{"id":"43a7068b-2959-41ab-9984-d173ea3da7bf","arxiv_id":"2502.01249","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":3,"one_line_summary":"KOBE-1 is a K7 dwarf hosting two sub-Neptune-minimum-mass planets with periods of 8.54 and 29.67 days.","lead":"Astronomers report the discovery of two new planets around the nearby K-dwarf star HIP 5957, using three years of high-precision radial velocity measurements from the CARMENES spectrograph. This is the first planetary system from the KOBE survey, which targets planets in the habitable zones of late K-dwarfs.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified","rationale":"I examined the RV analysis in §4.1, the confirmation logic in §5.1, and the lunar-exclusion appendix (B). The two-planet model is preferred over one-planet by ΔlnZ = 7.9–9.6 depending on the estimator, well above the standard threshold. The signals are recovered by three different RV extraction pipelines when the same NZP correction is applied (Appendix A), and their amplitudes are stable across four observing seasons (Appendix D). Stellar activity is ruled out by the absence of periodogram peaks at Pb and Pc in all activity indicators, no correlation of residuals with indicators, and a GP joint fit where the RV amplitude is consistent with zero. The only scenario that could invalidate planet c specifically is a 29.5-day source of RV variation. The paper's removal of nine Moon-contaminated points is conservative, and the contaminated subset shows no 29.7-day peak; the Cunha et al. (2013) upper limit of 0.3 m/s is an order of magnitude below the observed Kc. The evidence is therefore compelling. The main residual worry is whether the nightly zero-point corrections themselves contain a lunar-period common-mode signal that could be injected into all KOBE targets; the survey-wide control showing only KOBE-1 with the periodicity is reassuring, but a direct periodogram of the NZP series would close this. This does not change the verdict: the paper's central claim is well supported.","tokens_in":35779,"tokens_out":14007,"duration_ms":124893,"concrete_test":"Compute the GLS periodogram of the S-BART nightly zero-point (NZP) time series used to correct the KOBE-1 RVs, and check for a peak at 29.5 d with FAP < 1%. If the NZPs contain a lunar-period signal, re-fit the RVs with an NZP model that includes a 29.5-d sinusoid; if the 29.7-d planetary signal persists with Kc > 3 m/s, the lunar/systematic origin is excluded. If the NZP periodogram is clean, the current conclusion stands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that KOBE-1 hosts two RV-detected planets is supported by independent pipelines (S-BART, serval, SHAQ), strong Bayesian evidence (ΔlnZ ≈ 7.9–10 vs one-planet models), consistent semi-amplitudes across four seasons, absence of correlation with activity indicators, and a GP analysis showing zero RV-activity amplitude. The closest approach to a load-bearing concern is the 29.7-day signal's proximity to the lunar synodic month (Appendix B). This is convincingly excluded: nine Moon-contaminated epochs were removed, the contaminated subset shows no 29.7-day peak, other KOBE targets show no such periodicity, and the maximum lunar RV effect (0.3 m/s, Cunha et al. 2013) is 11 times smaller than Kc. A possible untested channel is a lunar-periodicity in the nightly zero-point corrections, but the survey-wide control argues against this. The reliance on the unpublished ECP protocol affects terminology ('confirmed' vs 'candidate') but not the underlying detection significance.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"KOBE-1 (HIP 5957) is a K7V dwarf monitored by CARMENES as part of the KOBE survey. Using 82 S-BART radial velocities over 1180 days, the paper reports two periodic signals at P_b = 8.5399 d with K_b = 3.75 ± 0.48 m/s and P_c = 29.671 d with K_c = 3.50 ± 0.45 m/s. The authors model the data with two circular Keplerians plus a quadratic trend and show that Bayesian evidence strongly favors this model over zero- and one-planet alternatives. The signals are stable across four observing seasons, are uncorrelated with activity indicators, and survive the exclusion of potentially Moon-contaminated epochs. The inferred minimum masses are 8.80 ± 0.76 M⊕ and 12.4 ± 1.1 M⊕, placing the companions in the planetary-mass regime with high probability. A single transit-like feature in TESS is analyzed but not claimed as a transit, and injection-recovery is used to set sensitivity limits inside the habitable zone. The paper closes with LIFE simulation-based predictions for future direct-imaging characterization.","tokens_in":35917,"tokens_out":11879,"duration_ms":115209,"significance":"This is a solid, carefully analyzed RV discovery and the first planetary system from the KOBE survey. Its main strength is cross-validation: the signals appear in independent extractions (S-BART, serval, SHAQ), in two Bayesian frameworks (kima and radvel with bayev), in seasonal amplitude checks, and in activity diagnostics including a Gaussian-process treatment. The authors are appropriately cautious about the ambiguous TESS feature and about the lunar alias, and the injection-recovery sensitivity analysis is reproducible with public tools. The system is scientifically valuable as a relatively quiet late-K dwarf hosting two sub-Neptune-minimum-mass planets, and the LIFE integration-time estimates are falsifiable predictions for future missions. I find no load-bearing technical error in the detection claim, and the presentation is generally transparent and conservative.","major_comments":[],"minor_comments":[{"comment":"The formal classification of the two signals as \"confirmed planets\" relies on the Exoplanet Confirmation Protocol of Lillo-Box et al. (in prep.), which is not publicly available; please cite a public version or state the adopted criteria explicitly so that the classification is independently checkable.","section":"5.1"},{"comment":"The abstract states that planets above 8.5 M⊕ are discarded within the habitable zone, whereas Section 5.3 reports sensitivity of about 8.5 M⊕ at 90-105 d for the conservative HZ and about 7.6 M⊕ at about 80 d for the optimistic HZ; this wording should be rephrased to avoid implying a uniform mass limit across the full HZ.","section":"Abstract and 5.3"},{"comment":"The lunar-contamination discussion is convincing, but a periodogram or RMS of the nightly zero-point time series itself would close the residual channel of a Moon-induced systematic in the NZPs; the current survey-wide control is reassuring yet indirect.","section":"Appendix B"},{"comment":"There is a typo: \"both √e cosω and √e cosω\" should read \"both √e cosω and √e sinω.\"","section":"4.1.2"},{"comment":"Minor typographical errors should be corrected: \"signal to noise natio\" (Section 2.1), \"photomery\" (Introduction), and \"for for their whole domain\" (Section 4.1.2).","section":"2.1 and Introduction"},{"comment":"The evidence comparison in Figure 1 is difficult to parse because multiple ΔlnZ values are plotted without a clear statement of the reference model in the caption; please make the reference model and the sign convention explicit.","section":"Fig. 1"},{"comment":"The title's \"sub-Neptune mass regime\" is stronger than Section 5.2's conclusion that the minimum masses are compatible with either the super-Earth or sub-Neptune populations; consider a wording such as \"super-Earth/sub-Neptune mass regime.\"","section":"Title and 5.2"}],"recommendation":"minor_revision","confidential_remarks":"The detection is robust and the manuscript fits A&A well. The main editorial point is the use of the unpublished Exoplanet Confirmation Protocol for the word \"confirmed\"; I suggest the editors encourage the authors to make the protocol criteria explicit in the text. No other concerns."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The KOBE-1 paper is a clean, careful RV discovery: two sub-Neptune-minimum-mass planets around a late K dwarf, the first system from the KOBE survey. Nothing methodologically new—GLS, Bayesian model comparison, GP activity modeling are standard—but the execution is thorough and the result is a genuine new data point for planet occurrence around late K dwarfs.\n\nWhat the paper does well: three independent RV extractions (S-BART, serval, SHAQ) agree; the two-planet model beats one-planet and no-planet models by strong Bayes factors; semi-amplitudes are stable across four seasons; activity indicators show no correlation, and a joint GP fit with an activity proxy finds zero RV amplitude. The activity analysis is particularly careful. The sensitivity limits are derived from injection-recovery, and the data are publicly available.\n\nThe soft spots are minor. The 29.7-day period sits close to the lunar synodic month, which is the obvious worry. The authors address it directly: nine Moon-contaminated epochs were removed, the signal persists in the cleaned set, the contaminated subset shows no 29.7-day peak, other KOBE targets show nothing at that period, and the maximum lunar-induced RV is 0.3 m/s—eleven times smaller than the signal. That is convincing, and I don't think the worry lands. The reliance on an unpublished confirmation protocol (ECP) for the word 'confirmed' is a citation issue, not a scientific one; the detection stands on its own. The single TESS transit-like feature is openly inconclusive, and they do not overclaim it. The masses are minimum masses, so the 'sub-Neptune regime' is a statement about m sin i, and the paper is appropriately careful about that.\n\nThe reader's take and the stress-test note both match my reading. The paper deserves a serious referee: it is a well-executed discovery from a new survey, with reproducible data and checks. I would accept it after minor revisions, mainly asking them to either cite a stable version of the ECP protocol or soften the 'confirmed' terminology. This is a useful paper for anyone working on K-dwarf exoplanets and RV survey strategies.","headline":"Solid two-planet RV detection from a new survey; the lunar-contamination worry is convincingly put to bed.","tokens_in":36717,"tokens_out":2391,"would_cite":true,"duration_ms":22375,"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":"The KOBE survey's first discovery is a K7 dwarf hosting two sub-Neptune-minimum-mass planets at periods of 8.54 and 29.67 days, confirmed by CARMENES radial velocities.","keywords":["exoplanets","radial velocity","K-dwarf stars","sub-Neptunes","KOBE survey","CARMENES","Bayesian model comparison","habitable zone"],"falsifier":"Continue monitoring KOBE-1 with CARMENES for two more seasons and recompute the 29.671-day signal after excluding every night with Moon illumination above 40% and angular separation below 80 degrees; if the signal's phase and amplitude do not remain coherent with the published ephemeris, or if a comparable 29.53-day signal appears in the RVs of other KOBE targets observed the same nights, the planetary interpretation for the outer signal is wrong.","tokens_in":35586,"feed_emoji":"🪐","tokens_out":7614,"duration_ms":64753,"temperature":0.7,"pith_summary":"This paper reports the first planetary system discovered by the KOBE blind radial-velocity survey: a late K-dwarf (KOBE-1, HIP 5957) hosting two planets with minimum masses of 8.80 and 12.4 Earth masses at periods of 8.54 and 29.67 days. The authors aim to establish that both signals are genuine planets, not stellar activity, instrumental systematics, or lunar contamination, by combining 82 CARMENES spectra over three years with Bayesian model comparison and activity-indicator checks. The result matters because K-dwarfs are prime targets for habitable-zone searches, and this system provides a concrete example of the compact, sub-Neptune-mass population such surveys can uncover. It also demonstrates a path to future atmospheric characterization: the paper argues the proposed LIFE nulling interferometer could detect both planets directly.","feed_headline":"Two sub-Neptunes found around K-dwarf KOBE-1","feed_subtitle":"CARMENES radial velocities confirm 8.8 and 12.4 Earth-mass planets; first discovery of the KOBE survey.","key_machinery":"The central machinery is a radial-velocity model with two circular Keplerians plus a quadratic long-term trend, whose parameters are sampled with diffusive nested sampling and with an MCMC ensemble sampler, with model comparison done through Bayesian log-evidence (Bayes factors). Around this core sit the S-BART template-matching RV extraction with nightly zero-point corrections, GLS periodograms to find candidate periods, activity indicators (bisector span, FWHM, differential line width, H-alpha and calcium indices) to reject stellar-activity origins, a Gaussian-process test using the CaIRT2 index as an activity proxy, and a Moon-contamination screening that removes and re-analyses potentially contaminated epochs. The same machinery is reused to compute detection limits in the habitable zone by injecting single Keplerians into the residuals and taking the 99% upper limits, which rules out planets above about 8.5 M_Earth in the conservative HZ.","core_discovery":"KOBE-1 is a relatively quiet K7V star at 23.9 pc that hosts two planets in the sub-Neptune minimum-mass regime. Planet b orbits with period 8.5399 days and semi-amplitude 3.75 m/s, giving m_b sin i_b = 8.80 ± 0.76 M_Earth; planet c orbits with period 29.671 days, semi-amplitude 3.50 m/s, giving m_c sin i_c = 12.4 ± 1.1 M_Earth. The two-circular-Keplerian model is strongly preferred over zero- and one-planet models (ΔlnZ = 9.7 and 7.9), the amplitudes are consistent across four observing seasons, and no activity indicator correlates with the RVs. Under the Exoplanet Confirmation Protocol scheme, the authors classify both as confirmed planets: the probability that either companion exceeds the 13 M_Jup planetary-mass limit is below 0.3%. The 29.7-day signal sits close to the lunar synodic month, but the paper excludes nine Moon-contaminated epochs and shows the signal persists, with a theoretical Moon-induced RV upper limit of about 0.3 m/s, eleven times weaker than the detected signal. A single TESS transit-like dimming in sector 17 is compatible within 1σ with a transit of planet c, but the paper does not claim it as a transit.","pith_inferences":["If the TESS dimming is later confirmed as a transit of KOBE-1 c, the planet would be a dense super-Earth at low irradiation, a combination that would be difficult to explain by photoevaporation and would test interior-structure models.","The near-coincidence between the 29.671-day period and the lunar synodic month suggests that other RV surveys should routinely apply the same Moon-contamination screening; a 29.5-day signal in a single target should be checked against other stars observed the same nights.","The KOBE-1 system, with two similar-amplitude planets and a quiet host, is a good benchmark for testing dynamical packing and formation scenarios of sub-Neptunes around low-mass stars.","If future seasons push the evidence for a third signal near 18.7 days (half the stellar rotation period), it would indicate that activity harmonics can lurk just below detection thresholds even in quiet K-dwarfs, strengthening the case for GP modeling in such surveys."],"forward_implications":["KOBE-1 b and c are confirmed planets with minimum masses of 8.80 ± 0.76 and 12.4 ± 1.1 Earth masses, orbiting at 8.54 and 29.67 days around a K7 dwarf.","The KOBE survey's observing strategy works: a blind CARMENES campaign with ~24 points per season can uncover multi-planet systems around late K-dwarfs.","No planets with minimum masses above about 8.5 Earth masses remain in the habitable zone of KOBE-1, so the system's HZ is empty at the current sensitivity.","The single TESS dimming, if it is a transit of planet c, would imply a radius of 1.69 ± 0.12 R_Earth and a dense super-Earth composition, but the data do not yet justify that claim.","The proposed LIFE nulling interferometer could detect both planets directly with integration times of about one hour (planet b) and four hours (planet c) in the baseline super-Earth scenario."],"supporting_citations":[{"why":"Provides the kima diffusive nested-sampling code used to infer the number of Keplerians and the Bayesian evidence for the two-planet model.","marker":"Faria et al. 2018"},{"why":"S-BART, the template-matching pipeline whose radial velocities and nightly zero-point corrections form the analyzed time series.","marker":"Silva et al. 2022"},{"why":"Defines the generalized Lomb-Scargle periodogram used to identify the 8.5 d and 29.7 d signals.","marker":"Zechmeister & Kürster 2009"},{"why":"Provides radvel, the MCMC Keplerian modeling package used for posterior sampling and model comparison.","marker":"Fulton et al. 2018"},{"why":"Supplies the upper limit on Moon-induced radial velocity (~0.3 m/s) used to dismiss lunar contamination of the 29.7 d signal.","marker":"Cunha et al. 2013"},{"why":"Defines the KOBE survey design and target sample from which KOBE-1 is drawn.","marker":"Lillo-Box et al. 2022"},{"why":"Establishes the injection-recovery method used to compute the 99% detection limits that exclude HZ planets above 8.5 M_Earth.","marker":"Standing et al. 2022"}],"fun_headline_variants":["First KOBE survey planets: two sub-Neptunes around K-dwarf KOBE-1","Two sub-Neptunes (8.8 and 12.4 Earth masses) around K-dwarf KOBE-1","KOBE-1: first planets from KOBE survey, two sub-Neptunes around K-dwarf","KOBE survey's first discovery: two sub-Neptune planets around K-dwarf"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The 29.7-day radial-velocity signal is a genuine Keplerian orbit rather than a lunar-synodic-month systematic, even though 29.7 days nearly equals the Moon's 29.53-day synodic period.","fun_headline_variants_meta":{"raw":{"variants":["First KOBE survey planets: two sub-Neptunes around K-dwarf KOBE-1","Two sub-Neptunes (8.8 and 12.4 Earth masses) around K-dwarf KOBE-1","KOBE-1: first planets from KOBE survey, two sub-Neptunes around K-dwarf","KOBE survey's first discovery: two sub-Neptune planets around K-dwarf"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000706,"raw_usage":{"total_tokens":3333,"prompt_tokens":1246,"completion_tokens":2087,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":862,"completion_tokens_details":{"reasoning_tokens":1976}},"tokens_in":862,"tokens_out":2087,"duration_ms":13354,"temperature":1.0,"reasoning_tokens":1976,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T15:53:03.824285+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Continue monitoring KOBE-1 with CARMENES for two more seasons and recompute the 29.671-day signal after excluding every night with Moon illumination above 40% and angular separation below 80 degrees; if the signal's phase and amplitude do not remain coherent with the published ephemeris, or if a comparable 29.53-day signal appears in the RVs of other KOBE targets observed the same nights, the planetary interpretation for the outer signal is wrong.","supporting_citations":[],"review_version":1}