The Vela Incident

Summary

An American satellite, built for exactly this purpose, recorded the optical signature of an atmospheric nuclear explosion over the southern ocean. It had recorded that signature forty-one times before. Every previous one had been a nuclear test.

No country has ever admitted to the forty-second. A White House panel concluded it probably was not nuclear, over the objections of the CIA, the Defense Intelligence Agency, the Department of Energy and the Naval Research Laboratory. The evidence has been accumulating against that panel for forty years, and the case is not closed because most of the documents are still classified.

What is documented

The detection. Vela 6911, one of a pair of satellites launched in 1969 to enforce the Partial Test Ban Treaty, carried optical sensors called bhangmeters. On 22 September 1979 they recorded a double flash.

Why a double flash matters. A nuclear explosion in the atmosphere produces a characteristic light curve: a brief intense flash, a dip as the expanding fireball's own opaque shock front hides the interior, and then a second, longer, brighter peak as the shock becomes transparent again. Nothing else in nature is known to produce it. The Vela satellites had detected it forty-one times, and on all forty-one occasions it had been a nuclear test, each subsequently confirmed by other means.

The satellite was old. Vela 6911 had been in orbit for more than ten years, past its design life, and its electromagnetic pulse sensor was no longer working. This is the strongest argument against the detection and it was the basis for what followed.

The Ruina Panel. The Carter administration convened a panel of academic scientists under Jack Ruina of MIT. It concluded that the signal was probably not from a nuclear explosion, and proposed that it was more likely a "zoo" event: sunlight reflected from a small meteoroid striking the satellite, or debris in its vicinity. This became the semi-official White House position.

The dissent, which was substantial. A CIA-sponsored panel concluded that the signals were consistent with an atmospheric nuclear detonation, and a CIA finding put the probability at above ninety per cent. Senior officials at the Department of Energy, the Department of Defense, the Defense Intelligence Agency and the CIA held that a nuclear explosion had taken place. Stansfield Turner, then Director of Central Intelligence, later said the Agency had no doubt it was a test.

Jack Varona, Vice Director of the DIA, described the White House investigation as a whitewash driven by political considerations and resting on flimsy evidence, and said the weight of the evidence pointed to a nuclear event.

Declassified documents record White House officials agreeing to hear dissenting views so that they could more safely ignore them.

The hydroacoustics. The Naval Research Laboratory analysed underwater acoustic data and produced a report of roughly three hundred pages concluding that a nuclear event had occurred near the Prince Edward Islands. A senior DIA official described the acoustic signals as unique to nuclear detonations in a maritime environment. Alan Berman, the NRL's director at the time, later stated that the hydroacoustic evidence strongly indicated a test at the time and place of the Vela flash, and that his report was not treated seriously by the Ruina Panel.

That report remains classified. The Navy has located it and a declassification review has been initiated.

The sheep. Lester VanMiddlesworth, a professor at the University of Tennessee, ran a long-standing surveillance programme measuring radioactive iodine in the thyroid glands of sheep slaughtered around the world. Australian sheep had shown no fission products since the French Pacific tests ended in 1974. In October and November 1979 they showed iodine-131, a short-lived fission product.

His raw notebooks were later recovered from the Nuclear Testing Archive in Las Vegas. Re-analysed directly from that raw data, the iodine-131 detection holds up. Meteorological modelling shows air moving from the Prince Edward Islands toward south-eastern Australia over the following days, and rain falling as it passed over the grazing land in question.

The ionosphere. The Arecibo Observatory recorded an ionospheric disturbance on 22 September resembling those associated with Soviet atmospheric nuclear tests in the early 1960s.

Carter's diary. On the night itself he wrote of an indication of a nuclear explosion near South Africa, and named three possibilities: South Africa, Israel using a ship at sea, or nothing. By February 1980 he recorded a growing belief among American scientists that the Israelis had conducted a test in the ocean off southern Africa.

Leading explanations

A low-yield nuclear test, most probably Israeli, possibly with South African assistance. This is the leading explanation and the weight of evidence behind it has grown, not shrunk, since 1979. De Geer and Wright, reanalysing the declassified optical, hydroacoustic and radionuclide data, put the yield at roughly two to four kilotons and concluded that the three independent data sets trace back to the same place and the same minute, which they argue would be an extraordinary coincidence if they did not share a cause.

The reasoning about who is by elimination. The five recognised nuclear powers had no need of a small clandestine test at sea. India, Pakistan and South Africa acting alone are each held to have lacked the capability or the logistics for that particular operation at that time. Gerard Smith, the American ambassador for non-proliferation, said he was never able to shake the thought that it was a joint Israeli and South African operation.

A "zoo" event. The Ruina Panel's finding: a micrometeoroid striking the ageing satellite and throwing off sunlit debris, producing a false double flash. This is a real failure mode, it is the reason the panel could reach its conclusion, and it is the alternative that has to be maintained. De Geer and Wright argue that the reanalysed optical data does not support it. It does not account for the hydroacoustics, the iodine-131, or the ionospheric disturbance, which is the burden it has always carried.

Political convenience. Not an explanation of the flash, but a documented factor in how it was handled. In September 1979 an Israeli nuclear test would have obliged the Carter administration to act, in the middle of a non-proliferation push and a fragile Middle East settlement. The declassified record shows officials who wanted the answer to be nothing.

Current status

Unsolved. The registry does not classify this as later explained, because nothing has been established. It does not classify it as debunked, because the evidence for a nuclear detonation is considerable and has strengthened.

Something happened over the southern Indian Ocean on 22 September 1979. The best available physical evidence says it was a nuclear explosion of a few kilotons. No state has claimed it, no state has been proven to have conducted it, and the documents that could settle it are in a vault.

Last reviewed: July 2026. Records are provisional. Where the evidence changes, the entry changes. Found an error? Tell us.