The tide does not read the app. Anyone who has stood on a harbour wall waiting for a metre of water that never arrived, or watched a slipway dry out forty minutes ahead of schedule, has met the gap between a prediction and the sea itself. The question is not whether to trust tide tables or tide apps. It is knowing what each one is really telling you, where the numbers come from, and what to do when they disagree while your engine is already warm.
The working answer is simple enough: treat the official prediction as your reference, use the app as your calculator, and carry enough paper and enough skill to keep going when the screen goes dark.
What a tide table actually predicts
An official tide table is published by a national hydrographic office and built from harmonic analysis: years of observed water levels at a standard port, broken down into constituents such as the principal lunar semi-diurnal, the solar one, and dozens of smaller terms. Add them back together for a given date and you get high and low water times and heights. That is a prediction, not a promise, and it assumes average weather.
Two details trip people up more than any other. First, heights are measured above chart datum, which differs between countries — lowest astronomical tide in much of Europe, mean lower low water in the United States. A height of 2.0 m in one system is not the same depth as 2.0 m in the other. Second, the times are quoted in a stated time zone, usually standard time, not daylight saving. Read the small print at the head of the table before you read the columns.
Where the apps earn their place
A good tide app is a genuine improvement on a column of numbers. It shows the curve rather than two daily points, so you can see at a glance how fast the water will move between them. It interpolates for you, converts units, remembers the ports you use, works weeks ahead, and often includes tidal stream atlases, sunrise, and the moon phase. Many apps draw on licensed hydrographic office data and are perfectly sound.
They are also honest about their limits, if you read the settings. The better ones name the standard port behind each prediction and show whether a secondary port correction has been applied automatically. That single detail tells you how much of the number is the office's work and how much is the app's interpolation.
Where apps quietly go wrong
Most app failures are not dramatic. They are small, plausible errors that look exactly like a correct answer.
- Datum drift. If you switch between a European almanac and a US-sourced app, or between a chart and a table built on a different reference, depths can be out by a metre or more in shallow water. Always establish which datum you are reading.
- Time zone and daylight saving. A device set to local summer time can display a tide an hour off what the printed table says. At a shallow bar with a narrow window, that hour is the whole plan.
- Silent secondary port corrections. The app applies a difference table you never see. Where the tidal range is sensitive to small changes in range or phase, that interpolation can wander by twenty or thirty centimetres without any warning on screen.
- Stale data. Cached predictions do not update when a harbour authority re-surveys, moves a sill, or changes a chart datum. Paper almanacs at least carry an edition date you can check.
- Skill fade. If you cannot work the tide out without the phone, you cannot check the phone. That is the real risk, and it grows quietly over a season.
Working a secondary port by hand
Most of us sail from places that are not standard ports, so the correction is the part worth practising. The method has not changed.
- Read high and low water times and heights at the standard port for your date.
- Apply the published time difference for the secondary port, checking the sign convention carefully — some tables give a correction to add or subtract, others give the secondary port's time relative to the standard port.
- Apply the height difference. This is usually tabulated against the standard port's range or against spring and neap conditions, so you will often interpolate between two figures.
- Interpolate between the corrected high and low water for the hour you actually care about, using a tide curve or the rule of twelfths.
An illustration, with invented numbers: suppose the standard port has high water at 10:42 and 3.4 m. The secondary port's differences are a time of minus 35 minutes and a height of minus 0.3 m at neaps, minus 0.4 m at springs. With a small range that day, you take the neap figure: high water at 10:07, height 3.1 m. Then check it on arrival. One caveat worth repeating: never carry a correction sideways from a neighbouring port, even a close one. Sills and channels have their own behaviour.
What the prediction leaves out
Harmonic predictions assume average meteorological conditions, and the sea rarely obliges. Low pressure lifts water roughly a centimetre per hectopascal, so a deep depression can add a third of a metre before wind is considered. A steady onshore breeze over a long fetch piles water into estuaries and behind sandbars. Heavy rain inland can hold a river mouth high for days. Long narrow inlets produce standing oscillations that make a nonsense of a smooth curve. Shallow water distorts the shape of the tide as well, giving double high waters, long stands, and the occasional bore.
In practice this means the predicted height is a starting point. Give yourself a margin on under-keel clearance, and treat any passage over a bar, a sill, or a drying bank as weather-dependent rather than table-dependent. If the wind has been blowing onshore for twelve hours, expect more water than the table says — and expect it to arrive earlier.
Building a backup that survives a dead battery
Redundancy at sea is unglamorous and effective. Five minutes of preparation covers the failure modes you will actually meet.
- Print or photocopy the tide table pages for your standard and secondary ports, plus the tidal stream pages for the passage. Keep them with the chart, not in a bag below.
- Write your own tide curve for the day on the chart or in the log, and mark the times you need: half tide, the bar window, the last chance to enter.
- Note the times in the time zone you will actually be using, and mark the date of any daylight saving change on the passage plan.
- Cross-check two independent sources before departure — a printed table and an app, ideally from different data providers. If they differ by more than a few minutes or a few centimetres, find out why before you weigh anchor.
- Record what you observe. Log the actual height at the sill, the stream turning against you earlier than forecast, the depth under the keel at a known mark. That record is worth more than any prediction for next time.
The two-minute pre-departure check
Before you start the engine, do three things. Confirm the datum and the time zone of the numbers you are using. Confirm the secondary port correction has actually been applied, by hand or on screen, rather than assumed. Then look at the weather and ask what a metre of surge or a day of onshore wind would do to your plan.
If the answer is comfortable, go. If it is marginal, move the departure, not the threshold. Tables and apps are both trying to describe the same sea, and neither of them has ever been aboard. The skipper supplies the judgement, and that is the one instrument that never runs out of battery.
Photo: aszak / Pixabay
