The High Seas Treaty

Two hundred nautical miles off Newfoundland, or roughly 370 kilometers, Canada's exclusive economic zone stops. The line cuts across the Grand Bank, leaving two pieces of it outside. Fishermen call them the Nose and the Tail, after the shape of the bank on a chart, and they matter because the water over them lies outside Canada's zone. Cod, redfish, Greenland halibut and shrimp cross in both directions, and there is no practical way to mark a boundary in open water. Captains know exactly where the line is, because their licenses and their quotas change when they cross it.

Farther east, about 560 kilometers off St. John's, an underwater plateau rises from the deep Atlantic to within 150 meters of the surface. This is the Flemish Cap, roughly 200 kilometers across, cut off from the Grand Bank by a channel more than 1,100 meters deep. A slow clockwise gyre sits over it and holds eggs and larvae on the bank instead of letting them drift off, so its cod and redfish are their own populations, fished since the 1950s. It is a self-contained place, and the water over it lies beyond Canada's exclusive economic zone.

That is the geography the High Seas Treaty works in. Its formal name is long, the Agreement under the United Nations Convention on the Law of the Sea on the Conservation and Sustainable Use of Marine Biological Diversity of Areas Beyond National Jurisdiction, and everyone calls it BBNJ, for Biodiversity Beyond National Jurisdiction. Governments adopted it on June 19, 2023, after nearly two decades of talks. Then it needed sixty countries to ratify before it took effect. Countries that ratify become Parties, and together they form the Conference of the Parties, the COP, which is the body that makes decisions under the treaty. Palau went first, on January 22, 2024. Sierra Leone became the sixtieth on September 19, 2025, which started a 120-day clock, and the treaty entered into force on January 17, 2026.

None of this was ungoverned before then. The Law of the Sea Convention, in force since 1994, is the closest thing the ocean has to a constitution, and it set the rules for navigation, resource use and protection of the marine environment. The seabed is governed separately from the water above it. Where it lies beyond any country's continental shelf, the International Seabed Authority administers the minerals in it, but a country's shelf can run well past 200 nautical miles, and Canada has filed a claim over roughly 1.2 million square kilometers of Atlantic seabed. Off Newfoundland the water can be high seas while the bottom underneath it is not. The International Maritime Organization regulates ships. Regional fisheries organizations set catch limits and close areas to fishing, and Canada belongs to seven of them, including the Northwest Atlantic Fisheries Organization, NAFO, which covers the high seas off Newfoundland. NAFO has closed four seamount areas and six sponge areas to bottom fishing, along with a set of coral closures, several of them on and around the Flemish Cap. What it is protecting is sponge, sea pens and cold-water coral, slow-growing animals that stand up off an otherwise flat seabed. Researchers who mapped the sponge estimated 231,140 tonnes of it, and calculated that the sponge inside the trawl tracks would be gone within a year if the fleet fished it directly. By 2016 the closures covered 380,511 square kilometers. Canada proposed one of them itself, a 239 square kilometer sea pen closure on the Flemish Cap.

So protected seabed already existed off Newfoundland, years before any biodiversity treaty. But those closures protect the bottom from fishing and nothing else. A fisheries organization can close the seabed to trawling and can do nothing about a ship passing overhead. The IMO can route ships and can do nothing about fishing. The Seabed Authority licenses mining and does neither. What did not exist was a way to protect a place for its biodiversity across everything happening in it. That gap is what the treaty was written to fill.

Governments have committed to protecting 30 per cent of the ocean by 2030. Two-thirds of the ocean lies beyond national jurisdiction, and about one per cent of that is protected now. Coastal waters on their own cannot carry a network that represents the range of habitats the ocean holds, because seamount chains, migratory corridors and open-water ecosystems sit mostly outside them.

The treaty's answer is a procedure for creating protected areas on the high seas. No area has been proposed under it yet, so take the Flemish Cap as a walk-through. Only a Party can start the process. A Party, or a group of them, submits a proposal to the secretariat under Article 19, giving the boundaries, what it is meant to conserve, and a draft management plan setting out the measures. The secretariat passes it to the Scientific and Technical Body, which screens it. Consultation follows, open to anyone with an interest, including Indigenous Peoples and local communities holding relevant knowledge, and including NAFO and the IMO, whose members already work there. The COP then decides, working from the final proposal, the consultation record and the scientific advice. Decisions are meant to be taken by consensus, and if that fails the COP votes by two-thirds that all efforts have been exhausted, then decides by a three-quarters majority. At the end of it the Flemish Cap would be a designated marine protected area with a management plan.

Now read the management plan. Suppose it says bottom trawling stops on the Flemish Cap. The treaty requires the COP to respect the authority of the bodies already governing activity in an area, and where a measure falls within the authority of one of them, the COP may only recommend it, not decide it. Fishing on the Flemish Cap is NAFO's, so the COP would write to NAFO and ask. NAFO began closing vulnerable high-seas habitat to bottom fishing in 2007, starting with four seamount areas, and closed a further group of coral and sponge areas including several around the Flemish Cap in 2010. It did all of that on its own authority and could close more tomorrow without any of this. The treaty can establish the area and adopt conservation measures of its own. What happens inside it still belongs to whoever governed it before. Fisheries bodies cover most of the high seas, and some already protect vulnerable habitats. Their powers remain tied to fisheries. The treaty adds a process for considering biodiversity protection across activities.

Article 23 adds a second limit. A Party that objects to a COP decision is not bound by it. The treaty sets out the grounds an objection can rest on, requires the Party to explain itself, expects it where practicable to do something equivalent instead, and makes it renew the objection every three years. It is a narrow door. It is still a door.

The same pattern runs through environmental impact assessment. The obligation falls on Parties, for activities under their jurisdiction or control, and the trigger is low. A Party must screen a planned activity when it may have more than a minor or transitory effect on the marine environment, or when nobody knows what the effects would be, and if that screening turns up reasonable grounds to believe the activity may cause substantial pollution or significant harm, a full assessment follows. The Party conducts it. The Party then determines whether the work may proceed, taking the assessment into account. Before authorizing it, the Party must determine that it has made all reasonable efforts to prevent significant adverse impacts, taking mitigation or management measures into account. The COP may offer advice, if asked. The treaty does not transfer the authorization decision to the COP. It also steps back where another body is already assessing. Ocean fertilization, for instance, sits under the London Protocol, which has its own assessment framework for marine geoengineering, and Article 29 provides that where an activity has been assessed under another instrument meeting the treaty's conditions, no separate BBNJ assessment is required. What survives in that case is publication. The report goes on the clearing-house mechanism and the activity has to be monitored, which is also what happens at the end of a full BBNJ assessment: the screening, the draft, the final report and the decision, with its conditions, all go up where other states can read them and comment.

Marine genetic resources are the one part of the treaty that was meant to produce payments from commercial activity. Organisms living at depth, around vents, in the cold and the dark, have evolved enzymes and compounds that work under conditions no laboratory organism tolerates, and a handful have become the basis of commercial products. So the treaty tries to attach a claim to material that belongs to no one. What it requires now is notification. A Party collecting genetic material beyond national jurisdiction has to file before the collection and again after it, and the material carries a standardized identifier so a sample can be traced to where it came from. Ireland filed the first of these notifications on July 16, 2026. Because the clearing-house is not running yet, the UN's ocean law division took it directly and gave it the reference BBNJ-MGR-TEMP-2026-001, a placeholder until the real system can issue a proper one.

The funding comes from governments. Developed Parties owe an annual contribution to the treaty's special fund, set at 50 per cent of what that Party is assessed for the COP's budget, meaning its share of the running costs. That is meant to be temporary, replaced once the COP adopts real terms. Until then, benefit sharing from marine genetic resources is calculated from a UN budget, with no connection to any organism, any patent, any product or any company, and no money has moved, because the budget it is calculated from has not been adopted. Alongside it sit a voluntary trust fund that pays to get delegates from developing countries into the room, and the Global Environment Facility trust fund, which has authorized up to USD 34 million for ratification and early action. A Force of Nature, the nature strategy announced March 31, 2026, commits CAD 24.4 million over five years and CAD 1.8 million a year after that to ratify the High Seas Treaty and advance the commitment to protect 30 per cent of the world's oceans. That is a budget line for legal work, scientific capacity and delegations, not for funding a marine protected area.

I looked for evidence that the treaty has changed a lending or investment decision and did not find any. That is not surprising this early. What the treaty has produced so far are obligations on states, and they are procedural: screen, assess, notify, publish. If it eventually reaches a borrower, the likely channels are the ordinary ones, a cost, a delay or a closed area, arriving as a rule from the body that regulates the activity rather than from the COP.

Canada signed the agreement on March 4, 2024. Two and a half years on, it has not ratified. As of September 2026 it is a signatory, not a Party. It is not bound by the treaty's obligations, and its main duty in the meantime is not to act in a way that would defeat the agreement's purpose. It cannot exercise the rights of a Party, including proposing an area or voting at a COP, unless and until it becomes a Party. Another Party could propose the Flemish Cap. Ratifying normally runs through Cabinet approval, then tabling in the House of Commons with an explanatory memorandum, then a waiting period of at least 21 sitting days, then an Order in Council and deposit at the UN. The policy allows exceptions to tabling, and the practice is that Canada does not ratify until it can implement a treaty in Canadian law. Fisheries and Oceans Canada leads the file and has said the aim is to ratify as soon as possible, without giving a date. The United Kingdom signed in September 2023, introduced its implementing bill two years later and deposited on July 10, 2026. Germany passed its legislation through the Bundestag in February 2026 and is not yet a Party either. The United States, Australia and New Zealand have also signed without ratifying.

Eight months after entry into force, the treaty has about ninety Parties and one working procedure. The first COP is expected to meet at the United Nations in New York in January 2027. Until then the Division for Ocean Affairs and the Law of the Sea is running things on an interim basis. The third and final preparatory session closed on April 2, 2026 without agreement on the COP's rules of procedure, the terms of reference for the subsidiary bodies, or the financial rules, and sent all three to the COP. The Scientific and Technical Body has not been elected, so there is nobody to review a proposal even if one arrived. Chile is developing a proposal for the Salas y Gómez and Nazca ridges and West African states are working toward a draft. None has been submitted, and no high seas protected area exists under this treaty.

This is the ordinary sequence, and roughly what finance following governance would predict. The reporting from the final preparatory session suggests much of the delay came from long-running disputes over voting rules and representation, carried into this forum from others. What COP1 has to settle in January 2027 is whether the institutions it builds can turn a biodiversity decision into measures that the fisheries, shipping and seabed bodies will carry through. On the Flemish Cap the question is plainer: whether an area protected under this treaty ever changes where a trawl goes.