Wednesday, October 7, 2026
They Buried
We Dug It Up
⌕ Search

8,600 Tonnes Of Sulphur Reach The Stratosphere Each Year. Nobody Aimed Any.

A mass balance, not an argument. The covert fleet would need 8,000 sorties a year and shows up in no excise return. The accidental version needs no fleet, and the one deliberate intervention on the books was aimed at lungs.

Analysis is written by news reporters and interprets facts they have verified. It is not opinion, and it takes no position on what should be done.

Thirty years of stratospheric aerosol depth drawn as a single strip. The three tall marks are volcanoes. Nothing in the trace was aimed at anything. Engraved for They Buried

That figure is ours. It is a mass balance rather than an argument, and the working is published with this story so that it can be taken apart.

We started at the other end. We wanted to know what a deliberate stratospheric aerosol programme would have to buy, and then whether anybody had bought it.

What the design studies actually specify

Three peer-reviewed design studies give payloads. We took the numbers out of their tables rather than their abstracts, because payload figures shrink on the way through a summary.

To offset something like a degree, the three converge on delivering on the order of four million tonnes of sulphur a year to about 20 kilometres. That is between 6,000 and 8,000 dedicated high-altitude sorties annually, every year, without a gap.

All three say, in their own text, that the aircraft does not exist. It would have to be designed, certified and built: a sustained lofter holding 20 km with about 20 tonnes in the tank. Nothing in service does that.

Where we looked for it

Fuel first, because fuel is taxed and tax is filed.

A fleet flying that schedule burns between 3.1 and 4.4 million tonnes of jet fuel a year, which is between one and one and a half per cent of world consumption. We pulled aviation fuel-duty and excise filings from four jurisdictions — three publish quarterly, the fourth released its returns to us in sixty-one days — and looked for the step.

There is no step. Uplift tracks scheduled traffic, in every quarter, in all four series.

Then airframes. We bought a commercial fleet database and read the tanker and high-altitude registries line by line, then the maintenance-hour returns that any fleet flying 8,000 sorties a year is obliged to generate. The fleet is not there, and a fleet that size is not a secret you can keep from a tax office and a hangar.

Figure The sulphur column, read from the top down
35 km — above all of it0Volcanic sulphur — 22 kmAbout 3,800 t·S in a median year.Nobody schedules a vent.−13kmWhere a programme would fly — 20 km6,000 to 8,000 sorties a year. Inno excise register we read.−15kmIN NO PUBLISHED RECORDTropical tropopause — 17 kmAbout 2,400 t·S of ship sulphur,lifted slowly. Now less.−18kmFire-storm tops — 16 kmAbout 500 t·S. One fire seasonbeat a decade of airliners.−19kmAirliner cruise — 11 kmAbout 1,900 t·S, duty-paid anditemised, above the tropopause.−24kmMarine boundary layer — under 1 kmThe one deliberate act. Sulphateremoved, on purpose, for lungs.−34kmDISPUTED
Depths are kilometres below the top of this plate, which sits at 35 km; the real altitude of each layer is in its label. Five of the six rows are arrivals nobody ordered, and they total about 8,600 tonnes of sulphur a year — roughly one five-hundredth of what a programme designed to offset a degree would have to deliver. The second row is where such a programme would fly. We looked for it in four countries' fuel-duty returns, in the tanker and high-altitude airframe registries and in maintenance-hour filings, and there is nothing there. The bottom row is the only deliberate act in the diagram, it happened at sea level, and it went the other way. They Buried, from three published design studies, four excise registers, a commercial fleet database and a commissioned mass balance

The same sum, run backwards

Then we ran the arithmetic in the other direction, and it produced a figure the desk had not gone looking for.

Volcanoes nobody scheduled put roughly 3,800 tonnes of sulphur above the tropopause in a median year. Ship stacks feeding the slow tropical upwelling contribute about 2,400. Airliners burning ordinary sulphur-bearing kerosene above the polar tropopause add about 1,900. Fire-storm tops account for perhaps 500.

Eight thousand six hundred tonnes. Never added up as a total until we paid a chemist to add it up, and owned by nobody.

Here is the ratio, and it belongs in the middle of this story rather than at the end. Those 8,600 tonnes are about one five-hundredth of what a programme designed to offset a degree would need to deliver. The accidental version is real, it has been running for decades, and it is not a quiet edition of the deliberate one. Anybody who tells you the fleet is already flying has not done the division.

The intervention that did happen

There is exactly one deliberate act of large-scale atmospheric modification in the recent record, and it was pointed the other way.

In 2020 the limit on sulphur in marine fuel fell from 3.5 per cent to 0.5. It was passed to stop people dying: the sponsoring assessment put the avoided burden at roughly a quarter of a million premature deaths and several million childhood asthma cases a year. It worked. Ship tracks over the North Pacific thinned visibly inside eighteen months.

It also took a great deal of reflective sulphate out of the marine boundary layer. Three reviewers we commissioned put the resulting change in forcing at between 0.04 and 0.09 watts per square metre: a small warming, concentrated over the shipping lanes, that no one had asked for.

What this desk takes from it

We have been putting sulphur into the stratosphere by accident for as long as we have had ships and jet engines, at a scale far too small to matter, and then in 2020 we took a much larger quantity of it out of the air at sea level on purpose, for lungs.

Dr Owusu-Lindberg objects to that sentence and her objection is printed in full below. She says the trade-off was raised in writing, weighed, and decided by people who chose the lungs, and that calling an unowned by-product an intervention flatters it. Read her rather than us.

What we cannot do is find the fleet. Not because it is well hidden. Because a fleet that size cannot be.

Disclosure. This newspaper paid £2,300 for a commercial fleet and maintenance database licence, and standard commissioning fees to Dr. Achterberg-Nwosu and to three reviewers of the marine forcing estimate, all in advance and none conditional on any finding. No payment was made to any interviewee. The Sky Desk has no aviation or shipping holdings and its editor's disclosure stands on his author page.

How Others Covered This

The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.

  1. The Hollow Post
    The Fleet That Does Not Exist Still Files Its Fuel Duty Somewhere

    Treated the absence of a fleet as evidence of a better concealment, and our excise search as the search they wanted us to run.

    Had that we did not

    Correctly noted that the design studies are real, published under named authors and openly funded, which many readers do not know.

    Left out

    That all three studies say in their own abstracts that the aircraft has not been built and would have to be designed from nothing.

  2. Aperture
    The Weather We Did Not Mean

    An essay on unowned consequence, and right about the moral shape of the thing in a way our piece is not.

    Had that we did not

    The ships, and the observation that an exhaust has no author and therefore no one to argue with.

    Left out

    The arithmetic, entirely. It is four thousand words about a quantity it never prints.

  3. They Buriedthis newspaper
    8,600 Tonnes Of Sulphur Reach The Stratosphere Each Year. Nobody Aimed Any.

    Costed the programme first, went looking for the purchase in tax and hangar records, then ran the same sum backwards.

    Had that we did not

    The full mass balance with its working, the four excise series, and the blind check by an atmospheric chemist who was told nothing about the story.

    Left out

    We built a headline on 8,600 tonnes and put the ratio — that this is about one five-hundredth of what a deliberate programme would need — in the middle of the piece. A reader who stops at our headline leaves believing the opposite of what we spent three weeks establishing, and the headline was chosen knowing that. — V. Ashcombe-Doyle, standards editor

Right of Reply

They Buried contacted Dr. Hyacinth Owusu-Lindberg, marine fuel policy; national delegation to the fuel-oil working group, 2014–2020 on 14 July 2026 in writing, with the full draft and the mass balance attached, and four weeks to answer. Replied 12 August 2026 by letter, asking that it run whole and unedited. It runs whole. Her correction has been made in the text and credited to her.

Your arithmetic is competent and I have no quarrel with the number. My quarrel is with your last two paragraphs, which I think you enjoyed writing more than you thought about.

Start with the correction, because you should not have needed me for it. You wrote that the warming appeared on no impact assessment. There were four assessments, not one, and a supplementary annex in 2018 discusses the climate effect over two paragraphs and then declines to put a number on it, which is a different thing from silence and a much more interesting one. Please stop saying nobody considered it. People considered it. They considered it and then declined to quantify it, and if you want a scandal that is where you should be digging, not in an excise register.

Now the sentence I object to. You write that we un-geoengineered on purpose, for lungs, as though the room had been handed a lever marked climate and had pulled it the wrong way by mistake. The trade-off was raised. It was raised in writing, by more than one delegation, and it was raised early. It was not overlooked. It was weighed, and the weighing was not close, because the counterfactual on the table was never keep the reflective sulphate. The counterfactual was keep killing people with it. Roughly a quarter of a million of them a year, disproportionately in port cities, disproportionately children, in a plume that came from a fuel so poor that it has to be heated before it will flow. If you want to describe a decision as a choice between a small warming and a large number of funerals, describe it that way, and then say which one you would have chosen, in the room, with your name on the paper.

And I would ask you to be more careful with the word intervention, because your own headline gets this right and your body then throws it away. Nobody can turn the ships up. Nobody can turn the volcanoes down. Nobody can be summoned to a committee to answer for the sulphur in a tank of kerosene. That is not an intervention. It is an exhaust, and the difference between an exhaust and an intervention is not scale, it is that an intervention has an author who can be argued with and stopped. You have spent six hundred words dissolving a distinction your headline states perfectly in three.

One thing I will grant you without reservation. The 8,600 tonnes is a real quantity, it has been arriving for as long as we have had ships and aircraft, and until you commissioned that sum nobody had written it down in one place, which is a poor reflection on my field and not on yours. I have circulated your working. Two people have already told me the volcanic term is too high. They are welcome to publish a better one.

Published unedited under our right-of-reply guarantee.

How was this story?

We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.

3,648 verdicts · 76.0% loved it

Readers' Letters 0

Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.

  1. No letters yet on this story. Yours would be the first.

Write to the desk