{"id":"f2ef2c76-95d7-4d31-8b69-c101867b8879","arxiv_id":"1908.08548","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"Single-transit K2 planet periods are inferred from Gaia-parallax based stellar densities; circular-orbit fits yield 15% fractional period uncertainty.","lead":"The authors use Gaia parallaxes to estimate host star densities and fit single K2 transits to infer orbital periods of planets that transit only once. When circular orbits are assumed, the method reaches 15% fractional period uncertainty, which could support follow-up of TESS single-transit candidates.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Headline period precision is an artifact of the circular-orbit assumption: with e free the same sample yields 94% fractional uncertainty, and four targets cannot be fit at all with e=0, so the 15% figure overstates what the method delivers for realistic single transiters.","rationale":"Reader's verdict CONDITIONAL is appropriate; I find no reason to reject the paper, but the central quantitative claim needs the caveat. The e=0 issue is more load-bearing than the reader's stated weakest assumption (ingress/egress sampling) because even perfectly sampled transits cannot yield a 15% period precision when eccentricity is unknown and the data are consistent with high e for one-third of the sample. The paper's own free-e result demonstrates this. The density-uncertainty underestimation (Section 2.1) is also concerning and would likely increase the e=0 uncertainty, but the e=0 model rejection is a logical precondition: a precision quoted under a model that cannot fit the data is not a property of the method. The paper is commendably transparent about both caveats, and the validation against known-period planets is useful; however, the abstract singles out the e=0 number as the main result. I would keep the verdict CONDITIONAL, with the condition being that the 15% claim be restricted to targets for which e=0 is supported by the data, or re-expressed through Bayesian model averaging.","tokens_in":22691,"tokens_out":11073,"duration_ms":105407,"concrete_test":"Compute the Bayesian evidence (or Δln Z) for e=0 versus free-e for each of the twelve single-transit fits, then recompute the 15+30−6% statistic three ways: (i) all twelve with forced e=0; (ii) excluding the four targets whose e=0 fits are rejected; (iii) with the four replaced by a Bayesian model average of the e=0 and free-e posteriors. If the median fractional uncertainty exceeds roughly 50% in (ii) or (iii), or if the e=0 credible intervals for those four exclude their free-e posteriors, the headline improvement applies only to a circular-orbit subset and should be reported as conditional.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline improvement (15+30−6% fractional period uncertainty with e=0) is the only quantitative claim that beats previous work, but it is an artifact of fixing a parameter that the data themselves reject for a significant fraction of the sample. The paper reports that four of the true single-transit targets (EPIC 201892470b, EPIC 204634789b, EPIC 228801451d, EPIC 248045685b) have high modal eccentricity and that it is 'unable to achieve good fits to these transits with e fixed to zero' (Section 4.2). The abstract nevertheless reports the 15% figure 'over the true single transits' without stating whether those four are excluded, included with a rejected model, or replaced. With e free, the same sample gives 94+87/−58%; the threefold improvement is therefore not a property of the transit data but a consequence of imposing circular orbits. The validation on 27 known-period planets does not calibrate the e=0 assumption: those fits leave e free, and the sample is dominated by short-period systems for which circularity is a reasonable approximation, not the long-period eccentric systems of interest. The paper is transparent about the e=0 condition, but the abstract and conclusions present 15% as the headline without the caveat that it applies only under a model assumption that fails for roughly a third of the targets.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a method for constraining the orbital periods of singly-transiting K2 planets by combining Gaia DR2 parallaxes with broadband photometry and Yonsei-Yale isochrones to estimate stellar densities, then using those density estimates as priors in single-transit light-curve fits. The method is validated by fitting individual transits of 27 planets with known periods as if they were single transits, and is then applied to 12 true single transiters. With eccentricity free, the reported fractional period uncertainty over the true single transiters is 94+87/-58%; with eccentricity fixed to zero, the reported value is 15+30/-6%. The paper also discusses choices of period prior, the Kipping (2018) prior, and the impact of missing or outlier in-transit data points.","tokens_in":22907,"tokens_out":4718,"duration_ms":50571,"significance":"If the headline precision claim held, the method would be a useful contribution to the long-period planet yield of K2 and TESS, where single-transit events are expected in large numbers. The paper has real strengths: it releases a public fitting code ('single'), it validates stellar densities against an independent asteroseismic sample, it is unusually transparent about limitations such as underestimated stellar-density uncertainties and failures of the e=0 model, and it does not use known periods as inputs in the period inference. However, the central quantitative claim that the method yields a roughly threefold improvement over previous work rests on the e=0 assumption, which the paper itself shows fails for a large fraction of the sample; the validation data also show poor period recovery for the longer-period, eccentric systems that are the target population. These issues substantially temper the significance of the claimed improvement.","major_comments":[{"comment":"The headline fractional period uncertainty of 15+30/-6% is reported in the abstract and conclusions without stating that it is conditional on fixing eccentricity to zero and that this assumption is rejected for a substantial subset of the sample. In the final paragraph of §4.2 the authors state that four of the targets (EPIC 201892470b, EPIC 204634789b, EPIC 228801451d, EPIC 248045685b) have high modal eccentricity and that good fits cannot be achieved with e=0. The abstract presents the 15% figure as if it applied to the true single transiters generally, while the e-free result for the same sample is 94+87/-58%. The comparison to previous work should either restrict the e=0 claim explicitly to the targets for which circular fits are adequate, or should be dropped from the abstract; otherwise the headline is misleading.","section":"Abstract and §4"},{"comment":"The asteroseismic comparison in §2.1 finds an extra scatter of sigma_extra = 0.0313 g/cm3, roughly equal to the mean quoted density uncertainty, and concludes that the error bars are underestimated by approximately a factor of two. However, §2.2 states that the split-normal prior on rho* is constructed from the uncertainties 'derived from the procedure described in 2.1' and gives no indication that sigma_extra was added in quadrature before the transit fits. If this extra uncertainty is not propagated into the period posteriors, the quoted period uncertainties are understated. The authors should state explicitly whether sigma_extra was included in the prior used for the fits, and if not, rerun or rescale the affected results.","section":"§2.1 and §2.2"},{"comment":"The validation section concludes that the method is robust 'as long as the individual transits we fit are well-sampled during ingress and egress,' but the long-cadence validation results in Table 1 show poor and biased period recovery for several planets in exactly the longer-period regime that motivates the paper. For example, K2-56b (P_known = 41.686 d) gives P_fit = 5.4+50.0/-0.6 d, K2-03c (P_known = 24.649 d) gives P_fit = 4.2+80.0/-0.6 d, and K2-32d (P_known = 31.719 d) gives P_fit = 9.1+60.0/-0.7 d. These posteriors are not centered near the true periods despite the transits being presumably well-sampled; a quantitative accuracy criterion or a dedicated long-period validation subset is needed before the robust claim can be supported.","section":"§3.1 and Table 1"},{"comment":"The paper rejects the Kipping (2018) prior for the true single transiters because validation fits with an arbitrary Pmin converge to P = Pmin, yet this is the prior that formally accounts for the single-transit selection effect. The failure on the validation sample is expected, since those transits are not true singles and the arbitrary Pmin is not physically meaningful. For the true singles, the paper shows in §4.1 that the K18 prior gives posteriors inconsistent with independently known periods for EPIC 246445793b and EPIC 211311380f, which suggests that the 12-parameter model is too flexible to be constrained by the long-cadence data under a strong prior. This undercuts the authors' stated reason for preferring the log-uniform prior, because the log-uniform prior is not the observationally motivated choice for true singles. The choice of period prior needs a stronger justification or an explicit sensitivity analysis.","section":"§4.1 and §4.2"},{"comment":"The manuscript contains an internal inconsistency about the number of true single transiters: §3.1 refers to 'the nine true single transits discussed in section 4,' §4 says 'twelve single transits,' Table 2 lists twelve targets, and §4.2 refers to 'four of our nine single transit fits.' Because the 15% e=0 statistic may depend on which targets are included, this mismatch must be resolved and the denominator of the reported fractional uncertainties must be stated unambiguously.","section":"§3.1, §4, Table 2"}],"minor_comments":[{"comment":"The text refers to 'Equation 2.2' for the definition of Pmin, but the equation is numbered (6) in the manuscript; the cross-reference should be corrected.","section":"§4.1"},{"comment":"The axis labels in Figure 1 appear garbled, with the density axis labels and the residual panel labels partly duplicated or missing ('rho_gaia+YY' and 'rho_aste' are misspelled or truncated). The figure should be regenerated with clean labels.","section":"Figure 1"},{"comment":"Posterior distributions are summarized by fitting split-normal distributions, but many posteriors (e.g., those for EPIC 201892470b and EPIC 211311380f) are strongly asymmetric and possibly bimodal; a short discussion of whether split-normal summary statistics are representative, or a plot of representative posteriors in the main text, would improve interpretability.","section":"§2.2 and Tables 1-2"},{"comment":"The claim that 'The nine true single transits discussed in section 4 are well-sampled during ingress and egress' is asserted without a quantitative criterion. In light of the demonstrated sensitivity to missing in-transit points for K2-140b and K2-32b, the authors should specify how many long-cadence data points fall within ingress and egress for each target, or provide a comparable metric.","section":"§3.1"}],"recommendation":"major_revision","confidential_remarks":"The strongest version of the paper's contribution is the method itself plus its careful, honest discussion of where it fails. The e=0 headline is conditional on an assumption that fails for four of the twelve true singles, and the validation does not demonstrate good recovery for the long-period, eccentric targets that are the actual population of interest. The internal inconsistency between 'nine' and 'twelve' single transiters should be resolved before any further review. None of these issues appears to be unfixable within the manuscript's scope, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nRead Sandford et al. The thing to know: the paper's real value is the validation experiment and the diagnostic that transit-shape uncertainty (a/R*) drives period precision, not the headline 15% number. That number is only reached with e fixed to zero, and the paper itself reports four of the targets cannot be fit with e=0. With e free, the sample-wide fractional uncertainty is 94+87/-58%, roughly previous-work level. The abstract does not make that conditionality clear enough.\n\nWhat's new: systematic use of Gaia DR2 parallax plus broadband photometry and Yonsei-Yale isochrones to build stellar density priors for 12 K2 single transiters, tested on 27 planets of known period treated as singles. The densities are checked against asteroseismology, the code is released, and the finding that density precision matters much less than a/R* precision is a genuinely useful caution for follow-up work.\n\nSoft spots, in proportion. First, the e=0 assumption is load-bearing for the 15% claim. Four of the single-transit targets have high modal eccentricity and cannot be fit with e=0, so the improvement over previous work is not a property of the data alone. Second, the paper is internally inconsistent about sample size: it says twelve single transiters, but Section 3.1 and Section 4.2 refer to nine. That needs fixing. Third, the authors identify underestimated density uncertainties (sigma_extra = 0.0313 g/cm3, comparable to their mean reported uncertainty) but do not appear to propagate that extra scatter into the transit fits. Their own diagnostic shows the density term is not dominant, so the practical impact is likely modest, but it should be acknowledged. Fourth, the validation sample is dominated by short-period planets where circular orbits are a reasonable approximation; it does not calibrate the e=0 assumption for the long-period eccentric systems the method is meant for. And the claim that the true single transiters are well-sampled during ingress/egress is not backed by a quantitative criterion, which matters given the K2-140b and K2-32b failure modes they document.\n\nOverall: the central method is credible and honestly reported; the headline precision is overstated. This deserves a serious referee and likely publication after revision that fixes the abstract, the 9/12 inconsistency, and ideally propagates the extra density scatter.","headline":"Useful methods paper with a real validation framework; the headline 15% period precision is conditional on circular orbits and does not hold for a large fraction of the sample.","tokens_in":23536,"tokens_out":2456,"would_cite":true,"duration_ms":25669,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["97.82.-k"],"model":"deepseek-v4-flash","headline":"Using Gaia parallaxes to anchor stellar densities cuts single-transit planet period uncertainty threefold.","keywords":["single-transit planets","stellar density","Gaia parallax","K2","orbital period estimation","transit fitting","eccentricity","long-period exoplanets"],"falsifier":"Watch for a second transit of any of the twelve targets. If future data place their periods outside the 1$\\sigma$ credibility bands given here, the accuracy claim fails; EPIC 211311380f is a sharp test, since the paper's modal period around 600 days sits well above the roughly 360-day period allowed by dynamical simulations from follow-up observations. A more direct check would be to test the four host stars where the Gaia-derived densities disagree with Osborn et al. (2016) using high-resolution spectroscopy and see which stellar density is correct.","tokens_in":22385,"feed_emoji":"🪐","tokens_out":8491,"duration_ms":76217,"temperature":0.7,"pith_summary":"A planet seen transiting only once cannot have its orbital period measured directly, but its period can be constrained from the transit shape if the host star's density is known. This paper combines Gaia parallaxes with stellar models and broadband photometry to measure the host-star densities of K2 single transiters, and verifies the densities against asteroseismology. When the density is used as a prior in single-transit fits to twelve true single transiters, the fractional period uncertainty is $94^{+87}_{-58}\\%$ with eccentricity free and $15^{+30}_{-6}\\%$ with eccentricity fixed to zero, roughly a threefold improvement over earlier work. Validation against 27 planets of known period shows the method is accurate when the transit is well sampled during ingress and egress, and that uncertainty in the transit-shape parameter $a/R_*$, not in stellar density, dominates the period error.","feed_headline":"Gaia stellar densities cut single-transit period error threefold","feed_subtitle":"With circular orbits assumed, fractional period error on K2 single transiters drops to about 15 percent.","key_machinery":"The central identity is Kepler's third law written for transiting planets, $P^2 = \\frac{3\\pi}{G}\\left(\\frac{a}{R_*}\\right)^3 \\rho_*^{-1}$, which converts a transit-shape measurement of the normalized semi-major axis $a/R_*$ and an independent stellar density $\\rho_*$ into the orbital period $P$. The paper couples this with a pipeline that measures $\\rho_*$ from the Gaia parallax distance, SED fits to APASS/2MASS/WISE photometry using BT-Settl-CIFIST models, and Yonsei-Yale isochrone mass/age estimation; it then fits each single transit with a nested-sampling code that treats eccentricity, limb darkening, and Gaussian-process detrending simultaneously. The analysis shows that the posterior uncertainty in $a/R_*$ --- controlled by how well ingress and egress are sampled --- is the dominant driver of $\\sigma_P/P$ in the validation sample, while the input $\\rho_*$ uncertainty is not.","core_discovery":"The paper's central claim is that a stellar bulk density $\\rho_*$ derived from a Gaia parallax, broadband photometry, and isochrone fitting is a reliable and powerful prior for single-transit period inference. Feeding that prior into a transit model and applying it to K2 long-cadence light curves yields period posteriors for true single transiters with fractional uncertainties of $94^{+87}_{-58}\\%$ when eccentricity is a free parameter and $15^{+30}_{-6}\\%$ when $e=0$ is assumed, compared with typical $\\sim 50\\%$ uncertainties in previous single-transit catalogues. The accuracy of the density prior is established by comparing densities computed this way to asteroseismic values for a sample of stars; the two agree with no significant bias. On 27 validation planets with known periods, treating each observed transit as a single transit recovers the true period whenever ingress and egress are well sampled in the data, and fails dramatically (as with K2-140b) when an in-transit point is missing. The paper therefore concludes that single-transit period estimation is limited by the precision with which the transit shape gives $a/R_*$, not by the stellar density uncertainty.","pith_inferences":["Since the paper shows $\\sigma_{a/R_*}$ drives $\\sigma_P$, jointly fitting multiple transiting planets of the same star should tighten $a/R_*$ more than any further improvement in stellar density precision; this is a natural next test.","An immediate extension would be to repeat the validation on TESS 2-minute cadence data for known multi-transit systems, treating each transit as single; the short-cadence results in this paper suggest fractional period uncertainties could shrink well below 15 percent.","The paper's caution about the baseline-informed K18 prior on long-cadence data suggests that such priors may be usable on high-cadence TESS light curves, where the transit model has more information to resist being overwhelmed; this is worth testing.","The four Osborn et al. (2016) disagreements indicate that archival catalogs of single-transit periods may be sensitive to stellar model assumptions; a uniform re-derivation of host star densities from Gaia parallaxes would be a worthwhile population-level check."],"forward_implications":["The threefold precision gain under $e=0$ means that any independent eccentricity constraint would make single-transit period estimates dramatically sharper.","Because $\\sigma_{a/R_*}$ dominates, single-transit surveys with shorter-cadence photometry or sharper limb-darkening priors should yield proportionally better period posteriors.","The K2-140b failure mode implies that single-transit catalogs should flag transits with missing or corrupted ingress/egress points, as these can shift the period by more than 3$\\sigma$.","The same Gaia-based density prior is directly applicable to future TESS single transiters, where it should help realize the predicted gains in long-period planet yield."],"supporting_citations":[{"why":"Supplies the core relation $P^2 = (3\\pi/G)(a/R_*)^3 \\rho_*^{-1}$ and the expected precision scaling that motivates density-based period estimation.","marker":"Yee & Gaudi (2008)"},{"why":"Provides the distance estimates from Gaia parallax used to convert observed photometry to surface flux and hence stellar radius.","marker":"Bailer-Jones et al. (2018)"},{"why":"Provides the BT-Settl-CIFIST model SEDs used to fit the broadband photometry and estimate stellar radius.","marker":"Baraffe et al. (2015)"},{"why":"Provides the Yonsei-Yale isochrones used to estimate stellar mass and age from radius and temperature.","marker":"Yi et al. (2001)"},{"why":"Supplies the asteroseismic stellar densities against which the Gaia+isochrone densities are tested for accuracy.","marker":"Silva Aguirre et al. (2015, 2017)"},{"why":"Provides the re-sampling prescription that folds K2's 30-minute long cadence into the transit model.","marker":"Kipping (2010)"},{"why":"Provides the reparametrization of impact parameter and radius ratio that makes single-transit posterior sampling efficient.","marker":"Espinoza (2018)"},{"why":"Provides the nested-sampling algorithm that explores the multimodal, degenerate single-transit posterior.","marker":"Feroz et al. (2009)"}],"fun_headline_variants":["Gaia densities shrink single-transit period error threefold","Single-transit K2 planets get 15% period errors with Gaia","Circular orbits assumed: Gaia cuts single-transit period error","Gaia stellar density prior improves K2 single-transit periods","Asteroseismology-checked Gaia densities yield 15% period errors"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The twelve true single transits are assumed to be well sampled during ingress and egress, with no missing or corrupted in-transit points that would bias the transit shape and therefore the inferred period.","fun_headline_variants_meta":{"raw":{"variants":["Gaia densities shrink single-transit period error threefold","Single-transit K2 planets get 15% period errors with Gaia","Circular orbits assumed: Gaia cuts single-transit period error","Gaia stellar density prior improves K2 single-transit periods","Asteroseismology-checked Gaia densities yield 15% period errors"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001263,"raw_usage":{"total_tokens":5213,"prompt_tokens":1026,"completion_tokens":4187,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":642,"completion_tokens_details":{"reasoning_tokens":4097}},"tokens_in":642,"tokens_out":4187,"duration_ms":29768,"temperature":1.0,"reasoning_tokens":4097,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:37:24.348288+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Watch for a second transit of any of the twelve targets. If future data place their periods outside the 1$\\sigma$ credibility bands given here, the accuracy claim fails; EPIC 211311380f is a sharp test, since the paper's modal period around 600 days sits well above the roughly 360-day period allowed by dynamical simulations from follow-up observations. A more direct check would be to test the four host stars where the Gaia-derived densities disagree with Osborn et al. (2016) using high-resolution spectroscopy and see which stellar density is correct.","supporting_citations":[],"review_version":1}