Safety and limits

The forecast reports the sea state itself: wave height, whether the water is short chop or swell, and the wind's surface signature on the official Beaufort scale. These are physical descriptions, not a safety rating, and calm words are not permission to go.

There is deliberately no comfort score. A kayak, a sailboat, a 60-footer and a center-console in a following sea all experience the same water differently, so any single "how it will feel" number would be a claim about your boat that this app cannot make. The translation from water to ride belongs to the captain.

Descriptors describe sea state. They do not account for your experience, your crew, your fuel, your engine, or anyone aboard who gets seasick.

Wave heights here are significant wave height — the average of the highest third of the waves, which is the standard marine measure and roughly what an experienced observer reports by eye. It is not the biggest wave you will see.

In any given sea, individual waves reach roughly 1.9 times the significant height. A forecast of 2 ft means occasional waves near 4 ft. Plan for the big ones.

• Thunderstorms. The score is derived from forecast wind. A summer squall arrives faster than any hourly forecast resolves it, and it can turn a 9 into unusable water in twenty minutes. A storm icon on a good-scoring day is a warning, not decoration.
• Where currents are unknown.Wind against tide is the Bay's signature hazard. Nodes with no nearby current station are scored as though the water were slack, which is optimistic at a constriction — exactly where it matters most. The day detail says so where it applies.
• Near shore and in the rivers. Fetch and depth are computed from a shoreline dataset and a depth model, not from a chart. Shoals, bars and channel edges are not represented. Some nodes are still marked as estimated rather than measured; where that is true, the forecast page says so.
• Outside boating hours. Daily ranges cover 8 AM to 10 PM local time. A calm at 4 AM is not part of the range, and the range says nothing about overnight conditions.
• Sea breeze and afternoon build. A Chesapeake day routinely runs glass at dawn and blown out by mid-afternoon. That is why the day shows a range and a best window rather than one number — but the forecast underlying it is still only as good as the gridded wind forecast it comes from.

When the National Weather Service has a Small Craft Advisory, Gale Warning or Storm Warning in effect, the score is capped — lowered to a ceiling regardless of what the physics says. A forecaster sees things this model does not.

The cap is applied on top of the physics and both numbers are shown, so you can always see what conditions alone would have scored. The cap only ever lowers a score; it never raises one.

Watches are shown but do not cap, because a watch describes a possibility rather than a condition.

Route times assume a planning speed of 18 knots and a straight track between points. They are a rough sense of timing, not a passage plan.

The planner will refuse a track that crosses land, but a track it accepts is not a safe or navigable course. It does not know about channels, shoals, bridges, restricted areas or traffic. Navigate from a chart.

Wind and marine alerts come from the National Weather Service gridpoint forecast. Current predictions come from NOAA CO-OPS. Buoy observations come from NDBC. Depths come from a NOAA elevation model, cross-checked against charted depths where those are known.

Sea state is derivedfrom wind and the water's geometry using published relationships — it is not copied from someone else's wave forecast. That is what makes each score explainable, and it is also why an error in the wind forecast becomes an error in the score.

When data is missing, it is shown as a gap rather than filled in. A blank is honest; a confident wrong number is not.

The weather icon is context only. Rain and cloud do not change the score, because they do not change how the water behaves — a sunny 20 knot day and an overcast 20 knot day are the same ride. They are shown precisely because the score ignores them: an 8 in the rain is still an 8, and you deserve to know which kind of 8 it is.