Eye in the Sky: The Story of the CORONA Spy Satellites

by Dwayne Day

  Another challenge that we faced was that the 24-inch Petzval f/3.5 vibrated a lot. To compensate for the problem, we decided to constantly rotate the lens. Fortunately, this adjustment decreased the vibration and we were able to get 180 to 200 lines of resolution per millimeter, which equaled 10 to 12 feet of ground resolution—a major improvement in performance.

  The next major improvement was the MURAL camera system, which consisted of two C‴ [“C triple prime”] cameras, one pointing forward, and one pointing backward, at 15 degrees each. This gave us a stereo angle of 30 degrees, and from that you could create a stereo model so that interpreters could calculate—in addition to being able to make the standard measurements in x and y—measurements in z, or in other words, they could get “height” information as well.

  The film take-up reels inside a CORONA bucket. Note the two film strips at top.

  The final modification of the CORONA camera was the Constant Rotator or J-3. We were still having an intermittent slip of the scan arm with the C‴ lens cell (the system used in the MURAL and J-1 cameras). So what we did was put a different drum on it so that the scan arm, as well as the lens, constantly rotated. This allowed the stereo coverage to work properly. It was the last of the CORONA series and was an extraordinarily successful and reliable system. Statistics show that it was a highly reliable camera system and a highly reliable vehicle.

  The ultimate improvement in the CORONA system was when we went to the J-1 and KH-4A’s dual film-bucket system. After that innovation we had a great deal of flexibility. For example, we could expose a bunch of film, load it into one of the buckets, and then reenter that bucket while still maintaining the satellite in orbit for another series of film. In other words, the system gave us twice as much coverage as the one-bucket system and the flexibility to photograph a different area than we had for the first bucket. In all, the J-1 increased CORONA’s orbital life to approximately 20 days and operated without failure for more than five years.

  The final problem I would like to talk about is the so-called corona effect on the film. From the very beginning, we had evidence of static electrical discharge. Sometimes it was dendritic tree-like patterns, other times it was bands of increased density, but we always had corona. Ed Purcell of Harvard, who was a member of the Itek Science Advisory Board and the Land Committee, made several suggestions for improving the situation. We tried them all. We even tried making the film roller out of static-free conducting rubber, but that didn’t consistently work. So what we finally did was put the rollers in a test rig; we then put them in a steam bath and installed them in a camera so that we could test them in an environmental chamber. If we got static discharge, we discarded them. But if we didn’t, we used them. And that’s the way that problem eventually got resolved.

  There were many people at Itek who contributed to the success of this camera, but it would be particularly unfortunate if I didn’t at least mention two names: one of them is John Wolfe, who was the project manager for most of the CORONA program, and the second one is Frank Madden, who was the chief engineer on the camera from the very early days of the balloons right through to the final J-3 design. Madden’s calm guidance, his penetrating intellect, his pragmatic approach, and his ability to come up with solutions to problems that appeared to be intractable were uncanny. He says that he thought the “corona” static problem was resolved by divine intervention. I note that it was his determination that eventually solved that problem.

  In summary, as Allen, Garwin, Levison, McMahon, and Plummer have noted, determination was an element that everyone who worked on CORONA possessed. Despite a number of early setbacks, the CORONA pioneers developed a distinct “can-do” attitude that helped them push themselves and make sure their plans became a reality. One of the most remarkable aspects of the CORONA program was the extensive cooperation that occurred between the government, the military, and industry. The teamwork between these three sectors assured the development of America’s first reconnaissance satellite. Ultimately, it seems that everyone involved with the program sensed that CORONA would be a key factor in deciding the outcome of the Cold War. But perhaps the best way to summarize the evolution of the CORONA program is to let some of the CORONA pioneers recount their ideas about the system’s development, for, as General Lew Allen notes:

  We were in a very unstable situation because we did not know what the Russians’ capabilities were. Consequently, I think all of the people involved felt very strongly that intelligence was critical and that knowledge of the Russians’ strategic capabilities was important in order to achieve stability.

  This program contained very challenging things. These challenges were all brand new for everyone involved, and it took an enormous amount of dedication and innovation to pull them all off. As one looks back on the program, it is remarkable that these people persisted even though a long series of failures occurred. I think it’s a remarkable story. It was a program of remarkable accomplishment, and remarkable cooperation between the Air Force and the CIA.

  James Plummer has similar admiration for the teamwork and cooperation that existed during the CORONA period and notes:

  The real work was done by the shop people, the technicians, and the design engineers. So if a person were to say, “Who’s responsible for CORONA?” I’d have to say it’s measured in the hundreds and hundreds of people. It is obvious that no single individual created this system. Industry worked with government. It was clearly a team effort.

  But perhaps Walter Levison most accurately captures the spirit and feelings of the men and women who worked on CORONA:

  CORONA was a great program. We worked very hard. We enjoyed being on it, and, it would not have been possible without the kind of environment that Dick Bissell, Joe Charyk, Jim Plummer, and the rest of the management staff provided for us. We all felt we were doing something important. We all felt they trusted us. And we all felt we were being amply rewarded for this job.

  10

  CORONA AND THE REVOLUTION IN MAPMAKING

  CORONA had a major impact on both American military and civil mapmaking during the height of the Cold War. In this chapter, four photogrammetrists and geodesists who worked with the CORONA program discuss how they used reconnaissance satellite imagery to create a global mapping system that defense officials could use to locate and target objects more accurately. They also describe how they used CORONA’s photographs to develop methods and technologies for revising civilian maps of the United States and other areas of the world.

  These four mapping experts’ comments range from their first experiences with the CORONA system to the profound effect that the satellite had on both the military and civilian mapping communities. Although each of them has a different perspective on CORONA, two common themes emerge in their discussions: the innovations they developed in order to exploit CORONA imagery during the mapmaking process, and the tremendous cooperation that occurred between the military, civilian, and industrial sectors when it came to exploiting CORONA’s mapping and targeting potentials. In all, these four mapping experts reveal that CORONA fostered a revolution in the sciences of geodesy, photogrammetry, and cartography, a revolution that would lead to the development of a sophisticated global mapping system for both civilian and military use.

  Elaine A. Gifford was a photogrammetrist at the National Photographic Interpretation Center (NPIC) during the latter part of the CORONA era. Gifford began her career at NPIC right after college.1 In her discussion, she gives a thumbnail sketch of the mapping process and recounts the kinds of questions that mapping experts who were working with the CORONA program had to address. Gifford notes:

  My involvement with the CORONA system began in 1965 when, as a brand-new employee of the CIA, I started my duties at the National Photographic Interpretation Center. I still remember the awe of it. Things were coming though NPIC’s door on a regular basis in high volume. The CORONA system was a real workhorse. I was really excited when I found out about all the photogrammetry and mapping
we were doing and saw the different kinds of capabilities that existed. It was a wonderful opportunity for me.

  However, I think it’s important to note that the CORONA system wasn’t initially designed for mapping purposes. It was initially there to “peer behind the curtain” and give us a sense of the Soviets’ military might. Nevertheless, it quickly evolved into supporting mapping capabilities.

  There were many complexities to the mapping process that we hadn’t dealt with before CORONA. There were horizon and stellar cameras, and other technologies that could help us fix the location of a satellite so that we could get good geodetic positions of points on the ground. Every day, intelligence people were coming in with new discoveries—new missile sites, new complexes, even new cities.

  We were mapping much of the Eurasian landmass on a daily basis in the mid-1960s. Requests for detailed information and intelligence came in to us regularly. People would ask us questions like, “Exactly where is this site located? What are its exact dimensions? Which way is this missile pointing?” We also did a lot of “line-of-sight” types of estimates at NPIC, which answered the question, “If I’m standing here, can I have a direct line-of-sight to another spot in this location?” In short, there were all sorts of detailed questions that challenged us to add value and dimension to what we were actually seeing on the film.

  It is also important to remember the context in which we had to deal with things in those days. For example, we didn’t have PCs [personal computers] or any other type of individual computer support. We did all of our centralized photogrammetric jobs on a mainframe instead. We’d get two or three jobs processed a day. We did most of them overnight. They were processed on punch cards. We also generated a lot of paper tape that we would then feed into plotters to get more detailed site maps. And we didn’t have handheld calculators in 1965; during that period, we had to look up trigonometry functions and use slide rules. Furthermore, as often happens, the ground resources lagged behind the overhead satellite system, so that when the satellite systems were coming into their own, there were a lot of folks on the ground who were doing the real yeoman’s work with new tools that were just evolving. They had to use those tools to get the full range of information that was available from the imagery that was coming through the door.

  William C. Mahoney, another CORONA photogrammetrist, began his career as an Army topographical officer during the 1940s and worked his way up to a position as a photogrammetrist at the Air Force’s Aeronautical Chart Information Center (ACIC) by the early 1960s.2 Mahoney recounts—in even greater detail than Gifford—the challenges mappers faced as they developed a global mapping system. He also elaborates about the processes and techniques mapmakers used to exploit CORONA’s information to its full potential.

  Imagery from a 1963 CORONA mission. A number of Hen Roost and Hen House phased-array radars located at the Sary Shagan antiballistic missile complex along Lake Balkhash are visible at the center of the image. Sary Shagan was a high-priority target for U.S. intelligence. Similar radar facilities were located throughout the Soviet Union, but not in the concentration seen at Sary Shagan.

  Before I went to work for the ACIC, I was studying analytical triangulation at Ohio State University. Up to that time, anybody who was interested in that field had been forced to make monoscopic measurements on photographs and then manually compute and record the complex records of what he or she measured. The process took a long time. The most advanced computers available at that time (1957) were severely limited. What we were doing at Ohio State was making analytical triangulation practical by using a first-generation IBM 650 digital computer. To do so, we seized upon an idea proposed by Dr. Paul Herget—which was based upon vector mathematics—to analytically extend a strip of photographs, similar to the model-by-model manner used to extend photogrammetric triangulation strips using an analog optical system. His concept did not require a high-capacity computer solving large arrays of least squares equations. To satisfy part of the requirements for my Ph.D. degree, I developed and demonstrated the first successful analytical triangulation system by applying Dr. Herget’s concepts and running a 400-mile frame photogrammetric strip over the United States using reconnaissance-type photographic materials.

  I accomplished this by basing the system on stereo-comparator measurements rather than the traditional mono-comparator photo measurements that photogrammetrists had relied on in the past. When I first started to apply the Herget method, the two-plate stereo comparator didn’t exist. We developed our own by converting a WILD A7, a first-order stereo-optical plotting instrument, into a stereo-comparator. This allowed us to bypass standard distortion-free mapping photographs, which had previously limited our use of such instruments.

  A map of the Sary Shagan antiballistic missile complex developed from CORONA imagery. A vast array of advanced tracking radars was located on the shore of Lake Balkhash and used to track Soviet reentry vehicles fired at Sary Shagan during tests of ABM systems.

  In June 1959, I was recruited to the ACIC (Saint Louis) by Tom Finney, who was the head civilian of the center. He had kept track of my work at Ohio State, and, as soon as he saw that I was getting ready to leave OSU, he approached me. The interesting part was that three days after I joined the ACIC, all of my paperwork had been completed so that I was fully cleared into every photo-intelligence collection system that existed at the time. Finney had done all of the preliminary work so that there would not be any time wasted getting me on board.

  Gold dredging and logging operation in Siberia, November 26, 1970. Such photographs now allow environmental scientists to gauge the extent of environmental degradation within the former Soviet Union. (Photo courtesy National Photographic Interpretation Center)

  The first challenge I faced at the ACIC was to get my OSU analytical triangulation system up and running and start using analytical photogrammetry to handle U-2 photographs. The main task was to extend photogrammetric control from areas of known ground control to position targets in the Soviet Union. Previous technologies had dealt with 6-inch, calibrated frame photographs taken while flying regular flight patterns. While it was obvious that the U-2 system had some useful mapping applications, it still presented photogrammetrists with challenges that they had never faced, including an extremely long focal length, a large frame format, uncalibrated images which covered a wide cross-field angle with no timing or attitude information, vertical to extreme tilt angles, and erratic flight lines which turned as much as 90 degrees or more within a few frames and covered very little or no ground control (either relative or absolute).

  As a result of our work with the U-2, we concluded that the MC&G (Mapping, Charting and Geodesy) community, with its equipment and training, was totally inadequate for handling extremely high-resolution photographs for mapping and target positioning purposes—I mean totally! But our work at the ACIC did provide us with a wonderful opportunity to apply our imaginations and our desires to experiment with new theories, techniques, and equipment so that we could exploit these materials for mapping.

  What was the MC&G’s mission? We were committed to closing the U.S./Soviet missile gap. We were using the new reconnaissance systems—the U-2 and CORONA—to detect and catalog missile targets. However, we had very little geodetic knowledge to tie these targets to a worldwide geodetic system for ICBM targeting. Our information consisted of a series of loosely coordinated Russian geodetic surveys for which many of the control stations were not photo-identifiable. Our Soviet bloc map collection was inadequate. Existing MC&G source data limited our ability to pinpoint Sino-Russian targets with more than a two- to three-mile degree of accuracy. It was even possible to make errors of up to 30 miles in some regions of Russia. In order for the MC&G to live up to its responsibility of helping to close the missile gap, we had to overcome these deficiencies with enough precision so that we could target our ICBMs against the Soviet sites.

  In order to hurl an ICBM from a launch point in the United States to a target in the Soviet Un
ion, three physical conditions would have to be met: (1) we would have to achieve an exact knowledge of the three dimensional geometric shape of the world “geoid”3 by referring to a common datum encompassing the launch and the target points; (2) if a vehicle was launched, we would have to account for the way gravity influenced its trajectory throughout its entire flight; and (3) we would have to determine the precise geodetic position of both the launch site and the target.

  The world geodetic community had already started to address the first of these three conditions by tying together worldwide geodetic surveys, both on the ground and from space. It was only possible to tie the system together from space because of CORONA. We needed a photographic collection system that was geared to meet MC&G requirements. To get it, we had to “piggyback” on the Intelligence Community without compromising the intelligence collection function. We had to upgrade CORONA and augment it with precise timing, scan calibration, and attitude readout in order to enable the MC&G’s processes to control the geometry and distortions of pan geometry. MC&G had to develop a whole new analytical photogrammetry technology from scratch. It needed sophisticated mathematical models, computer power, measuring equipment, rectifiers, enlargers, and printing devices compatible with CORONA’s high resolution and geometry. Old tri-met templates and inadequate optical stereo plotters were all the ACIC had to start with. We had to design and develop exploitation equipment and computer software for production application. Finally, and most important, we needed a trained workforce in order to build it. The workforce that we had in our plants, and I am counting the Army, Navy and Air Force, was using World War II technology. We had to wean them off of existing Multiplex/Kelsh conventional mapping techniques and train them to handle the new technologies based on CORONA’s collection data. The development of MC&G capabilities both instigated and followed the evolution of the CORONA, ARGON, and LANYARD systems.