Water Column

Shaded by rain

2026-09-08T00:00:00Z

The number

In its fortieth year, the Australian Institute of Marine Science went back to 121 reefs and counted. Hard coral cover in the north rose from 30.0 to 35.1 percent. In the center, from 28.6 to 31.6. In the south it slipped, 26.9 to 26.4. All three regions now sit above or just below their long-term average.

That is a better year than anybody had a right to expect.

AIMS Long-Term Monitoring Program annual summary 2025/26, released 10 August 2026.

The reason

The sea was warmer than average again. What changed the outcome was a late monsoon: cloud over the reef through the worst weeks of summer, and sea temperatures pulled down with it. The institute's own words are that this likely limited the severity and extent of bleaching.

A reef was saved by rain. Not by zoning, not by water quality policy, not by anything anybody did on purpose.

A reef was saved by rain. Not by anything anybody did on purpose.

AIMS, 10 August 2026. The word 'likely' is the institute's.

One reading

Cover is up in two regions of three and the reef is back around its long-term average. That is the measurement, it was taken the same way it has been taken for forty years, and no other quantity in this argument was counted at all.

On this reading the reef has demonstrated something real about itself. Given one cool summer, it grew back toward where it was, and that capacity is in the data rather than in a model.

The second reading

Dr Mike Emslie, who leads the monitoring program, said something easy to read past: while many reefs did not lose coral cover, bleaching likely prevented the continued increases that would normally be expected.

Read that carefully and the good year turns into a smaller year than it should have been. The reef did not lose ground; it also did not gain what an unbleached summer would have handed it. That missing growth is real and it is not in the survey, because a survey counts what is there rather than what would have been.

One thing underneath the headline belongs to this reading, and it is in the survey. In the north — the region that posted the largest gain — crown-of-thorns starfish numbers are rising in a pattern the institute reads as the start of the fifth outbreak wave since the 1960s.

Emslie quoted in the AIMS release, 10 August 2026. Reef distribution and crown-of-thorns status from the same summary.

What would settle it

The disagreement is not about coral. Both readings accept every number in the survey. They disagree about the counterfactual — what this year would have produced without the heat — and no monitoring program measures a year that did not happen.

What would settle it is a run of summers cool enough to show the reef's unimpeded growth rate, which is the one experiment nobody can schedule. Failing that, the fifth outbreak wave will answer a different question fairly soon, and the answer will arrive in the same regions that just posted the gain.

Eight months

2026-08-18T00:00:00Z

November

Royal Decree 96/2025 was signed on 11 November 2025. It created a marine buffer zone reserve around the Hallaniyat Islands, off the Dhofar coast, where the Arabian Sea runs deep and cold and turtles come ashore to nest.

The decree refers to an annex setting out the reserve's boundaries. The annex was not published with it, so the area of the thing is not a number this piece can give you, and a figure has been circulating that nobody here has read on a government page.

Royal Decree 96/2025, Sultanate of Oman, 11 November 2025.

June

On 30 June 2026, seven months and nineteen days later, the Caroline Bezengi ran aground inside it. A sanctioned Suezmax with roughly a million barrels of crude aboard, bound for Asia.

The National, 16 August 2026.

Two measurements

By 10 August Oman's Environment Authority had put a figure on the slick: about 390 square kilometers, extending some 449 kilometers along its trajectory, coming within about 7 kilometers of shore at the nearest point. It stated there were no indications of impact on critical infrastructure, including seawater desalination plants.

Greenpeace, tracking the same water by satellite, had it at 45 square kilometers on 26 July, 150 on 2 August, and 600 on 4 August.

Both figures are here because printing one alone would be choosing a side by omission. They are not the same kind of number: one is a government's assessment of its own waters and one is a campaigning organization's read of a satellite, reported by a newspaper. They disagree by a factor of roughly one and a half, on a question that is in principle answerable from orbit, and nobody has reconciled them.

They disagree by a factor of roughly one and a half, on a question answerable from orbit.

Environment Authority figures via Oman News Agency, 10 August 2026. Greenpeace track via Muscat Daily, 8 August 2026.

Why nobody could get to it

The khareef is the reason the salvage cannot proceed and it is also the reason Dhofar has a tourist season. For three months every summer the southwest monsoon drags cloud and drizzle over a coast that is otherwise desert, the hills turn green, and Salalah fills with visitors who have come to see rain.

The same weather makes the sea off those islands unworkable. "Adverse climatic conditions of the autumn monsoon season have significantly complicated technical spill response operations," said Mohammed Abdullah Al Rawahi, Oman's Director General of Maritime Affairs, in the flattest available language. Ed Wollaston, who runs global response at the maritime risk firm Ambrey, put it as an extremely challenging situation compounded by the weather.

A salvage team boarded in mid-August, seven weeks after the grounding.

Al Rawahi via Oman News Agency, 10 August 2026. Wollaston in The National, 16 August 2026.

The decision in front of somebody

The oil has reached the mainland. Reporting puts it along roughly 40 kilometers of coast near Ras Madrakah, with Masirah Island potentially exposed, though the day it first came ashore is not something anybody has printed clearly enough to repeat.

What has to be decided now is not really about this hull. A reserve was declared in November and a vessel nobody would insure was allowed to sit broken inside it from the end of June, through a monsoon that was always going to arrive on schedule. Whatever is written into the next decree about that water, the interesting question is what it says about vessels rather than about turtles.

Landfall extent as reported by The National, 16 August 2026.

The sea that sets its record in June

2026-07-28T00:00:00Z

The surface

The Mediterranean averaged 24.3 degrees in June 2026. That is the warmest June anybody has recorded there, and it beat two benchmarks at once: 23.95 set in 2025, and 23.94 set in 2003.

The benchmark that fell this June was set in 2003, the summer of the European heat wave, and it had stood for twenty-three years as the outer limit of what this sea could do. It was beaten in June. Not in August, which is when a Mediterranean summer usually decides what it is going to be, but in the month before it.

Copernicus Marine Service, first-half 2026 ocean state, published 1 July 2026.

Where it was hottest

On 29 June the surface off southern France, and around Corsica, Sardinia and the Italian peninsula, ran as much as 8 degrees above the 1991 to 2020 average for the date.

Eight degrees is a number that sounds like an error. It is a one-day satellite reading of a thin surface layer rather than a change in the body of the sea, and that distinction matters enormously for what happens next — but a thin surface layer is exactly where the light is, where the swimmers are, and where a great deal of what lives in this sea is anchored.

European Space Agency, on Copernicus Marine data for 29 June 2026.

The part that does not behave

Between January and June, roughly 98 percent of the Mediterranean saw marine heatwave conditions of some kind. About 80 percent of it saw conditions rated strong, severe or extreme.

Now read that against the two years before. In 2025 the strong-or-higher figure was 88 percent. In 2024 it was 95.

So the extent of the worst heatwaves has been falling for two years while the temperature record broke anyway. That is not what a warming sea is supposed to look like from the outside, and it is the most interesting thing in the whole report. Intensity and area have come apart. A smaller share of the sea is being cooked harder, and the averages keep climbing regardless.

A smaller share of the sea is being cooked harder, and the averages keep climbing regardless.

Copernicus Marine Service, 1 July 2026.

The bed

What answers for a hot summer here is the part that cannot leave. Gorgonians, sponges and the seagrass Posidonia oceanica are fixed to the bottom, and Copernicus links Mediterranean heatwaves to recurrent mass mortality among exactly those communities. Posidonia meadows respond to thermal stress with higher mortality and slower growth, and the meadows have declined by as much as 34 percent over fifty years under climate change and everything else people do to a coast.

A meadow is not scenery. It holds the sediment that holds the beach, it is where a great deal of the sea's young fish begin, and it grows at a rate measured in centimeters a year. Anything a bad June takes out of it is not coming back inside a human career.

Copernicus, European State of the Climate, on Mediterranean biodiversity.

What the record is for

A record is a number in a press release, and this one will be beaten. The reason to write it down is that it moved the date. For twenty-three years the outer limit of this sea belonged to a specific and terrible August. It now belongs to a June, and the month that used to decide the summer has not had its turn yet.

The coast is clear, and the job is not done

2026-06-30T00:00:00Z

The unit

At the height of it South Australia was weighing the damage in kilograms of dead animals collected off its beaches per week. About 9,500 of them, every week, hauled off the sand by people with gloves and bags.

By the end of April the figure was under 50.

That is a real recovery and it is also a strange number to be governed by. Nobody designs a monitoring program around the mass of what washes up, because the mass of what washes up is a function of wind and tide and how many volunteers turned out on Saturday. It became the headline number because it was the one thing a person could see. The bloom had been running in Gulf St Vincent and Spencer Gulf since March 2025.

Collection weights as reported by ABC News, 27 April 2026.

Two right answers

On 27 April the Premier of South Australia, Peter Malinauskas, said the coast was effectively clear, and that the bloom had essentially cleared up. On the same day Professor Mike Steer, who runs the state's research and development institute, said the job was not done.

Neither man was spinning. A premier answers for beaches, and the beaches were clean. A research director answers for stocks, and the stocks had not come back. What had ended was the visible part.

A premier answers for beaches. A research director answers for stocks.

Both quoted by ABC News, 27 April 2026.

What stayed shut

From 1 May, southern calamari closed across both gulfs. Commercial, recreational and charter, all of it, after stock assessments found severe declines. The closure notice itself sits behind a page that answers every request with a refusal, so what is on the record here is an outlet's report of it rather than the instrument.

Oysters and mussels had already been shut for as long as eight months. That is the part of a bloom that outlives the bloom, and the mechanism is not complicated: brevetoxin ends up in the animal, not in the water, so cell counts can fall to nothing while the shellfish stay unsellable. A fishery does not reopen when the water is clean. It reopens when the flesh tests clear, and those are different dates.

Closure scope and duration as reported by ABC News, 27 April 2026. The state closure notice was not accessible.

Ten kilograms

Jarrod Day fishes commercially out of Port Lincoln. During the bloom he was landing around 10 kilograms where he would usually land at least 50.

One fisher's fortnight is not a fishery, and his number is his own rather than a stock assessment. It is worth putting beside the other numbers anyway, because it is denominated in a way a person can feel and nothing else here is. Nine and a half tonnes off a beach is an abstraction. Forty kilograms missing from a day's work is a mortgage.

Jarrod Day, quoted by ABC News, 27 April 2026.

The line item nobody has

Here is what the ledger cannot show. There is a figure for what was collected and a figure for what a boat landed, and there is no figure at all for the eight months in between — for a grower who kept paying lease fees on racks that could not be harvested, for the processor who lost a season's contracts, for the deckhand who took other work and did not come back.

A bloom is priced by what it kills, because that is what can be counted. The thing that actually costs the most is the interval, and nobody has an instrument for an interval.

Update, 9 July 2026

Nine days after this issue, researchers published what the organism actually does. Karenia cristata killed half the invertebrates in the test at five cells per milliliter - a concentration so low it is hard to picture, and by the researchers' account more lethal to invertebrates than anything else tested the same way.

That reframes the accounting above rather than contradicting it. The premier was right that the beaches were clear and the research director was right that the job was not done, and this is part of why: at five cells per milliliter, a bloom does not have to be visible to still be doing something. Nothing in the ledger changes. The interval just got easier to understand.