{
 "dataset": {
  "archive": "figshare",
  "id": "24639285",
  "label": "FIGSHARE:24639285"
 },
 "page": "datasets/figshare/24639285/",
 "generated": "2026-11-05T06:30:02+00:00",
 "caution": "A failure to reproduce is not a finding that the paper is wrong. It may be a failure of the attempt.",
 "outcome_counts": {
  "reproduced": 12,
  "partially_reproduced": 3,
  "not_reproduced": 0,
  "blocked": 1
 },
 "outcome_counts_with_outside_data": {
  "reproduced": 0,
  "partially_reproduced": 0,
  "not_reproduced": 0,
  "blocked": 0
 },
 "outcome_counts_elsewhere": {
  "reproduced": 0,
  "partially_reproduced": 0,
  "not_reproduced": 0,
  "blocked": 0
 },
 "outcome_counts_links": {
  "reproduced": 1,
  "partially_reproduced": 0,
  "not_reproduced": 0,
  "blocked": 0
 },
 "attempts": [
  {
   "id": "figshare-24639285-tarduno-2024-01",
   "dataset": {
    "archive": "figshare",
    "id": "24639285",
    "version": "1",
    "doi": "10.6084/m9.figshare.24639285.v1"
   },
   "key": "figshare:24639285",
   "label": "FIGSHARE:24639285",
   "dataset_version": "1",
   "paper": "10.1038/s41597-024-03626-0",
   "paper_title": null,
   "date": "2026-10-27",
   "kind": "agent",
   "accountable": "CatalystNeuro",
   "agent": {
    "harness": "Claude Code",
    "model": "claude-opus-5-5",
    "protocol": "pilot-12"
   },
   "runner": {
    "platform": "GitHub Actions",
    "url": "https://github.com/catalystneuro/encore/actions/runs/40572290491"
   },
   "tier": "agent_verified",
   "deprecated": "",
   "deprecated_on": null,
   "entry": "figshare-24639285-tarduno-2024",
   "entry_title": "Replicate Apollo sample magnetizations (Tarduno, Cottrell and Zhou 2024)",
   "description": "Tests the findings of Tarduno, Cottrell and Zhou (2024) on Figshare item 24639285, which holds the thermal paleointensity measurements of 22 Apollo 64455 glass subsamples that failed the selection criteria and the replicate measurements of twelve single silicate crystals. It recomputes the fit statistics of Table 2, the magnetization thresholds of Table 1, the consistency of replicate directions and the alteration checks.",
   "notebooks": [
    {
     "path": "reproduce_tarduno_2024.ipynb",
     "title": "Reproducing Tarduno, Cottrell and Zhou (2024) from Figshare 24639285",
     "executed": "notebooks/figshare-24639285-tarduno-2024/reproduce_tarduno_2024.html"
    }
   ],
   "outcomes": [
    {
     "claim": "glass-specimen-set",
     "statement": "The deposit holds the measurements of the 22 Apollo 64455 glass subsamples that failed the paleointensity selection criteria, named as in Table 2, and replicate measurements for the twelve single crystals of Table 1.",
     "source": "Background & Summary, Glass, paragraph 2; Methods, paragraph 2; Table 2; Data Records",
     "quote": "Plots for all the data not meeting paleointensity criteria1 (n=22) are provided26 with photomicrographs of specimens that were measured",
     "statistics": [],
     "kind": "link",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "All 22 glass specimens of Table 2 are sheets of 64455lunar_cryodata.xlsx (Table 2's s11 is sheet ss11; sheet g2 labels its rows 64455g2), and all 12 crystals of Table 1 have a workbook, each with steps, moments, declinations and inclinations. No other specimen is in the deposit. The 10 glass specimens with AF steps (ss19, ss22, ss27, ss34, ss38, ss41, ss43, ss45, ss46, ss48) are exactly those with the second mark in Table 2, and the other 12 have none, as the claims file's reading of the marks has it. 71055g1 and 71055g3 have one NRM measurement each, the other crystals two or three. Four glass sheets (g2, ss8, ss39, ss48) hold scaled copies of one measurement in place of replicates (see findings).",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "glass-success-rate",
     "statement": "Of the Apollo 64455 glass subsamples given thermal paleointensity experiments, 12% passed the selection criteria (3 of 25, 13 run without and 12 with AF pre-treatment).",
     "source": "Background & Summary, paragraph 3; Methods, paragraph 2",
     "quote": "In the thermal studies1 of Apollo 64455 (Methods), the success rate was 12%",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "22 failed specimens in the deposit give 3 / 25 = 12.0%. 12 have no AF steps and 10 do; with the Methods' one and two successful specimens this gives 13 and 12.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "table2-fit-statistics",
     "statement": "For the nine failed glass specimens with a line fit (ss19, ss43, ss22, ss41, ss6, ss9, ss10, ss30, ss34), the number of points, temperature range, MAD, R2 and pTRM check differences are as in Table 2, which lists the criteria each failed.",
     "source": "Methods, Paleointensity Selection criteria, paragraph 2; Table 2",
     "quote": "Reasons for de-selection are provided in Table 2",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "partially_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "With the mean vector of each step's repeated measurements, n matches Table 2 for all nine specimens, and 23 of 29 values match under one of the named readings (79%, the criterion asks 80% for reproduced and 50 to 80% for partially reproduced), MAD 8 of 9, R2 6 of 9, pTRM checks 9 of 11. Measured values are MAD under reading (a), a free line, and R2, in the claims file's order; the last value is the fraction of the 29 values that match. MAD reading (a) alone gives all 8 matching MADs (median difference 0.2 deg, six within 0.3 deg); readings (b) and (c) give 2 and 0. Outside tolerance under every reading are ss10's MAD (33.3 against 42.9 deg) and R2 (0.23 against 0.15), ss9's R2 (0.89 against 0.74), ss34's R2 (0.03 against 0.06, tolerance 0.006), and the pTRM checks of ss19 at 250 C (23.7 to 105.9% against 30%) and ss30 at 450 C (8.2 to 43.8% against 10%). The matching pTRM checks need different readings for different specimens; no single reading matches more than 5 of 11. With the first measurement of each step, 11 of 29 values match (38%), which would not reproduce the claim. Against the stated thresholds, reported and not judged, ss30 and ss34 have R2 below 0.9 and n of 4 or more, as their single failure \"R2\" says; ss6 has R2 0.89 here, so its listed R2 failure follows here. Counting ss6's R2 as a miss, since its 250.1 step averages two groups of directions, gives 22 of 29 (76%). With one pTRM reading held fixed the score is 17, 16, 19 and 14 of 29 (readings a_i, a_ii, b_i, b_ii; 48 to 66%), partial or lower under every fixed reading. None of the nine is a synthetic sheet. Within a step, ss10's field-off steps and its 100, 175 and 250 C field-on steps scatter by 20 to 113 deg (resultant over n 0.79 to 0.98), and ss9's 360 and 550 C steps by 57 and 30 deg, so how these steps are combined moves the MAD and R2 that miss. The deposit's Supplementary Figure 1, read from its vector drawing, plots TRM gained from the mean vectors and NRM as the mean of the rows' moments for 12 of its 14 non-synthetic specimens (median difference 0.1% or less; ss27 and ss34 fit no reading), including ss6's averaged 250 C step, and 11 of its 12 pTRM checks at the check's TRM referred to the field-off step before it (reference i). With NRM read that way, R2 matches for 7 of 9 (ss10 0.152 against 0.15; ss9 0.916 and ss34 0.032 still miss) and 24 of 29 values match under any reading (83%); holding pTRM reference (i), as the figure plots the checks, gives 8 of 11 pTRM checks and 23 of 29 (79%). The outcome is kept, as the figure's NRM reading crosses the 80% bar only together with a pTRM reference the same figure does not use.",
     "deviations": [
      "The paper does not say how the repeated measurements of a step (2 to 19 per step) enter Table 2. Assumed here that they are averaged as vectors, the usual treatment of repeated measurements and the reading under which the MADs of six specimens come within 0.3 deg of Table 2. The first measurement of each step is shown as the other reading and gives a different outcome. The deposit's Supplementary Figure 1 supports averaging; it takes NRM as the mean of the rows' moments, not the moment of the mean vector, which changes R2 by at most 0.08 here and is shown as a second reading.",
      "MAD is Kirschvink's principal component measure under the claims file's three readings, and the pTRM check difference is taken on the TRM-gained axis, where Figures 1 and 2 plot the checks (legends, \"red triangles represent partial thermal remanent magnetization checks\"). The paper gives no formula for either; it names PmagPy and \"standard paleointensity definitions\" (Technical Validation).",
      "ss6's 250.1 step holds 19 rows in two groups of directions (inclination near 71 and near -13 deg); all are averaged, as for every step. Ten other steps of six glass sheets also hold two groups of directions more than 45 deg apart (s11 360.1, 550 and 550.1, ss10 250 and 360, ss18 100, ss24 NRM and 550, ss29 175.1, ss36 450.1) and are averaged the same way."
     ],
     "assumed": true,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "table2-no-fit",
     "statement": "The other thirteen failed glass specimens (g2, ss8, s11, ss18, ss24, ss27, ss29, ss36, ss38, ss39, ss45, ss46, ss48) gave no fit, most because they did not acquire TRM in a regular fashion and some because their directions trend out of the origin.",
     "source": "Methods, Paleointensity Selection criteria, paragraph 2; Table 2",
     "quote": "We note that some specimens did not acquire TRM in a regular fashion, or showed directions that trended out of ... the origin of their respective orthogonal vector plots",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "Under reading (a), TRM gained not increasing at every step, all 11 specimens marked \"no regular TRM acquisition\" are irregular (9 of 11 under reading b, a fall of more than 15%); the same with first measurements. Without the four synthetic sheets (g2, ss8, ss48 among the 11), 8 of 8 under (a) and 6 of 8 under (b). ss36 and ss39 have field-off vectors that move away from the origin over the last step of their ranges (ss36 1.77 to 6.00, ss39 2.31 to 9.67, in 1e-9 emu). Reading (a) counts any fall; the median standard error of the mean TRM gained is 1 to 6% of the TRM gained for the non-synthetic specimens. ss29's only fall, 0.6%, is below its median standard error of 2.5%; counting it as regular leaves 10 of 11 under (a). The trend out of the origin rests on ss36 and on ss39, a synthetic sheet. The pTRM checks of these specimens, reported only, match Table 2 under some reading for 6 of 10; ss36 has no pTRM check at 450 C in its sheet although Table 2 prints 25%, and ss27 has 5 paired steps from 100 to 550 C where Table 2 gives 6.",
     "deviations": [
      "Repeated measurements averaged as in table2-fit-statistics; first measurements give the same counts."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig1-linear-components",
     "statement": "Glass subsamples ss43 and ss19, which failed the reliability criteria, show hints of linear directional components on orthogonal vector plots, from which nominal paleointensities can be calculated with the 18 uT laboratory field.",
     "source": "Background & Summary, Glass, paragraph 2; Figure 1",
     "quote": "some of these data reveal linear directional components, and nominal paleointensities can be calculated (Fig. 1)",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "With averaged replicates and MAD reading (a), ss43 has MAD 10.3 deg and R2 0.90 (Table 2 10.0 and 0.90) and ss19 MAD 30.1 deg and R2 0.96 (30.1 and 0.94). Nominal paleointensities, |slope| x 18 uT, are 9.6 uT for ss43 and 23.0 uT for ss19 (8.5 and 23.7 uT with first measurements); the figure's values were not available to compare. With first measurements ss43's R2 is 0.99, outside 0.90 plus or minus 0.09. With NRM as the mean of the rows' moments, as the deposit's Supplementary Figure 1 plots it for other specimens, R2 is 0.909 and 0.956 and the nominal paleointensity of ss43 9.2 uT.",
     "deviations": [
      "Rests on the averaging of repeated measurements assumed in table2-fit-statistics."
     ],
     "assumed": true,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig2-changing-field",
     "statement": "Glass subsamples ss41 and ss22 show field-off directions that change over a temperature interval, consistent with a changing ambient field during magnetization, and curved NRM lost against TRM gained plots.",
     "source": "Background & Summary, Glass, paragraph 2; Figure 2; Table 2",
     "quote": "Other specimens show a changing vector, consistent with a changing ambient field (Fig. 2)",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "With averaged replicates, ss41 (500 to 580 C) turns 33.3 deg between its first and last field-off directions against a line MAD of 24.1 deg, and a great circle fits with MAD 15.1 deg; ss22 (250 to 500 C) turns 107.8 deg against a line MAD of 35.4 deg, great-circle MAD 32.0 deg. Both readings hold for both specimens, also with first measurements. R2 is 0.384 and 0.378 (0.38 plus or minus 0.038); with first measurements it is 0.16 and 0.44, outside tolerance. Both readings are weak tests. With no changing field and 3 vectors, as in ss41's range, reading (a) holds for 98% of random directions and 78% of points along one line with noise, and reading (b) for 56% and 51% (median line MAD 21.1 and 16.7 deg, against ss41's 24.1); with 4 vectors, as in ss22's range, (a) holds for 95% and 66% and (b) for 56% and 51% (median line MAD 28.1 and 23.6 deg, against ss22's 35.4). With NRM as the mean of the rows' moments R2 is 0.382 and 0.370. The curved Arai plot in the statement is not in the criterion and is not judged. A circle fit to the normalized points gives curvature -0.41 for ss41 and -1.12 for ss22, against -2.59 for ss19 and 0.67 for ss43, which the paper shows as linear, so with 3 or 4 points it does not separate them; ss22's NRM does not fall monotonically over its range (1.00, 0.65, 0.82, 0.28 of the first).",
     "deviations": [
      "Rests on the averaging of repeated measurements assumed in table2-fit-statistics.",
      "The great circle is fit through the origin (Kirschvink's plane MAD), a standard choice; the paper gives no fitting method."
     ],
     "assumed": true,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "glass-directions-not-random",
     "statement": "The field-off directions of the failed glass specimens, as seen on orthogonal vector plots, are not random, implying the presence of an ambient field.",
     "source": "Background & Summary, Glass, paragraph 2",
     "quote": "The directions, as viewed on orthogonal vector plots, are not random and imply the presence of an ambient field",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "partially_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "Watson's test rejects randomness of the thermal field-off directions (one per step, 2 to 9 steps) in 8 of 22 specimens with averaged replicates and 9 of 22 with first measurements, between a quarter and a half. Within the Table 2 ranges, 8 of 22 (7 with first measurements). Without the four synthetic sheets, 7 of 18. ss38 (k 27.9, p 0.0001), ss45, ss9, ss8, ss19, ss41, s11 and ss6 reject; ss39 and ss48 have two thermal steps each.",
     "deviations": [
      "The p-value of Watson's test is computed exactly by simulation (200000 sets of uniform directions, fixed seeds), a standard form of the test the claims file names."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "glass-af-no-change",
     "statement": "For nearly all glass samples given AF pre-treatment, AF cleaning did not significantly change the NRM direction or magnitude.",
     "source": "Methods, paragraph 2",
     "quote": "For nearly all samples, AF cleaning did not significantly change the NRM direction or magnitude",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "Under both readings 8 of the 10 AF-treated specimens show no significant change; the direction after the last AF step is within 2.8 deg of the NRM and the moment within 1.1%. ss38 keeps its direction (2.4 deg) but its moment falls to 0.49 of the NRM at 20 mT, and ss48, a synthetic sheet, turns 17.5 deg and falls to 0.67. The same counts with first measurements.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "table1-max-intensity",
     "statement": "For each of the twelve single crystals, the maximum field that cannot be excluded, given the magnetometer threshold of 8e-11 emu and the moment acquired at 590 C in a 20 uT field, is as in Table 1, between 0.20 and 0.51 uT.",
     "source": "Background & Summary, Single crystal measurements, paragraph 4; Table 1; Technical Validation, Example of threshhold detection limit",
     "quote": "Considering a magnetometer threshold value (see Methods) the maximum field that cannot be excluded by the data can be calculated from the data for each crystal. These values are presented in Table 1",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "Under reading (b), M590 the moment of the single 590.1 measurement, all 24 values are within tolerance and the 12 M590 values equal Table 1 to its four digits; the max intensities run from 0.200 to 0.500 uT. Reading (a) with averaged 590.0 replicates also gives 24 of 24, and with the first replicate 23 of 24. Table 1's max intensities equal the formula on its own M590 column rounded to two places for 4 of 12 crystals, and rounded up for 10 of 12; the exceptions are 14053g1 (0.4995, printed 0.51) and 12053g2 (0.2005, printed 0.20). The equal M590 of 12021g3 and 12040g3 (6.292e-09 emu) is in the data, two 590.1 measurements with different directions.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "baseline-range",
     "statement": "The maximum fields that the single crystal data cannot exclude define a baseline of about 0.2 to 0.5 uT below which fields cannot be evaluated with these data.",
     "source": "Background & Summary, paragraph 7",
     "quote": "they nevertheless define a baseline of ~0.2 to ~0.5 \u03bcT below which fields cannot be evaluated with the present data",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "The twelve values of table1-max-intensity under reading (b) run from 0.200 uT (12040g1) to 0.500 uT (14053g1); under reading (a) from 0.203 or 0.204 to 0.501 or 0.503 uT.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "example-14053g1-threshold",
     "statement": "For crystal 14053g1, the threshold field detection limit computed from the magnetometer sensitivity (8e-14 A m2) and a TRM at 590 C of 8e-12 A m2 in a 20 uT field is 0.2 uT.",
     "source": "Technical Validation, Example of threshhold detection limit",
     "quote": "Threshold field detection = Magnetometer sensitivity (8 \u00d7 10\u221214 A m2) / TRM590o (8 \u00d7 10\u221212 A m2) \u00d7 Lab Field (20 \u03bcT) = 0.2 \u03bcT",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "partially_reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "Matches under reading (b) only, the misprint reading the criterion names. For 14053g1 the data give a TRM at 590 C of 3.203e-12 A m2 and 0.500 uT, not the example's 8e-12 and 0.2; for 12040g1 they give 8.002e-12 A m2 and 0.200 uT, the example's numbers. The values recorded are 12040g1's. The criterion calls the claim reproduced as written only when reading (a) matches, so a match under (b) alone is recorded as partially reproduced (changed from reproduced after verifier round 1).",
     "deviations": [
      "The example's crystal name is read as a misprint for 12040g1, as the criterion's reading (b) allows; as written, for 14053g1, the example does not match the data."
     ],
     "assumed": false,
     "disagreed": true,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig3-nrm-consistent",
     "statement": "For most crystals the replicate NRM measurements have a consistent direction, as shown on stereonets for 12040g1 and 14053g2.",
     "source": "Background & Summary, Single crystal measurements, paragraph 2; Figure 3a and 3d",
     "quote": "These show that for most of the crystals there is consistency of the NRM direction (Fig. 3a,d)",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "Under reading (a) the NRM replicates have a higher Fisher k than the 590.0 replicates in 10 of 12 crystals, including 12040g1 (k 981) and 14053g2 (k 206); under reading (b) Watson's test rejects randomness of the NRM replicates in 8 of 12. 71055g1 and 71055g3 have one NRM measurement and count as not consistent. With two or three replicates per crystal, k and the test rest on few directions (12040g3 NRM k 1.3, 12053g1 k 9.8).",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig3-590-inconsistent",
     "statement": "After demagnetization at 590 C in zero field, the replicate directions of the crystals are inconsistent, consistent with a null magnetization state.",
     "source": "Background & Summary, Single crystal measurements, paragraph 2; Figure 3b and 3e; Methods, Thermal measurements, paragraph 5",
     "quote": "but this consistency is not seen after demagnetization at 590 \u00b0C (Fig. 3b,e)",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "Under reading (b) Watson's test does not reject randomness of the 590.0 replicates in any of the 12 crystals (p 0.34 to 0.99, k 0.8 to 1.8); under reading (a) their k is below the NRM's in the 10 crystals with more than one NRM measurement. The 590.0 moments average 1.16e-10 to 5.68e-10 emu per crystal, and all 37 single 590.0 measurements are above the 8e-11 emu threshold.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "fig3-alteration-check-inconsistent",
     "statement": "The replicate directions after the zero-field alteration check at 590 C are also inconsistent.",
     "source": "Background & Summary, Single crystal measurements, paragraph 2; Figure 3c and 3f",
     "quote": "The directions are also inconsistent after the alteration check at 590 \u00b0C (Fig. 3c,f)",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "Watson's test does not reject randomness of the 590.3 replicates in any of the 12 crystals (reading b); their k is below the NRM's in 10 of 12 (reading a). Mean 590.3 moments are 2.14e-10 to 5.12e-10 emu.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "crystal-14053g1-400c-stable",
     "statement": "Crystal 14053g1, measured after heating to 400 C during initial tests, held a stable remanence, which was removed by treatment at 590 C.",
     "source": "Background & Summary, Single crystal measurements, paragraph 3",
     "quote": "One crystal (14053g1) was measured after heating to an intermediate temperature (400 \u00b0C) during initial tests, and found to be stable",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "The deposit holds three 400 C zero-field measurements of 14053g1, with one 400.1 and two 400.3 rows. The 400 C replicates are grouped (k 78, alpha95 14 deg, Watson p 0.0003) and the 590.0 replicates are not (k 1.0, p 0.79), so reading (a) holds. Under reading (b) the 400 C mean is 9.3 deg from the NRM mean, outside the NRM's alpha95 of 4.0 deg (two NRM measurements). The moment falls from 1.24e-09 emu (NRM) to 8.2e-10 (400 C) and 1.2e-10 (590 C).",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "no-alteration-recording-efficiency",
     "statement": "Alteration during the laser heating of the single crystals was not significant; apparent recording efficiencies far above 100% were not observed and the alteration checks did not fail.",
     "source": "Background & Summary, paragraph 5",
     "quote": "If alteration had resulted in the formation of new magnetic minerals, apparent recording efficiencies \u226b 100% should have be observed (but these were not), and alteration checks should fail",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "reproduced",
     "blocked_by": null,
     "notebook": "reproduce_tarduno_2024.ipynb",
     "notes": "Under reading (b) every crystal passes; the 40 uT TRM is 0.77 to 1.17 times twice the 20 uT TRM (0.79 to 1.14 with averaged 590.0 and 590.3 replicates), against the cutoff of 2. Under reading (a) 11 of 12 pass with first replicates and 10 of 12 with means; 14053g1's 590.3 moment exceeds its 590.0 moment by 9.8e-11 to 1.21e-10 emu, and 14053g3's by 9.9e-11 with means. The paper's words are recording efficiencies far above 100%; the largest here is 117% of linear recording. The 590.0 replicate moments of one crystal scatter by 0.6e-11 to 6.3e-11 emu (sd), below reading (a)'s allowance of 8e-11 emu; 14053g1's 590.3 excess is about 6 to 8 times its 590.0 sd of 1.6e-11 emu, so its failure under reading (a) is not replicate noise. The 590.0 moments themselves, 1.0e-10 to 6.0e-10 emu with random directions, are 1.3 to 7.5 times the threshold.",
     "deviations": [
      "The cutoff of 2 on the ratio is the claims file's, not the paper's; the largest ratio, 1.17, is far from it. The cutoff fails a crystal only when the 40 uT TRM exceeds four times the 20 uT TRM, although the criterion's text says \"about twice\". Without the factor of two, TRM40/TRM20 is 1.54 to 2.35 and three crystals are above 2 (12053g1 2.35, 14053g3 2.08, 12053g2 2.06), so a cutoff of 2 on that ratio would fail reading (b) for them and no reading would pass every crystal. This is asked of a reader in questions/figshare-24639285.yaml; the claim is tested as the file has it."
     ],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    },
    {
     "claim": "interpretation-impact-no-dynamo",
     "statement": "Together with ref. 1, the data are consistent with impact plasma magnetization of the 64455 glass and with the absence of a long-lived lunar core dynamo recorded in the 3.9 to 3.2 Ga crystals.",
     "source": "Abstract; Background & Summary, paragraphs 2 and 7",
     "quote": "These 64455 field strengths1 match those predicted by the charge separation mechanism",
     "statistics": [],
     "kind": "finding",
     "inferred": "",
     "outcome": "blocked",
     "blocked_by": "other",
     "notebook": null,
     "notes": "The claims file says this interpretation is not tested, as it rests on ref. 1 and on models of impact plasmas, and is judged from four other claims. Their outcomes here are fig2-changing-field reproduced, glass-directions-not-random partially reproduced, fig3-590-inconsistent reproduced and table1-max-intensity reproduced.",
     "deviations": [],
     "assumed": false,
     "disagreed": false,
     "data": "dataset",
     "where": null,
     "proposed": null
    }
   ],
   "findings": [
    {
     "kind": "other",
     "statement": "In four glass sheets of 64455lunar_cryodata.xlsx (g2, ss8, ss39, ss48) the rows given as repeated measurements of each step are one row multiplied by factors between 0.95 and 1.03. In g2, ss8 and ss39 the same factor multiplies the moment, the declination and the inclination of a row; in ss48 each column has its own factors. Every step of these sheets is built this way and no step of the other 18 sheets or the 12 crystal files is. These rows are not measurements, and the scaling of a declination has no physical meaning (one ss8 declination becomes 361.7 deg). The deposit does not say how they were made.",
     "found": null,
     "status": null
    },
    {
     "kind": "format_error",
     "statement": "Values outside their ranges or format in 64455lunar_cryodata.xlsx; sheet ss19, 10 mT step, an inclination of -136 deg (likely -13.6); sheet ss8, NRM, a declination of 361.692 deg; sheet ss18, row 41, a 250 C row with no moment, declination or inclination; sheet ss48 has an extra column CSD before the moment, so its columns differ from the Data Records' description.",
     "found": null,
     "status": null
    },
    {
     "kind": "metadata_insufficient",
     "statement": "The glass sheets hold 2 to 19 repeated measurements per step, and neither the deposit nor the paper says in words how they were combined for Table 2 and Figures 1 and 2. The mean vector reproduces most of Table 2; the first measurement does not. The deposit's Supplementary Figure 1 plots means (see the finding on it). Eleven steps of seven glass sheets hold two groups of directions more than 45 deg apart, nine of them inside the specimen's Table 2 range; sheet ss6's 250.1 step holds 19 rows in two groups (inclination near 71 deg and near -13 deg). NRM is labelled both 0 and 0.0 in ss27 and ss38.",
     "found": null,
     "status": null
    },
    {
     "kind": "methods_unavailable",
     "statement": "Questions for the paper's authors. (1) Were the repeated measurements of each step averaged as vectors for Table 2 and Figures 1 and 2, or was one measurement used? (2) Is MAD a free (unanchored) line fit to the field-off vectors, without the three-point smoothing? A free fit to averaged replicates gives 8 of the 9 printed MADs. (3) How is the pTRM check percentage defined (denominator, and which field-off step the check is referred to)? No single definition tried here gives more than 5 of the 11 printed values. (4) Which measure supports \"consistent\" and \"inconsistent\" for the crystal replicates? (5) Were Table 1's max intensities rounded up? Ten of twelve equal the formula rounded up, not rounded. (6) Is Table 1's M590 the moment of the 590.1 measurement alone, or the vector difference of the 590.1 and 590.0 steps that the Methods define as the TRM? Supplementary Figure 1 answers part of (1) to (3) for the specimens it plots (mean vectors for TRM gained, mean of the moments for NRM, checks referred to the field-off step before them); Table 2's numbers for ss19, ss43, ss22 and ss41 are in the paper's Figures 1 and 2.",
     "found": null,
     "status": null
    },
    {
     "kind": "paper_inconsistency",
     "statement": "Table 1's M590 equals the moment of the single 590.1 (field-on) measurement to its four digits for all 12 crystals, while the Methods define the TRM as the vector subtraction of the field-on and field-off steps (Methods, Thermal measurements, paragraph 3). The vector difference with the 590.0 replicates differs from Table 1 by 0.2 to 4.2% with their mean and by up to 11% with the first replicate.",
     "found": null,
     "status": null
    },
    {
     "kind": "other",
     "statement": "The deposit's \"SupplFig1 - other Thelliers March 2024.pdf\" plots, for 12 of its 14 specimens that are not synthetic sheets, TRM gained from the mean vectors of the repeated rows and NRM as the mean of their moments, to 0.1% or better. For the four sheets whose rows are scaled copies (g2, ss8, ss39, ss48) its points differ from every combination of the deposited rows by 0.5% or more (median), so these sheets' deposited rows are not the values plotted. One ss27 point is labelled 210 C, a step with no field-on measurement in the sheet.",
     "found": null,
     "status": null
    },
    {
     "kind": "paper_inconsistency",
     "statement": "The Technical Validation example of the threshold for 14053g1 uses a TRM of 8e-12 A m2 and gives 0.2 uT; the data and Table 1 give 3.2e-12 A m2 and 0.50 to 0.51 uT for 14053g1, and the example's numbers are those of 12040g1. The text before Figure 3 names 12041g1, which is not a specimen of the deposit; the Figure 3 legend has 12040g1.",
     "found": null,
     "status": null
    },
    {
     "kind": "paper_inconsistency",
     "statement": "Table 2 gives ss36 a pTRM check at 450 C (25%), but its sheet has no pTRM check step, and gives ss27 6 points from 100 to 550 C where its sheet has 5 paired steps (100, 305, 400, 500, 550 C).",
     "found": null,
     "status": null
    },
    {
     "kind": "metadata_insufficient",
     "statement": "The laboratory field of the glass experiments differs between sheets, 18 uT (10 sheets), 15 uT (11) and 14.4 uT (g2), given in a footnote of each sheet. The paper states only the 18 uT of Figure 1.",
     "found": null,
     "status": null
    },
    {
     "kind": "other",
     "statement": "The crystals' moments after zero-field heating to 590 C (37 measurements, 1.0e-10 to 6.0e-10 emu per measurement) are all above the 8e-11 emu magnetometer threshold the paper uses for its bound on paleointensity. The directions of these measurements are random by Watson's test in every crystal.",
     "found": null,
     "status": null
    },
    {
     "kind": "metadata_error",
     "statement": "The item's description lists files named \"SI Figure 1 - Other Thelliers n=18.pdf\" and \"SI Figure 2 - All Stereonets copy.pdf\", while the files are \"SupplFig1 - other Thelliers March 2024.pdf\" and \"SuppFig2 - All Stereonet.pdf\". The item's title differs from the paper's.",
     "found": null,
     "status": null
    }
   ],
   "reconciliation": null,
   "verifications": [
    {
     "date": "2026-10-27",
     "kind": "agent",
     "by": "Independent agent run by CatalystNeuro (same harness, separate context, did not write the notebook)",
     "reviewed": "Round 1. The first executed notebook with the conclusion cells and the summary removed, the helper module and the claims file, with the paper. The verifier checked the archive through the Figshare API and tested the helper functions on synthetic inputs. Its report gave its own outcome for each claim; 16 of 17 agree with the record, and example-14053g1-threshold does not.",
     "agreed": 16,
     "of": 17,
     "disagreements": [
      {
       "claim": "example-14053g1-threshold",
       "outcome": "partially_reproduced",
       "reason": "For 14053g1 the TRM is 3.2e-12 A m2 and the threshold 0.50 uT, so the example as written fails. For 12040g1 the numbers are 8.0e-12 A m2 and 0.20 uT, so it matches only if the crystal name is a misprint."
      }
     ]
    },
    {
     "date": "2026-10-27",
     "kind": "agent",
     "by": "Independent agent run by CatalystNeuro (same harness, separate context, did not write the notebook)",
     "reviewed": "Round 2. The full executed notebook after the round-1 revision, with its conclusions and summary, the helper module, the README, the attempt record and the claims file, with the paper. The verifier checked every result sentence, the summary, the README and the record notes against the printed numbers. It agreed with all 17 outcomes.",
     "agreed": 17,
     "of": 17,
     "disagreements": []
    }
   ]
  }
 ],
 "entries": [
  {
   "name": "figshare-24639285-tarduno-2024",
   "type": "reproduction",
   "title": "Replicate Apollo sample magnetizations (Tarduno, Cottrell and Zhou 2024)",
   "description": "Tests the findings of Tarduno, Cottrell and Zhou (2024) on Figshare item 24639285, which holds the thermal paleointensity measurements of 22 Apollo 64455 glass subsamples that failed the selection criteria and the replicate measurements of twelve single silicate crystals. It recomputes the fit statistics of Table 2, the magnetization thresholds of Table 1, the consistency of replicate directions and the alteration checks.",
   "datasets": [
    "figshare:24639285"
   ],
   "papers": [
    "10.1038/s41597-024-03626-0"
   ],
   "derived_from": [],
   "notebooks": [
    {
     "path": "reproduce_tarduno_2024.ipynb",
     "title": "Reproducing Tarduno, Cottrell and Zhou (2024) from Figshare 24639285",
     "executed": "notebooks/figshare-24639285-tarduno-2024/reproduce_tarduno_2024.html"
    }
   ]
  }
 ]
}