How to read a marine forecast: wind, gusts, waves and the hours between

In this piece

1. Wind direction is where the wind comes from
2. Knots, and the conversions you will meet
3. Mean wind and gusts in a marine forecast
4. Wave height forecasts: what significant height means
5. Why wave period matters as much as height
6. UTC or local time?
7. Hourly or three-hourly
8. Reading one day hour by hour
9. What a forecast line leaves out

To read a marine forecast, take each hour in the same order: the direction the wind comes from, its mean speed in knots, the gusts, then the wave height and the direction the waves come from. Check whether the times are UTC or local. Then step back and read the whole day as a curve: when the wind starts to build, when it peaks, how far the gusts sit above the mean, and when the sea goes down again, which is always later than the wind.

A lot of forecasts that “got it wrong” were read wrong. A forecast of 14 knots that turned out to mean 14 knots gusting 26, from the opposite direction to the one you assumed, three hours earlier than you thought, was a perfectly good forecast. Each of those numbers has a precise meaning, and each is easy to misread.

Wind direction is where the wind comes from

A northerly blows from the north. A wind of 200 degrees comes from a little west of south and pushes you toward a little east of north. Every marine forecast uses this convention, and nearly all of them use it for waves too: waves from the northwest are rolling toward the southeast.

Currents are the exception. A current is named for where it goes, so a south-going current runs the same way as a northerly wind. When you read a tidal atlas and a wind forecast side by side, that is the difference that catches people out.

Two more words worth knowing. The wind veers when it shifts clockwise, from north to northeast, and backs when it shifts anticlockwise, from north to northwest. A forecast of a southwesterly “veering northwest” later in the day is usually describing a cold front going through.

Knots, and the conversions you will meet

One knot is one nautical mile an hour, or 1.852 km/h. Marine forecasts use knots; land forecasts in much of Europe use km/h or metres per second, and many official bulletins give Beaufort forces. Two rough conversions will get you through. Metres per second doubled comes out a little above the knots (10 m/s is 19.4 knots). Kilometres an hour halved comes out a little below (40 km/h is 21.6 knots). For forces, our Beaufort scale in knots has every band.

Mean wind and gusts in a marine forecast

The headline wind figure is a mean. The international standard is the average over ten minutes, measured 10 m above the surface, and a model’s figure is an average in space as well, across a whole grid box. The gust is the short peak inside that average, a burst lasting a few seconds.

The ratio between the two tells you what kind of wind it is. Over open water in a steady breeze, gusts typically run 20 to 40 percent above the mean: 15 knots gusting 18 to 21. Downwind of high land, along cliffs and under shower clouds the air is far more turbulent, and gusts of one and a half to two times the mean are normal.

So read the two numbers together. 14 knots gusting 18 is a pleasant force 4. 14 knots gusting 26 is the same mean with force 6 punches in it, and the size of the gap is a warning on its own: something, usually high ground upwind or a shower cloud, is stirring the air. Plan the passage on the mean, and reef for the gusts.

The wind on deck is not the forecast wind

The masthead sensor on most cruising yachts sits higher than 10 m, so in the same wind it reads a little more than the forecast. It also reads apparent wind, the true wind combined with the breeze made by your own motion. Sailing upwind at 6 knots in 14 knots of true wind, the instrument shows 18 or 19. Running downwind at the same speed, it shows 8 or 9. That is why a run in a forecast force 5 can feel like a lovely afternoon right up to the moment you turn and head into it.

Wave height forecasts: what significant height means

The wave height in a forecast is the significant wave height: the average height, trough to crest, of the highest third of the waves. It was chosen because it comes close to what an experienced observer reports when asked to estimate the sea. It is not the biggest wave, and it is not the typical one either.

In a sea with a significant height of 2 m, roughly one wave in seven is bigger than 2 m, about one in a hundred is around 3 m, and over a few hours you should expect one or two close to 4 m. Those are the waves that fill a cockpit, and a 2 m forecast already includes them.

The figure usually combines the wind sea, raised by the local wind, with any swell that has travelled in from a distant blow. A forecast of 1.5 m on a light day may be almost all swell; on a windy day it may be steep wind waves with a cross swell underneath. The direction arrow describes the main part of it.

Near land the height depends on fetch, the stretch of open water the wind has blown across. A force 6 across a mile of bay raises a short chop. The same wind blowing for a day over a few hundred miles of open sea builds the 3 m waves of the Beaufort tables. Wave models work on grids as well, and in a small bay or a narrow channel they can be well out.

Why wave period matters as much as height

The period is the time between one crest and the next, and it decides the shape of the wave. In deep water, a wave’s length in metres is about 1.56 times the square of its period in seconds. A 5 second wave is about 39 m from crest to crest. A 12 second swell is about 225 m.

Put 2 m of height on each. The 5 second wave is a short, steep face that a 12 m yacht climbs, slams off and meets again a few seconds later, all afternoon. The 12 second swell is a long hill you ride up and down and barely notice at sea, until it reaches a shoal, a harbour entrance or an anchorage open to it. There it breaks, or turns into the slow roll that keeps a whole bay awake.

Current changes the shape too. Wind blowing against a tidal stream or the steady current of a strait shortens and steepens the waves, which is why tidal races such as Portland and the Alderney Race are dangerous with wind against tide and much quieter at slack water. The forecast will not tell you this; the tidal atlas and the pilot book will.

SafeBoat shows wave height and direction but not wave period, so take the period from a dedicated wave forecast or the national marine bulletin; many GRIB sources carry it too. A height without a period tells you how big the waves are and nothing about how they will feel.

UTC or local time?

GRIB files, many official bulletins and some websites give times in UTC, often written with a Z. Most apps convert to your phone’s time, but check, because the mistake costs hours. In summer Greece and Turkey are three hours ahead of UTC, so a peak at 12:00 UTC is a peak at 15:00 on deck. Turkey stays at UTC+3 all year; Greece goes back to UTC+2 in winter.

Check the issue time while you are at it. A bulletin issued at 18:00 yesterday knows less than this morning’s model run, and a forecast saved on a phone that lost signal two days ago is older still.

Hourly or three-hourly

Many forecasts step in three-hour intervals, which is fine on a settled day. When something is changing fast, such as a front, a squall line or a sea breeze setting in, three hours is long enough to hide the peak entirely: 15 knots at 12:00 and 16 knots at 15:00 can sit either side of 25 at 13:30. Hourly steps show the shape.

Hourly does not mean accurate to the hour, though. A front forecast two days ahead can arrive three or four hours early or late. Treat the timing as a window and leave slack on both sides.

Reading one day hour by hour

Here is a typical summer day in the central Aegean, in local time. You are in a bay and want to go 22 nm north, which puts the wind on the nose.

Read down the columns before you read across. The wind is light at dawn, builds from mid-morning, peaks between 16:00 and 18:00 and eases after dark, but only to force 4. That is the daily rhythm of a summer northerly in these waters, and the direction barely moves all day.

The gusts sit about a third above the mean all day, a normal gust factor for open water. Nothing in the numbers is hiding a nasty surprise. Close under a high island the afternoon could be a different story, with gusts nearer twice the mean.

The sea lags the wind. It peaks at 18:00, two hours after the wind, and at 22:00, with the wind back down to 12 knots, there is still 1.2 m of short sea running.

Now the decision. Upwind into a short sea, a typical cruising yacht makes perhaps 4 knots toward where she is going, so 22 nm is five to six hours. Leave at 06:00 and you arrive around noon, with the last hour in force 4 and a building chop. Leave at 10:00 and you spend the afternoon beating into force 5 and a short sea of well over a metre, a long, wet day for the sake of a lie-in. The evening looks tempting on the wind column alone, but the sea is still up and you would be finding a strange anchorage in the dark.

So we would go at first light, motor-sailing the first hour or two if there is too little wind to make progress, and be anchored by early afternoon. Going the other way, south with the wind behind, the afternoon would be the sail of the week.

Last, check that the models agree on that shape. If one has the wind building at 08:00 and another at noon, the window is shorter than it looks. Our piece on ECMWF, GFS and ICON goes into how to weigh them against each other, and passage planning picks up from there.

What a forecast line leaves out

Some of the weather that matters most does not fit in a row of numbers. Thunderstorms show up as an icon or a word, but the gusts under one can be far above anything in the gust column. Fog and poor visibility live in the written bulletin. Local effects, the wind bending round a cape or funnelling down a valley, are in the pilot book and in the heads of the people who sail there. Read the official marine forecast for those, every time, after the numbers.

Then try this tonight. Open tomorrow’s forecast for where you are and answer four questions from it: when does the wind start to build, when does it peak, how far above the mean are the gusts, and when does the sea go down. If you can answer all four, you have read the forecast.