Daily Routing Brief — 2026-06-09
Generated 2026-06-09T13:46:37.187893+00:00 · Newport→Bermuda 2026 · T−10 days · « all briefs
Regime: PRACTICE + OCEAN-TRACKING — T−10 days. No atmospheric model reaches race day yet (~7-day horizon). The codes below are crew practice, not a June-19 forecast. The genuinely predictive signal this far out is the Gulf Stream — track its evolution below.
PRE-RACE PRACTICE ONLY — not for use during the race
These are pre-race research / practice reps. Per RRS 41 (2026 NOR) a boat shall not receive routing assistance from shore — including AI-generated — while racing. This page stops updating before the June 19 start; do not open it from the boat once racing.
Headline
Current forecast window most resembles 2018 (W-LA/W-RC/W-PL, score 2.55).
- Winner that year: Grundoon (Columbia 50) — 112h corrected (extreme light-air distortion)
- What won: Find Stream core in light air; exploit current as equalizer; don't thrash in the park
- Winning tactics: T-CF T-PF T-CL
- Lesson: Light-air collapses corridor differences but never reverses them. Current core matters more than position.
- Best corridor that year: RHUMB
Per-period classification (current window)
Start anchor: 2026-06-09T00:00:00+00:00 · horizon +167 h
| period | code | conf | GFS | ECMWF | ICON |
|---|---|---|---|---|---|
| P1 start / coastal | W-BH | HIGH (100%) | 8.9/17.9 | 9.8/22.5 | 8.4/25.1 |
| P2 Gulf Stream crossing | W-BH | HIGH (100%) | 11.2/14.2 | 10.0/17.3 | 10.1/16.5 |
| P3 Bermuda approach | W-PL | MEDIUM (67%) | 8.9/17.1 | 7.4/19.8 | 7.8/16.9 |
Codes are atmospheric only and link to the analog framework. The GFS/ECMWF/ICON columns are mean wind / max gust (kn). Confidence = model agreement. See the method.
AI medium-range cross-check (ECMWF-AIFS)
A separate read from the data-driven ECMWF-AIFS model — strongest in the 4–10 day range on the large-scale pattern, with less run-to-run swing. A third opinion vs the core GFS/ECMWF/ICON vote above, not blended into its confidence and not for short-term or routing detail. (GraphCast is offline on Open-Meteo right now, so this is AIFS-only; AIFS carries no gust field.)
| period | AIFS read | vs core |
|---|---|---|
| start / coastal | W-BH | matches core |
| Gulf Stream crossing | W-BH | matches core |
| Bermuda approach | W-BH | core says W-PL |
When AIFS agrees with the core models on the medium-range pattern, lean into that regime; when it diverges, treat the period as genuinely uncertain and don't over-commit. Source: Open-Meteo (ECMWF-AIFS, ecmwf_aifs025_single).
Trend (day over day)
vs 2026-06-08 (T−11):
- Headline analog changed: 2024 → 2018.
- Period codes: P1 W-PF→W-BH; P2 W-LA→W-BH; P3 W-BH→W-PL.
- Model agreement: 78% ↑ 89%.
- New Gulf Stream snapshot: 2026-06-08 → None.
- Crossing SST: 24.3→22.8 °C (76→73 °F) ↓, north-wall jump -0.5°C (-1°F).
- Peak corridor wave: 2.0→2.0 m (7→7 ft) →.
Convergence: race day not yet in model horizon — convergence tracking begins automatically once it is (~T−7).
Analog ranking (period-aware)
| rank | year | score | P1/P2/P3 | best corridor |
|---|---|---|---|---|
| 1 | 2018 | 2.55 | W-LA/W-RC/W-PL | RHUMB |
| 2 | 2016 | 2.55 | W-PF/W-CC/W-PL | RHUMB |
| 3 | 2012 | 2.55 | W-PF/W-BH/W-BH | RHUMB |
2014 excluded as seed data. Years link to their winner-route writeups.
Gulf Stream (observed — the real signal this far out)
No local Stream snapshot in this build — see the live sources below. (A dated snapshot is embedded here after a local brief:refresh.)
Full interactive planning map — all six past corrected-time winners' tracks over the corridor (toggle layers).
Data sources
- Forecast (wind · gust · pressure — GFS·ECMWF·ICON): Open-Meteo · full list in the Sources & Tools catalog.
- Gulf Stream SST: NOAA OPC NCOM · Rutgers RUCOOL · NASA SPoRT.
- Gulf Stream currents: NOAA OPC RTOFS · AOML altimetry.
- Expert read: Frank Bohlen's Gulf Stream notes.
Sea state & ocean (numeric — Open-Meteo Marine)
Near-term (next 24 h) means along the corridor. SST and waves are reliable; ocean current is a COARSE model that understates the Gulf Stream (2–4 kn) and cannot see rings or meanders — trust the Stream imagery for structure.
| waypoint | SST | current | wave |
|---|---|---|---|
| start | 14.7 °C / 58 °F | 0.3 kn @ 125° | 0.8 m (3 ft) / 6 s |
| shelf | 23.3 °C / 74 °F | 0.4 kn @ 276° | 1.1 m (4 ft) / 6 s |
| stream | 22.8 °C / 73 °F | 2.2 kn @ 193° | 2.0 m (7 ft) / 7 s |
| post_stream | 22.4 °C / 72 °F | 0.8 kn @ 264° | 1.8 m (6 ft) / 6 s |
| approach | 23.8 °C / 75 °F | 0.4 kn @ 156° | 1.2 m (4 ft) / 6 s |
North wall: shelf→crossing SST change is only +-0.5 °C / +-1 °F — front is weak/diffuse or displaced from the rhumb waypoint in this coarse model.
Sea state: peak corridor wave ~2.0 m (7 ft).
Race-day horizon
Race day is T−10 — beyond the ~7-day model horizon. The classifier correctly declines (no fabricated race-day codes). Convergence tracking begins automatically when race day enters the window (~T−7).
Sources: forecast — Open-Meteo (GFS/ECMWF/ICON, no key). Analogs — framework + winner routes + tactics. Gulf Stream — scripts/gulfstream_monitor. Method — forecast analog matcher. Pipeline — scripts/daily_brief.py.