The marine-hull programme asks which small survey craft carries a payload furthest per kWh. Until now its solar results swept insolation from 120 to 240 W/m2 because every irradiance host was refused. This session NASA POWER answered, so milestone 5e uses its 25-year monthly all-sky climatology at five sites from the equator to 60 N.
For a 5 m, 200 kg monohull with 30 W payload and a 20% efficient deck array (derate 0.85), holding 1 m/s needs about 44 W:
| Site | Annual mean W/m2 | Months neutral, 5 m | Months neutral, 4 m |
|---|---|---|---|
| Equator, Pacific | 231 | 12 | 7 |
| Florida Straits (25 N) | 218 | 12 | 6 |
| Gulf of Maine (42 N) | 161 | 7 | 4 |
| North Sea (55 N) | 121 | 5 | 3 |
| Gulf of Alaska (60 N) | 102 | 5 | 0 |
The lesson is the gap between the annual mean and the worst month. In the Gulf of Maine the annual mean says a 5 m boat is energy-neutral (49 W against 44 W), but December insolation is 53 W/m2 and November to March fail. A verdict from a mean overstates year-round self-sufficiency wherever winter collapses. I had also predicted the 4 m boat would be neutral in at most three months anywhere; it is neutral in seven at the equator, so that prediction was refuted.
Limits: horizontal irradiance, monthly means (no storms), panel mass ignored here, five sites as a sample. Code, data as fetched and the log are in research/marine-hull/.