Weather Modification: The Documented Part

Start with what is licensed

The argument about weather modification is almost always conducted as though the question were whether it happens. It is not, and has not been for fifty years.

In the United States, anyone attempting to modify the weather must tell the federal government. The duty is in the Weather Modification Reporting Act of 1972: notify the Secretary of Commerce at least ten days before starting, report again afterwards, and the record is public. As of July 2025, ten states had reported activities, every one of them with a permitting or licensing process of its own. In 2024, agencies across nine states were spending or planning to spend at least $20 million on cloud seeding to put more rain and snow into western basins.

That is the documented part, and it is unglamorous: a few per cent more precipitation out of clouds that were going to produce some anyway, bought by water districts, flown by contractors, filed on a form.

It is also, by the government's own account, badly recorded. In February 2026 the Government Accountability Office reported that NOAA is not fully meeting its responsibilities under the Act — that it has no written guidance for reviewing the reports or maintaining the database, and that an estimated 78 per cent of initial reports contain at least one error, including missing maps. The activity is legal, reported and audited. The audit says the reporting is unreliable.

That finding is worth sitting with, because it is the kind of fact that both sides of this argument handle badly. It is not evidence of a secret programme. It is evidence that the public record of the open programme is patchy, which is precisely the condition in which people reasonably start to assume there is something behind it.

The weapon was real, and it was denied

The strongest card the conspiracy account holds here is not a theory. It is a programme, and the history is exact.

From 1967 to 1972 the US Air Force seeded clouds over the Ho Chi Minh Trail, to extend the monsoon and soften the roads. When Senator Claiborne Pell asked the Defense Department about it in 1971, the department waited four months and then declined to discuss the matter on national security grounds. It acknowledged the programme to the Senate in closed session in 1974, and the transcript was released.

So: weather modification was used as a weapon of war, in secret, and the denial came first. Anyone who tells you the subject is inherently absurd has not read the Popeye file.

What followed is the part the conspiracy account tends to leave out, because it cuts the other way. The disclosure produced a treaty. The Environmental Modification Convention was adopted by the General Assembly on 10 December 1976, by ninety-six votes to eight with thirty abstentions, opened for signature at Geneva on 18 May 1977, and entered into force on 5 October 1978. It prohibits military or other hostile use of environmental modification techniques having widespread, long-lasting or severe effects. It is of unlimited duration.

The documented ancestor of the weather-weapon claim therefore has a documented ending: it was exposed, and it was banned by a treaty that is still in force.

What the physics will not do

Between the licensed activity and the alleged one sits a question about scale that is usually skipped, and the record answers it directly.

The United States spent twenty years trying to weaken hurricanes. Project Stormfury ran from 1962 to 1982, openly, with federal funding and published results, on the hypothesis that seeding the eyewall with silver iodide would broaden it and cut peak winds. NOAA's own account is that it was unsuccessful and was discontinued, and that no attempt at hurricane intensity modification has been made since. The reason is physical: hurricanes hold far less supercooled water than the hypothesis required.

NOAA's current position, in terms: no technology exists to create, destroy, modify, strengthen or steer a hurricane.

The same scale problem disposes of the HAARP claim from a different direction. The facility is real, was built by the military, was described publicly from the beginning, was transferred to the University of Alaska Fairbanks in 2015 and holds open houses. It heats a small patch of ionosphere about sixty miles up. Weather happens in the bottom ten.

And the earliest row on the register shows how old the ambiguity is. In October 1947, under Project Cirrus, a hurricane off the south-eastern coast was seeded with dry ice, changed course, and came ashore near Savannah. The project was suspended amid the threat of litigation. Nobody could establish whether the seeding had anything to do with it, and the reconstruction afterwards suggested the storm was already turning. Seventy-nine years later that is still the structure of every argument of this kind: an operation, a weather event, and no way to connect them.

Why a licensed flight is not an explanation

Which brings this to the live version, and to the reason the distinction matters practically rather than philosophically.

When a flood or a fire follows a seeding flight, the flight is real, the permit is real, and the inference is the thing that has to be argued rather than assumed. A cloud-seeding operation adds a modest percentage to precipitation from clouds already producing it. A catastrophic flood involves orders of magnitude more water than seeding has ever been shown to generate, which is why the question to ask is never was there a flight but how much water does the flight account for.

That is a quantity question with a published answer, and it is the same question the panics ledger keeps running into: a small measured effect, and a very large inference built on top of it. The archive holds the specific cases — the Kerr County flood, the water-war claims — and the general answer does not settle any of them by itself. What it does is tell you which figure decides it.

The part that is genuinely open

The live end of this subject is not chemtrails. It is solar geoengineering, and the honest description is that it is a research field with almost no experiments.

The most-cited field experiment, a Harvard stratospheric project, ended in March 2024 without ever flying. NOAA's position is that solar geoengineering is theoretical and not taking place at scale anywhere in the world, while NOAA itself studies the stratosphere in order to inform decisions about the risks. Meanwhile China runs the largest programme of the conventional kind in the world, announced in official plans with square-kilometre targets: more than 5.5 million square kilometres of rainfall enhancement by 2025, and 1,104 rockets fired from twenty-one sites around Beijing before the 2008 opening ceremony.

And the regulatory question is open in a way that will produce a dated answer. A petition is before NOAA asking that the 1972 reporting duty be extended to private solar radiation modification — which would mean the next company to put sulphur in the stratosphere has to file a form, in a public database that the GAO has just said is 78 per cent error-strewn at the first step.

What this page does not establish

It does not tell you that any particular flood, fire or drought was natural. It tells you what is documented, what scale the documented things operate at, and which number would have to be argued to connect one to the other.

It does not cover dispersal of substances for purposes other than precipitation, because no programme of that kind appears in the reporting record — and the absence of a filing is weak evidence either way, as the audit above makes plain.

And it does not settle whether the treaty has teeth. ENMOD has been in force since 1978 and has never been invoked in a dispute, which can be read as deterrence or as a dead letter. Both readings are available, both are held by serious people, and this page has no way to choose between them.