Back To Black; Declassification Of Space Photography

Jeffrey T. Richelson - Bulletin of the Atomic Scientists - May 1, 2001

On February 22, 1995, President Bill Clinton signed an executive order directing the declassification of more than 800,000 images produced between 1960 and 1972 by the Corona, Argon, and Lanyard space reconnaissance programs. During the previous administration, President George Bush had accepted the recommendation of Robert Gates, then director of Central Intelligence, and authorized the Defense Department to acknowledge the existence of the National Reconnaissance Office (NRO), which had been established in September 1961 to coordinate the development of space and aerial reconnaissance systems and overflights of denied territory.

The ultimate impact of those actions was to make available tens of thousands of pages of material related to the history of U.S. space reconnaissance. Beyond the images released to the National Archives (a good number of which appear on the Federation of American Scientists web site), the NRO declassified more than 40,000 pages of documents relating to Corona, Argon, and Lanyard--including documents about the reconnaissance systems, management of the reconnaissance program, and the 1960s battles between the NRO and CIA over reconnaissance issues. In June 1998, details concerning the first U.S. ferret satellite--initially designated GRAB (Galactic Radiation and Background) and then Dyno--were declassified. Then, in 2000, the CIA released several thousand pages of imagery interpretation reports from the 1960s and early 1970s based on the "take" from Corona and other U.S. imagery satellites. In response to Freedom of Information Act requests, the NRO has released several thousand pages of documents covering topics such as its creation, early evolution, declassification, and recent restructuring.

As a result, scholars and journalists now have the opportunity to explore the history and evolution of the U.S. space reconnaissance effort and its contribution to U.S. national security in a manner that was virtually unimaginable a decade ago.

Now, however, further declassification may be blocked by differences within the U.S. intelligence community over what to release, as well as by the recommendations of a recent commission headed by Cong. Porter Goss, chairman of the Permanent Select Committee on Intelligence, and former Sen. Robert Kerrey.

Wanted: More secrecy

A review of the NRO--particularly with regard to questions of structure, mission, and vision--is nothing new. Outside reviews were conducted in 1989, 1992, and 1996. But unlike the earlier reviews, the November 2000 Report of the National Commission for the Review of the National Reconnaissance Office: The NRO at the Crossroads was the result of congressional action, rather than being undertaken at the request of the director of the NRO.

The report revealed a yearning for greater secrecy on the part of its authors. Among its recommendations was the creation of an "Office of Space Reconnaissance" within the NRO to work on advanced systems. According to the report, the office "would operate from facilities separate from other space reconnaissance activities, and it would be covered by a new security compartment. The purpose would be to establish effective secrecy to shield the technologies and collection techniques under development." The authors imply that the current headquarters, with its "Byeman" security compartment (the Byeman control system is used to protect information related to NRO collection systems), is not sufficient to maintain secrecy, and that the Advanced Systems and Technology Directorate, established as a result of the 1996 review, is not adequately performing its mission.

The commission also observed that: "Until 1992, the NRO was surrounded by a wall of secrecy. The environment kept foreign intelligence services from gaining a comprehensive understanding of U.S. space reconnaissance capabilities. The absence of information on NRO spacecraft attributes, sensors, and its approach to the development of new technology hampered those who intended to use cover and denial and deception techniques to counter U.S. space reconnaissance."

Earlier in its report, the commission suggested, without providing examples, that the "widespread knowledge of the NRO's existence and public speculation on how NRO satellites are used" had been helpful to terrorists' and other potential adversaries' efforts to develop denial and deception techniques "to thwart U.S. intelligence efforts." And at a November 15 news conference marking the release of the report, Bob Kerrey, who only a few years earlier had called for much greater openness on the part of the NRO, observed that "We've got to get back to black." The NRO commission thus effectively repudiated the wisdom of even declassifying "the fact of" the NRO's existence.

But the commission's conclusions are highly questionable. The pre-1992 golden age of secrecy is mythical, and current NRO security as well as new NRO programs have succeeded in making its operations more opaque than they were before declassification. Furthermore, increasing the secrecy around the NRO may hamper its ability to support U.S. national interests.

Forgetting the past

A review of the historical record demonstrates that foreign intelligence services managed to obtain significant information about U.S. satellite reconnaissance systems before 1992.

In 1968, the United States launched the first high-altitude communications intelligence satellite, code-named "Canyon," which at the time was mistakenly reported to be an early warning satellite. Five more successful Canyon launches followed through 1977. The quantity of data the satellites produced was so vast that the British and Canadian signals intelligence agencies were recruited to do much of the translation and analysis. Among the Britons involved in the effort was Geoffrey Prime, who would be unmasked as a Soviet agent several years after his 1977 resignation from British intelligence. As a result, Soviet intelligence discovered the role of Canyon more than two decades before its true function was publicly disclosed in the West. (Its existence and mission are still officially classified.)

Even more damaging were disclosures by Christopher Boyce and William Kampiles. In 1975 Boyce worked in TRW, Inc.'s "black vault," where he monitored communications relating to CIA-TRW satellite projects, which included Rhyolite, a geosynchronous satellite that could intercept the telemetry from Soviet missile tests as well as a broad range of communications. That year, Boyce, aided by childhood friend Andrew Daulton Lee, began selling roll after roll of film containing photographs of documents from the black vault, including documents on Rhyolite, to the KGB.

Kampiles was a disgruntled CIA employee who left the agency in November 1977 and sold the Soviets a copy of the KH-11 Technical Manual, which described the capabilities of the KH-11 (also known as "Kennan"), then America's newest photographic reconnaissance satellite capable of returning images in real-time.

That there were other penetrations remains a serious possibility. The identity of some of the projects compromised by former National Security Agency employee Ronald Pelton remain undisclosed, and could well include Vortex, the successor to Canyon. In addition, the names of several Soviet agents working for U.S. defense contractors have not been publicized. They include "Zenit," a TRW scientist recruited by Boyce.

The commission's implicit claim that there is a better outside understanding today of NRO operations than in earlier years is also debatable. A combination of leaks and disclosures due to the uncovering of the activities of Boyce and Kampiles provided a detailed map of NRO programs, including code-names and missions, in the mid-1980s to early 1990s.

Because a number of new programs reached the deployment stage in the 1990s--including advanced imagery and geosynchronous signals intelligence satellites and what appear to be some satellites with special missions and unexpected orbits--the picture of NRO activity has become much fuzzier than it was before 1992.

Authorized disclosures

An examination of the material released by the NRO or other elements of the government since September 1992 does not support the conclusion that the releases could be employed by terrorists or others to defeat U.S. intelligence collection priorities.

The NRO has revealed precious little concerning current or future satellite systems. It permitted the CBS television network to photograph two satellites under construction (widely believed to be an imagery relay satellite and a radar imagery satellite), acknowledged awarding a contract to Boeing to develop satellites for its Future Imagery Architecture, and announced its decision not to go ahead with the Integrated Overhead SIGINT [Signals Intelligence] Architecture-2.

But while the NRO has also acknowledged its satellite launches, which cannot be concealed in any event, it has provided no information on the names, numbers, capabilities, or targeting of current U.S. reconnaissance satellites, including the advanced KH-11 or Onyx radar imagery satellites, the new high-altitude signals intelligence spacecraft launched since 1994, or the new ocean surveillance/ferret satellites. The only "information" to leak out was an arm patch, given to the crews that launched NRO radar imagery satellites, showing the satellites at particular inclinations. [See "In Brief," November/December 2000 Bulletin.] However, these inclinations had already been determined by civilian space observers, armed with nothing more than binoculars and telescopes. (The Defense Department has posted degraded satellite images of the targets of U.S. military action on its web site, but it can hardly be surprising to Iraq or Yugoslavia that targets were photographed before they were bombed.)

Further, a veil of official secrecy still covers all systems other than Corona, Argon, and Lanyard. The last of those satellites flew in 1972. Still officially secret are the Hexagon and Gambit imagery systems, whose declassification by 2000 was envisioned in President Clinton's 1995 executive order; long-defunct signals intelligence satellites like Canyon; and one system whose only launch occurred in 1964. A Freedom of Information Act request for documents concerning that system--a radar imagery satellite designated "Quill"--was denied on the same day it was received at the NRO.

Unauthorized disclosures

Other sources of information on U.S. space reconnaissance systems are research organizations and the press. One press account even suggested that the commission's desire for a stricter security environment was a reaction to information posted on the Federation of American Scientists (FAS) web site that was maintained, until late last year, by John Pike.

According to the November 17, 2000, issue of Inside the Air Force, the VAS web site "contains information about many classified NRO programs, a situation that has irked some senior Defense Department officials." The site contains detailed drawings and descriptions of a variety of NRO programs, which to the uninitiated (including many journalists) may seem like a gold mine of information. However, as many inside and outside the U.S. government realize, over the years much of the information emanating from FAS with regard to NRO programs has been incorrect and based on considerable speculation.

One former CIA analyst, who prefers to remain anonymous, has observed that the FAS site contains "good drawings of imaginary spacecraft." Indeed, the site has also contained details of imaginary reconnaissance programs, such as the supposed Space-Based Wide Area Surveillance System (SBWASS). For many years, FAS claimed that these air force and navy surveillance systems were in operation (one using radar, the other infrared), with the first satellites having been launched in 1989 ( spp/military/program/surveill/sbwass.htm). But a declassified U.S. Space Command history covering 1990 and 1991 clearly demonstrates that there were no such systems in operation at the time (nor any guarantee that there would be in the future). Indeed, the total funding for SBWASS-related programs in fiscal year 1991 was a mere $ 4.6 million, to be used for a software development program designated Have Gaze.

A number of after-the-fact revelations concerning the targets of U.S. imagery satellites have appeared in the press, particularly in articles by Bill Gertz in the Washington Times. In an attempt to provide evidence to support claims about Chinese, Russian, and other nations' activities, one or more sources have revealed the contents of U.S. satellite imagery to Gertz.

But these revelations do not stem from the 1992 declassifications. Pre-1992 Gertz articles discussing the intelligence gleaned from satellite imagery included ones that discussed developments in Algeria, Cuba, Iraq, Israel, Laos, North Korea, and the Soviet Union. Long before 1992 a number of other newspapers also carried articles discussing the results of satellite imagery of China, Libya, Russia, the Soviet Union, Syria, and other countries--a product of the decades-long practice of national security officials leaking information to support their evaluations of events and threats.

Furthermore, casual suggestions that such revelations are a significant factor in facilitating foreign denial and deception--such as India's ability to conduct surprise nuclear tests in 1998--are questionable. Despite all the revelations before December 1995, the United States was still able to detect Indian test preparations. Not surprisingly, after being confronted by the U.S. ambassador with U.S. satellite images of those preparations, the Indians took steps to evade coverage leading up to their 1998 tests.

The light of day

Thus, the evidence does not support the commission's depiction of a golden era of NRO secrecy before the 1992 declassification--or a flood of damaging disclosures since then. The most serious hostile intelligence penetrations of NRO programs occurred more than a decade before declassification. Meanwhile, the material released by the NRO has been of value to scholars and journalists and not adversaries of the United States. A history of the management of the National Reconnaissance Program from 1960 to 1965 or a photograph of a Soviet missile site in 1968 is of value to historians, but not to Iraqi officials concerned with denial and deception.

The call for greater secrecy may not only limit the disclosure of information of historical value, such as the Gambit and Hexagon programs or early high-altitude signals intelligence programs, whose declassification is long overdue. It may also hamper the actual work of the NRO and its value to its customers.

One reason for the 1992 declassification was to facilitate NRO'S support of a broad set of customers and interests ranging from the military in the field to environmental science. A supersecret "Office of Space Reconnaissance" within the NRO, isolating developers of future systems from current operators and future customers, might well decrease the probability that the NRO will produce systems that fill critical intelligence needs.

Nor are the commission's recommendations in line with the greater openness that has been practiced by a number of U.S. intelligence agencies in recent years, including the NRO. A preferable course would be to follow the recommendation that Senator Kerrey himself made in 1996, when he suggested to the Senate Select Committee on Intelligence "that not only the top line [of the NRO budget], but several other lines of the budget, not only could but should [be disclosed] for the purpose of giving the taxpayer-citizens confidence that their money is being well spent."

Jeffrey T. Richelson is a senior fellow with the National Security Archive in Washington, D.C., and the author of America's Space Sentinels: DSP Satellites and National Security (1999).

  • See Also: The William Kampiles Case