The sea will not stay where you treated it.
Everything about coastal work follows from exchange. Water arrives from next door twice a day, which is why what we contract is a zone with a boundary and a season, not a visit to a site.
The refusal comes first
We do not work in the open ocean, and the reason is not modesty.
Some of what goes wrong in seawater is circulation. Circulation is not a site, and no fleet that has ever been built or costed will touch it.
Two seagliders spent eight months working the Gulf of Oman and mapped an oxygen minimum zone described as larger than Scotland, intensified since the 1990s, in water that had gone almost unsampled for half a century. We reached that through a research summary rather than the paper, so treat the description as indicative — and the square-kilometre figures quoted alongside it elsewhere are not in the summary, which is why they are not here.
That feature is a property of an ocean basin. It moves vertically between seasons. It is not remediable by anybody, at any price, and a company willing to imply otherwise in a tender document would tell you everything you need to know about the rest of its claims.
What is workable is the other kind of coastal water: semi-enclosed, engineered, or simply poorly exchanged. Bays and creeks. Lagoons, including the ones somebody dredged into existence. Port and marina basins. The approach in front of a desalination intake. The lagoon inside a reef. Water somebody can point at and say that stretch is mine to answer for.
The variable that sets the price
How fast the basin replaces itself decides everything else.
An exposed headland with a strong tidal stream flushes itself constantly. It rarely has a problem, and if it does, treatment would be washed out before it did any measurable work. A creek with one narrow opening, or a resort lagoon with a sill across its mouth, may hold the same water for weeks. That is where the money is, and it is also where the failures are, because those two facts are the same fact.
So the first question at a site read is not what is wrong. It is how long the water stays. Flushing time sets the treatment interval, the fleet size, the gas volume and the price, and it does so before anyone has looked at a single water-quality reading.
The second question is what arrives from next door. A zone is never sealed. A bloom four miles outside the boundary is inside it on the following tide, an outfall upstream of the zone is functionally inside the zone, and a neighbour's bad season becomes your bad season without anyone signing anything. This is why the base subscription covers proactive protection and adjacent Deep Strikes rather than callouts — intercepting an event in the water next door is cheaper than treating it after it has become yours.
And the third is draught. Pale water over sand and reef decides the vessel tier long before the treatment specification does, which is a boring constraint that has settled more programme designs than any of the interesting ones.
Salt changes the chemistry
Seawater caps the oxidant duty, and the cap is not negotiable.
Fresh water carries no meaningful bromide. Seawater does, and ozone reacts with bromide roughly eighty-three times faster than with chloride; the hypobromite that results can go on to form bromate. Formation depends on ozone concentration, contact time, pH and temperature, and each of those can be constrained — at a cost in disinfection efficiency that has to be accounted for rather than hidden.
The practical effect on a programme is that dose and contact time stop being panel settings and become monitored limits with a permanent reporting obligation. On a reef site the ceiling is lower still, and where a site will not tolerate an oxidative duty at all, it does not get one.
This is also why the pond-scale freshwater result described on the lakes page does not simply transfer to a bay. Different chemistry, different limits, different monitoring. The full treatment of the constraint sits on the systems pages, and the unresolved parts of it are listed on the evidence page rather than left for a reviewer to discover.
What sits on a shared sea
Nobody owns coastal water, so the buyer is whoever has something on it.
A lake has an owner and a river has a committee. A bay has neighbours — and the one who writes the cheque is usually the one whose asset stops working first.
Desalination intakes
Gulf desalination is not a convenience: World Bank figures put GCC states at nearly half of global desalinated output while holding under one per cent of the world's population. A bloom that never reaches a beach can still take a plant offline.
Oman, July 2023. Nama Water Services reported a red tide bloom significantly disrupting Barka's main plant, and reserve wells and plant interconnection carried the load while seawater quality recovered. That account comes from regional reporting, so treat the detail as indicative.
Aquaculture and fisheries
Pens concentrate oxygen demand into exactly the water with the least margin, and the loss is not gradual. A warm still night is the whole event.
Operators in this sector price a bad night better than anyone else we deal with, which makes them the easiest counterparty to write viability parameters with and the hardest to impress with anything vague.
Ports and industrial basins
Poor exchange, warm discharge water, ballast operations and a berth that has to stay working. The failures here are compliance failures as often as ecological ones.
And when something is spilled, the numbers stop being ecological entirely. BP's cumulative pre-tax charge for Deepwater Horizon reached US$67.0 billion by the end of 2018, as filed with the SEC.
Coastal tourism and development
Nothing breaks. The bookings simply stop, and the following season starts from a worse position than the one that was actually damaged.
Reef-adjacent development adds a second obligation, usually written into an environmental statement years before anybody worked out who would deliver it.
What a zone cannot hold back
Drawing a boundary on water is only honest if you say what crosses it anyway.
Heat is the clearest case. NOAA Coral Reef Watch measures accumulated thermal stress in degree heating weeks, and reef-wide bleaching with mortality among heat-sensitive corals becomes likely at 8. A regional marine heatwave five hundred kilometres wide does not care about a service-zone boundary, and cooling and shading work inside a lagoon is a local intervention against a global forcing. We say that in the site read, in writing, before anybody signs.
Acidification is the same argument at a slower speed. Ocean chemistry is not a zone-scale variable, and nothing in a contracted season alters it.
What a boundary does hold is a duty: a defined volume, a defined period, an agreed set of readings, and a named institution outside this company holding the same live feeds our crews steer by. That last part is the one we would keep if we had to give up all the rest. A supplier with the only instruments, the only data and the only report has told a coastal authority a story it cannot check.
Reef-specific duties, their ceilings and their published limits are set out by ReefHealer; the regional evidence base for the Gulf and the Red Sea, including how uneven it is, sits with Marine Prosperity.
Systems that work this water
One hull, several duties, and a licence behind all of them.
Alarivean operates the systems rather than owning them. Which duties a coastal zone actually needs is decided at the site read, not in a catalogue.
Answered without hedging
Three about coastal work
Do you work in the open ocean?
No. Features like the Gulf of Oman's oxygen minimum zone are circulation rather than sites, and nothing that can be put on a hull touches them.
The platform is for semi-enclosed and engineered water, where exchange is slow enough that treated water is still near where it was treated when somebody comes to measure it.
How is a service zone boundary decided?
Exchange rate first, ownership second. We want a boundary the water respects and an asset holder who can point at it.
Where those two disagree — and on a busy coastline they often do — the honest answer is a smaller zone with a real boundary rather than a larger one drawn to match a map.
Has any of this been demonstrated at coastal scale?
Not yet, and we would rather write that sentence ourselves than have a reviewer write it about us. The published validations of this technology class are pond-scale and ballast-tank-scale.
Our own field record is a Florida permit for a bounded red tide trial, documented independently by the Sarasota Bay Estuary Programme and by the nonprofit START. The estuary programme's position is that open-water efficacy has not been demonstrated, and they want to be the ones holding the instruments when it is tested. So do we.
Start with the boundary
Which stretch of coast is yours to answer for?
The basin or the shoreline, roughly how long the water sits in it, what goes wrong and in which months, and which asset stops earning when it does. Bring your own scientists to the site read and let them push on the answers.