Six stressors, sorted by how much notice you get.
Not by how frightening they are. Warning time is the column that decides your contract structure, your rota and whether a mitigation exists at all — so it goes first, and everything else hangs off it.
Environmental briefings describe stressors by what they are. An operator needs them described by how much time there is between the first signal and the loss.
Start with the table. Everything below it is elaboration for the rows that turned out to matter.
| Stressor | Warning you actually get | Dominant cost type | In-water treatment |
|---|---|---|---|
| Harmful algal bloom | Days, if somebody is watching ocean colour and sampling downcoast | Production and stock | Applies inside a bounded volume |
| Hypoxia | Hours to days from instrumented profiles; none without them | Stock write-off | Applies directly, at depth |
| Thermal load | Days from forecast, but the response window is short | Derating and consumption | Applies only where the hot spot is local |
| Biofouling load | Weeks — it accumulates rather than arrives | Capital brought forward | Applies at the intake approach |
| Spill | None | Liability and disclosure | Recovery phase only, never response |
| Ballast transfer | Not applicable — it is a compliance state, not an event | Regulatory and permanent | Applies, with the strongest published evidence here |
Two of these give you days. Two give you hours. Two give you nothing, and belong in a different conversation entirely — which is why they are dealt with briefly at the end rather than padded out to match.
01 — Harmful algal blooms
The largest single-event losses on record, and they arrive on a current.
Advection is what makes this one awkward. The bloom forms offshore or downcoast and arrives on a current, so site-boundary thinking fails: nothing inside your fence caused it and nothing inside your fence stops it. Biomass loads screens and pretreatment. Toxins contaminate stock and shellfish. Then the material dies, decomposes, and strips the oxygen out of the same water — which is why row two of that table so often follows row one by about forty-eight hours.
Aerosolised toxin from some species reaches guests and staff onshore without anybody entering the water at all.
The concentrated losses sit in aquaculture. A 2016 bloom in Chile's Patagonian fjords killed roughly 40,000 tonnes of farmed fish, an estimated loss near US$800 million, and the peer-reviewed synthesis puts Japan's annual harmful-bloom aquaculture losses above US$1 billion. Wild fisheries take it more slowly and more publicly: $97.5 million of Dungeness crab landings on the United States West Coast in 2015, plus about $40 million of tourism spending tied to razor clam closures, and $10.3 million off Texas oyster landings from a single 2011 event.
Does in-water work touch it? Yes, inside a bounded volume. Bloom biomass in an intake approach, a pen block or a basin is a finite organic load in a finite volume, and that is a treatable object. Whether it comes back next month depends on nutrients arriving from land, which no vessel programme changes.
Chilean and Japanese figures come from a peer-reviewed synthesis of the global event database and are held at medium confidence.
02 — Hypoxia, with thermal load in front of it
The one that kills stock overnight and leaves no residue.
These are one problem on two clocks. Heat arrives first and is mistaken for weather; oxygen arrives second and is mistaken for disease.
Warmer water holds less dissolved oxygen and stratifies harder, so a heat spike is an oxygen crash with a few days of delay built into it. Warm, fresher water caps colder, saltier water. Exchange stops. Everything decomposing below the interface consumes what oxygen is left, and a stocked pen concentrates that demand in precisely the water with the least margin.
The failure is not gradual, which is what catches people. Stock holds, holds, holds, and then a still warm night takes the column under and the write-off is floating by morning. For an intake the same condition arrives sideways: anoxic bottom water is chemically aggressive and biologically loaded, and drawing it changes what your pretreatment is being asked to do.
Discharge-warmed margins, shallow basins and enclosed lagoons hold heat longer than the open coast beside them. Cooling assets therefore manufacture part of their own exposure, which makes the internal accounting awkward and the engineering interesting.
Does in-water work touch it? Directly, and this is the strongest case on the page — oxygen delivered into the layer that lost it moves a number on an instrument you already own. Surface aeration on a stratified column is largely decorative, which is the argument for putting the delivery array on a hull that can work at depth. Heat is conditional: breaking a thermal cap in a shallow basin or a lagoon treats the cause, and where the warm water is basin-wide nothing on a hull changes the driver.
Set the alarm against the right number
NOAA's 2 mg/L is the definition of hypoxic water, not a safe operating floor. The Delaware estuary review most often quoted for low-oxygen thresholds in fact places lethal-level requirements for its sensitive species between roughly 4 and 6.4 mg/L, and suggests 5 as adequate for many species with 6 to 7 protective. The distance between those two numbers is your entire margin, and plenty of sites are managing to the wrong end of it.
03 — Biofouling and organic load at intakes
The stressor that eats capital rather than production.
Suspended organics, bloom material and biofilm arrive through the intake and settle on every surface that offers one. Reverse-osmosis membranes are the expensive case; strainers, screens, condenser tubes and heat exchangers are the frequent one.
Trade-press accounts of the 2008–09 Cochlodinium bloom name plants disrupted at Sohar and Barka 1 in Oman, Fujairah 2 in the UAE and Shuwaikh in Kuwait, with one of them replacing one hundred per cent of its reverse-osmosis membranes because of severe biofouling. Note the geography: one bloom, four plants, three sovereign borders. Membrane replacement is capital brought forward, and once it has been brought forward it does not go back.
This is also where the economics are easiest to evidence, because you already know the replacement cost and the cleaning frequency. Treating the approach is a different job from treating the plant: it happens upstream of the screens, in open water, and fixed infrastructure is the wrong shape for it.
Those plant disruptions come from desalination trade press rather than an operator filing or a regulator's record, so we treat them as indicative rather than established.
The last two rows
Two rows that belong on a different page, and here is why.
Neither is a seasonal water-quality exposure. One gives no warning at all; the other is not an event. Both get a short entry rather than a padded one.
Spills and hydrocarbon release
Bunkering, transfer, collision, structural failure — or a third party's incident drifting into your water, which is the one operators plan for least and encounter most. Frequency is low and expected value is not, because the tail is long: response and containment, then downtime, then a liability position that outlives the operating team.
Deepwater Horizon released an estimated 4.9 million barrels over 87 days before the well was capped in July 2010, a figure we carry from general reference sources rather than a primary agency document. BP's cumulative pre-tax charge for the incident had reached US$67.0 billion by the end of 2018, and that one is in a filing with the United States Securities and Exchange Commission.
Does in-water work touch it? Partially, and only afterwards. Oxidation and oxygen life-support in affected water are a recovery capability. Anyone offering in-water treatment as a spill response strategy is overselling — it belongs alongside the statutory response, never instead of booms, skimmers and a plan.
Ballast water and invasive transfer
The only row here with a number written into international law. The IMO's Ballast Water Management Convention entered into force on 8 September 2017, and the D-2 standard limits discharge to fewer than ten viable organisms per cubic metre at 50 micrometres and above, and fewer than ten per millilitre between 10 and 50 micrometres, with the phase-in for existing ships running to 8 September 2024.
For a port authority the exposure is reputational and regulatory rather than operational: a receiving water that acquires an established invasive has been permanently changed, and the port is assessed against it from then on. For a shipowner it is compliance capital and port-state control risk. Neither party owns the water in between.
Does in-water work touch it? Yes, and with the strongest published evidence anywhere in our field. Testing run through the Lake Superior Research Institute found a commercial nanobubble ozone system effective against bacteria, algae and zooplankton in ballast water, with no statistically significant effects on growth, survival or reproduction in organisms exposed to the discharge. That work names another party's equipment, not ours.
Sources for this page: NOAA Fisheries · Reviews in Aquaculture · NOAA NCCOS · DRBC · WaterWorld · Deepwater Horizon reference · BP plc Form 20-F, FY2018 · IMO · NOAA NCCOS · Alarivean.
Your stressor, your asset
Which row is yours?
Send the asset, the failure mode and the months it happens in. Alarivean answers with a read on warning time, treatable volume and what the intervention would and would not change.