The administration of a biological hazard
Somebody has to sign the closure notice.
A harmful algal bloom is a scientific event that arrives as an administrative one. An advisory to draft. A beach to close, a plant to re-rate, a harvest to suspend, and a public that wants a date for when it ends.
Why this is hard to govern
Four authorities, one bloom, no shared budget line.
In most jurisdictions the same event is watched by an environment agency, judged by a public-health laboratory, closed by a fisheries or municipal authority, and absorbed by a water utility that answers to none of the other three. Each has its own trigger, its own legal instrument and its own committee. None of them owns the water.
That division is not a scandal. It is how competence is organised, and it mostly works for chemical spills and sewage failures, where the source is a point and the responsible party has a name.
A bloom does not behave that way. It is diffuse, it is biological, it responds to weather, and its cause was distributed across ten thousand hectares of catchment over thirty years. The instrument that would fix it sits in an agriculture ministry. The instrument that has to answer the phone this week sits somewhere else entirely.
So the practical question for a programme is not who is to blame. It is which functions are covered, which are assumed, and which have no owner at all.
Programme anatomy
The functions a bloom programme has to cover.
Names and ministries differ by country. The functions do not. This is the checklist we use when reading somebody's existing programme before saying anything about intervention.
| Function | Where it usually sits | What it produces | How fast it acts |
|---|---|---|---|
| Surveillance | Environment agency, often with a university or institute contract | Cell counts, chlorophyll imagery, a sample archive | Days to weeks; faster inside a declared season |
| Toxin analysis | Public-health or food-safety laboratory | Toxin concentrations, reported by matrix | Days, and the queue lengthens exactly when it matters |
| Advisory | Health directorate | A public notice carrying a named signatory | Hours, once a defensible number exists |
| Closure | Fisheries authority, municipality or port | Suspension of harvest, bathing or navigation | Hours to days |
| Drinking-water response | The utility, under a regulator's standing conditions | Treatment change, blending, or an alternative source | Immediate, and expensive |
| Public communication | Whoever answers the telephone first | The version of events the public will remember | Immediate, whether you are ready or not |
| Nutrient reduction | Agriculture, planning and wastewater regulation | Catchment load targets, permit conditions, subsidy rules | A decade or more, if it holds through elections |
| In-water intervention | Usually unassigned | Nothing, in most programmes we have read | Not scheduled |
The final row is the reason this site exists. Everything above it is well established practice. The last line of the table is where the argument starts.
The ledger
What a documented bloom has cost, when somebody bothered to count.
Treasury officials ask for a number and receive an adjective. These are the events where a number exists in a public agency record, which is a much shorter list than the events that happened.
They are mostly United States figures for one unglamorous reason: NOAA publishes them as a series. Comparable series for the Gulf, the Mediterranean and East Asia are thinner, and the absence of a figure is not the absence of a loss.
| Event | Documented impact | Source |
|---|---|---|
| Maryland, 1998 | $81 million in lost seafood sales after a Pfiesteria event | NOAA Fisheries |
| Maine, 2005 | $7.2 million in direct harvester losses; $49.46 million in total lost income | NOAA Fisheries |
| Texas, 2011 | $10.3 million drop in oyster landings | NOAA Fisheries |
| Indian River Lagoon, 2011 | $235–470 million potential loss from seagrass damage — a modelled range, not a settled figure | NOAA Fisheries |
| Toledo, Ohio, 2014 | Roughly 500,000 residents advised not to drink tap water; $10.05 million in lost shoreline property-value services | NOAA Fisheries |
| United States West Coast, 2015 | $97.5 million in lost Dungeness crab landings, plus about $40 million in tourism spending lost to razor clam closures | NOAA Fisheries |
| Worldwide, 2000–2020 | Fish-killing blooms associated with more than US$8 billion in losses across the global event database | Reviews in Aquaculture, 2024 |
A University of Central Florida study put the 2018 Florida red tide season at $2.7 billion across the state, split roughly between the south-west and south-east coasts. We reached that figure through secondary reporting of the published paper rather than the paper itself, so we treat it as indicative and would not put it in a submission without checking the original.
The uncomfortable evidence
The Baltic did the hard thing, and the dead water stayed.
This is the single most important result for anyone writing a bloom or hypoxia strategy, and it is the one least likely to appear in the strategy.
Nine countries have run the HELCOM Baltic Sea Action Plan for two decades: agricultural regulation, wastewater investment, national input ceilings, multilateral negotiation. By the standards of environmental policy it is the most serious attempt anyone has made.
HELCOM's own indicator, published in October 2025, puts the result at an 11 per cent cut in normalised total nitrogen input and 32 per cent in phosphorus, measured against a 1997–2003 baseline. The Baltic Proper and the Gulf of Finland — the two sub-basins that most needed it — still fail their maximum allowable input for both nutrients.
Meanwhile the hypoxic area went from under 10,000 square kilometres before 1950 to more than 60,000 since 2000. Sediment holds phosphorus and gives it back under low oxygen. Stratification is stronger in a warming sea. The system's memory is measured in decades, and it is not finished remembering.
That is the argument for adding in-water capability to a programme, and it is the whole of the argument. Not that catchment work fails. That catchment work is slow, and the fisheries, beaches, intakes and public confidence being spent in the meantime are also public assets. Alarivean exists to work in that interval.
Before the first meeting
What in-water intervention is not.
Every director we have met has the same four suspicions, and they are all reasonable. Stating the limits first is cheaper than being caught by them in year two of a programme.
- It is not a substitute for catchment nutrient reduction, and a ministry that presents it as one will be held to that in a decade.
- It does not reduce the monitoring obligation. It increases it, because now there is a treatment to verify as well as a bloom to track.
- It is not a permit-free activity. Your regulator is still your regulator, and an operator who is vague about that is telling you something.
- It is not yet demonstrated at open-water scale. The published validations of this class of technology are pond-scale and tank-scale, and the field record is a Florida trial permit rather than a portfolio.
The evidence position is set out in full on seabreather.com, including the parts that do not help the sale.
The row with no owner
Assign the intervention. Then assign the people who watch it.
The table further up this page has a row with no owner. Give that row an owner and a second gap opens the same afternoon, and a strategy that closes the first without closing the second will not survive its first bad night.
Every other function in a bloom programme stands behind an institution with something to lose. The laboratory is accredited. The advisory carries a signature and a professional register behind it. The utility answers to an economic regulator that can fine it. That is what makes their outputs usable at an inquiry, and none of it happened by accident — it was built, function by function, over decades.
An intervention supplier arrives holding none of that. So the question a directorate is really asking is not whether the treatment works. It is sharper than that: if this goes wrong, who finds out, how fast, and are they somebody we are not paying?
Alarivean's answer is designed into the programme rather than fastened to the outside of it. Relevant local scientific expertise is engaged continuously, with access to every water-quality feed the operator works from and to whatever those institutions can bring alongside it, and with human and environmental safety ranked ahead of everything else in the risk controls. The company insists on this arrangement. An agency should insist harder.
What it changes in the document
The intervention line stops reading as a procurement and starts reading as a function with the same shape as the rest of the programme: an actor, an overseer, a data flow, and a defined route by which a problem becomes somebody else's knowledge.
In drafting terms that is one clause naming the oversight institution and its access, sitting immediately beside the clause naming the supplier. Not an annexe.
What it does not change
Your regulatory position, in any respect. Permitting, discharge conditions and the closure decision stay exactly where they sat before, and oversight substitutes for none of them.
Nor does it make the technology proven. It makes the evidence about the technology worth reading — a smaller claim, and a considerably more useful one.
The clauses this implies, in the form we would want to read them in somebody else's draft, are on the nutrient policy page.
Related reading in this network
Immediate, Significant, Scaled
Send us the programme, not the press release.
The water body, the season, the monitoring you already run and the decision you are currently unable to make. Alarivean returns an assessment of what in-water capacity would change in that jurisdiction — and where it would change nothing.