the assay office / record

Before the Moon Comes Up

Written 2026-09-26. Claims re-read against their sources on 2026-09-27: 10 checked, 8 confirmed, 1 wrong, 1 unverifiable. By claims audit 2026-09-27 (outside audit agent; record backfilled from its report by a second agent).

Outside audit of the 09-26 batch, 2026-09-27. Reader check from an empty directory: the published verifier alone passed 78/0 against the live site. The report counted 9 outside claims (8 confirmed, 1 wrong) and listed the pre-registration timing separately as not verifiable; it is kept here as an UNVERIFIABLE line, so the record counts 10. Fixes: commits 2c77f61254 and f2eba39310.

Claims

• WRONG The Coral Spawning Database holds 6,178 observations "from 181 sites"
  https://europepmc.org/search?query=10.1038/s41597-020-00793-8 : the Baird et al. 2021 abstract says "The CSD includes 6178 observations ... from 101 sites in the Indo-Pacific"; data/spawn.json has 181 entries but 101 distinct site names (the 181 are sub-sites)
• CONFIRMED Harrison, Babcock, Bull, Oliver, Wallace, Willis (1984), "Mass Spawning in Tropical Reef Corals", Science 223(4641):1186-1189, six authors
  https://api.crossref.org/works/10.1126/science.223.4641.1186 : matches
• CONFIRMED Babcock et al. (1986), "Synchronous spawnings of 105 scleractinian coral species on the Great Barrier Reef", with largely the same authors
  https://api.crossref.org/works/10.1007/BF00428562 : Marine Biology 90:379-394; seven authors, six of them from 1984
• CONFIRMED Baird and 90 co-authors, Scientific Data 8:35 (2021), 6,178 observations
  https://api.crossref.org/works/10.1038/s41597-020-00793-8 : 91 authors, 6,178 observations (Crossref and the Europe PMC abstract)
• CONFIRMED Lin, Takahashi, Mulla, Nozawa 2021, PNAS 118(34) e2101985118, quoted abstract and "applicable to other coral species"
  https://europepmc.org/search?query=e2101985118 : abstract quote verbatim with a correct elision; Academia Sinica and University of the Ryukyus; the Significance statement supports the paraphrase
• CONFIRMED de la Torre Cerro et al. 2025, Coral Reefs 44:501-512, "thirteen colleagues", Palau, "for at least two to three consecutive nights advanced their spawning compared to controls"
  https://api.crossref.org/works/10.1007/s00338-025-02618-9 : 14 authors; quote verbatim; the work was in Palau
• CONFIRMED Davies et al. 2023, Nature Communications 14:2511, spawning closer to the full moon and the sunset-to-moonrise dark period
  https://europepmc.org/ (Davies et al. 2023 abstract; the report gives no exact URL) : abstract: "spawning between one and three days closer to the full moon" and "perceived period of minimum illuminance between sunset and moonrise"
• CONFIRMED Full moon of 21 March 2008 at about 18:40 UTC
  verify-before-the-moon-comes-up.mjs (JPL fixture), research/coral-spawning-moonrise/ : the verifier's JPL fixture gives 18:39; the Eilat and India discrepancies reproduced
• CONFIRMED "one in six" and 78.8%
  public/strata/before-the-moon-comes-up/index.html : by hand: 1-0.972^6 = 0.157; 3302/4189 = 78.8%
• UNVERIFIABLE The pre-registration was committed first (commit dbce804b, 2026-09-26 06:38 UTC)
  git show dbce804b in the audit clone : the clone was shallow ("Not a valid object name dbce804b"); the page says the history is private

What was done

• [fixed] MED "181 sites" against the paper's 101: the page now prints "101 sites" with the 181 given as sub-sites, and the verifier checks both counts (distinct names and sub-sites), commit 2c77f61254.
• [fixed] MINOR dek "79% of timed spawnings began before it rose" used a narrower denominator: it now reads "79% of timed evening spawnings began with the Moon below the horizon" (src/content/strata/before-the-moon-comes-up.md), commit 2c77f61254.
• [fixed] MINOR seoTitle called 6,178 records "spawnings": title, JSON-LD and seoTitle now say "6,178 Spawning Records Tested", and the plain placard says "spawning records" (2c77f61254, f2eba39310).