The sea that makes its own fertilizer

Hasle, 18 August

Hasle is a small harbor on the west coast of Bornholm, the Danish island out on its own in the southern Baltic. In the third week of August the water in the basin stopped looking like water.

"I've never seen anything like it," the harbor master, Denni Bjerregaard Chemnitz, told TV2 Bornholm in Danish; his words here and below are translated. "And generally here on this west side, it's the worst I've seen." The harbor bath had already closed. Hammerhavn, round the north point, was worse.

Then the part that is not about swimming. Sailors at Hasle reported that their sea filters were blocking, because the algae were dense enough to starve an engine of its cooling water. "We have people reporting that their sea filters are clogging up because the algae are so dense," Chemnitz said. "And that can cause problems with engine cooling."

A bloom that can stop a diesel is a bloom with real mass behind it, and mass is what separates a Baltic summer from a nuisance.

Source. TV2 Bornholm, 18 August 2026. https://www.tv2bornholm.dk/bornholm/tyk-algesuppe-giver-lystsejlere-problemer-09e1d

A bloom that can stop a diesel is a bloom with real mass behind it.

Where the phosphorus is coming from

Ten days before Chemnitz was standing on his quay, one of the Copernicus Sentinel-2 satellites had already photographed the thing. On 8 August the bloom was developing in the Arkona Basin, between southern Sweden, the Danish island of Møn and the German island of Rügen, with bright green filaments tracing eddies tens of kilometers wide, drawn out by currents and wind.

Why this sea, and so reliably? The Baltic is brackish, nearly enclosed and stacked. River water floats on salt water that came in through the Danish straits, and the boundary between them, the halocline, stays put. Below it the water is old.

HELCOM measures what that costs. Its oxygen debt indicator reports that oxygen debt below the halocline has increased in all basins since the early 1900s, most sharply from 1990 onward, and that neither the Bornholm Basin nor the Baltic Proper meets the threshold for good environmental status. In the Baltic Proper the debt has been rising at 1.49 milligrams per liter per decade since 1990.

And then the sentence in that same indicator that turns the whole argument around. Anoxic periods cause the release of phosphorus from sediment.

Iron holds phosphorus in the mud only while there is oxygen to keep it bound. Take the oxygen away and the mud gives it back. So the dead bottom fertilizes the living surface, every summer, out of a stock nobody put there this century.

The organisms best placed to use surplus phosphorus are the ones that do not need anybody to bring them nitrogen. Nodularia and its relatives fix their own from the air. Which means the great lever of European water policy, cutting the nitrogen going in, reaches almost everything in this sea except the thing on the front page.

Sources. Copernicus image of the day, EU Space Support Office, 12 August 2026. https://eu-space.europa.eu/components/earth-observation-copernicus/image-of-the-day/algal-bloom-swirls-across-western-baltic-sea HELCOM oxygen debt core indicator, April 2023. https://indicators.helcom.fi/indicator/oxygen-debt/

The dead bottom fertilizes the living surface.

Forty years of doing the right thing

It would be easy, and wrong, to read that as nobody having tried.

HELCOM's nutrient-input indicator reports normalized total nitrogen input to the Baltic down 11 percent and total phosphorus down 32 percent between the 1997 to 2003 reference period and 2022. That is nine countries, several decades and a very large number of farmers, and it is real.

The same indicator reports that the Baltic Proper and the Gulf of Finland still failed their maximum allowable input for both nutrients. Over the longer view the hypoxia has gone the other way: under 10,000 square kilometers before 1950, above 60,000 since 2000.

This is what a slow system looks like from the inside, and it is not a failure of effort. The inputs came down. The water has not noticed yet, because the sediment is still handing back a century of phosphorus and the halocline is still in the way.

Per Andersen, senior consultant at the Department of Ecoscience at Aarhus University, was asked in August what could be done. He answered in Danish and is translated here. "There are optimal conditions in the Baltic Sea for cyanobacteria," he said, and then: "The only way we can reduce the risk of algal blooms and widespread oxygen depletion is to lower nutrient loading."

He is right, and the indicator explains why it takes so long to show.

Sources. HELCOM inputs of nutrients indicator, October 2025. https://indicators.helcom.fi/indicator/inputs-of-nutrients/ Per Andersen, TV 2, 22 August 2026. https://vejr.tv2.dk/2026-08-20-ekstraordinaer-sommer-giver-algevaeksten-frit-spil

How anybody knows, and who is counting this year

The satellite is the cheap way to watch a bloom, and it has one honest limitation built into it. Darker blue water marks where the algae have not reached the surface. Cyanobacteria manage their own buoyancy, so the same population can be a spectacular green spiral at breakfast and invisible by four in the afternoon once the wind has mixed it down. Darker does not mean gone.

Which is why the ship counts and the shore reports matter, and why Karen Timmermann of DTU Aqua was careful with her wording when asked what the green was: current algal blooms in the Baltic Sea including Bornholm, Køge Bay and the Sound are quite certainly cyanobacteria. Quite certainly. From a satellite you cannot get closer than that.

For years the clearest public count on this coast was Finland's. The Finnish Environment Institute published a cyanobacteria review every Thursday through the summer: a national agency, a fixed weekday, a map people planned their weekends around.

On 1 January 2026 the National Supervisory Authority for Welfare and Health, the Regional State Administrative Agencies and the Centres for Economic Development, Transport and the Environment all ceased their operations and became a single new Finnish Supervisory Agency. Environmental permits and supervision moved with them.

The Institute's own page records what happened to the review. In summer 2026 the Finnish Supervisory Agency monitors the presence of cyanobacteria and informs about the algae situation every second week until mid-August, and the Institute's traditional weekly cyanobacterial reviews will not be published.

A machinery-of-government change moved a small public series, as reorganizations do, and it happened to be the series that counts a thing which keeps its own calendar. Chemnitz was standing on the quay at Hasle on the eighteenth. What was lost is resolution, in the month resolution was worth having.

Sources. Finnish Supervisory Agency, 2 January 2026. https://lvv.fi/en/-/the-finnish-supervisory-agency-started-regional-state-administrative-agencies-centres-for-economic-development-transport-and-the-environment-and-the-national-supervisory-authority-for-welfare-and-health-no-longer-exist Finnish Environment Institute, blue-green algae monitoring. https://www.syke.fi/en/services/monitoring-and-inventories/cyanobacterial-reviews Karen Timmermann, TV 2, 22 August 2026.

The reporting went fortnightly and stopped in mid-August. The bloom kept its own calendar.

What reasonably healthy would look like here

Not the Baltic of 1900. Nobody is offering that, and a target borrowed from a different century is not a target.

HELCOM has already written down something checkable: an oxygen debt below the halocline that meets the threshold for good environmental status in basins that currently do not, starting with the Bornholm Basin and the Baltic Proper. That is a number, in a named place, published by the body the nine countries answer to. Reach it and the sediment stops handing phosphorus back at the rate it does now, and the summers change on their own.

It is worth saying what that implies about time. The inputs have been coming down for thirty years and the oxygen debt has been going up for a hundred, so the honest expectation is a sea that keeps blooming for a while after the right thing is being done, and a public record good enough to tell the difference between a bad summer and a trend.

What would settle that is a count nobody has to guess at: a weekly series, on the same weekday, for the whole of the season, from a body that will still exist next year.

In June, four researchers gave a sense of where this goes. Mohanad Abdelgadir, Bengt Karlson, Elin Dahlgren and Malin Olofsson used measured concentrations of the toxin nodularin to model where Nodularia spumigena turns up, driven by salinity, temperature, phosphate and the nitrate-to-phosphate ratio. They put it today in the Eastern and Western Gotland Sea, the Northern Baltic Proper and the southern Bothnian Sea. By 2100 their model moves it north, into the Åland Sea and slightly into the Bothnian Bay: water that is fresher, colder and has mostly been spared so far.

Source. Biogeosciences, 30 June 2026. https://bg.copernicus.org/articles/23/4321/2026/