{"id":"18bd0844-5032-4fcf-b6e9-8d6574210d56","arxiv_id":"2608.11328","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"TOI-3850 hosts two Saturn-mass planets, a transiting warm Jupiter at 14.48 days and a non-transiting companion at 29.85 days, just wide of the 2:1 mean motion resonance.","lead":"Astronomers report a second planet in the TOI-3850 system, a non-transiting Saturn-mass world whose gravity perturbs the transits of a warm Jupiter and shows up in radial velocity data. The pair sits just outside a 2:1 orbital resonance, a configuration that favors disk-driven migration and makes the inner planet a high-priority target for atmosphere studies.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Mass–eccentricity prior and post-hoc RV cut: reported M_c and near-circular architecture may be prior-driven; no sensitivity fit without these choices is shown.","rationale":"The reader's conditional verdict is appropriate, and the flagged exclusion and inclination prior are real concerns. My stress-test pass, however, places the most weight on the mass–eccentricity prior, because the paper itself names the degeneracy in Section 4.3 and then imposes a prior that restricts eccentricities to values consistent with the reported near-circular solution without showing an unconstrained comparison. The RV semi-amplitude K_c anchors the minimum mass near 90 M_Earth, so the mass claim is not fragile to eccentricity alone; but a joint TTV+R V solution can still trade mass against eccentricity and mutual inclination, and the absence of any sensitivity analysis leaves the central characterization conditional. The independent TTV signal and the dynamical stability tests are genuine supporting evidence, which is why I would not move the verdict away from conditional. A single rerun with relaxed eccentricity and inclination priors and with the excluded RV point restored would directly test whether the reported values are data-driven.","tokens_in":29716,"tokens_out":15762,"duration_ms":153362,"concrete_test":"Rerun the joint RV+TTV fit with all 14 MAROON-X RVs and replace the (h,k) priors with a uniform eccentricity prior on [0,0.5] for both planets; in a second run, also replace the i_c prior U(70,90) with U(0,90) plus b>1. Compare the marginal posteriors of M_c, e_b, e_c, and P_c against Table 3. The concern is settled if M_c shifts by less than about 1 sigma and the P_c posterior remains single-moded near 29.85 days; if M_c moves by more than 1 sigma or a second mode appears near P_c approximately 28.17 days, the reported parameters are prior-dependent and the paper would need to soften its headline mass and architecture claims.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 4.3 explicitly acknowledges the known mass–eccentricity degeneracy in TTV fits and then suppresses it with Gaussian priors N(0,0.1) on (h,k)=(sqrt(e)cos(omega), sqrt(e)sin(omega)) for both planets (Table A4). At 95% this restricts e to roughly <0.06, so the reported e_b=0.018±0.008 and e_c<0.015 may be prior-dominated rather than measured. Since TTV amplitudes for a near-2:1 pair can trade M_c against e_c/e_b, the headline M_c=90±15 M_Earth and the 'near-circular' architecture are not shown to be robust to this choice. The companion inclination prior U(70,90) plus the hard b>1 cut (Table A4) is a second prior that could interact with this degeneracy; no sensitivity analysis with a broad or free inclination is reported. Nor is the one excluded MAROON-X point (Section 2.4.2) tested in a refit, so its influence on K_b and K_c is unknown. Any of these choices could alter the claimed true dynamical masses or the wide-of-2:1 branch assignment, although the existence of a second perturbing planet is independently supported by the large TTVs and is not the main vulnerability.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports the discovery of a compact two-planet system around the G0 dwarf TOI-3850, based on TESS photometry, ground-based transit follow-up, MAROON-X radial velocities, and TTV analysis. The authors identify a transiting warm Jupiter, TOI-3850 b (P_b = 14.484 days, M_b = 112 ± 20 M_Earth, R_b = 12.07 ± 0.09 R_Earth), and a non-transiting Saturn-mass companion, TOI-3850 c (P_c = 29.85 days, M_c = 90 ± 15 M_Earth, e_c < 0.015), on near-circular, nearly coplanar orbits just wide of the 2:1 mean motion resonance. The central claim is that the TTV and RV data jointly require the second planet and that both masses are true dynamical masses. The paper also presents dynamical stability tests, a prediction that TOI-3850 c may begin transiting on decadal timescales, and an assessment of the system's formation history and follow-up potential.","tokens_in":29924,"tokens_out":5901,"duration_ms":67457,"significance":"If correct, this is a valuable addition to a small sample of compact multi-giant-planet systems near a first-order resonance, and it provides a rare case where a non-transiting companion's mass is claimed to be dynamically determined rather than only an M_p sin i lower limit. The paper is thorough in its data collection and modeling: it uses five TESS sectors, extensive ground-based photometry, speckle imaging, MAROON-X RVs, and a joint RV+TTV fit with publicly available tools. The authors provide full prior tables, model comparison tables, convergence checks (R-hat < 1.01, emcee chains longer than 50 autocorrelation times), SPOCK stability classifications, and direct N-body integrations. The two-planet interpretation is independently supported by the large TTV amplitude and super-period and by the RV model comparison favoring a two-planet solution over a one-planet solution. The main weaknesses are that the reported masses and near-circular architecture rest on specific priors and on the exclusion of one RV point, with no sensitivity analysis demonstrating that the conclusions are robust to those choices.","major_comments":[{"comment":"The Gaussian priors on (h,k) = (sqrt(e) cos(omega), sqrt(e) sin(omega)) with standard deviation 0.1 restrict eccentricities to roughly e < 0.06 at 95% confidence, so the reported e_b = 0.018 ± 0.008 and e_c < 0.015 may be prior-dominated rather than measured. Because the TTV signal near the 2:1 commensurability can trade mass against eccentricity, the reported M_c = 90 ± 15 M_Earth and the 'near-circular' architecture are not demonstrated to be robust to this choice. I request a sensitivity test with broader priors (for example, uniform in e up to 0.3, or N(0,0.3) on the eccentricity components) and a report of how the posterior masses and eccentricities shift.","section":"§4.3, Table A4"},{"comment":"A single MAROON-X spectrum is excluded because it lies 5-sigma from the posterior solution, but no fit that includes this point is presented. With only 14 RVs and a companion semi-amplitude K_c = 17.5 ± 2.9 m/s, the outlier could plausibly influence K_b and K_c. Please provide a joint RV+TTV fit that includes the excluded point, or at least quantify the change in the fitted parameters and BIC when it is included, and describe an a priori criterion for outlier rejection rather than a post-hoc one.","section":"§2.4.2"},{"comment":"The companion inclination is constrained by the prior U(70,90) and the hard cut b > 1, rather than by a fit that marginalizes over the full allowed parameter space. Although the reported i_c = 87.18 ± 1.29 deg lies inside the prior, the paper shows no test of whether the TTV data alone constrain this angle; without such a test, the statement at the end of §4.3 that both masses are 'true dynamical masses rather than the usual RV minimum mass' is not supported. A fit with a broader inclination prior (or a prior derived from the non-transit constraint) should be reported.","section":"§4.3, Table A4"},{"comment":"The choice of the wide-of-resonance exterior 2:1 branch over the shortward branch is based on a BIC difference of only 1.7 in the RV-only comparison, which is not statistically significant. The subsequent joint RV+TTV fit assumes this branch without re-testing the branch assignment. Since the system's location wide of the 2:1 MMR is a central claim, the joint fit should either compare the two branches or demonstrate that the TTV phases break the degeneracy.","section":"§4.2, Table A3"}],"minor_comments":[{"comment":"The phrase 'W arm Saturn-mass Planets' in the title and in the first line of the draft appears to be a typo for 'Warm Saturn-mass Planets'.","section":"Title, §1"},{"comment":"The BIC values in Table A3 are not accompanied by the number of free parameters or the number of data points for each model; adding these would allow readers to verify the quoted BIC values and assess the model comparison.","section":"Table A3"},{"comment":"The RV-only model comparison uses Gaussian priors on the companion period that are seeded by the TTV super-period; the text should state more explicitly that the detection of a second Keplerian signal is nonetheless independent of the TTV amplitude, since the one-planet versus two-planet comparison does not use the TTV data directly.","section":"§4.2"},{"comment":"The prediction that '50% of samples reach a transiting geometry for TOI-3850 c within the next ~27 years' should be accompanied by a statement of how the assumed inclination prior affects this fraction, since the U(70,90) prior with b > 1 shapes the distribution of first transit epochs.","section":"§5.2.2"}],"recommendation":"major_revision","confidential_remarks":"This is a solid detection paper, but the headline masses and near-circular architecture depend on priors and on an outlier cut that are not sensitivity-tested. The requested analyses are feasible with the existing code and data, so I recommend major revision rather than rejection. The BIC-based branch selection (Delta-BIC = 1.7) should be presented as tentative unless the joint fit confirms it."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Liora —\n\nThe two-planet detection is probably right, but the masses and near-circular architecture are softer than the abstract implies. The TTV signal (1.2-hour amplitude, 513-day super-period) and MAROON-X RVs independently support a second companion; that’s the core result and it holds. The paper is a careful discovery effort: TESS plus a lot of ground-based photometry, difference imaging to rule out BEBs, speckle, SED/isochrone stellar parameters, activity checks, BIC comparison over 18 architectures, and stability tests (SPOCK and REBOUND). Credit where it is due — it is a solid contribution to the sparse warm-Jupiter-with-companion population.\n\nThe stress-test note lands on the right weak spots. The N(0,0.1) priors on (sqrt(e)cosω, sqrt(e)sinω) effectively constrain e ≲ 0.06 at 95%, so the reported e_b = 0.018 and e_c < 0.015 are very likely prior-driven, not measured. In a near-2:1 TTV fit, mass and eccentricity trade; without a run with wider eccentricity priors, M_c = 90±15 M⊕ is conditional. The companion inclination prior U(70,90) plus the hard b>1 cut is similarly load-bearing: claiming “true dynamical masses” at the end of Section 4.3 is premature. A sensitivity test with a broader inclination prior and a free/broader eccentricity prior is needed. Also, one MAROON-X point is excluded as a 5-sigma outlier (Section 2.4.2); that is fine to remove, but a refit including it should be reported. And the BIC gap between the wide and short 2:1 branches in Table A3 is only ΔBIC ≈ 1.7 — weak support on its own for the “wide” branch; the joint RV+TTV fit, not the RV-only comparison, should be the decisive evidence.\n\nNone of this kills the detection; the existence of planet c is robust. But the headline numbers should be treated as prior-conditioned until the robustness checks are shown. The paper deserves a serious referee — this is a real system in a rare configuration, and the issues are fixable. I would send it to peer review and ask for sensitivity analysis on eccentricity and inclination priors, the outlier refit, and a data/code availability statement. For a reading group, it is a good case study in how prior choices interact with TTV/RV joint fits.","headline":"Two-planet detection is likely real and the paper is careful, but the headline masses and near-circular orbits are prior-influenced; needs robustness checks before I would quote them.","tokens_in":30688,"tokens_out":4348,"would_cite":true,"duration_ms":75379,"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":"TOI-3850 hosts two giant planets: a transiting warm Jupiter and a hidden Saturn-mass companion just wide of the 2:1 resonance.","keywords":["exoplanet systems","warm Jupiters","transit timing variations","mean motion resonance","radial velocity","Saturn-mass planets","planetary migration","TOI-3850"],"falsifier":"Retake MAROON-X radial velocities at the phase of the excluded 5-sigma outlier (the epoch near BJD 2461103.95); if a velocity of similar magnitude repeats, the two-Saturn dynamical solution and the quoted masses are not uniquely supported.","tokens_in":29457,"feed_emoji":"🪐","tokens_out":9665,"duration_ms":81247,"temperature":0.7,"pith_summary":"TOI-3850, a G0 dwarf observed by TESS in five sectors, showed a single transiting planet candidate whose arrival times wobbled by about an hour on a 513-day cycle. The paper argues that this wobble, combined with radial velocities from MAROON-X, requires a second planet: a transiting warm Jupiter, TOI-3850 b (period 14.484 days, mass $112\\pm20\\,M_\\oplus$, radius $12.07\\pm0.09\\,R_\\oplus$), and a non-transiting Saturn-mass companion, TOI-3850 c (period 29.85 days, mass $90\\pm15\\,M_\\oplus$), on near-circular, nearly coplanar orbits just wide of the 2:1 mean motion resonance. If the interpretation is right, the system is one of only about eleven known compact two-giant architectures, and its location just outside resonance supports a formation history dominated by disk-driven migration rather than high-eccentricity migration. The same joint fit yields both masses as true dynamical masses rather than minimum masses, and N-body integrations predict that TOI-3850 c may begin transiting within roughly three decades.","feed_headline":"Two Saturn-mass planets found near a 2:1 resonance around TOI-3850","feed_subtitle":"One transits every 14.5 days; the other, hidden, pulls it into hour-long timing wobbles.","key_machinery":"The central mechanism is the transit timing variation signal: the measured deviations of the inner planet's mid-transit times from a constant ephemeris, with an amplitude of about one hour and a super-period of 513 days. That super-period is the signature of a nearby companion near a first-order mean motion resonance, and it fixes the possible period branches through the relation $P_{\\rm super}=|J/P_{\\rm outer}-(J-1)/P_{\\rm inner}|^{-1}$. The paper then fits the radial-velocity and TTV data simultaneously in a single N-body dynamical model, with the outer planet restricted to a non-transiting geometry. The joint fit is the load-bearing inference: it converts an observed timing wobble into the companion's mass and orbit.","core_discovery":"Working from TESS photometry, ground-based transit light curves, MAROON-X radial velocities, and transit timing variations, the paper identifies two giant planets around TOI-3850. The transiting inner planet has period $P_b=14.484$ days, radius $R_b=12.07\\,R_\\oplus$, and mass $M_b=112\\pm20\\,M_\\oplus$; the unseen outer planet has period $P_c=29.85$ days and mass $M_c=90\\pm15\\,M_\\oplus$, with eccentricity below 0.015 and a mutual inclination of about $4.6^\\circ$ relative to the inner planet. The period ratio $P_c/P_b\\approx2.06$ places the system just wide of the 2:1 mean motion resonance, and resonant-angle integrations indicate the planets are likely near, but not locked in, exact resonance. Because the joint RV+TTV fit constrains the outer planet's inclination, the paper concludes that both reported masses are dynamical masses, not minimum masses. Roughly half of the posterior samples have the outer planet reaching a transiting geometry within about 27 years.","pith_inferences":["If the future-transit prediction holds, the system offers a rare natural experiment: the same two planets can be observed both as a TTV/RV pair and later as a double-transit system, so the mutual inclination and resonant state could be measured independently.","The stability of the reported masses could be tested immediately by taking more radial velocities at the phase of the one excluded 5-sigma observation; a repeating signal there would break the two-Saturn solution.","The super-period branch-selection method used here could be applied to other single-candidate systems with large TTVs, potentially turning lone-transiting-planet surveys into a population census of unseen companion architectures."],"forward_implications":["The system becomes a rare benchmark: among more than 6000 confirmed exoplanets, only eleven systems are known with multiple giant planets inside 0.2 AU.","Both planets' masses are stated as true masses; for TOI-3850 c, the joint RV+TTV fit removes the usual $\\sin i$ ambiguity, a step that works because the geometry is nearly coplanar.","The just-wide-of-resonance, low-eccentricity architecture is evidence for disk-driven migration and against high-eccentricity migration, providing a direct test for formation models.","About half of the posterior configurations predict that TOI-3850 c will reach a transiting geometry within about 27 years, so the hidden planet could become measurable from its transits.","TOI-3850 b has a transmission spectroscopy metric higher than 87% of warm Jupiters, making it a strong near-term target for atmospheric follow-up."],"supporting_citations":[{"why":"Supplies the N-body transit-timing integrator used to produce model transit times in the joint RV+TTV fit.","marker":"K. M. Deck et al. 2014"},{"why":"Provides the TTV super-period relation that fixes possible period branches near first-order resonances and the mass-eccentricity degeneracy treatment.","marker":"S. Hadden & Y. Lithwick 2016"},{"why":"Documents the MAROON-X spectrograph, the source of the radial velocities that independently confirm the companion signal.","marker":"A. Seifahrt et al. 2018"},{"why":"The independent faint-star TESS search that first identified TOI-3850.01 as a candidate planet.","marker":"M. Kunimoto et al. 2022"},{"why":"Establishes that TTV solutions for non-transiting planets are degenerate, motivating the exploratory posterior search.","marker":"G. Boué et al. 2012"},{"why":"Recent demonstration of the same TTV posterior degeneracy used to justify the two-step sampling approach.","marker":"C. Lammers & J. N. Winn 2026"},{"why":"The resonance-capture migration framework invoked to interpret the just-wide-of-2:1 architecture as disk-driven migration.","marker":"M. H. Lee & S. J. Peale 2002"},{"why":"Supplies the SPOCK stability classifier used to show that most posterior configurations are dynamically stable.","marker":"D. Tamayo et al. 2020"}],"fun_headline_variants":["Two warm Saturn-mass planets near 2:1 resonance discovered","Hidden Saturn-mass planet revealed by hour-long timing wobbles","TOI-3850 hosts two giants; one may transit within decades","Near-resonant pair: one transits, one undetected, both tug","TTVs expose non-transiting Saturn-mass companion around TOI-3850"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result stands on the assumption that the one MAROON-X measurement flagged as a 5-$\\sigma$ outlier is not real, and that the prior forcing the outer planet to be non-transiting (inclination $70^\\circ<i<90^\\circ$, impact parameter $b>1$) is not doing work the data cannot; if either is wrong, the claimed companion mass and the statement that both masses are true dynamical masses would no longer follow.","fun_headline_variants_meta":{"raw":{"variants":["Two warm Saturn-mass planets near 2:1 resonance discovered","Hidden Saturn-mass planet revealed by hour-long timing wobbles","TOI-3850 hosts two giants; one may transit within decades","Near-resonant pair: one transits, one undetected, both tug","TTVs expose non-transiting Saturn-mass companion around TOI-3850"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000233,"raw_usage":{"total_tokens":1615,"prompt_tokens":1190,"completion_tokens":425,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":806,"completion_tokens_details":{"reasoning_tokens":330}},"tokens_in":806,"tokens_out":425,"duration_ms":4970,"temperature":1.0,"reasoning_tokens":330,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:13:15.195600+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Retake MAROON-X radial velocities at the phase of the excluded 5-sigma outlier (the epoch near BJD 2461103.95); if a velocity of similar magnitude repeats, the two-Saturn dynamical solution and the quoted masses are not uniquely supported.","supporting_citations":[],"review_version":1}