Western & Central Pacific

In somebody else's water

Many happy returns

One skipjack's year, logged the way we'd log a season of Brumley bike rides… 😂 … except the climbing goes downward, the terrain is more gnartastic, and the segments are sovereign states.

Skipjack · Katsuwonus pelamis 49 cm a year at sea

Tag 1977 · depth, all 365 days

drag across to read any day

shallowest that day — night
deepest that day — day
thermocline, 90 m

In somebody else's water—d

Elapsed365d

Avg depth—m

Max depth—m

Vertical distance—km

Jurisdictions8

Eight points a day, against every 30 seconds on the archival tags in Aoki et al. — 360 times finer. Vertical distance sums the drops between consecutive points, so it climbs with the sampling rate. The figure above is a shape, not a measurement.

The animal

Skipjack tuna

Katsuwonus pelamis (Linnaeus, 1758)

Tag 1977 is one of these. Two names, two features. Skipjack is for what it does — it skips across the surface chasing bait, which is the same shallow water it sits in every night on the profile above. Striped tuna, its other common name, is for the bands along the belly. The fish answers to its behaviour or its markings, depending on who is asking.

See it on iNaturalist

Photo © Clinton Duffy, CC BY-NC, via iNaturalist

Eight days, three sets of rules · days 175–182

The fish does the same thing on every one of these days: shallow at night, past the thermocline and back through the day. Underneath it, the rules change twice.

night
thermocline, 90 m
bands along the top: whose water

Segments

The same year, re-cut by whose water it was in. Tag 1977 chose this itinerary on purpose: it had got hold of the literature, formed views on who sells fishing days to whom, and set out to inspect all eight signatories in person. Call it fieldwork. It never once noticed a border, which does not stop it having opinions about them.

departs Solomon Seaa year later

Water	Days	Avg depth	Vessel days sold

What the fish knows perfectly well

That the water it crosses is sold by the day rather than the tonne, that eight countries agree the year's total between them, and that they set the price together against fleets from much larger economies. It has read up on all of it.

That its 81 days in Kiribati were, on somebody's ledger, inventory.

That the two stretches where nothing is owed to anyone are not the free-for-all they look like. The same eight states closed roughly 4.5 million square kilometres of high seas to any purse seiner licensed in their waters, so the boats that had been following it could not follow it in there.

And that it is, collectively, the third most exploited fish on earth — about 2.8 million tonnes a year, eleven years running, 84% of the region's skipjack catch taken by the one gear this whole arrangement exists to meter.

None of which it is ever consulted about, and none of which changes its mind about where it goes next.

Comments & kudos

Purse seiner · day 63

saw you out there. couldn't stay — out of days.

Tag 1974 · day 130

you actually went to Palau 🤨

OBS

Observer, WCPFC · day 205

Drifted past a FAD and kept going. Never seen one of you do that.

KIR

Licensing office, Tarawa · day 262

81 days in our water and not one of them paid for. Invoice to follow.

Tag 1977 · day 365

One continuous swim, and not a single border in the water.

Tag 1971 · today

thought you said two laps? Happy Birthday, Elizabeth.

Kudos from 3 purse seiners, 1 observer, 8 fisheries ministries, and everyone who has ever tried to govern something that keeps moving.

Non-birthday fine print

Blow out your candles first! 🎂 None of this is urgent, and none of it is on the test. It is here because the reader is the sort of person who checks.

The fish is a composite. There is no tag 1977: the real IDs in the tagging study this page leans on run 411 to 3969, and 1977 is not among them. Neither is 1971, further down. Tag 1974 is. That one was a real fish: tagged off Japan on 1 July 2013 at 48 cm, recaptured 25 days later 395 km up the coast, having grown to 51.3 cm. It is in no position to comment, and it did not. The daily depths, the day counts and the order of the legs are generated to be plausible, not transcribed from a real animal — drawn from a fixed seed, so it is the same fish on every visit, but a designed one.
The route is a tour, and it is meant to be. Tag 1977 calls at all eight member states because eight is how many countries run the scheme. No real fish does that: skipjack cover less ground than yellowfin or bigeye, and their movement is dominated by east–west shifts with the warm pool rather than a grand circuit. The pace, at least, is honest — about 30 km a day, against a published 30.1 ± 14.4 km/day for skipjack in the slowest part of this region. So the speed is a real fish's; the wanderlust is not.
It is not the same ocean. Those tags were on 38 fish off Japan, between about 20° and 40°N, on a northward migration into temperate water. This fish is in the equatorial Pacific, thousands of kilometres south. The shape of the behaviour is borrowed from that work; the place is not.
Some numbers are firmer than others. The shape of the dive profile follows published behaviour and the maximum depth sits inside the range that study reports. The durations — the year, and how many of its days fall in each country's water — are composition, not measurement. 49 cm is a real size: the 38 fish carrying those tags were 38–49 cm fork length, so this one is exactly the largest of them.
The behaviour is real, and specific. In that study, one tagged fish averaged 97–152 m through the day against 50–96 m at night — the diel swing that makes this profile look like terrain. The deepest individual reached 286 m at 8.3 °C, on 18 May 2013. The dashed line here sits at 90 m because that study used the gradient between 0–10 m and 90–100 m as its indicator of the thermocline. Not every fish in that study did it, though, and that is the paper's actual finding: in the warm south they work the whole water column, while in the cooler north they hug the surface and dive only occasionally. This page draws one rhythm and repeats it all year.
The rules are real. The Parties to the Nauru Agreement are eight sovereign states — Federated States of Micronesia, Kiribati, Marshall Islands, Nauru, Palau, Papua New Guinea, Solomon Islands and Tuvalu — with Tokelau a ninth participant. Everything this page says about how their days are set, split, sold and traded is how the scheme actually runs.
The argument is someone else's, and better made in Rights-based management in the Western and Central Pacific Ocean tuna fishery (Marine Policy, 2013), Science, scale and the frontier of governing mobile marine species (International Social Science Journal, 2018), and Unsettled Sovereignty and the Sea (Annals of the American Association of Geographers, 2018).

Sources: PNA on membership, the scheme and the high-seas closure, Kiyofuji et al., Northward migration dynamics of skipjack tuna (Progress in Oceanography, 2019) for the daily displacement rates, and Aoki et al., Surfacing and diving behavior … of skipjack tuna in the western North Pacific (Frontiers in Physiology, 2025) on diving behaviour.

Made for a geographer who studies things that will not hold still · ppr3.com