---
slug: counted-twice
passed: 2026-11-11
by: claude-gifted-feynman-8p7zpl, one agent on oversight/claims-pass.md over all 120 entries of assay/sources/counted-twice.json at b6d9f093ea plus 10 claims it added (the title, the plain card, the image text, the nearby and relates notes, the juniper's ring 776, the verifier's header); the author re-read the source of every finding before any edit
checked: 130
confirmed: 122
wrong: 5
unverifiable: 0
observed: 3
stale: 0
---
# Claims pass: Counted Twice

Written 2026-11-11, before the layer was published. Live page not checked: it is not deployed yet
(https://artwaste.land/strata/counted-twice/ and /checks/verify-counted-twice.mjs 404; /strata/set-by-another-clock/
and /strata/every-tree-wrote-the-same-bad-year/ 200 as controls). Reader check on local copies of b6d9f093ea (git show
into an empty directory: index.html, data.json, engine.mjs, app.mjs, verify-counted-twice.mjs), run with AW_DIR set to
that directory: exit 0, 285 passed, 0 failed; with --mutate exit 0, 295 passed, ten plants caught. In the repository
build-data.mjs --check exit 0, and the verifier with --online exit 0, 404 passed, the six kept data files
byte-identical to what NOAA, Copernicus and Europe PMC serve. Every quotation was tested against fresh copies of the
sources (the kept Sinnl and Adolphi-Muscheler texts are clipped where wide lines were cut, so the agent made its own
extraction of the publishers' PDFs); every figure was recomputed with the agent's own spreadsheet reader; the ten DOIs
were checked at Crossref; eleven sentence and headline plants were run in scratch copies. Robots policy before every
fetch: intcal.org refused (its robots.txt answered 503), Wiley reference only (its full text answered 403), everything
else allowed.

## Claims
- CONFIRMED | In 774 a storm of particles, most likely from the Sun, put a burst of carbon-14 into the air and of beryllium-10 into the snow. | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4639793/fullTextXML | "these peaks were most likely produced by extreme solar events"; the hedge is kept in the dek, the plain summary and the body
- CONFIRMED | GICC05, the count of annual layers published in 2006 | https://api.crossref.org/works/10.1029/2005JD006921 | Vinther et al., JGR 111, D13102, 2006; NOAA header "Published_Date_or_Year: 2006"
- CONFIRMED | put that layer in 768, with a maximum counting error of one year | https://www.ncei.noaa.gov/pub/data/paleo/icecore/greenland/summit/ngrip/vinther2006-gicc05-holocene-ngrip-20yr-noaa.txt | NGRIP1 at 249.124 m gives 1231.80 b2k, the year 768, MCE 1 (rows 1220 and 1240 both 1), recomputed
- CONFIRMED | The same ice was counted again for a chronology published in 2022, tied only to eruptions in 1783 and 1362 | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | "no previous knowledge about age was initially used to fine-tune the timescale, except for Laki and Oræfajökull" (§3.3) and Table 3 note a
- CONFIRMED | and the layer came out in 774 | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | Table 3 "774 ± 4.1 CE"; sheet 3 average 1225.352 b2k gives 774
- CONFIRMED | the gap opens where the old count had trusted two eruptions it had misnamed | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | Conclusions, "mismatch of the Hekla (1104 CE) and the Vesuvius (79 CE) eruptions"
- CONFIRMED | comes from what Florian Mekhaldi and colleagues call | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4639793/fullTextXML | verbatim
- CONFIRMED | In 2012 Fusa Miyake and colleagues measured it in the annual rings of Japanese cedar, one or two rings at a time | https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=EXT_ID:22699615%20OR%20EXT_ID:26153860%20OR%20EXT_ID:23612289&format=json&resultType=core | "annual rings of Japanese cedar ... with 1- and 2-year resolution"
- CONFIRMED | and reported | https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=EXT_ID:22699615%20OR%20EXT_ID:26153860%20OR%20EXT_ID:23612289&format=json&resultType=core | verbatim in the 2012 abstract
- CONFIRMED | Fusa Miyake | https://api.crossref.org/works/10.1038/nature11123 | "Fusa Miyake" for doi 10.1038/nature11123
- CONFIRMED | In 2018 U. Büntgen and colleagues measured | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6127282/fullTextXML | verbatim in the abstract
- CONFIRMED | and the two spikes, they wrote, | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6127282/fullTextXML | verbatim in the abstract
- CONFIRMED | The likeliest cause, from beryllium and chlorine in polar ice, is the Sun | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4639793/fullTextXML | verbatim; 36Cl used "in pair with 10Be"
- CONFIRMED | as another layer here begins | https://artwaste.land/strata/every-tree-wrote-the-same-bad-year/ | the live page opens with it
- CONFIRMED | Every Tree Wrote the Same Bad Year | https://artwaste.land/strata/every-tree-wrote-the-same-bad-year/ | the live page's second sentence
- CONFIRMED | a juniper from the Sierra Nevada of California and an oak from Dorestad in the Netherlands | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/supplementaryFiles | MOESM1 §S6 "Juniper ... Sierra Nevada Mountains, California" and "Oak ... Dorestad Haven, The Netherlands" (PDF sha256 d7c8c166... as recorded)
- CONFIRMED | published by A. Scifo and colleagues at Groningen in 2019 | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/fullTextXML | first affiliation "University of Groningen, Centre for Isotope Research"; 2019
- CONFIRMED | Each point is one ring, with its one-sigma error. | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/supplementaryFiles | Fig. S4a and S4c captions, "1-sigma error bars"
- CONFIRMED | The juniper’s rings 762 and 763 were measured together and are drawn as one bar. | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/supplementaryFiles | B1 GrM 16805 "762-763", -18.63 ± 1.95
- CONFIRMED | In the juniper the radiocarbon rises | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/supplementaryFiles | -18.29 ± 1.47 to -3.76 ± 1.98, so +14.53 ± 2.47, recomputed
- CONFIRMED | its largest rise between any two neighbouring rings | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/supplementaryFiles | the largest step between neighbouring rings, and the largest step of any kind, in B1
- CONFIRMED | In the oak it rises | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/supplementaryFiles | +8.93 ± 3.58, then +8.44 ± 2.63 into 776, recomputed
- CONFIRMED | Ring 774 reads like the rings before it in both trees, and ring 775 does not. | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/supplementaryFiles | B1 774 at -18.29 lies within the earlier rings' -18.63 to -24.92; B2 774 at -17.56 lies within -16.32 to -20.20
- CONFIRMED | spikes like this are often drawn out, and Mekhaldi’s group found the rises in beryllium and carbon | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4639793/fullTextXML | verbatim
- CONFIRMED | Büntgen’s network puts the start of the event in | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6127282/fullTextXML | verbatim in the abstract
- CONFIRMED | Mekhaldi’s group measured it there and at the other pole | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4639793/fullTextXML | verbatim
- CONFIRMED | Some of what falls with the snow follows the seasons (sodium, for one, | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim
- CONFIRMED | GICC05, the Greenland Ice Core Chronology 2005, was made on three cores, DYE-3, GRIP and NorthGRIP | https://api.crossref.org/works/10.1029/2005JD006921 | Vinther's abstract (Crossref's copy matches NOAA's)
- CONFIRMED | B. M. Vinther and colleagues described it in 2006 | https://api.crossref.org/works/10.1029/2005JD006921 | verbatim in the publisher's abstract
- CONFIRMED | and After the verification, the main dating is transferred to all records using the volcanic reference horizons as tie points. | https://api.crossref.org/works/10.1029/2005JD006921 | verbatim in the publisher's abstract
- CONFIRMED | GICC05’s table in the NOAA archive gives, for every twentieth year, the depth at which that year begins in each core and the count’s maximum counting error there | https://www.ncei.noaa.gov/pub/data/paleo/icecore/greenland/summit/ngrip/vinther2006-gicc05-holocene-ngrip-20yr-noaa.txt | depth "at sample start" for each core, iceage_err "maximum counting error", 20-year rows
- CONFIRMED | the count’s maximum counting error there | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim (§1.4)
- CONFIRMED | an uncertain layer being one of the | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim
- CONFIRMED | At the depth of the beryllium layer that error is | https://www.ncei.noaa.gov/pub/data/paleo/icecore/greenland/summit/ngrip/vinther2006-gicc05-holocene-ngrip-20yr-noaa.txt | MCE 1 at the layer in all three cores
- CONFIRMED | Giulia Sinnl and colleagues, looking back at it in 2022, put the published figure plainly | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim
- CONFIRMED | On that count the layer is | https://www.ncei.noaa.gov/pub/data/paleo/icecore/greenland/summit/ngrip/vinther2006-gicc05-holocene-ngrip-20yr-noaa.txt | 768, recomputed
- CONFIRMED | Sinnl’s group counted the same ice again | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim in the abstract
- CONFIRMED | with automated counting of annual layers on multiple parallel proxies and manual finetuning, and almost no historical dates | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim in the abstract
- CONFIRMED | and almost no historical dates | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim; "almost no historical dates" is supported by §3.3
- CONFIRMED | Laki erupted in 1783 | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | "between June 1783 and February 1784 CE"
- CONFIRMED | one core, DYE-3, whose measurements start below Laki, was tied to Öræfajökull in 1362 instead | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | "For DYE-3, however, we tie the timescale to the Öræfajökull eruption of 1362 CE because the DYE-3 ECM measurements start below the Laki layer"
- CONFIRMED | Of the dated events in their table, | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | Table 3 note a
- CONFIRMED | The new count, GICC21, puts the beryllium layer in | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | 774
- CONFIRMED | The NorthGRIP1 core from | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | 55 to 400 m, inside both tables
- CONFIRMED | NorthGRIP1 had no record fine enough to count in, and both chronologies gave it ages carried over from the other cores | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | sheet 1 text; the last NorthGRIP1 year in sheet 2 is 1800 b2k
- CONFIRMED | both chronologies gave it ages carried over from the other cores | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | "From 1813 until 3835 b2k, NorthGRIP ages were transferred from DYE-3 and GRIP"
- CONFIRMED | Below about | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | 1800 b2k at 349.075 m (sheets 2 and 4)
- CONFIRMED | the year at the hairline between two of them is read by straight-line interpolation | public/strata/counted-twice/engine.mjs | engine.mjs `interp`; the verifier's C checks pass
- CONFIRMED | so it can be a year out either way | verify-counted-twice.mjs | the verifier's B check holds at all 14 layers (reader run)
- CONFIRMED | GICC21’s is the age its table gives the tie point, averaged over the cores | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | sheet 1 text and sheet 3's "Average Age" column
- CONFIRMED | The last row is the layer the paper calls Aniakchak, | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | Table 3 "Aniakchak, Alaska (formerly attributed to Thera, Santorini)"
- MINOR | Every layer GICC21’s table of tie points names, with the year each count gives it. | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | the 14 rows match sheet 3's 14 named rows, and every cell was recomputed (see finding 4 on Samalas). The Samalas row reads 1258 by the supplement’s rule on 741.0005 b2k, where the sheet’s year column prints 1257 and Table 3 prints 1259 ± 2; Bárðarbunga is 877 ± 3.9 in Table 3. See finding 4
- CONFIRMED | Sinnl’s group published the difference between the two counts for every year of the new one | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | sheet 5 has 3,836 rows, one per year from 0 to 3835 b2k
- CONFIRMED | Sinnl’s group built their transfer function for this | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim (§4)
- CONFIRMED | with GICC21’s own one-sigma uncertainty around it | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | sheet 5 column 2, "to be interpreted as 1 σ"
- CONFIRMED | Thick line: GICC21’s transfer function, a weighted mean over the cores of each year’s offset | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | "a weighted mean of the offset"
- CONFIRMED | Adolphi and Muscheler’s 2016 estimate against the tree-ring timescale, with its 95% range | https://www.ncei.noaa.gov/pub/data/paleo/icecore/greenland/summit/ngrip/gicc05adolphi2016.txt | columns "lower/upper 95.4% confidence" (these are bounds, not widths)
- CONFIRMED | At 775 GICC05 calls the ice | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | rows 1224 to 1226 b2k are all 6
- CONFIRMED | GICC21’s own uncertainty there is | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | 4.197, shown as 4.2
- CONFIRMED | Years by which GICC05 dates a layer older, against the year, from 1800 back to 1836 BCE. | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | 3835 b2k is 1836 BCE
- CONFIRMED | Samalas, the eruption of 1257 | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | Table 3 historical age 1257 CE
- CONFIRMED | the two counts are never more than | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | the maximum /offset/ between Laki and Samalas is exactly 3
- CONFIRMED | Then the line climbs: on average about | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | 2.80, 5.93 and 7.00, recomputed
- CONFIRMED | at the layer GICC05 had taken for Vesuvius | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | 10.0
- CONFIRMED | GICC05’s error band is | https://www.ncei.noaa.gov/pub/data/paleo/icecore/greenland/summit/ngrip/vinther2006-gicc05-holocene-ngrip-20yr-noaa.txt | at most 2 below 2000 b2k (NOAA's column itself goes back from 2 to 1 at 1540 b2k; the page reads the table correctly)
- CONFIRMED | Sinnl’s group traced the gap to two of GICC05’s pins | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim
- CONFIRMED | GICC05 had matched its cores to each other by | https://api.crossref.org/works/10.1029/2005JD006921 | verbatim
- CONFIRMED | and tied some of those layers to eruptions whose dates were known from history | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | "many well-known historical eruptions tied the chronology together"
- CONFIRMED | The first was a prominent peak attributed to the great eruption of Hekla, in Iceland, in 1104. | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | "attributed a prominent ECM peak in GICC05 to the massive Hekla eruption (1104 CE, Iceland)"
- CONFIRMED | When Coulter and colleagues looked in 2012 for Hekla’s ash near that peak, | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim; Coulter et al. 2012, JGR 117, D21303, is in Sinnl's references
- CONFIRMED | Sinnl’s group found the mark it had left on the count | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim
- CONFIRMED | The second was the deepest pin, the eruption of Vesuvius in 79. | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim; "deepest pin" fits "considered exact for the part younger than" that peak
- CONFIRMED | And the count agreed with it | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim; "The authors" are "the authors of GICC05"
- CONFIRMED | years of history | verify-counted-twice.mjs | 1362 - 79 = 1283
- CONFIRMED | In each of the three cores GICC05 counts about | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | GICC05 1283.29 / 1283.18 / 1283.04 and GICC21 1273.12 / 1273.30 / 1273.04 (NorthGRIP1 / GRIP / DYE-3), recomputed
- CONFIRMED | In 2022 Gill Plunkett and colleagues analysed ash from the same acid layer | https://cp.copernicus.org/articles/18/45/2022/cp-18-45-2022.pdf | the abstract; Sinnl adds that the tephra came from NEEM-2011-S1 at the same acidity peak
- CONFIRMED | In 2022 Gill Plunkett and colleagues analysed ash from the same acid layer | https://cp.copernicus.org/articles/18/45/2022/cp-18-45-2022.pdf | verbatim in the abstract, hedge kept
- CONFIRMED | GICC21’s table calls the layer UE 88, gives its age on another ice chronology as about 88, and dates it on its own count to | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | Table 3 "∼ 88 CE^d" (NS1-2011) and "89 ± 5.4 CE"; sheet 3 gives 89
- CONFIRMED | Sinnl’s group again, on what the maximum counting error left out | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | verbatim, two sentences
- CONFIRMED | who matched the slow wander of beryllium-10 in Greenland ice to the slow wander of carbon-14 in tree rings, decade by decade | https://www.ncei.noaa.gov/pub/data/paleo/icecore/greenland/summit/ngrip/gicc05adolphi2016.txt | the NOAA description and the paper's abstract
- CONFIRMED | with no annual spikes at all | https://cp.copernicus.org/articles/12/15/2016/cp-12-15-2016.pdf | verbatim
- CONFIRMED | Their abstract | https://cp.copernicus.org/articles/12/15/2016/cp-12-15-2016.pdf | verbatim, in the abstract
- CONFIRMED | at 775 it says GICC05 is | https://www.ncei.noaa.gov/pub/data/paleo/icecore/greenland/summit/ngrip/gicc05adolphi2016.txt | at 1175 BP, -6 (95%: -11 to -1), so 6 (1 to 11)
- CONFIRMED | And at the old pin | https://cp.copernicus.org/articles/12/15/2016/cp-12-15-2016.pdf | verbatim
- CONFIRMED | which would imply an age of AD 90 ±6 for the GRIP tephra layer | https://cp.copernicus.org/articles/12/15/2016/cp-12-15-2016.pdf | verbatim; the page drops the leading "Assuming that the tree-ring chronologies are correct at this time", a premise the page states openly elsewhere
- CONFIRMED | They checked themselves against the two spikes, which their decadal data could not see | https://cp.copernicus.org/articles/12/15/2016/cp-12-15-2016.pdf | verbatim; in context, set against the spikes' -7 ± 2
- CONFIRMED | The second spike, in 993 by the trees | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6127282/fullTextXML | "boreal spring of 993"; "from AD 992 to 993" (Miyake 2013)
- CONFIRMED | is in GICC21’s table too, at | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | Table 3
- CONFIRMED | and it was Not used as tie point across ice cores | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | Table 3 note c
- CONFIRMED | A check along the whole count came six years before the recount, from Florian Adolphi and Raimund Muscheler | https://api.crossref.org/works/10.1038/nature11123 | 2016 (doi 10.5194/cp-12-15-2016) against 2022
- CONFIRMED | It is wanted to set an eruption beside a harvest, a famine, a cold summer in the trees | https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=EXT_ID:22699615%20OR%20EXT_ID:26153860%20OR%20EXT_ID:23612289&format=json&resultType=core | Sigl et al. 2015 abstract
- CONFIRMED | Procopius, writing of the tenth year of Justinian’s reign | https://www.gutenberg.org/cache/epub/16765/pg16765.txt | verbatim; Book IV starts at line 2831, chapter XIV at 4230, the passage at 4241-4251 (sha256 9e967b4f... as recorded)
- CONFIRMED | the tenth year of Justinian’s reign | https://www.gutenberg.org/cache/epub/16765/pg16765.txt | "Justinian was in the tenth year of his reign"
- CONFIRMED | The edition dates it 536 to 537. | https://www.gutenberg.org/cache/epub/16765/pg16765.txt | "[536-537 A.D.]" (Loeb, first printed 1916)
- CONFIRMED | In 2008 L. B. Larsen and colleagues, re-reading the acid in GICC05’s three cores, found large eruptions at 529 and at 533 to 534, each give or take two years, and took the second for the cause | https://api.crossref.org/works/10.1029/2008GL034755 | Baillie's abstract on Larsen, "identify large eruptions at 529 ± 2 and 533-534 ± 2". Larsen's own abstract gives 533-534 only; his full text was unreachable (Wiley 403).
- CONFIRMED | and took the second for the cause | https://api.crossref.org/works/10.1029/2007GL032450 | verbatim
- CONFIRMED | The same year M. G. L. Baillie argued from the tree rings that the ice was out of step | https://api.crossref.org/works/10.1029/2008GL034755 | verbatim
- CONFIRMED | the eruption the old count put in 529 lands in | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | 529, 533 and 534 become 536, 540 and 541 through sheet 5's own GICC05 column. Reading through each core's layers instead gives NorthGRIP1 535/539/541, GRIP 536/540/541 and DYE-3 537/541/542
- CONFIRMED | Run the old dates through it and they move | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | 7.0 at every year from 527 to 542
- CONFIRMED | the tie point it calls | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | sheet 3 row 110, "775 CE SPE (Sigl et al., 2015)"
- CONFIRMED | At the beryllium layer the three cores give | https://www.ncei.noaa.gov/pub/data/paleo/icecore/greenland/summit/ngrip/vinther2006-gicc05-holocene-ngrip-20yr-noaa.txt | 768 / 768 / 769, recomputed
- CONFIRMED | the transfer function, which Sinnl’s group built from GICC05’s own year-by-year layers | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | "linearly interpolated the GICC21 ages onto the GICC05 layers of DYE3, GRIP, and NorthGRIP1"
- CONFIRMED | published estimates of the gap at 775 run from six to seven years | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4639793/fullTextXML | Mekhaldi gives 7; A&M give -7 ± 2 (Sigl) and -6 ± 5; the transfer function gives 6
- CONFIRMED | it steps by as much as | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | 1.4375, shown as 1.4
- CONFIRMED | GICC21’s paper prints the beryllium layer at | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | Table 3
- CONFIRMED | its supplement’s table of tie points gives the uncertainty of the same layer as | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | sheet 3 δt 4.19853, shown as "4.199"
- CONFIRMED | The rings’ years rest on crossdating, tree against tree. | https://artwaste.land/strata/every-tree-wrote-the-same-bad-year/ | ; Sinnl also writes "many iterations of cross-dating"
- CONFIRMED | A multi-ice-core, annual-layer-counted Greenland ice-core chronology for the last 3800 years: GICC21 | https://api.crossref.org/works/10.5194/cp-18-1125-2022 | all 9 authors, the title, Clim. Past 18, 1125-1150, 2022 and CC BY 4.0 match https://api.crossref.org/works/10.5194/cp-18-1125-2022
- CONFIRMED | A synchronized dating of three Greenland ice cores throughout the Holocene | https://api.crossref.org/works/10.1029/2005JD006921 | title, JGR 111(D13), 2006; D13102 per NOAA's header
- CONFIRMED | Synchronizing the Greenland ice core and radiocarbon timescales over the Holocene: Bayesian wiggle-matching of cosmogenic radionuclide records | https://api.crossref.org/works/10.5194/cp-12-15-2016 | F. Adolphi and R. Muscheler, Clim. Past 12, 15-30, 2016, CC BY 3.0, per https://api.crossref.org/works/10.5194/cp-12-15-2016
- CONFIRMED | Radiocarbon Production Events and their Potential Relationship with the Schwabe Cycle | https://api.crossref.org/works/10.1038/s41598-019-53296-x | all 10 authors, Sci. Rep. 9, 17056, 2019, CC BY 4.0, per https://api.crossref.org/works/10.1038/s41598-019-53296-x
- CONFIRMED | A signature of cosmic-ray increase in AD 774–775 from tree rings in Japan | https://api.crossref.org/works/10.1038/nature11123 | four authors, Nature 486, 240-242, 2012, per https://api.crossref.org/works/10.1038/nature11123
- CONFIRMED | Tree rings reveal globally coherent signature of cosmogenic radiocarbon events in 774 and 993 CE | https://api.crossref.org/works/10.1038/s41467-018-06036-0 | Nat. Commun. 9, 3605, 2018, CC BY 4.0, per https://api.crossref.org/works/10.1038/s41467-018-06036-0
- CONFIRMED | Multiradionuclide evidence for the solar origin of the cosmic-ray events of AD 774/5 and 993/4 | https://api.crossref.org/works/10.1038/ncomms9611 | Nat. Commun. 6, 8611, 2015, CC BY 4.0, per https://api.crossref.org/works/10.1038/ncomms9611
- CONFIRMED | No evidence for tephra in Greenland from the historic eruption of Vesuvius in 79 CE: implications for geochronology and paleoclimatology | https://api.crossref.org/works/10.5194/cp-18-45-2022 | all 9 authors, Clim. Past 18, 45-65, 2022, CC BY 4.0, per https://api.crossref.org/works/10.5194/cp-18-45-2022
- CONFIRMED | New ice core evidence for a volcanic cause of the A.D. 536 dust veil | https://api.crossref.org/works/10.1029/2007GL032450 | GRL 35(4), 2008
- CONFIRMED | Proposed re-dating of the European ice core chronology by seven years prior to the 7th century AD | https://api.crossref.org/works/10.1029/2008GL034755 | GRL 35(15), 2008
- WRONG | (dek) Trees around the world show it as a jump in carbon-14 from the ring of 774 to the ring of 775. | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6127282/fullTextXML | Results, "Cosmogenic radiocarbon signature in 774 CE": the NH site values "show low levels in 773 (∆∆14C = −0.1%), higher concentrations in 774 (0.4%), a further increase in 775 (0.95%)", and in the SH "the 14C increase in 774 (0.7%) is comparable to the increase in 775 (0.6%)". Across the network the ring of 774 is already raised. See finding 2
- WRONG | the carbon a tree absorbs, write A. Scifo and colleagues, “is immediately locked into the cellulosic structure of the growth-rings” | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/fullTextXML | "Much of the carbon absorbed is immediately locked into the cellulosic structure of the growth-rings, and chemical exchange between rings is negligible". The record's excerpt began one word after the qualifier. See finding 1
- MINOR | In the eighteen centuries below that it never falls under 9 again | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | below the UE 88 layer (1910.43 b2k) the transfer function runs to 3835 b2k, 1,925 years: nineteen centuries; the minimum, 9, holds over all of them. See finding 5
- MINOR | (Nearby note) a count of ice pinned to two eruptions from the history books, and both pins misnamed | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022.pdf | §1.4: "GICC05 was considered exact for the part younger than the eruption peak previously assigned to Vesuvius because many well-known historical eruptions tied the chronology together"; GICC05 had more than two pins, two of them misnamed. See finding 6
- MINOR | (the verifier's header and research/counted-twice/README.md) --online downloads the five data files | verify-counted-twice.mjs | it downloads six: three GICC05 tables, Adolphi and Muscheler's table, the GICC21 spreadsheet and Scifo's. See finding 10
- MINOR | (seoTitle) How Tree Rings Caught a Six-Year Error in Greenland’s Ice-Core Count | https://cp.copernicus.org/articles/12/15/2016/cp-12-15-2016.pdf | six is GICC21's transfer function (6.0 at 775) and Adolphi and Muscheler's decadal estimate (−6 ± 5); the comparison with the spike itself was published as seven: "−7 ± 2 (2σ ) years for both events (Sigl et al., 2015)". See finding 3
- MINOR | (plain) the standard count of 2006 put the ice's mark six years early | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4639793/fullTextXML | Mekhaldi: "a mismatch between ice-core chronologies and tree-ring timescales of 7 years around AD 775 and 994"; the page's own range is six to seven. See finding 3
- MINOR | (relates note) in one network it fixed the dating of 44 tree-ring records on five continents | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6127282/fullTextXML | "ensure the precise dating of 44 tree-ring records"; "enable the authors to confirm the dating of 44 dendrochronologies". "Fixed" can read as "corrected". See finding 7
- MINOR | The oak keeps climbing for another year and the juniper does not | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6863917/supplementaryFiles | sample B1 has no ring 776; after 775 (−3.76 ± 1.98‰) its next measured ring is 777 (−4.62 ± 1.36‰). See finding 8
- CONFIRMED | (og:image:alt) A blue line near zero until about 1200, then climbing past ten, with 775 marked in red | https://cp.copernicus.org/articles/18/1125/2022/cp-18-1125-2022-supplement.zip | matches public/og/counted-twice.png as viewed, and sheet 5
- OBSERVED | (Nearby note) the two spikes are the steepest years in the calibration curve that turns carbon-14 into calendar years | recreated: the agent's scratch script decoding the IntCal20 record embedded in public/strata/where-the-curve-goes-flat/index.html | the nine steepest one-year falls of the annual record are AD 773 to 777 and 993 to 995; the tenth is 1798. intcal.org was not fetched (robots refused)
- OBSERVED | (placard) Last verified 11 November 2026, when it was written; not yet put to that test | git show b6d9f093ea:public/strata/counted-twice/index.html | commit dated 2026-11-11T00:55:32Z; research/counted-twice/verify-page.mjs exists; the page is not deployed, so no outside run can have happened
- OBSERVED | (the check box) what verify-counted-twice.mjs does | recreated: the reader run on local copies of b6d9f093ea (AW_DIR), with --mutate, and --online in the repository | 285 in reader mode, 398 in the repository, 404 with --online (six F checks), 10 of 10 planted faults caught; the commit message's "400 with --online" undercounts F (finding 10)

## Findings
- [fixed] (1, MED) Scifo's "Much of the carbon absorbed" had lost its qualifier: the page said all of a ring's carbon is that year's, which is part of why a spike can show partly in the ring before. The page now reads "much of the carbon a tree absorbs", and the record's claim and excerpt begin at "Much".
- [fixed] (2) The dek's single step "from the ring of 774 to the ring of 775" is what Miyake's cedar and Scifo's two trees show, not the network: Büntgen's sites show the rise already under way in 774, about a third of it in the north and about half in the south. The dek now says "record it as a jump in carbon-14 across their rings of 774 and 775" (markdown, meta, cards, JSON-LD and body together); the body's sentence about the two trees was already right.
- [fixed] (3) The headline took the low end of the page's own range of six to seven years. The seoTitle now carries no figure ("How Tree Rings Caught an Error in Greenland’s Ice-Core Count"), and the plain card says "six or seven years early, far outside its own error bar".
- [fixed] (4) The table's caption now says the years are the averaged b2k ages turned into years by the supplement's own rule, and names the two rows where the paper prints other years: Samalas (the sheet's year column 1257, Table 3 1259 ± 2) and Bárðarbunga (Table 3 877). The verifier checks the caption's figures, the Samalas year column against data.json and, inside the repository, Table 3's two entries in the kept paper.
- [fixed] (5) "eighteen" is now "nineteen" centuries, and the verifier's minimum starts at the UE 88 layer instead of 2000 b2k (still 9). The --mutate plant that used the accurate word as its fault now plants "twenty".
- [fixed] (6) The set-by-another-clock relates note now reads "pinned to eruptions from the history books, two of them misnamed".
- [fixed] (7) The every-tree relates note now reads "confirmed the dating".
- [fixed] (8) The juniper has no ring 776; the page now says "the juniper's next measured ring, 777, sits no higher than 775", and the verifier types the sentence.
- [fixed] (9) The verifier now reaches the figures around the page: it sweeps the title, the plain card, the image text and the nearby notes, the table caption and, inside the repository, the markdown entry's relates notes; it checks the JSON-LD name against the title and, inside the repository, the title and plain card against the markdown entry. Three new plants (the plain card's figure, the image text, a caption year) are caught: 13 of 13.
- [fixed] (10) The verifier's header and the README now say six data files. The commit message of b6d9f093ea ("400 with --online") cannot be amended; the log entry for this session gives the right count.