How tide predictions work

Tide tables are one of the oldest working forecasts in science: NOAA can tell you the time of high tide in Bridgeport or San Diego years in advance, to within a few minutes. That's possible because the tide isn't weather. It's astronomy filtered through geography, and both of those are predictable.

The forces are simple; the coastline isn't

The moon and sun pull on the ocean, and the Earth rotates underneath the resulting bulges. If the planet were covered in uniform deep water, that would produce a tidy wave you could compute from orbital mechanics alone. It isn't — every basin, shelf, bay, and river mouth responds to that forcing differently. Water sloshes, resonates, and piles up. That's why Boston sees a nine-foot range while Key West sees under two, and why the Gulf of Mexico gets one high tide a day where the Atlantic coast gets two.

The consequence: you can't predict the tide at a place from first principles. You have to measure that place.

Harmonic constituents: a fingerprint for every station

The trick, worked out in the 19th century by Lord Kelvin and refined ever since, is that however complicated a coastline's response is, it's complicated in a periodic way. Every astronomical cycle that drives the tide — the 12.42-hour principal lunar cycle (called M2), the 12-hour solar cycle (S2), the daily lunar declination cycles (K1, O1), and dozens more — shows up in the local water level as a sine wave with a fixed size and time offset.

NOAA runs water-level gauges for years at each station, then decomposes the record into those component waves. The result is a set of harmonic constituents — typically 37 of them — each with an amplitude (how big) and a phase (how delayed) that together act as the tidal fingerprint of that exact spot. Making a prediction is then just running the clockwork forward: add the sine waves back up for any future date. You can inspect the raw constituents for any station through the Perigee API and MCP server.

Reference and subordinate stations

Maintaining a permanent gauge everywhere is expensive, so NOAA publishes full harmonic predictions for a core set of reference stations and derives thousands of subordinate stationsfrom them — a nearby creek mouth might be listed as “Bridgeport high tide plus 14 minutes, times 0.9.” Most of the ~3,500 prediction stations Perigee lists by stateare subordinate stations. They're accurate for times and heights of high and low water, which is what they're for.

Predictions vs. reality

A tide prediction is the astronomical tide only. The actual water level on any given afternoon also carries everything the atmosphere is doing:

• Wind — a sustained onshore blow stacks water against the coast; offshore wind drains it.
• Barometric pressure — low pressure lets the sea surface rise roughly a centimeter per millibar; storms ride in on that.
• Rivers and rain — discharge raises levels in estuaries well above the predicted curve.

That gap between predicted and observed is exactly what a live gauge shows you. Perigee's station pages plot both lines — the harmonic prediction and the six-minute observations — so the difference isthe weather. When the observed curve runs well above the prediction, take note: that's surge.

What “zero” means

Tide heights are measured from a tidal datum, and in the US that's MLLW — mean lower low water, the average of each day's lower low tide over a 19-year epoch. Nautical chart depths use the same zero, which is the point: predicted height plus charted depth equals water under your keel, roughly. It also means negative tides are real and useful — more on that in how to read a tide chart.

See it live: pick your coast from the state directory or find the gauge nearest you.