Detection, sensing, forecasting
A satellite measures reflectance. A fluorometer measures fluorescence. A microscope measures what settled in a chamber under one analyst's eye on one afternoon. Everything a monitoring programme publishes is an inference, and knowing which inference you are reading is most of the skill.
The inference chain
The thing anybody actually cares about is at the end of that chain. Harm to a person, a fish, a membrane or a beach. Nothing in orbit and very little in the water measures harm, so every operational product is built by chaining inferences and hoping the couplings hold.
An ocean-colour sensor records radiance at the top of the atmosphere. Correct for the atmosphere and you have water-leaving radiance. Apply a band ratio and you have an estimate of chlorophyll-a concentration. Chlorophyll is a proxy for phytoplankton biomass, biomass is a weak proxy for the abundance of any single species, and species abundance is a weak proxy for toxin.
Each arrow is defensible on its own. Multiply the uncertainties and you get the reason a chlorophyll anomaly is a reason to sample rather than a reason to close a beach.
This site is an account of where those couplings are strong, where they are weak, and which instrument to reach for when the answer matters.
The observation stack
A monitoring programme is a portfolio, not a purchase. The failure mode we see most often is a programme that buys one layer well and leaves the layer that would have made it interpretable unfunded.
| Instrument class | What it physically measures | What you infer from it | What it cannot resolve |
|---|---|---|---|
| Satellite ocean colour | Radiance in a handful of visible bands, over a wide swath | Chlorophyll-a, turbidity, surface bloom extent and drift | Species, toxin, anything under cloud, anything below the first optical depth |
| Hyperspectral imaging | Radiance in many narrow, contiguous bands | Pigment assemblage, and sometimes a dominant group | Toxin concentration; and it still cannot see through cloud |
| Airborne and drone survey | The same optics, flown under the cloud, on your schedule | High-resolution extent along a specific stretch of coast | Continuity — you get the day you paid for and nothing between |
| Moored sensor and buoy | Temperature, salinity, dissolved oxygen, fluorescence, turbidity | Condition and rate of change at one point, continuously | Anything a kilometre away; and fluorescence is not cell count |
| Glider and profiler | The same variables, but through depth and along a track | Structure — where the layer sits, how thick it is, where it is going | Real-time coverage of a wide area; endurance is finite |
| Sample, microscope, assay | Cells in a settling chamber; toxin in a matrix | Species identity, cell density, toxin concentration | Everywhere and everywhen you did not take a bottle |
Nothing in the last column is a defect. It is the specification. A programme gets into difficulty when a product's limitation is treated as a temporary shortcoming to be fixed by better software.
Honest limits
Chlorophyll-a is common to almost all phytoplankton, so a strong anomaly tells you biomass and nothing about whether the dominant organism is harmless or produces a neurotoxin. Toxin production also varies with strain and conditions within a single species. The bottle is not optional.
Optical sensors do not see through cloud, and a polar-orbiting instrument returns on its own schedule rather than yours. A monsoon or a dust season can remove a fortnight of imagery from exactly the period you needed it. Coverage statistics quoted in a procurement rarely mention clear-sky coverage, which is the only kind that counts.
Near shore, the signal carries suspended sediment, dissolved organic matter and reflection from the bottom, and a pixel spanning water and shore mixes all of it. Standard open-ocean algorithms were not built for that, and the places where blooms hurt most are precisely the places where those algorithms are weakest.
Anyone selling a monitoring system that answers all three is describing an ambition. The remote-sensing page sets out what current products genuinely deliver and where the honest error bars sit.
What instrumentation revealed
In 2018 a University of East Anglia team working with Sultan Qaboos University published a survey of the Gulf of Oman made with seagliders — slow, autonomous, and cheap enough to lose, which is the whole point of them. Two of them worked the area for eight months, reaching water that ships had not sampled in nearly half a century because of piracy and geopolitics.
Where they expected some oxygen they found an area larger than Scotland with almost none. The lead author's summary was that the situation was worse than feared and the zone vast and growing. Nothing on the shoreline would have shown you that, which is the corrective for anyone who reads an unmonitored sea as a healthy one.
They also found that the depleted layer rises and falls between seasons, squeezing fish into a thin band near the surface. That is a structural finding, and structure is precisely what a satellite cannot give you.
We have this through a research summary rather than the journal paper, which we could not retrieve. So we quote only what the summary states, and we do not repeat the square-kilometre and micromole figures that circulate with it — we have not been able to check them against the original.
The other half of the problem
Detection gets the attention. Verification is where monitoring programmes and treatment suppliers both tend to go quiet.
If a water body is treated, somebody has to establish what changed, against what baseline, over what area, and for how long. That is a harder measurement problem than detecting the bloom was, because the counterfactual is unobservable and the water keeps moving.
There is published work to build on. NOAA's National Centres for Coastal Ocean Science validated an ozone nanobubble aeration system on an eight-acre pond near Fort Myers Beach in 2018, reporting complete elimination of algae within forty-eight hours with proper reoxygenation and no apparent harm to aquatic life. A separate evaluation in 2020 tested a nanobubble ozone system on ship ballast water and found it effective against algae, bacteria and motile zooplankton without a statistically significant adverse residual-toxicity effect in the receiving water.
Both are bounded environments with a controllable boundary. Neither names the vessel this network operates, and neither establishes anything about open coastal water, where dilution and exchange make the same measurement much harder. That is a limitation of the evidence and not a criticism of the studies.
The discipline is the same one this whole site argues for. Measure it, state the uncertainty, and do not let a mechanism stand in for a result.
Custody of the record
On a slide those two look identical. They are not remotely the same object, and what separates them is whether anybody outside the company doing the work could see the raw stream while it was still arriving.
Everything else here concerns the limits of instruments. This concerns the limits of provenance — harder to repair afterwards, and almost free to get right at the start.
A time series that reaches a reviewer as a processed export has already passed through decisions nobody can now inspect. Which stations were in. Where a gap was filled and by what method. Which calibration applied on which day, and when the sonde was last wiped. None of those choices has to be dishonest for the result to be useless. They only have to be invisible.
So the remedy is structural rather than analytical. Alarivean's programmes engage local scientific institutions continuously and give them access to the live feeds the operator is working from, with the freedom to run their own sensors beside them and keep whatever those produce. The company requires this rather than tolerating it, for a reason that fits in one line: a treatment record with a single custodian cannot be audited, only believed.
The test for an analytics team is one question long. If the operator's servers went dark tonight, would anyone else still hold this season's observations? Where the answer is no, what you have is not a dataset. It is a claim about one.
Adjacent work in this network
Immediate, Significant, Scaled
Station records, chlorophyll time series, cell counts, the years the bulletins were wrong. Alarivean returns a read on what your current stack can and cannot support — including where an additional instrument would tell you nothing new.