{"id":"bbcd23b6-a536-46d6-ac82-eed9c1166b92","arxiv_id":"2505.10376","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"TESS full-frame images can recover stellar rotation periods longer than 10 days, matching ZTF periods for 66% of a 272-star sample and 81% after a signal-strength cut.","lead":"This paper tests whether TESS satellite images can measure the rotation periods of cool stars that are longer than about 10 days. It reports that 66% of the test stars (81% for strong signals) match periods measured independently by the ground-based ZTF survey.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 81% match rate depends on a power threshold chosen by scanning the same sample; in-sample optimization likely inflates the headline, and the paper does not validate the threshold out of sample.","rationale":"The reader's weakest assumption was that ZTF periods are correct, which is a valid concern about the external validity of the validation. My concern is complementary and, I think, more directly load-bearing for the quantitative headline: the 81% match rate is obtained by optimizing the power threshold on the same sample used to report it. This is a well-known form of selection bias and can be settled with a simple split-half or cross-validation test. The core claim that long TESS periods are measurable is still supported by the threshold-free 66% rate and by the worked examples, including a 50-day match, so I do not think the paper should be rejected. A conditional acceptance is appropriate, with the requested revision being an out-of-sample or cross-validated estimate of the power-threshold match rate. The paper is otherwise transparent about its limitations, which supports the CONDITIONAL verdict rather than a harsher one.","tokens_in":19992,"tokens_out":10177,"duration_ms":110723,"concrete_test":"Split the 272 sources into two halves (or use 5-fold cross-validation). On the training half, repeat the threshold scan from Figure 9 and select the threshold that maximizes match rate. On the held-out half, apply that threshold and compute the match rate. If the held-out match rate is substantially below 81% (for example, below 70%), the headline number is inflated by in-sample overfitting. Also report the distribution of selected thresholds across folds; if the chosen threshold varies widely, power >= 0.02 is not a stable cut.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing weakness is the in-sample selection of the TESS LS power threshold that produces the 81% headline. In Section 4 and Figure 9, the authors scan thresholds from 0 to 0.3 in increments of 0.01 and choose 0.02 because it is the smallest threshold that increases the match rate, evaluated on the same 272-star sample. This is a posteriori optimization: the reported 81% (137/170) is the training-set match rate at a threshold selected to maximize that rate, so it is an optimistic, biased estimate rather than an unbiased prediction of what a user would obtain by applying power >= 0.02 to new TESS light curves. The scan over 31 thresholds also inflates the chance of finding a favorable cut by chance. The 66% overall match rate is threshold-free and therefore not affected, so the core feasibility claim survives, but the abstract's '81%' overstates the reliability of the power-based fidelity cut. The ZTF ground-truth assumption is a related but distinct issue; the threshold concern is more concrete and directly testable.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper asks whether TESS prime-mission FFI light curves in the northern continuous viewing zone can yield reliable stellar rotation periods longer than 10 days, including periods beyond the 27-day sector length. The authors select 272 K and M dwarfs with ZTF r-band periods longer than 10 days from six years of ZTF data, de-trend TESS Cycle 2 FFI cutouts with the unpopular pipeline, and search with a modified Lomb-Scargle periodogram that includes per-sector offset terms. They report that 179/272 (66%) of TESS periods match ZTF periods within 10%, rising to 137/170 (81%) after applying a TESS Lomb-Scargle power threshold of 0.02, and they present individual worked examples including a ~50-day match. The conclusion is that long rotation periods can be measured from TESS FFI light curves.","tokens_in":20216,"tokens_out":5544,"duration_ms":53918,"significance":"If the result holds, it is practically important: TESS all-sky data would substantially expand the sample of slow rotators useful for gyrochronology and stellar activity studies. The paper's strengths are its transparent methodology, a machine-readable table of the 272 stars and their periods, the offset-corrected periodogram formulation with a toy demonstration, and explicit discussion of failure modes such as 1:2 harmonics and the window-function alias. The core feasibility claim is supported by the threshold-free 66% recovery rate and by the worked examples, even though the 81% headline is inflated by in-sample threshold selection and the ZTF \"ground truth\" is not independently verified.","major_comments":[{"comment":"The power threshold of 0.02 is selected by scanning thresholds from 0.0 to 0.3 in steps of 0.01 on the same 272-star sample with the explicit objective of increasing the match rate; consequently, the 81% (137/170) figure is an in-sample optimization result, not an unbiased estimate of the reliability of a power>=0.02 cut for new targets. This is load-bearing because the abstract presents the 81% as the paper's headline reliability claim. Please validate the threshold with cross-validation or an independent sample, and either report the threshold-free rates (66%, or 82% when 1:2 harmonics are counted) as the primary claims or clearly label the 81% as exploratory.","section":"Section 4, Figures 9 and 10"},{"comment":"The validation treats ZTF r-band periods as the ground truth, but no independent verification of those periods is provided, and Section 4 itself states that \"it is difficult to identify the conditions under which ZTF periods are erroneously wrong\" and shows a case (TIC 353875094) where the TESS period is likely correct and the ZTF period is not. If a substantial fraction of the 272 ZTF periods are harmonics or aliases, the reported match rates measure mutual consistency rather than TESS period accuracy. Please quantify this risk, for example by examining the 32 g-band mismatches, cross-matching against literature periods or activity benchmarks, and showing that the main conclusions are robust to removing or reclassifying such sources.","section":"Sections 2.4 and 4"},{"comment":"The paper's own limitation statement in Section 5 says the match rates should not be interpreted as the fraction of sources with reliable TESS periods, but rather as the fraction recovered for sources with relatively strong periodic signals. Since the abstract presents 66% and 81% without this caveat, the paper risks overgeneralizing the rates to the general TESS sample; please move this qualification into the abstract and the discussion of Figures 6 and 10.","section":"Section 5 and Abstract"}],"minor_comments":[{"comment":"The column labeled \"dec dr3\" is described as \"Gaia DR2 Declination\" in the table notes; this appears to be a typo and should read \"Gaia DR3 Declination.\"","section":"Table 1"},{"comment":"The paragraph preceding Figure 5 contains a garbled sentence that introduces TIC 320504531 and then continues \"for TIC 198459831, the same source presented above\"; please clarify which source is actually shown in Figure 5.","section":"Section 3.3, Figure 5 text"},{"comment":"Match rates are reported without binomial uncertainties; for example, 179/272 = 65.8% with a 1-sigma binomial uncertainty of about 2.9%. Adding confidence intervals would help readers compare the rates quantitatively.","section":"Section 4, Figures 6 and 10"},{"comment":"The sentence about g-band consistency reports 240 matching sources and 32 mismatches, of which 5 had no downloadable g-band light curves; please clarify whether those 5 are included among the 32 mismatches or counted separately.","section":"Section 2.4"},{"comment":"The phrase \"sources with a with maximum TESS Lomb-Scargle power\" contains a duplicated word and should be corrected.","section":"Section 1"}],"recommendation":"major_revision","confidential_remarks":"The core feasibility claim is credible and the paper is within the journal's scope. The main issue is the overstatement of the 81% figure due to in-sample threshold selection, which is fixable with cross-validation or a de-emphasized presentation, and the ZTF ground-truth assumption needs stronger validation. I do not see grounds for rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The central result is real: with unpopular de-trending and an offset-corrected Lomb-Scargle periodogram you can recover rotation periods longer than 10 days, and even longer than a 27-day TESS sector, from prime-mission FFIs of NCVZ stars. The threshold-free recovery rate of 66% (179/272) is the number I trust. The 81% figure is the soft spot: the power threshold of 0.02 was picked by scanning thresholds on the same 272-star sample, so it is an in-sample optimization. It probably overstates what a user will get on new light curves. That should be presented as a descriptive property of the high-power subset or validated out-of-sample, not as a predictive fidelity cut.\n\nThe paper has real merits. It is the first classical-periodogram recovery test with ZTF ground truth in the northern CVZ, complementing Claytor et al.'s CNN work. The design-matrix trick of fitting per-sector flux offsets simultaneously with the sinusoid is clean and clearly explained via the least-squares formulation; the toy example in Figure 4 makes the point. The worked cases, including a 50-day recovery (TIC 165552443), are convincing, and the authors are unusually honest about limitations: they flag the strong-signal selection bias, they flag the ZTF ground-truth assumption, and they even show a case (TIC 353875094) where ZTF appears to be the wrong one. That is good practice.\n\nThe soft spots are moderate. Besides the in-sample threshold, the ZTF periods are the only reference and are not independently verified; the paper acknowledges this but cannot fully escape it. Match rates have no uncertainties, which matters because N=272 is small. There is no code release, though the inputs are public. Minor: Figure 5's text initially names TIC 320504531 then switches to 198459831; a leftover. Also the 43 sources on the 1:2 harmonic line are counted as non-matches, but the paper reasonably notes that if you count them as aliases the upper-limit match rate is 82% (222/272).\n\nWho is this for? Anyone building TESS rotation catalogs or doing gyrochronology with K/M dwarfs. It is an incremental capability paper, not a paradigm shift, but it is exactly the kind of careful calibration work the field needs before trusting large TESS samples of slow rotators. I would send it to peer review and expect a useful paper after the threshold discussion is reframed and the match-rate uncertainties are added.","headline":"The threshold-free 66% recovery rate is the real result; the 81% with a power cut is tuned in-sample and should be read as optimistic.","tokens_in":20781,"tokens_out":2320,"would_cite":true,"duration_ms":21568,"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":"TESS can recover stellar rotation periods longer than the 27-day sector length from full-frame image light curves, matching ground-truth ZTF periods for 66% of a 272-star sample.","keywords":["stellar rotation","TESS","ZTF","Lomb-Scargle periodogram","continuous viewing zone","M dwarfs","K dwarfs","FFI light curves"],"falsifier":"A reader could settle this by injecting synthetic sinusoids with known periods between 15 and 60 days into real TESS FFI cutouts from the northern continuous viewing zone, adding realistic scattered-light systematics, and running the same de-trending and offset-corrected period search; if most injected periods beyond 27 days are not recovered within 10%, the central claim would be undercut.","tokens_in":19806,"feed_emoji":"🛰️","tokens_out":7262,"duration_ms":65174,"temperature":0.7,"pith_summary":"This paper argues that TESS full-frame image light curves can yield reliable stellar rotation periods longer than 10 days—even beyond the 27-day span of a single TESS sector—for K and M dwarfs in the northern continuous viewing zone. Using six years of ZTF r-band photometry as a reference set of 272 slow rotators, the authors report that 179 of 272 TESS periods (66%) agree with ZTF to within 10%, and that restricting to sources with TESS Lomb-Scargle power above 0.02 raises the agreement to 81% (137 of 170). If correct, the result opens TESS's all-sky coverage to the study of slow rotators, the stars most relevant to gyrochronology and magnetic-activity relations.","feed_headline":"TESS can measure star spins longer than its 27-day sector","feed_subtitle":"66% of TESS periods match ground truth from ZTF, rising to 81% with a power cut, opening slow rotators to survey.","key_machinery":"The carrying mechanism is an offset-corrected Lomb-Scargle periodogram: the standard sinusoidal design matrix is augmented with one boxcar column per TESS sector, so the flux offset in each sector is fitted at the same time as the sinusoid rather than subtracted in advance. This removes the artificial boundary jumps that would otherwise suppress signals longer than a sector. The other half of the machinery is the unpopular de-trending step, an L2-regularized linear regression that models each target pixel's systematics using flux from distant pixels in the same FFI cutout, chosen to retain astrophysical variability while removing scattered-light trends.","core_discovery":"On the paper's own terms, the discovery is that slow stellar rotation is recoverable from TESS prime-mission FFI light curves once each sector is de-trended with the unpopular pipeline and the period search fits per-sector flux offsets jointly with a sinusoid. On a reference sample of 272 K and M dwarfs with ZTF periods longer than 10 days, the TESS period matches the ZTF period within 10% for 66% of the sample. About 43 of the 93 non-matches are TESS periods at half the ZTF value, a known Lomb-Scargle harmonic failure; if those are counted as matches the upper-limit agreement is 82%. Applying a TESS Lomb-Scargle maximum-power threshold of 0.02 leaves 170 sources, of which 137 (81%) match, and examples such as TIC 165552443 show recovery of a ~50-day period, almost twice the sector length.","pith_inferences":["Because the paper itself presents a source where TESS's shorter period is likely correct and ZTF's longer period is wrong, the true TESS recovery rate for slow rotators may be higher than 66%.","The sector-offset design-matrix trick is a general device: any multi-epoch photometric survey with per-epoch flux offsets could adopt the same simultaneous-fit periodogram for long-period searches.","A shape-aware or harmonic-aware prior could rescue many of the 1:2 mismatches without a power cut, since those cases retain genuine periodicity at the fundamental frequency.","If the power threshold selects against incompletely removed systematics, de-trending hyperparameter choices will directly control catalog completeness at long periods, a testable prediction for the upcoming catalog."],"forward_implications":["TESS continuous-viewing-zone data can support systematic searches for slow rotators instead of being limited to periods under the 13.7-day orbital timescale.","Catalog users can apply a Lomb-Scargle power threshold around 0.02 to trade sample size for period fidelity, roughly 81% ZTF agreement versus 66%.","Periods up to at least 50 days, nearly double a 27-day sector, are measurable, as illustrated by the ~49-day recovery for TIC 165552443.","Many apparent failures are harmonic aliases rather than wrong detections, so counting 1:2 matches raises the upper-limit agreement to 82% (94% after the power cut)."],"supporting_citations":[{"why":"Defines ZTF, the survey whose r-band light curves supply the ground-truth rotation periods.","marker":"Bellm et al. 2019"},{"why":"Defines TESS and the sector/CVZ geometry that set the 27-day baseline and the target selection.","marker":"Ricker et al. 2015"},{"why":"Provides the least-squares interpretation of the Lomb-Scargle periodogram and the harmonic and window-function failure modes used to interpret mismatches.","marker":"VanderPlas 2018"},{"why":"Supplies the unpopular de-trending pipeline used to remove per-sector systematics while preserving stellar variability.","marker":"Hattori et al. 2022"},{"why":"Provides the TESS Candidate Target List from which the K and M dwarf targets in the NCVZ are selected.","marker":"Stassun et al. 2018, 2019"},{"why":"Kepler rotation catalog used as a sanity check for the shape of the ZTF period distribution.","marker":"McQuillan et al. 2014"},{"why":"A complementary TESS long-period study in the southern CVZ whose CNN-based methodology this flux-space approach is compared with.","marker":"Claytor et al. 2024"}],"fun_headline_variants":["TESS cracks slow rotator problem: 66% match with ZTF","TESS extends stellar rotation periods beyond 27-day sector","TESS measures >10-day star spins: 81% match after cut","TESS slow rotators: ZTF-confirmed periods beyond orbit limit"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The comparison treats the ZTF r-band periods as the true rotation periods, so if many of those reference periods are harmonics, aliases, or non-rotation signals, the reported match rates would not measure how well TESS recovers real rotation periods.","fun_headline_variants_meta":{"raw":{"variants":["TESS cracks slow rotator problem: 66% match with ZTF","TESS extends stellar rotation periods beyond 27-day sector","TESS measures >10-day star spins: 81% match after cut","TESS slow rotators: ZTF-confirmed periods beyond orbit limit"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000745,"raw_usage":{"total_tokens":3383,"prompt_tokens":1066,"completion_tokens":2317,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":682,"completion_tokens_details":{"reasoning_tokens":2239}},"tokens_in":682,"tokens_out":2317,"duration_ms":18449,"temperature":1.0,"reasoning_tokens":2239,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T21:09:57.417384+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A reader could settle this by injecting synthetic sinusoids with known periods between 15 and 60 days into real TESS FFI cutouts from the northern continuous viewing zone, adding realistic scattered-light systematics, and running the same de-trending and offset-corrected period search; if most injected periods beyond 27 days are not recovered within 10%, the central claim would be undercut.","supporting_citations":[],"review_version":1}