks heavier than 0.03 planetary masses usually form a single dominant moon, while disks between 0.003 and 0.03 planetary masses tend to form two. The payoff is a quantitative bridge from impact-generated disk masses to observable moon masses, including an estimate that an Earth-Moon-like system around a 1.6-Earth-radius planet could be detectable with 10 parts-per-million photometry. The authors flag that their perfect-merging collision model may overestimate satellite masses and that planetary tides, omitted from the runs, matter for the lowest-mass disks, so the clean scaling is bracketed by these caveats.","feed_headline":"Biggest moon weighs about 20% of its birth disk's mass","feed_subtitle":"A 30-run N-body study finds the largest satellite tracks disk mass almost linearly, with one moon dominating high-mass disks.","key_machinery":"The argument is carried by 3D N-body ensembles of a self-gravitating solid-particle disk with surface density $\\Sigma(a) = \\Sigma_0 (a/R_{\\rm C})^{-\\beta}$ and 50,000 equal-mass particles, evolved with a hybrid collision rule: a collision is accretive only if the pair's Jacobi energy (the binding energy in the planet's tidal field) is negative and the particles' radii are within $0.7$ mutual Hill radii; otherwise the collision is elastic, and merging is perfect with no fragmentation or tidal stripping. Satellite growth is then regulated by inner Lindblad resonances, the orbital resonances through which a satellite shepherds surrounding disk material and, in response, migrates outward. The load-bearing output is the power-law fit $M_{\\rm l,sat} = (0.20 \\pm 0.16)\\,M_{\\rm disk}^{0.98 \\pm 0.16}$, which the authors connect directly to the analytical mass-angular-momentum prediction for accretion from a massive disk.","core_discovery":"The paper's central finding is an empirical scaling law extracted from ensembles of 30 simulations for each disk model: the mean mass of the largest satellite follows $M_{\\rm l,sat} = (0.20 \\pm 0.16)\\,M_{\\rm disk}^{0.98 \\pm 0.16}$, so the typical largest satellite carries about one fifth of the initial disk mass, with run-to-run scatter of the same order as the mean. The multiplicity of the system is set by disk mass: for $M_{\\rm disk} \\geq 0.03\\,M_{\\rm C}$ a single gravitationally dominant satellite forms in most simulations, while for $0.003 < M_{\\rm disk} < 0.03\\,M_{\\rm C}$ systems typically end with two satellites, the inner one being 10-50% as massive as the outer. Flatter surface-density profiles (lower $\\beta$) give more massive satellites, while steeper profiles funnel more mass onto the planet. Interpreting Solar System cases through this law, Moon-like satellites require disks around $0.03$-$0.1\\,M_{\\rm C}$, Charon-like masses would need disks $\\gtrsim 0.3\\,M_{\\rm C}$ (strongly suggesting Charon did not accrete from an impact-generated disk), and Phobos-like moons require extremely low-mass disks. The paper also estimates that a disk near the $0.03\\,M_{\\rm C}$ threshold around a 1.6-Earth-radius planet orbiting a Sun-like star could yield a detectable Earth-Moon-like system at 10 parts-per-million photometric precision.","pith_inferences":["If the near-linear scaling extends beyond the fitted range, observers could invert a measured exomoon mass into a parent-disk mass; the paper itself warns the fit is only justified inside the simulated span, roughly $0.003$-$0.1\\,M_{\\rm C}$.","The embedded reviewer note in the reference list concedes that calling $\\beta=3$ a typical value was incorrect; since the disk-mass scaling was measured almost entirely at $\\beta=3$, applying it across the full range of impact-generated profiles is an extrapolation, not a tested result.","Switching on fragmentation, tidal stripping, and planetary tides would likely lower $M_{\\rm l,sat}$ (the paper says so), which would push the JWST-detectability threshold upward; a straightforward extension is to rerun the 30-run ensembles with those processes included.","The two-satellite systems formed between $0.003$ and $0.03\\,M_{\\rm C}$ should leave distinctive transit-timing and transit-duration variations; searching for such correlated signals in Kepler or JWST light curves is a testable observational corollary."],"forward_implications":["Disks with $M_{\\rm disk} \\gtrsim 0.03\\,M_{\\rm C}$ are expected to leave a single gravitationally dominant satellite, with any additional satellites at least about 100 times less massive in more than half of the runs.","Over the studied range, the mean largest-satellite mass stays close to $0.2\\,M_{\\rm disk}$, so knowledge of an impact-generated disk mass directly estimates the typical moon mass, and vice versa.","Moon-like satellites around terrestrial planets are compatible with disks of $0.03$-$0.1\\,M_{\\rm C}$, matching the disk masses inferred for the proto-lunar impact.","Charon-like satellites require disks at least around $0.3\\,M_{\\rm C}$, which is far above the masses giant-impact simulations give for Pluto, so accretion from such a disk is not the likely Charon formation path.","A disk near $0.03\\,M_{\\rm C}$ around a 1.6-Earth-radius planet could produce an exomoon detectable at 10 parts-per-million photometric precision, which suggests that the absence of exomoon detections so far partly reflects instrumental limits."],"supporting_citations":[{"why":"Supplies the analytical mass-angular-momentum prediction that the fitted power law is compared against.","marker":"(Ida et al., 1997)"},{"why":"Provides the N-body method, initial eccentricity and inclination distributions, and the Jacobi-energy accretion criterion the simulations inherit.","marker":"(Kokubo et al., 2000)"},{"why":"Establishes the 0.03 M_C transition between single- and multiple-satellite formation that this study's statistics confirm and quantify.","marker":"(Hyodo et al., 2015)"},{"why":"Supplies the ring-disk feedback picture (inner Lindblad resonance confinement and sequential satellite formation) used to interpret the two-satellite pathway.","marker":"(Crida and Charnoz, 2012)"},{"why":"Justifies the adopted beta=3 profile and documents how particle number affects the simulated dynamics.","marker":"(Sasaki and Hosono, 2018)"},{"why":"Gives the small Pluto-Charon disk masses used to conclude that Charon needs disks of at least roughly 0.3 M_C and likely did not accrete from such a disk.","marker":"(Canup, 2011)"},{"why":"Shows massive rocky exomoons can form by giant impact around super-Earths, framing the observational motivation.","marker":"(Barr and Syal, 2017)"},{"why":"Provides the 1.6-Earth-radius threshold above which impact disks become vapour-rich, justifying the planet size used in the detection estimate.","marker":"(Nakajima et al., 2022)"},{"why":"Supplies the 10 parts-per-million photometric precision adopted for the exomoon detectability calculation.","marker":"(Bean et al., 2018)"}],"fun_headline_variants":["Satellite size tracks disk mass linearly","Biggest moon: 20% of disk mass","Disk mass decides moon multiplicity","High-mass disks produce single moons","Moon mass scales with disk mass"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that every gravitationally bound collision merges the two bodies perfectly, with no fragmentation, no tidal stripping, and no mass loss, while the planet's tidal forces are not computed in the N-body integration; the authors explicitly warn this may overestimate satellite masses and note that tides become important for disks below about $0.005\\,M_{\\rm C}$.","fun_headline_variants_meta":{"raw":{"variants":["Satellite size tracks disk mass linearly","Biggest moon: 20% of disk mass","Disk mass decides moon multiplicity","High-mass disks produce single moons","Moon mass scales with disk mass"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000211,"raw_usage":{"total_tokens":1496,"prompt_tokens":1109,"completion_tokens":387,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":725,"completion_tokens_details":{"reasoning_tokens":327}},"tokens_in":725,"tokens_out":387,"duration_ms":4425,"temperature":1.0,"reasoning_tokens":327,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T19:37:12.976901+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same 30-run ensembles at $M_{\\rm disk}=0.005\\,M_{\\rm C}$ and $\\beta=3$ with planetary tides switched on ($k_2/Q \\sim 10^{-3}$-$10^{-2}$): if the fraction of runs that still produce a second satellite with at least 10% of the largest satellite's mass falls below roughly 50%, the claimed two-satellite regime for $0.003$-$0.03\\,M_{\\rm C}$ does not survive the physics the paper itself identifies as missing. A parallel check: include collisional fragmentation at $M_{\\rm disk}=0.03\\,M_{\\rm C}$; if the mean $M_{\\rm l,sat}$ drops by more than about a factor of two, the fitted coefficient of about 0.2 is an artifact of perfect merging.","supporting_citations":[{"cited_title":"Evolution of a circumterrestrial disk and formation of a single Moon","cited_arxiv_id":null,"evidence_quote":"Provides the N-body method, initial eccentricity and inclination distributions, and the Jacobi-energy accretion criterion the simulations inherit."}],"review_version":1}