The number, the notice and the signature.

There is no international harmful algal bloom threshold. There are national and state ones, a small number of World Health Organization guidelines, and a large amount of local practice set by a committee in the nineteen-nineties and standing unchanged since.

1. Three different objects, all called a threshold

A guideline value is a concentration in a matrix, derived from toxicology, above which harm is expected over a defined exposure. It is a health number and knows nothing about your treatment works.

WHO publishes three for microcystins, all derived from studies using microcystin-LR and applied to total microcystins. The short-term drinking-water value indicates, in WHO's description, the extent to which the lifetime value may be exceeded for up to two weeks while treatment is augmented. A document carrying only the lifetime value lacks the one a duty engineer needs at four in the morning, and will produce a do-not-drink notice where WHO's own guidance does not require one.

An action level is what your organization does at a given reading, and it belongs to you. It sits below the guideline value by a margin that reflects your sampling frequency, your laboratory turnaround and how fast your source water can change. A utility sampling weekly needs a wider margin than one sampling daily. Setting that margin is the directorate's management decision.

A trigger is the observation that starts a process — a satellite anomaly, a duty operator's report of discolored water, a cell count crossing a screening level. Triggers start testing. They close nothing on their own.

Programs get into difficulty when all three sit in the same column of the same table.

2. Drinking water

Cyanotoxins in raw water give the shortest chain and the worst consequences. Toledo in August 2014 is the reference event: microcystin in treated water, and roughly half a million residents told not to drink from the tap.

The science was there. The gap lay between an environment agency's source-water monitoring and a utility's treatment decision at four in the morning, on a result hours old and describing different water.

Four things belong in a drinking-water annex and are frequently missing. A named action level for raw water, lower than the finished-water guideline. A written treatment response for each level, agreed with the regulator in advance. A standing arrangement for who may issue a do-not-drink notice outside office hours. And an explicit statement of which WHO value the notice is judged against, lifetime or short-term, because the two are twelve-fold apart and the difference is a city's water supply.

3. Shellfish, where the number lives in the tissue

Filter feeders concentrate toxin from water that would harm no swimmer. The exposure route is the meal, and the number that matters is in the tissue.

Domoic acid, from Pseudo-nitzschia blooms, causes amnesic shellfish poisoning: vomiting, diarrhea, headache, cramps and disorientation at low doses; trouble breathing, seizures, coma, permanent loss of short-term memory and death at higher ones. Onset can lag the meal by nearly forty hours, which makes case-finding hard. Contaminated shellfish look, smell and taste normal.

California's health department names older people and people with kidney disease or diabetes as at greater risk of severe illness, and adds pregnant women, infants and children as groups who should take advice before eating shellfish at all. A notice that says only "the elderly" has under-warned three of those five.

So the closure rests on tissue testing, and the laboratory queue sets how quickly a fishery reopens. Cell counts decide when to test. Rarely, on their own, when to close.

Only toxin classes with a published figure or published clinical description behind them are tabulated. Saxitoxin, ciguatera, brevetoxin and the rest are real and important, and stay untabulated for want of a source that would stand up to a reviewer. Half-remembered limits are worse than none.

4. Recreational water

Bathing advisories are the most visible part of a program and the least standardized. Jurisdictions differ on whether the trigger is a cell count, a toxin concentration, a scum observation or an officer's judgment, and on whether the notice is advisory or prohibitive.

For cyanotoxins, WHO's recreational value is a starting point: a health number for incidental ingestion while swimming. It is no scum-avoidance rule, and says nothing about the dermal and inhalation exposures that dominate complaints on a busy beach.

No further bathing figures are published. Marine toxins in bathing water are handled differently in almost every jurisdiction we have read, and a figure lifted from a neighboring country without its sampling regime, species assemblage and legal framing is a threshold nobody can defend at an inquiry.

What travels between jurisdictions is the process: a defined observation trigger, a defined confirmation step, a named signatory, a published notice explaining the risk in plain language, and pre-agreed rescission criteria.

5. The step that goes unrehearsed

Every program document we have read describes how an advisory is issued. Perhaps a third describe how it comes off.

A closure that outlives the hazard costs as much as one never justified, and it is harder to defend, because the operator can see clear water. Rescission criteria written mid-season are written under pressure from people with a commercial interest in the answer.

Write them in the winter. Consecutive clear results, a minimum number of stations, a minimum elapsed time, and a named person who signs the withdrawal.

Two dates matter downstream and the second is this one. The hour a notice takes effect opens the count of interrupted trading days for anybody inside the line; the hour it is withdrawn closes that count. Where no liability regime attaches to water condition, those interrupted days and the harm to plant and stock behind them are the two headings an operator's whole event has to fit into, so the interval between the two signatures is doing arithmetic on other people's books whether or not the directorate intends it to. Criteria fixed in advance, and sampling read by a laboratory with no fee riding on the outcome, are what let a notice come off on the day the water clears and stand up if anybody asks why.

6. The decision chain, in order

1. Trigger Something is observed: an anomaly, a report, a count over a screening level. This step is cheap and should be generous. A false trigger costs one sample; a missed one costs a season.
2. Confirm A sample is analyzed by a laboratory whose method and detection limits are stated in the program document. An unwritten method makes the result indefensible.
3. Classify Which organism, at what abundance, and — for the exposure routes that need it — how much toxin in which matrix. This is where the queue lives, and where most programs lose their days.
4. Decide Against the action level your organization set in advance, in writing, with a name against it. That level is neither the guideline value nor the mood of the meeting.
5. Publish A notice with a named signatory, a plain statement of what is and is not affected, and the criteria for lifting it. The last part stops the telephone ringing.
6. Rescind Against the criteria agreed before the season. Then record what the event cost; the ledger is what moves a budget.

7. Where the chain runs out

Every step in the chain is about knowing and telling. None of it changes the water.

That is the boundary of an advisory regime. In-water intervention is a scheduling question in nutrient policy: a standing instrument measured in weeks, beside catchment measures measured in decades. It is contracted as an ongoing subscription and runs through the year, so a stressor is never allowed to accumulate into the event a directorate has to procure its way out of.

It removes no line from the chain. It adds one, because a treated water body needs verification as well as surveillance — and the treatment supplier should be the party asking for that clause and paying for the verifier.