{
 "schema": "nephos.method.v1",
 "slug": "boggs-clamming-shelter-island-1878",
 "title": "Frank Myers Boggs, Clamming at Shelter Island (1878), Long Island Museum",
 "urls": [
  "/studies/boggs-clamming-shelter-island-1878"
 ],
 "written": {
  "summary": "The museum supplies the year and subject. Nephos reads the breaking crests, spray, tilted boats, long rakes and shirtsleeves as a windy warm-season working scene. That reading defines a May–October test window, not an execution date.",
  "items": [
   {
    "quote": "turbulent waters",
    "source": "Long Island Museum object record",
    "url": "https://artsandculture.google.com/asset/clamming-at-shelter-island-frank-myers-boggs/ggFsh8U6s5HH0g"
   },
   {
    "quote": "several light squalls with some rain fall",
    "source": "Moriches Signal Service register, May 1878, supplementary remarks for the 12th, image 00131.",
    "url": "https://www.ncei.noaa.gov/nerms/api/document/AE45035B1FA6453E9BB2CED04280F3C5/download"
   }
  ]
 },
 "records": [
  {
   "name": "Moriches Signal Service wind observations, May–October 1878",
   "detail": "184 daily rows at 7 a.m., 2 p.m. and 9 p.m. Force 4 is labelled Strong wind on the printed historical key. 2 unreadable days remain separate.",
   "url": "https://www.ncei.noaa.gov/nerms/api/document/698C44F1158C41B8BF1600BC9337A211/download"
  },
  {
   "name": "Gabreski Airport, modern regional analogue",
   "detail": "17 sufficiently complete seasons, 2004–2020; 3,065 complete three-hour dates. This is not an official 1991–2020 normal.",
   "url": "https://www.ncei.noaa.gov/pub/data/noaa/isd-lite/"
  }
 ],
 "gaps": [
  "No dated sitting, preparatory study or documented studio reworking has been verified.",
  "No local Shelter Island wind, tide or wave observation is linked to this canvas.",
  "The museum accession is unconfirmed in the opened holder records; no open image reproduction licence is verified."
 ],
 "tests": [
  {
   "claim": "Does the visually proposed warm season include recurring strong wind?",
   "result": "Moriches records strong wind on at least 15 of 184 May–October 1878 days (8.2%), with 11 of those days in May or October. This supports recurring regional strong wind within Nephos’s warm-season reading; it does not identify the painted day or count the clammers’ rough-water days."
  },
  {
   "claim": "How sensitive is the tally to qualified and unreadable entries?",
   "result": "15 confirmed days; including every unresolved day would give 17 of 184 (9.2%). This bounds the sampled record only, not gusts between observations."
  },
  {
   "claim": "Is there a usable modern seasonal comparison?",
   "result": "At one modern station, 452 of 3,065 complete dates (14.7%) reach 6.8 m/s at one of three fixed EST hours. The different sites and observing methods prevent a climate-trend inference."
  }
 ],
 "finding": "Moriches records strong wind on at least 15 of 184 May–October 1878 days (8.2%), with 11 of those days in May or October. This supports recurring regional strong wind within Nephos’s warm-season reading; it does not identify the painted day or count the clammers’ rough-water days.",
 "verdicts": [
  {
   "claim": "Strong wind recurs in the proposed season",
   "verdict": "Partially agrees",
   "reason": "The threshold is defined by the form and counted over the full proposed season. This verdict concerns regional occurrence only."
  },
  {
   "claim": "Documentary season",
   "verdict": "Unknown",
   "reason": "The warm-season scene is Nephos’s visual reading; no documentary month or day is established."
  },
  {
   "claim": "Local water and painted day",
   "verdict": "Unknown",
   "reason": "Moriches is regional; wind alone does not determine a bay’s waves or a sitting."
  }
 ],
 "limits": [
  "The comparison window is a test of the visual hypothesis, not evidence of a six-month visit.",
  "Estimated force is not Beaufort force or a measured wind speed. Three readings can miss brief squalls.",
  "Qualifiers and unreadable cells remain explicit. No blanket calm or dry-day filling is used.",
  "Gabreski uses another site, instruments and time convention; the modern percentage cannot establish a wind trend since 1878.",
  "Newspaper archive access failed; this is not evidence that reports do not survive.",
  "Child’s 1884 account describes general practice, not alterations to this canvas.",
  "Daily force readings were manually transcribed and reread against the original scans."
 ]
}