Reading the gap between two islands before we commit to crossing

Part 01

Wind against current in a narrow sound

The sounds between our islands are narrow, shallow at the ends and deep in the middle, and water moves through them far harder than the tidal range suggests. Tides here are small, often under twenty centimetres, so for our first year we treated current as something that happened to paddlers in other countries. Then we crossed a sound on an August afternoon with a fresh southwesterly behind us and found a metre of steep, square water piling up in the middle of a gap we had paddled through flat two weeks earlier. The current was not tidal at all. Two days of westerly wind had pushed water into the fjord system and it was draining out through every narrow place it could find.

What we watch now is the direction of the last two days of wind rather than the number on the tide table. After a long westerly, the sounds run out hard and any wind with east in it will stand the water up. After a week of easterlies the same gap is lazy. You can see it from the launch if you know where to look: a line of disturbed water across the narrows, slightly paler, sometimes with a few metres of foam on the downwind side. From a kayak seat that line is invisible until you are a hundred metres from it, which is why we walk up the rocks behind the slipway before loading, not after.

The number we settled on is unromantic. If the wind is against the run of the water and the forecast mean is over eight metres per second, we do not cross an open gap wider than about a kilometre. Below that we go, in a tight group, and we accept that the middle third will be unpleasant. Between eight and eleven we look for a route that crosses in two shorter hops using whatever rock sits in the middle, even if it adds three kilometres. Above eleven we paddle the sheltered side of something and go home earlier than planned.

Part 02

The fetch problem nobody mentions in the car park

Forecast wind speed tells you almost nothing on its own, because the sea you actually sit in depends on how far that wind has travelled over open water before it reaches you. Seven metres per second with a hundred metres of fetch is a ripple. The same seven metres per second with forty kilometres of open water to the west is a short, steep metre swell with breaking tops, and it arrives at your beam while you are turning around a headland with rock on the other side. We have been surprised by this more times than the arithmetic deserves, because the wind in the car park feels the same in both cases.

The map exercise that cured us takes ten minutes at the kitchen table the night before. Take the forecast direction, draw a straight line from each planned exposed section back upwind, and see where it hits land. Under about five kilometres of fetch, the sea will be choppy but workable. Between five and twenty it becomes a real swell that a loaded boat handles fine and an empty one handles badly. Over twenty kilometres in anything above moderate wind, we choose a different coast. Writing the fetch figures on the trip sheet beside the forecast turned a vague nervousness into something two people could argue about with numbers.

The other half of the problem is refraction around the outer skerries. Swell that arrives from the west bends around the islands and closes in behind them, so a channel that looks sheltered on the chart gets waves from two directions at once, a few seconds apart. That is the water that has made most of our group seasick, and it is the water in which our least confident paddler capsized in 2021. We now treat the lee side of an outer island as a place to pass through rather than a place to stop and eat lunch.

Part 03

The turn-back rule and who is allowed to use it

For the first two seasons our decisions were made by whoever sounded most confident, which in practice meant whoever had the newest boat. It produced three trips that were genuinely frightening for at least one person, and the person in question said nothing until we were back at the cars. Since the autumn of 2020 the rule is that any member of the group can end a crossing before it starts, without arguing a case and without being asked to justify it later over coffee. Nobody has abused it. In four seasons it has been used eleven times, and on seven of those days the wind came in above forecast by early afternoon.

The rule works because it is exercised at a fixed point rather than in the middle of the water. We stop at the last piece of shelter before the open section, get out if there is somewhere to land, and each person says one sentence about whether they want to cross. Doing it out loud matters. Standing on a rock and saying 'I would rather not' is easy in a way that shouting it across twenty metres of broken water is not. Once we are committed, the group holds together and the decision is not reopened until the far side.

The part we still get wrong is the return leg. Everyone judges the outward crossing carefully and then treats the way home as a formality, even though the wind has usually built by then and everybody is four hours more tired. Two of our worst hours on the water were return legs across gaps we had crossed comfortably in the morning. The fix we are trying now is to set a time at the launch, before anyone gets in a boat, after which we take the long sheltered way back regardless of how good the day has been.

Part 04

October 2022, the crossing we should not have started

The forecast was six metres per second from the southwest, easing. Water was eleven degrees, air fourteen, and five of us set out to cross a gap of about two and a half kilometres to reach an island with a good landing on the north side. We started at half past two, which was already late for October, and within fifteen minutes the wind was clearly not easing. By the middle of the crossing it was gusting to something none of us could estimate honestly, and the sea had gone from swell to short breaking chop that filled cockpits through the spray deck seam.

Nobody capsized, which was luck rather than skill. What actually saved the crossing was that the group stayed inside a boat length of each other and nobody tried to sprint for the shelter of the island, which would have strung us out across four hundred metres of water. We landed on the wrong side of the island, walked the boats over a low saddle of rock, and paddled home in the lee with headtorches on for the last twenty minutes. The write-up that evening runs to two pages and is the most useful thing in the log.

Three errors, in order of seriousness. We started an exposed crossing after two in the afternoon in October, when the light and the temperature are both running out. We believed the word 'easing' instead of watching the water. And we had no agreed abort plan for the second half of the crossing, so if conditions had got worse there was no answer to the question of where we were going. All three are cheap to fix and none of them are about paddling skill, which is the pattern in most of our bad days.

Part 05

What goes on the sheet the night before

The trip sheet is one side of A4 in a plastic sleeve and it has not changed much since 2019. Forecast wind direction and mean speed for three times of day, the fetch figure for each exposed section, water temperature from the last trip, sunset time, the two alternative landings, and a note of who is bringing the spare paddle and who has the first aid roll. It takes about twelve minutes to fill in and it is the single habit that has improved our judgement most, mostly because writing a guess down makes it easy to check afterwards whether the guess was any good.

We compare the sheet to reality the same evening. Forecast against felt, planned landing against actual landing, planned time against actual time. Over five seasons the pattern is boringly consistent. Our forecasts are usually about right for the morning and about two metres per second optimistic for the late afternoon, our distance estimates are fine, and our time estimates are twenty per cent short because we never count the standing about. Knowing our own bias is worth more than any amount of reading about somebody else's coast.