{"id":"6e638c4f-6a80-4a54-908a-ecf623e938ab","arxiv_id":"2505.22951","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Two microlensing events are shown to be caused by roughly Jupiter-mass planets, with one event exhibiting a newly identified four-way solution degeneracy.","lead":"This paper reports two new giant planets found by gravitational microlensing, where the planet's signal is a weak pair of brightness spikes near the peak of the star's light curve. The authors also describe a new type of model ambiguity, a degeneracy, that can complicate such detections and that future space surveys should resolve.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"MOA-2022-BLG-091's planetary origin rests on five points with no reported binary-source or outlier rejection; a 1L2S or systematic interpretation is not excluded.","rationale":"The reader's weakest assumption points at sparse data and systematics; I agree and want to make the test concrete. The central claim is not just 'there is an anomaly' but 'the anomalies in both events are of planetary origin.' For KMT-2024-BLG-1209 there is a measured finite-source radius and a well-sampled caustic; the planetary interpretation is much better supported. For MOA-2022-BLG-091, the data are five points, one of which defines the first spike. A four-solution degeneracy with Δχ²<6.7 means the 2L1S parameter space is shallow; it does not by itself rule out non-planetary models. The standard microlensing literature usually checks 1L2S when anomalies have multiple excursions; its absence here is a missing control rather than an inconsistency. A single re-reduction or model run could settle whether the claim survives. My recommendation is to keep the reader's CONDITIONAL verdict, with the explicit condition that the authors perform and report the 1L2S fit and the outlier-removal fit for MOA-2022-BLG-091 before the two-event conclusion is accepted as robust.","tokens_in":15386,"tokens_out":6700,"duration_ms":73988,"concrete_test":"Use the archived MOA-2022-BLG-091 photometry (or request it from the authors) and: (1) fit a 1L2S model with free binary-source parameters and report Δχ² = χ²_{1L2S} − χ²_{2L1S,best}; if |Δχ²| is within a few units, the 'planetary origin' claim is not unique. (2) Refit the 2L1S model with the single KMTA point near HJD∼2459655.3 removed; if the best-fit q changes by more than ~30% or the four solutions reorder/merge, the pre-peak anomaly does not robustly determine the planet mass ratio. (3) Check that point's pixel-level residuals against neighboring reference stars to rule out a PSF or bad-pixel artifact. If 1L2S is strongly rejected and q is stable to removal, the concern is settled.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The combined conclusion in §8 that 'the anomalies in both lensing events are of planetary origin' inherits its weakness from MOA-2022-BLG-091. In §4 the pre-peak anomaly is a single KMTA point and the post-peak anomaly is four KMTC points; the normalized source radius is only an upper limit (ρ < 1.5×10^{-3}), so no resolved finite-source caustic crossing is actually measured. The four 2L1S solutions are mutually degenerate (Δχ² < 6.7, Table 2), which already indicates that the anomaly is weakly constraining. The paper never reports a 1L2S (binary-source) model or a test with the single KMTA point removed. Two flux deviations bracketing the peak could in principle be produced by a binary source passing near a single lens, or by one or two outlier points. Because neither alternative is shown to be excluded, the claim that the anomaly is planetary is not uniquely established for this event; the Bayesian masses and separations in Table 5 then inherit this model assumption. This is not an attack on the method or the authors; it is a missing falsification in an otherwise standard analysis.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper analyzes two microlensing events, MOA-2022-BLG-091 (KMT-2022-BLG-0114) and KMT-2024-BLG-1209 (OGLE-2024-BLG-0777), whose peak-region light curves show similar weak caustic-crossing anomalies. The authors perform a standard 2L1S grid-plus-MCMC modeling search and find, for the first event, four degenerate solutions attributed to a newly proposed cusp-crossing degeneracy, and for the second event, the previously known inner-outer degeneracy. Using Bayesian analysis with a Galactic model, they estimate host masses near 0.75 M_Sun, planet masses of about 2-4 M_Jup, distances around 4 kpc, and disk-dominance probabilities near 87%. The paper concludes that both anomalies are of planetary origin and that the MOA-2022-BLG-091 degeneracy should be resolvable by the Roman RGES survey.","tokens_in":15614,"tokens_out":5174,"duration_ms":51758,"significance":"If the planetary interpretation holds, the paper adds two giant planets around early K-type main-sequence stars and identifies a new class of weak caustic-crossing anomalies. The analysis of KMT-2024-BLG-1209 is reasonably supported: the caustic crossings appear to be resolved (Table 3), and the inner-outer degeneracy check via Eq. (1) is quantitative and appropriate. The paper is also honest about its degenerate solutions and presents the physical parameters with credible intervals. However, the evidence for MOA-2022-BLG-091 is substantially weaker: the anomaly is defined by a single KMTA point and four KMTC points, the four 2L1S solutions are separated by Delta-chi-squared < 6.7, and only an upper limit on the normalized source radius is obtained. Binary-source models and outlier-rejection tests are not reported, so the planetary origin of that event is not uniquely established. The proposed new degeneracy is interesting but needs robustness checks before it can be regarded as established.","major_comments":[{"comment":"The planetary-origin claim in Section 8 ('the anomalies in both lensing events indicate a planetary origin') rests, for MOA-2022-BLG-091, on a sparse anomaly: one KMTA point before the peak and four KMTC points after it. With Delta-chi-squared less than 6.7 among four 2L1S solutions and only an upper limit on the normalized source radius, the data do not uniquely constrain a resonant-caustic crossing. The paper does not report a 1L2S (binary-source) model, nor a test with the single KMTA point removed or with the KMTC points checked for systematics. The authors should model these alternatives and quantify the robustness of the four 2L1S solutions; if these alternatives cannot be excluded, the conclusion for this event should be weakened to a candidate planetary anomaly and the physical parameters presented as conditional on the 2L1S model.","section":"Section 4, Fig. 1, Table 2"},{"comment":"The Bayesian posteriors for MOA-2022-BLG-091 are treated on the same footing as those for KMT-2024-BLG-1209, even though the former event has only a lower limit on theta_E (Table 4). Because Eq. (5) is applied with the one-sided constraint theta_E,i > theta_E,min, the median and 16-84 percentile ranges quoted in Table 5 are largely prior-dominated. The text should state this explicitly, show the sensitivity of the posteriors to the adopted mass function and Galactic model, and avoid presenting the physical parameters of the two events as equally secure in the abstract and Section 8.","section":"Section 7, Eq. (5), Table 5"},{"comment":"The claim of a newly identified four-fold degeneracy for MOA-2022-BLG-091 is not yet fully supported. The four solutions are formally separated by less than 6.7 in chi-squared, but with sparse sampling and an unconstrained source radius, the apparent separation into four distinct alpha values could reflect under-sampling rather than a true four-fold cusp-crossing degeneracy. The authors should demonstrate that the distinct minima are stable under removal of individual anomaly points and under alternative photometric reductions, and should explain which features in the data select these four specific trajectories rather than a continuous family.","section":"Section 4, Fig. 2"}],"minor_comments":[{"comment":"The phrase 'as as positive deviations' should be corrected to 'as positive deviations'.","section":"Section 1"},{"comment":"The event MOA-2022-BLG-091 is referred to as KMT-2022-BLG-0114 in the opening sentence, which is inconsistent with the naming convention stated in Section 2 that the MOA designation is adopted throughout.","section":"Section 8"},{"comment":"The uncertainty on u0 for solution D is given as 0.84, which appears to be a decimal typo; the other rows suggest 0.084 was intended.","section":"Table 2, Sol D"},{"comment":"The sentence 'A planet-induced caustic take different shape' should be 'A planet-induced caustic takes different shapes', and the abstract should use 'space-based' rather than 'space based'.","section":"Section 3"},{"comment":"The notation <s> is defined as the geometric mean in the text, but for readability it would help to also state in words that <s> = sqrt(s_in * s_out) immediately after the equation.","section":"Eq. (1)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the scope of A&A and the modeling methodology is standard. The main concern is evidentiary: the planetary interpretation of MOA-2022-BLG-091 is under-supported by the sparse photometry and the absence of alternative-model tests. I would support publication after the requested robustness tests are added and the conclusions for that event are appropriately qualified."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Punchline: this is a competent, honest microlensing analysis that adds two giant planet candidates and one genuinely new degeneracy, but the MOA-2022-BLG-091 planetary claim is less secure than the abstract suggests. The stress-test note lands. The first anomaly in that event is a single KMTA point, the second is four KMTC points, rho is only an upper limit, and the four 2L1S solutions are degenerate at Delta chi-squared < 6.7. The paper does not report a 1L2S model or a fit excluding that one point, so a binary-source or systematic interpretation has not been falsified. That is a missing check, not a fatal flaw, but it should be fixed before the \"planetary origin\" conclusion is stated as firmly as it is in Section 8.\n\nWhat is new and good: two new planet detections (MOA-2022-BLG-091Lb, KMT-2024-BLG-1209Lb), a clearly described four-way degeneracy for the first event, and a clean confirmation of the known inner-outer degeneracy in the second via the <s> = s-dagger relation. The modeling is standard but careful: grid plus MCMC, Delta chi-squared maps, parameter tables for all local solutions, honest upper limits. The source color/magnitude calibration and Bayesian posteriors are transparent; tE and theta_E are used as constraints, not outputs, so there is no circularity. Citations are appropriate, including the Han self-citations to similar-signal papers.\n\nSoft spots beyond the missing 1L2S test: the Bayesian masses and distances (2-4 M_J, ~4 kpc) depend on adopted Galactic priors and, for MOA-2022-BLG-091, on the theta_E lower limit. Those numbers should be labeled model-dependent in the abstract. Also, the new degeneracy is interesting but it is a modeling ambiguity, not a new physical phenomenon. The KMT-2024-BLG-1209 detection is considerably firmer: the caustic spikes are better sampled and rho is measured.\n\nBottom line: this deserves peer review, not desk rejection. I would ask the referee to require a 1L2S model and an outlier-robustness test for MOA-2022-BLG-091, plus a data release. With that, it is a useful contribution to the microlensing census and relevant to Roman RGES forecasts.","headline":"Two new giant-planet microlensing candidates, one solid and one that needs a 1L2S/outlier check before the planetary claim is secure.","tokens_in":16636,"tokens_out":3782,"would_cite":true,"duration_ms":39790,"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":"Two weak caustic-crossing signals in microlensing events reveal giant planets two to four times Jupiter's mass.","keywords":["gravitational microlensing","exoplanet detection","caustic crossing","degeneracy","giant planets","Galactic disk","K-type stars","Roman RGES"],"falsifier":"A re-reduction of the MOA-2022-BLG-091 images with an independent photometric pipeline, or any archival coverage of the same anomaly region, that shows the single deviant KMTA point and the four KMTC points to be noise or artifacts would eliminate the claimed planetary interpretation; likewise, a future high-cadence observation of a similar weak-caustic event whose spikes are resolved as non-caustic would challenge the identification.","tokens_in":15131,"feed_emoji":"🪐","tokens_out":5388,"duration_ms":45856,"temperature":0.7,"pith_summary":"The paper analyzes two microlensing events whose light curves show nearly identical weak anomalies near their peaks: brief, low-amplitude spikes consistent with the source crossing a caustic but with almost no distortion while the source is inside it. Through binary-lens modeling, the authors conclude that both anomalies are planetary in origin, produced when the source crosses the resonant caustic of a giant planet nearly perpendicular to the planet-host axis. The planets are estimated to be roughly two to four times Jupiter's mass, orbiting early K-type main-sequence stars in the Galactic disk at about 4 kiloparsecs. The analysis also uncovers a previously unrecognized four-solution degeneracy in one event, tied to the uncertain angle of the source trajectory, that is expected to be broken by future space-based microlensing surveys.","feed_headline":"Two faint lensing blips reveal giant planets 2-4 times Jupiter","feed_subtitle":"Weak caustic-crossing signals hide a new degeneracy that the Roman space telescope should break.","key_machinery":"The central object is the resonant (six-cusp) caustic formed when a giant planet sits near the Einstein ring of its host, together with the source trajectory's incidence angle $\\alpha$. A trajectory crossing the cusps that are perpendicular to the planet-host axis produces the weak caustic-crossing spikes with minimal in-caustic deviation; different $\\alpha$ values give rise to the new four-way degeneracy. For KMT-2024-BLG-1209, the argument is carried by the inner-outer degeneracy relation $\\langle s\\rangle = \\sqrt{s_{\\mathrm{in}} s_{\\mathrm{out}}} = s^{\\dagger} = \\sqrt{u_{\\mathrm{anom}}^2 + 4} + u_{\\mathrm{anom}}$, which the paper verifies numerically and uses to show the two solutions are intrinsically indistinguishable in caustic structure.","core_discovery":"The central claim is that the anomalies in MOA-2022-BLG-091 and KMT-2024-BLG-1209, which consist of weak caustic-crossing spikes with minimal in-caustic deviation, are produced by planetary companions rather than by single-lens or stellar-binary configurations. Modeling yields four nearly equally good solutions for the first event and inner-outer degenerate solutions for the second; all solutions have mass ratios q ~ 3-5 x $10^{-3}$, i.e., giant planets, with the source crossing a single resonant caustic at an incidence angle close to right angle. Bayesian estimates place both systems in the Galactic disk at about 4 kpc, with 2-4 Jupiter-mass planets beyond the snow line of early K-type hosts. The paper also claims a new degeneracy type for MOA-2022-BLG-091, arising by chance from similar source trajectories relative to cusps, as opposed to the intrinsic inner-outer degeneracy of the other event.","pith_inferences":["The paper's identification of a chance degeneracy tied to $\\alpha$ implies that other published planetary microlensing detections based on sparse caustic spikes could carry similar hidden ambiguities; re-checking archived events for four-way solutions may turn up additional degenerate cases.","If the new degeneracy is as common as the two similar anomalies suggest, the Roman survey's cadence will not only resolve this event but also statistically calibrate how often ground-based sampling underdetermines planet-host geometry.","The similarity between the two anomalies hints at a selection effect: events viewed almost perpendicular to the planet-host axis are the most likely to show weak, easily missed caustic crossings, so the true rate of such giant planets near the Einstein ring may be higher than current ground-based detection statistics imply.","One could test the planetary interpretation further by checking whether the implied relative proper motion and source color/radius are consistent with a single epoch of adaptive-optics or HST imaging resolving the lens-source blended light."],"forward_implications":["Both systems add two giant planets (roughly 2-4 Jupiter masses) to the census of microlensing planets in the Galactic disk, each beyond its host's snow line.","The newly identified trajectory-angle degeneracy implies that similar weak-caustic events analyzed with sparse ground-based data may harbor multiple solutions and should not be treated as unique parameter determinations.","The MOA-2022-BLG-091 degeneracy is predicted to be lifted by the future Roman RGES survey's continuous high-cadence sampling, providing a testable forecast.","The inner-outer degeneracy in KMT-2024-BLG-1209 is intrinsic to the caustic structure and will remain unresolved even with better data, so the planet's separation is only known up to the $\\sqrt{}$-relation.","The common morphology of the two anomalies suggests a recognizable class of planetary signals - weak caustic crossings near the peak of moderately high magnification events - that future searches can target."],"supporting_citations":[{"why":"Establishes gravitational microlensing as a method to detect planets, providing the foundation for interpreting the events.","marker":"Mao & Paczyński 1991"},{"why":"Shows that planetary signals arise from caustic perturbations, forming the theoretical basis for the anomaly interpretation.","marker":"Gould & Loeb 1992"},{"why":"Review of caustic shapes and planetary lensing configurations used to relate the observed anomalies to the resonant caustic geometry.","marker":"Gaudi 2012"},{"why":"Origin of the inner-outer degeneracy that frames the KMT-2024-BLG-1209 solution ambiguity.","marker":"Gaudi & Gould 1997"},{"why":"Supplies the $\\langle s\\rangle = s^{\\dagger}$ relation used to confirm the inner-outer degeneracy in KMT-2024-BLG-1209.","marker":"Hwang et al. 2022"},{"why":"Together with Hwang et al. 2022, provides the degeneracy relation and context for interpreting the planetary mass-ratio distribution.","marker":"Gould et al. 2022"},{"why":"Adopted lens mass function used in the Bayesian estimation of physical lens parameters.","marker":"Jung et al. 2021"},{"why":"Adopted Galactic model used in the Bayesian estimation of lens mass and distance.","marker":"Jung et al. 2022"},{"why":"Provides the Roman RGES survey cadence and yield predictions that underpin the claim that the new degeneracy will be resolved by space-based observations.","marker":"Penny et al. 2019"}],"fun_headline_variants":["Weak caustic-crossing signals net two giant planets","Subtle lensing anomalies reveal giant planets in the disk","Two gas giants found via weak microlensing signals","Giant planets lurk behind faint caustic crossings","Chance degeneracy in one planet find may be settled by Roman"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"For MOA-2022-BLG-091 the planetary interpretation hangs on a handful of photometric points - one KMTA point and four KMTC points - being genuine caustic crossings; if those points are corrupted by systematics, the planetary parameters and the new degeneracy would not survive.","fun_headline_variants_meta":{"raw":{"variants":["Weak caustic-crossing signals net two giant planets","Subtle lensing anomalies reveal giant planets in the disk","Two gas giants found via weak microlensing signals","Giant planets lurk behind faint caustic crossings","Chance degeneracy in one planet find may be settled by Roman"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001219,"raw_usage":{"total_tokens":5103,"prompt_tokens":1123,"completion_tokens":3980,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":739,"completion_tokens_details":{"reasoning_tokens":3898}},"tokens_in":739,"tokens_out":3980,"duration_ms":28282,"temperature":1.0,"reasoning_tokens":3898,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:55:49.032436+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A re-reduction of the MOA-2022-BLG-091 images with an independent photometric pipeline, or any archival coverage of the same anomaly region, that shows the single deviant KMTA point and the four KMTC points to be noise or artifacts would eliminate the claimed planetary interpretation; likewise, a future high-cadence observation of a similar weak-caustic event whose spikes are resolved as non-caustic would challenge the identification.","supporting_citations":[{"cited_title":"1991, ApJ, 374, 37 Mróz, P ., Ryu, Y .-H., Skowron, J., et al","cited_arxiv_id":null,"evidence_quote":"Establishes gravitational microlensing as a method to detect planets, providing the foundation for interpreting the events."},{"cited_title":"S, 2012, ARA&A, 50m 411","cited_arxiv_id":null,"evidence_quote":"Review of caustic shapes and planetary lensing configurations used to relate the observed anomalies to the resonant caustic geometry."}],"review_version":1}