Official Australian tide tables, and how to use them
Who publishes tide predictions for each state, what the products cost, and when you are legally obliged to use the official one.
The Atmosphr tide guides
Tides, explained
before they are tabled.
Australian tide predictions are produced by one organisation, published against a datum most people never see named, and routinely misread. These guides fix that.
01 · Start here
Which tide question do you have?
Eight guides, each answering one question properly rather than eight pages answering the same question badly.
How to read a tide chart
The four things a row tells you, the two it does not, and the rule of twelfths including where in Australia it stops working.
Tide datums explained: chart datum, LAT, MSL and AHD
What the height number is measured from, and why you cannot compare it to a survey level or a bridge clearance without an offset.
Spring tides and neap tides
The fortnightly cycle, why the name has nothing to do with springtime, and why the biggest tide lags the full moon.
King tides in Australia
A popular term rather than a technical one. What actually produces the year's highest predicted tides, and why the water often does not match.
Why most places get two high tides a day, and where they do not
Parts of northern Australia get one. If a tide page tells you there are always two, it was not written for this country.
How tide predictions are made, and why the real sea level differs
Harmonic analysis is astronomically reliable and meteorologically blind. Wind, pressure and surge do the rest.
Tides and tidal currents are not the same thing
High water does not mean slack water. At a bar crossing that distinction is a safety issue.
02 · Geography
Australia does not have one kind of tide
Most tide explainers are written for the North Atlantic, where the tide is reliably semidiurnal: two highs and two lows a day, of roughly equal size. Australia is not like that, and the differences are large enough to change how you read a table.
Semidiurnal
Two highs and two lows each day, of broadly similar height. This is the pattern along much of the east coast and across much of the north west shelf. It is the pattern the rule of twelfths was designed for.
Diurnal
One high and one low each day. The Gulf of Carpentaria is the best known Australian example. A table for a diurnal port has about half the rows you may be expecting, and a page that assumes four events a day will simply be wrong there.
Mixed
Two highs and two lows, but noticeably unequal, so that the two highs on the same day differ in height. This is common across southern and western Australia. The inequality is not an error in the table.
The South Australian dodge tide
In the South Australian gulfs the semidiurnal and diurnal components can very nearly cancel, producing a period of little tidal movement at all. It is known locally as the dodge tide, and it is distinctive enough that the Bureau of Meteorology publishes dodge tide information alongside its South Australian tide tables.
Range varies enormously
Australian tidal range runs from well under a metre in parts of the south west to more than ten metres in the Kimberley, where King Sound near Derby has among the largest tides in the southern hemisphere. A rule of thumb learned in one part of the country can be badly wrong in another.
This matters for more than curiosity. Tide type determines how many rows a table has, whether the rule of twelfths applies, and how a spring tide actually presents. The full explanation is here.
03 · Honesty
What Atmosphr's ocean panel is, and is not
The weather app includes an ocean conditions panel showing wave height, swell, sea temperature, current and a modelled sea level. That modelled sea level is not a tide prediction, and we label it that way on the panel itself.
It comes from a global numerical ocean model at a resolution of roughly eight kilometres, and it is referenced to global mean sea level rather than to a local tide datum. A harmonic tide prediction and a modelled sea level are different products built for different purposes. Treating one as the other is the single most common way to get a wrong answer from otherwise correct data.