by Dwayne Day
MOTIVE FOR SECRECY
Eisenhower proposed his Open Skies concept at Geneva in July 1955, only to have it rejected by Khrushchev. The Soviets protested publicly and loudly when American reconnaissance balloons began to drift across the USSR in 1956. Vigorous diplomatic protests were made following each U-2 overflight, although the Russians never admitted to their own people that Soviet territory was being systematically violated. Eisenhower was determined not to confront the Soviets with the reality of our ongoing reconnaissance program. He refused to contradict John Kennedy’s claim of a missile gap during the presidential campaign of 1960, even though he knew from both U-2 and (after August 1960) CORONA flights that the claim was wrong. He tried in every way to legitimatize overhead reconnaissance and hoped gradually to gain Soviet acceptance of it. This was a slow process, but it was finally successful. Overhead reconnaissance is now enshrined in each of our strategic AVMS limitation treaties with the former USSR, where it is described as “National Technical Means.”
EARLY SATELLITE PLANNING
In 1945 the U.S. Air Force commissioned RAND to examine the feasibility of launching satellites for military purposes. RAND’s early studies explored the idea of reconnaissance from space and identified the polar orbit necessary. RAND technical people, working with a growing group of Air Force officers, advocated satellite reconnaissance and did a great deal to establish the concept as a reasonable one. American ballistic missiles then being designed offered the means for lifting such satellites into orbit. RAND’s work was focused primarily on the requirements of the Strategic Air Command (SAC), which was then the dominant voice in the Air Force. The SAC need was the information required for poststrike bomb damage assessment. This would allow targets that were missed by the first wave of SAC bombers to be retargeted for subsequent attacks. The resolution required for this task was not great, since one would be trying to locate large craters in relation to cities and military bases. On the other hand, this information would be needed promptly. These requirements were addressed in the satellite system proposed by RAND in 1954.2 Its solution was based on a television satellite with a resolution of approximately 100 feet, operating 300 miles above the earth. These proposals became the blueprint for the photographic reconnaissance component of what came to be called the WS-117L program.
REFOCUSING THE SATELLITE PROGRAM
The accelerated development of American and Soviet ICBMs gave special urgency to Eisenhower’s fear of a surprise attack. With only a 30-minute flight time and no realistic prospect for defense, the emerging reality of a Soviet ICBM force put the possibility of surprise attack at the top of the president’s concerns. The shock of Sputnik crystallized both public and presidential concern in October 1957. It also made people in high places think seriously about satellite reconnaissance.
In 1956, the CIA’s U-2 program had been expected to operate only for a year or two. In 1957, its second year of overflights, missions were few in number and limited in coverage. They could not survey large areas of the USSR to look for deployed ICBMs, and it was unrealistic to rely on them for guiding national security policy in the nuclear missile era.
There was a sense of extraordinary urgency in getting good pictures of the entire USSR; a satellite reconnaissance system was the obvious way to do so. Eisenhower asked the President’s Board of Consultants on Foreign Intelligence Activities (PBCFIA) to review the Air Force program and make recommendations. The PBCFIA’s report in late 1957 was skeptical that WS-117L could provide the needed capability. Because its resolution was more than 100 feet, it would be too crude to provide strategic intelligence. The satellite’s resolution was limited by the small focal length of its camera and the narrow bandwidth of its television downlink. The program was running late and encountering technical difficulties.3 In an effort to develop budgetary support for it, the Air Force had started to publicize the WS-117L program.
The U-2 prototype at its desert airstrip. This was an enormously successful project and established the precedent of close CIA-Air Force cooperation on development of reconnaissance programs. (Photo courtesy of Chris Pocock)
Killian and Land were frustrated by what they saw of the Air Force program. They strongly recommended simplifying and accelerating satellite reconnaissance activity. They wanted to start a new program based on film return from orbit, focusing on peacetime national intelligence objectives rather than on reconnaissance after a nuclear exchange. Eisenhower agreed. He was concerned with preventing nuclear war, not waging it.
Killian and Land wanted to streamline both the program and its technical management system. They urged the president to assign the leadership for a new system to the CIA, to be supported by selected elements of the Air Force. Richard Bissell would lead a small group of CIA and Air Force people, thus emulating the successful partnership that had created the U-2. This recommendation was made for several reasons. The CIA had demonstrated an ability to maintain tight security during the development phase of the U-2 project. Bissell and his CIA people had shown a remarkable capacity for moving rapidly from concept to operation. They had demonstrated an ability to make and implement decisions quickly. In addition, they were quite open to suggestions from Land and his colleagues.
The presidential decision to proceed with what became CORONA on this basis was made just eight weeks after the PBCFIA report was submitted.4 The goal was to achieve a resolution of 25 feet or better within one year. Elements of the WS-117L program that promised early capability were transferred to the new program. As it turned out, what was imagined to be an interim system became the backbone of U.S. intelligence collection capability for the next 12 years.5
A second reconnaissance program was approved by President Eisenhower at about the same time. The CIA and Bissell were authorized to develop an aircraft as a successor to the U-2 that would fly at three times the speed of sound. This piloted reconnaissance system would complement the CORONA satellite. It would provide greater coverage flexibility and greater resolution than could be obtained from Earth orbit. The program was called OXCART. It became operational in 1966, but was never used over the USSR because of the pledge to stop overflying Soviet territory with airplanes which Eisenhower made following the downing of Gary Powers’s U-2 in May 1960. Follow-on versions of this remarkable reconnaissance airplane were operated by the U.S. Air Force as the SR-71 until 1990.6
DIFFICULT DESIGN CHOICES
Richard Bissell chose Air Force Brigadier General Osmond Ritland to be his deputy for the CORONA project. The two had worked together on the U-2 program and had the utmost confidence in one another. A program office was assembled in Los Angeles in early 1958, comprised of about five Air Force officers commanded by Colonel Lee Battle. It was supported by a small group of CIA officers in Washington reporting directly to Bissell.
The program group decided to use the Thor intermediate-range ballistic missile as the first stage of a rocket combination to place CORONA in Earth orbit. This was logical because the Thor had been flying successfully since 1957 and thus was further along in development than Atlas or Titan. Thor was then in large-scale production and was being deployed to operational sites in England. More important, a Thor launch training site being established at Point Arguello in California7 could fire directly south, and thus place a satellite in polar orbit. Douglas Aircraft Company built the Thor rocket and became a charter member of the CORONA team.
The CORONA payload needed to be placed in what came to be called “reconnaissance orbit.” This is a retrograde 90-minute orbit that passes over the North and South Poles. It has a perigee of approximately 100 miles and an apogee of 240 miles. The rotating Earth turns beneath this fixed orbital plane, presenting a new swath of territory on each pass. The Thor would burn out at an altitude of 70 miles, well short of the speed required to reach this orbit. An additional rocket stage was thus needed to lift CORONA.
The Agena upper-stage vehicle had been in development at Lockheed for two years, but had not yet flown.8 It was five feet
in diameter and used a 16,000-pound rocket engine that burned storable propellants: hydrazine (UDMH) and nitrogen tetroxide. Agena was the logical choice to provide the additional velocity needed to reach Earth orbit. Lockheed thus became the second team member.
Bissell and Ritland decided to combine the second-stage and the orbital-spacecraft functions. This meant that the camera and film-recovery systems would remain attached to the Agena even after it had exhausted its fuel. Agena would have to provide precise attitude control, battery power, and thermal protection for the reconnaissance payload for several days, and later for several weeks. The choice of a camera system would determine the performance required from this redefined Agena.
The most important decision facing Bissell and Ritland was the type of camera to be used. RAND and Lockheed had done some work on a film-recovery system based on a spinning spacecraft to provide stability and to achieve the scan. It used a relatively short (12-inch) focal-length camera design by Fairchild, which had no provision for orbital image motion compensation.
An experienced reconnaissance camera design team had recently left Boston University to form the Itek Corporation. The team proposed a 24-inch panoramic camera, like those used to take pictures of large university and high school classes. The film would be held stationary in a cylindrical platen as a rotating “telescope” (i.e., a lens assembly) scanned a slit image over it. The next section of film would then be moved forward along the platen while the telescope returned to the starting position and the process was repeated. Scanning the rotating telescope over a 70-degree arc would map out a strip 10 miles by 120 miles on the ground. The specific design was based on the cameras Itek had built previously for covert balloon flights.9 This design offered significant future opportunities for improving resolution. It was entirely mechanical, but included fixed-image motion compensation set for the planned orbital speed and altitude. The telescope lenses were diffraction limited and required precision optical glass. Bissell chose this design with the overriding objective of obtaining high-definition peacetime intelligence. With this choice, the 25-foot resolution goal was easily met; it would eventually improve to six feet.
The camera decision shifted a large burden to the Agena spacecraft. It did so at a time when the United States had virtually no experience in building such a vehicle. The panoramic camera design required that the spacecraft be stabilized around all three axes with an active control system.10 An attitude control system using gyros, infrared horizon scanners, and cold gas jets eventually gave an accuracy of 0.20 degrees in pitch, roll, and yaw. This active control was augmented by horizon, star, and framing cameras that recorded the instantaneous vehicle attitude for later use by photo-interpreters. This decision required redesign of the spacecraft and accounted for many early development problems.
Three decades of space missions have used the basic Agena developed for CORONA. The LANYARD spotting system and ARGON mapping missions could not have been completed without it. In a very real sense, CORONA pioneered all subsequent satellite reconnaissance programs.
A time-lapse exposure of a HYAC balloon reconnaissance camera as the swing arm travels along the curved film platen, exposing the film. The 12-inch HYAC camera served as the model for the 24-inch CORONA camera. (Photo courtesy of the National Reconnaissance Office)
A fine-grain film was needed to fulfill the resolution promised by the Itek camera. Eastman Kodak had developed such film for the U-2 program. For CORONA, Eastman Kodak developed an acetate-based 70-millimeter film which was approximately three millimeters thick. It was relatively slow but gave 280 line pairs per millimeter over the entire field of view, at high contrast.11 This compares with the best reconnaissance film used in World War II, which gave 10 to 50 line pairs per millimeter.
There was concern that scintillation induced by the turbulent atmosphere might limit CORONA resolution. Image motion of one or two arc seconds was consistently measured by astronomical telescopes.12 At a slant range of several hundred miles, this effect could produce image smearing of three or four feet. That was below the CORONA threshold, but scintillation was considered carefully for follow-on satellite systems.13 Fortunately, image quivering is considerably reduced when the observer is well removed from the turbulent region rather than being immersed in it, as terrestrial telescopes are.
When the CORONA program began, American ballistic missile programs had already developed ablating nose cones that could withstand the enormous heat loads generated during ballistic reentry. A capsule had never been returned from orbit, but it was clear that the heating involved in orbital reentry would be substantially less severe than a missile warhead because the satellite return trajectory is quite shallow compared to a ballistic path. The General Electric Company had developed ablation technology for U.S. ICBMs and was assigned the task of developing the film-return capsule for CORONA. The center space of the capsule held a large take-up reel for the 70-millimeter film that would pass through the Itek camera. A small rocket motor was fitted to the rear of the capsule. Agena would re-orient itself at the end of the mission and separate the capsule that now contained the exposed film.14 The capsule would “spin up” for attitude stability, and then fire its rocket motor to reduce its speed by 1,300 feet per second.15 This was enough to send it back to earth after traveling a quarter turn in orbit.
To begin the recovery operation, a microwave command signal was sent to the Agena as it came over the North Pole heading toward the equator. If all worked according to plan, the CORONA capsule would splash down near Hawaii in an impact area of 150 by 400 miles. The maximum heating rate during reentry would occur at 350,000 feet altitude, where the ablation heat shield would reach temperatures of 4,000 degrees Fahrenheit. A parachute would deploy when the capsule reached 50,000 feet, and would slow the descent rate to about 30 feet per second. A fleet of Air Force C-119s were deployed from Hawaii to snatch the descending capsule. These planes each towed a long nylon loop with which the air crew attempted to snare the parachute and then reel it into the aircraft.16 Ships and helicopters were also deployed to recover the capsule if it was missed and fell into the sea. It took a great deal of trial and error to make aerial recovery a routine operation.
A two-page work statement dated April 25, 1958, reflects these daring decisions. Subsystems were rapidly built up by teams of engineers and technicians working under pressure comparable to wartime conditions. The cameras, film, and reentry capsules were integrated at a secret facility near Palo Alto, California. The completed CORONA payload was then taken to the Vandenberg Air Force Base where it was mated to the Thor-Agena rocket. The first CORONA mission was ready for launch on February 28, 1959—less than 10 months after program go-ahead.
DEVELOPMENT PROBLEMS
The first twelve CORONA missions were failures. It is important to remember that the country had almost no experience in developing satellites before 1958. The pioneering role fell on CORONA, not the much less ambitious scientific satellite program carried out for the International Geophysical Year. In the current state of technical accomplishment, we tend to forget how desperately inexperienced we all were in those early days. Vanguard had been a profound embarrassment. Thor and Atlas had gone through similar problems when they began. Titan was then experiencing repeated failures. I had worked on ballistic missile programs in Los Angeles; our monthly program review meetings were properly called “Black Saturdays.” Upper stages were a new challenge. Our group had tried and failed three times to build an upper stage for Thor which could send a payload to the moon.
Getting the film capsules to land in the desired recovery areas also proved to be difficult. We found that we still had a great deal to learn about snagging the parachutes of descending film capsules. The acetate-based 70-millimeter film broke several times during orbital operations; this represented a genuine threat to the program. Fortunately, for other applications Eastman Kodak developed a polyester-based film which solved that problem.
One of the greatest problems was the pressur
e to continue launching at a rapid pace. This was driven by the extraordinary urgency of getting firm evidence of Soviet strategic deployments—evidence that could only come from CORONA. During the first year, despite major technical problems we launched about once a month. This short interval between launches did not give the engineers enough time to analyze and fix problems before the system was launched again. (It is useful to recall that the space shuttle was grounded for almost three years after the Challenger accident. That luxury was not available to the men and women working on CORONA.)
These problems were further compounded by the fact that the satellite had a razor-thin weight margin and thus could carry very little test instrumentation. This meant that the engineers usually did not have enough diagnostic data to correct the problems with confidence. The choice each time was simple: carry film or carry instruments. We almost always chose film. What is remarkable is that Bissell and Ritland pressed on despite these failures, and that Eisenhower continued to support them.
IMPACT ON INTELLIGENCE
The thirteenth CORONA mission was launched on August 10, 1960, and became the first completely successful flight. However, it carried no film because the program office had decided to fly a full load of diagnostic instruments. The fourteenth CORONA mission was launched a week later and, after seventeen passes over the USSR, returned 16 pounds of film.17 This mission produced a cornucopia of data and gave more coverage of the USSR than all prior U-2 flights combined. For policymakers and intelligence analysts alike, it was as if an enormous floodlight had been turned on in a darkened warehouse.
CORONA photography quickly assumed the decisive role that the Enigma intercepts had played in World War II.18 When the American government eventually reveals the full range of reconnaissance systems developed by this nation, the public will learn of space achievements every bit as impressive as the Apollo moon landings. One program proceeded in utmost secrecy, the other on national television. One steadied the resolve of the American public; the other steadied the resolve of American presidents.