Documents Show Significant Disagreement with Presidential Panel Concerning Cause of September 22, 1979 Vela "Double-Flash" Detection

National Security Archive Electronic Briefing Book No. 190
For more information contact:
Jeffrey T. Richelson - 202/994-7000

Washington, DC, May 5, 2006 - Many U.S. government officials and scientists disagreed with the findings of a presidential panel that the double flash signal picked up by a U.S. nuclear detonation detection satellite (Vela 6911) in late September 1979 was possibly not a nuclear test, according to a number of studies posted today by the National Security Archive.

The signal appeared to come from a 3,000 mile area that included the South Atlantic, Indian Ocean, tip of Africa, and part of Antarctica. A presidential panel concluded in May 1980 that the signal was more likely an artifact of a meteoroid hitting the satellite and sunlight reflecting off particles ejected as a result of the collision.

In addition to the report of the presidential panel, the posting includes reports produced by the DCI's Nuclear Intelligence Panel (completely redacted), and scientists and analysts at Los Alamos, SRI International, Sandia, the Intelligence Community, the Defense Intelligence Agency, Mission Research Corporation, and the Aerospace Corporation. Included are several reports which concluded that a nuclear test was the most probable explanation of the Vela detection and/or specifically questioned the presidential panel's explanation.

Many of the reports were obtained under the Freedom of Information Act by Archive Senior Fellow Jeffrey T. Richelson, while conducting research for his new book Spying on the Bomb: American Nuclear Intelligence from Nazi Germany to Iran and North Korea (W. W. Norton).

Late in the evening of September 21, 1979 at Patrick Air Force Base, Florida, technicians from the Air Force Technical Applications Center (AFTAC), the organization responsible for running the U.S. Atomic Energy Detection System, conducted a routine readout of a Vela satellite, designated Vela 6911, which had been launched on May 23, 1969.

The first two Velas had been launched in 1963. Vela 6911, which orbited the earth at an altitude of 67,000 miles, carried a variety of equipment to detect the numerous signatures associated with atmospheric nuclear detonations. In addition to sensors to detect gamma rays, x-rays, and neutrons, the Vela satellite also carried two bhangmeters - sensors which could detect the light flashes associated with a detonation, which included an initial brief but intense flash, and a subsequent, longer lasting flash. (Note 1)

In conducting their readout the AFTAC technicians saw a double humped signal that corresponded to the double flash associated with a nuclear explosion. In the 41 previous occurrences when a Vela satellite detected such a double flash (including the 12 Vela 6911 detections), subsequent data confirmed that a nuclear detonation had actually occurred. The signal Vela 6911 had apparently detected came from a remote region of the world, for the territory in view of its bhangmeters encompassed 3,000 miles in diameter - the southern tip of Africa, the Indian Ocean, the South Atlantic, and a bit of Antarctica. The detection took place at about 3:00 a.m. local time, September 22.

The detection raised the possibility that some nation, particularly South Africa or Israel, or the two in collaboration, had conducted a covert test. South Africa was believed to have been preparing for a nuclear test in August 1977 before Soviet and U.S. satellites detected the preparations, and diplomatic pressure caused the South Africans to deny any such plan. Israeli-South African cooperation had been reported in a variety of media sources, although the specifics were often obscure. (Note 2)

There was a discrepancy in the bhangmeter readings with regard to the second flash. Because the bhangmeters were not equally sensitive it was not expected they would produce identical numerical values. But it was expected that the ratio between the two would be the same from one detonation to the next. In the case of the September 22 detection the ratio was not what was expected from previous experience.

Given the importance of determining if a test had taken place, and who had conducted a test if it had occurred, the U.S. government devoted a considerable effort to trying to gather and evaluate evidence in order to produce definitive conclusions. One component of this effort involved searching the data already collected by a variety of U.S. data collection systems at the time of the incident. Those systems included satellites such as the Defense Support Program (DSP), Satellite Data System (SDS), and Defense Meteorological Satellite Program (DMSP) satellites - all of which carried sensors that could detect some of the signals of a nuclear explosion. DSP satellites operated in geosynchronous orbit, 22,300 miles above the earth, and carried sensors that could detect the infrared (heat) signature of a nuclear detonation, bhangmeters, an x-ray locator, and an atmospheric fluorescence detector.

Other sensors with the potential to have collected relevant data included two underwater acoustic arrays - the Sound Surveillance System (SOSUS) and Missile Impact Location System (MILS), whose primary missions, respectively, were to monitor Soviet submarines and to determine where missile test warheads splashed down. 

In addition to searching for data that might have been collected by such sensors, an effort was made to gather data that could not be collected passively - such as the debris associated with a very low-yield detonation. While AFTAC sent specially-configured aircraft to try to gather debris from the region of the apparent blast, the CIA sent some of its personnel into various nations in the region to gather the leaves from trees - leaves that might contain the radioactive residue of an explosion. Such efforts apparently were futile (although in September 1980 a professor who had been studying sheep thyroids around the world reported that iodine-131 [a fission product] had been detected in the thyroids of sheep slaughtered in Melbourne, Australia in November 1979, but not subsequently).

Various elements of the government, particularly the Naval Research Laboratory, also sought out, or were presented with, data that had been collected as the part of scientific, non-military research. Included were data from the Arecibo Ionospheric Observatory, as well as from civilian weather satellites such as Nimbus and Tiros. Two scientists working at Arecibo detected a traveling ionospheric disturbance moving in an unusual northward direction at the time of the Vela detection.

The data accumulated by U.S. and allied intelligence, military, and civilian agencies, as well as scientific institutions, were examined by a variety of analysts and organizations - an ad hoc presidential panel, a DCI panel, the Central Intelligence Agency, Defense Intelligence Agency, national laboratories such as Los Alamos and Sandia, as well as organizations under contract to the Department of Energy and AFTAC.

The conclusions of the presidential panel (the Ad Hoc Panel) were reassuring, as they suggested that the most likely explanation of the Vela detection was a meteoroid hitting the satellite - in part because of the discrepancy in bhangmeter readings. Others who examined the data, including DIA, the national laboratories, and contractors reached a very different conclusion - that the data supported the conclusion that on September 22, 1979 Vela 6911 had detected a nuclear detonation. (Note 3)

Document 1: Christine Dodson, Staff Secretary, National Security Council, Memorandum
for: The Secretary of State [and others], Subject: South Atlantic Nuclear Event, w/att: untitled discussion paper, October 22, 1979. Secret
Source: Freedom of Information Act Request

This paper was prepared at a time when it was assumed that Vela 6911 had detected a nuclear event - and the limitations of U.S. knowledge about the South African program permitted the belief that South Africa was the most likely culprit. The paper explores the nonproliferation stakes involved, the impact public disclosure would have on foreign policy efforts in Africa, the pros and cons of approaching the South African government, the effect on various nuclear negotiations with South Africa, informing the Soviet Union, possible U.N. sanctions, and the implications for public perceptions of the ability to verify a comprehensive test ban treaty.

Document 2: Guy E. Barasch, Los Alamos Scientific Laboratory, Light Flash Produced by an Atmospheric Nuclear Explosion, November 1979. Unclassified
Source: Freedom of Information Act Request

In this analysis Barasch examined the possibility that the flashes detected by the Vela satellite had been the result of a natural phenomenon. Vela bhangmeters had been triggered hundreds of thousands of times by lightning, cosmic particles, and direct sunlight. He concludes that naturally occurring signals would not be confused with signals from a nuclear detonation, whose light signature is "unmistakable." In particular, he dismisses the possibility that the Vela signal came from a lightning "superbolt," lightning that is over a hundred times more intense than typical lightning and usually occurs over water when cold polar air moves in over warm, moist oceanic air.

Document 3: Director of Central Intelligence, The 22 September 1979 Event, December 1979. Secret
Source: Freedom of Information Act Request by Natural Resources Defense Council

This study begins, as was requested by the National Security Council, with the assumption that the September 22, 1979 Vela event was a nuclear detonation. It discusses the possibility that the detonation could have occurred due to an accident, and noted the Defense Intelligence Agency's suggestion that the Soviet Union might have had reasons to conduct a covert test in violation of its treaty commitments. But the majority of the study is concerned with three possibilities to explain the incident - a secret test by South Africa, a secret test by Israel, and a secret test by South Africa and Israel together.

Document 4: Christine Dodson, Staff Secretary, National Security Council, Memorandum for: The Secretary of State, Subject: Discussion Paper for Mini-SCC, January 7, 1980, Classification Redacted
Source: Freedom of Information Act Request

This paper was prepared for a meeting of the "mini"-Special Coordination Committee. The SCC was established by President Jimmy Carter's January 20, 1977 Presidential Directive 2, "The National Security Council System," and was assigned responsibility for matters such as covert operations, arms control evaluation, and crisis management. The paper considers the options available in dealing with three issues: what the United States should say publicly and privately about the results of its analysis of data concerning the Vela incident, whether the U.S. should continue to press for a nuclear agreement with South Africa, and what position the U.S. should take on nuclear sanctions against South Africa. 

Document 5: George N. Oetzel and Steven C. Johnson, SRI International, Vela Meteoroid Evaluation, January 29, 1980. Secret
Source: Freedom of Information Act Request

This report resulted from a rush evaluation of the theory that the Vela double-flash signal was the result of a meteoroid. The only conclusion they could offer was that proposed scenarios which involved two meteoroids were likely to produce the Vela signal only once in one billion years. They also noted that meteoroid data from the Pioneer space probe suggested other models for meteoroids producing the signal, but the limited time available for the study prevented their reaching firm conclusions. 

Document 6: Chairman, Joint Atomic Energy Intelligence Committee, Memorandum for: Director of Central Intelligence, Deputy Director of Central Intelligence, Subject: Judgments of the DCI's Nuclear Intelligence Panel on the 22 September 1979 Event, February 14, 1980. Secret w/att: Judgments of the DCI's Nuclear Intelligence Panel on the 22 September 1979 Event, n.d. Secret
Source: Freedom of Information Act Request

Shortly after the Vela incident, Director of Central Intelligence Stansfield Turner asked his Nuclear Intelligence Panel, chaired by Donald Kerr, who had served in the Carter administration as acting director of defense programs at the Energy Department, to examine the data relating to the incident. According to the February 14, 1980 memorandum to the DCI, the panel reviewed data from the event on February 11-13, 1980. The facts and discussion contained in the attached report were completely redacted from the report before it was released in response to a Freedom of Information Act Request. According to Seymour Hersh's The Samson Option, Kerr stated "We had no doubt it was a bomb." (Note 4)

Document 7: Henry G. Horak, Los Alamos Scientific Laboratory, Vela Event Alert 747, May 1980. Secret
Source: Freedom of Information Act Request

Among the issues addressed by Horak in his paper is why the bhangmeters on two Defense Support Program (DSP) satellites which could receive signals from the South Atlantic area - DSP 6 and DSP 7 - did not trigger. He offers two explanations: the event did not take place within the satellite's field of view or the signal was weakened by transmission through clouds and was not strong enough to reach the necessary brightness threshold for detection. He concluded that after looking at all the data associated with the signal, there was "strong evidence that a nuclear explosion actually produced Vela Alert 747." 

Document 8: G.H. Mauth, Sandia National Laboratories, Alert 747, May 1, 1980. Secret
Source: Freedom of Information Act Request

In his report Mauth notes that of all the satellites equipped with bhangmeters, which included the Vela, DSP, and Satellite Data System spacecraft, only Vela 6911 detected the double flash associated with a nuclear detonation. He also disputes, based on laser calibration tests, the hypothesis in the Mission Research Corporation study (Document 11) that the different bhangmeter readings could be explained by a malfunction. However, based on his analysis of all the data available to him he concluded that the Vela signal was "fully consistent with those expected from a low-yield atmospheric [nuclear detonation]."

Document 9: Office of Science and Technology Policy, Ad Hoc Panel Report on the September 22 Event, May 23, 1980. Secret
Source: Freedom of Information Act Request 

Following the Vela detection President Jimmy Carter asked Dr. Frank Press, his science adviser, to establish a panel of outside experts. The Ad Hoc Panel that Press formed was chaired by Dr. Jack Ruina, a former head of the Defense Department's Advanced Research Projects Agency, and included Richard Garwin, Luis Alvarez, and Wolfgang Panofsky among its members. The panel was asked to review classified and unclassified data that could help determine whether the Vela signals had been the result of a nuclear detonation, to consider the possibility that the signal was a "false alarm" resulting from a satellite malfunction, and to investigate whether the signal was the result of one or more natural phenomena.

After three meetings, which included hearing presentations from the Defense Intelligence Agency, the Naval Research Laboratory, and other organizations, the panel completed its work and issued its report. Based on the discrepancy in bhangmeter readings, and the construction of what the panel considered a plausible alternative - which involved a meteoroid colliding with the Vela satellite and ejecting some particles which reflected sunlight into the field of view of the bhangmeters - the panel concluded that the Vela signal was probably not the result of its having detected a nuclear detonation. 

Document 10: John E. Mansfield and Houston T. Hawkins, Defense Intelligence Agency, The South Atlantic Mystery Flash: Nuclear or Not?, June 26, 1980. Secret
Source: Freedom of Information Act Request

The authors of this paper included a Harvard Ph.D. in physics (Mansfield) and a lieutenant colonel (Hawkins) who had recently joined DIA as head of the Nuclear Energy Division's Nuclear Weapons Branch. The authors noted much of the search for corroborating evidence of a nuclear detonation on September 22 had failed, and that while there were a number of signals that might be considered as corroborating evidence, those signals were weak, embedded in noise, or represented a phenomenon that was not well understood.

Among the topics they examined were the absence of detected radioactivity, the traveling ionospheric disturbance, and the possibility that one or more micrometeoroids caused the Vela signal. They proceeded to offer an explanation of how radioactivity might not have been detected despite a detonation, calculated that the probability of northward ionospheric disturbance occurring at the same as the Vela event was extremely low (not more than 0.02), and concluded, based on the calculation of Stanford Research Institute scientists (see Document 5), that the probability of a single meteoroid causing the Vela signal was less than one in one hundred billion.

When news of the study appeared in The New York Times and The Washington Post, the White House released an only mildly-redacted copy of the Ad Hoc Panel Report.

Document 11: D.S. Sappenfeld, D.H. Sowle, and T.H. McCartor, Mission Research Corporation, Possible Origins of Event 747 Optical Data, August 1980. Secret
Source: Freedom of Information Act Request

This paper, prepared under contract to the Air Force Technical Applications Center, was an expanded version of an earlier paper of the same title that had been completed in December 1979. The authors reached the same conclusion that they had nine months earlier - that the likelihood that Vela had detected a nuclear detonation was much higher than the probability of any nonnuclear explanation for the triggering of the bhangmeters. They also concluded that the data indicated a surface detonation and that there were serious flaws in the Ad Hoc Panel's meteoroid theory. 

Document 12: E.W. Hones Jr., D.N. Baker, and W.C. Feldman, Los Alamos Scientific Laboratory, Evaluation of Some Geophysical Events on 22 September 1979, April 1981. Unclassified
Source: Freedom of Information Act Request

The authors report on their analysis of data obtained by the TIROS-N weather satellite on September 22, 1979. They had sought to determine whether an electron precipitation event detected by the satellite could have been related to a surface nuclear burst (SNB). They concluded that the event was "unusually large" but not unique. On the other hand, they found no data that were inconsistent with the occurrence of a SNB. In addition, the authors note that "a patch of auroral light that suddenly appeared in the sky near Syowa Base, Antarctica a few seconds after the Vela event can be interpreted (though not uniquely) as a consequence of the electromagnetic pulse of an SNB."

Document 13: C.J. Rice, Aerospace Corporation, Search for Correlative Data, January 15, 1982. Secret
Source: Freedom of Information Act Request

This study was performed by the Aerospace Corporation, whose mission was to provide technical support and analysis to NRO and the Air Force, for the Air Force Systems Command. The author focused on the infrared data obtained by two DSP satellites (Flights 6 and 7) whose footprints overlapped that of Vela 6911. Ultimately, no confirmation of a test could be found. While at least one signal merited special attention, it was insufficiently intense and was considered very unlikely to represent a nuclear event.

Document 14: E.M. Jones, R.W. Whitaker, H.G. Horak, and J.W. Kodis, Los Alamos National Laboratory, Low-Yield Nuclear Explosion Calculations: The 9/22/79 VELA Signal, May 1982
Source: Freedom of Information Act Request

In their study, the four Los Alamos scientists discounted a variety of explanations for the discrepancies in the bhangmeter readings, including atmospheric absorption of the signal or cloud cover. They went on to present their own model for the September 22 event, a model that remains largely classified but may have relied on the effect of surface bursts on bhangmeter readings. It does seem likely that the authors shared the belief of many of their colleagues at Los Alamos and Sandia that the September 22 event was a nuclear test. They wrote that "our model is consistent with the apparent absence of nuclear debris, the collection of which is required by some analysts for absolute confirmation of an atmospheric detonation." 

Document 15: Gerald S. Wright, Air Force Technical Applications Center, History of the Air Force Technical Applications Center, 1 January 1979 - 31 December 1980, Volume I, Narrative (Excerpt), May 17, 1982. Secret
Source: Freedom of Information Act Request

This extract from an AFTAC history describes some of the collection effort that followed the Vela detection. According to AFTAC historian Gerald Wright the detection "set off one of the most extensive air sampling operations in recent years." 

1. This section is based on Jeffrey T. Richelson, Spying on the Bomb: American Nuclear Intelligence from Nazi Germany to Iran and North Korea (New York: W.W. Norton, 2006), pp. 283-316.

2. For example, see James Adams, The Unnatural Alliance (London: Quartet, 1984).

3. The Naval Research Laboratory also conducted a study, which has never been declassified. The study concluded that the Vela 6911 signal most probably resulted from a nuclear detonation.See Richelson, Spying on the Bomb, pp. 306-310.

Contents of this website Copyright 1995-2006 National Security Archive. All rights reserved.
Terms and conditions for use of materials found on this website.