SP-4218 To See the Unseen


- Chapter Four -

Little Science/Big Science



[87] Lincoln Laboratory was not the only center where planetary radar took root. Cornell University had its Arecibo Observatory; JPL had its Goldstone facility. At each center, radar astronomy developed in the shadow of military, space, radio astronomy, and ionospheric Big Science. In fact, without those Big Science activities, planetary radar astronomy would not have had instruments to carry out research and, in short, would not have existed.

In 1961, when the first successful detections of Venus took place, virtually the sole funder of planetary radar astronomy in the United States was the military. The one exception was JPL's Goldstone facility, which NASA funded. Ten years later, the NSF took over the role of prime underwriter of the Arecibo Observatory from ARPA, and NASA agreed to support a major S-band upgrade of the facility's radar. As a result, NASA became the de facto patron of planetary radar astronomy at Arecibo, Goldstone, and Haystack. NASA supported planetary radar at those three centers through a variety of financial arrangements. Only at Arecibo, however, did NASA formally agree to support a planetary radar facility, as well as the research conducted with it. That agreement, moreover, was an obvious departure from its policy formulated in the wake of the Whitford Report.

The shift from military to civilian sponsorship at Arecibo, just as in the case of Haystack, was not in response to the Mansfield Amendment. Under the Kennedy Administration, the military, mainly the Office of Naval Research, already had started transferring research laboratories, especially nuclear physics facilities, to the NSF. The budgetary reforms introduced under Defense Secretary McNamara, whose first major reform was to make the DoD's budget reflect the military missions for which it was responsible, probably provided the initial impetus to those transfers.1

The emergence of NASA as the patron of planetary radar astronomy is obvious only in hindsight. Throughout the 1960s, NASA refused to fund radar construction, except for the Deep Space Network. The NSF was the prime underwriter of astronomy facilities, but did not support planetary radar research. Consequently, during the 1960s, planetary radar astronomers depended on the kindness of Big Science, whether the radio astronomers at Haystack, or the NASA space programs at Goldstone, or ARPA (the military sponsor of Arecibo), for its instruments.

But in 1971, NASA broke with its established policy and paid for S-band radar equipment and underwrote the research conducted with it at Arecibo. The result was not just a new NASA policy but also the creation of a permanent institutional and financial home for planetary radar astronomy that the field lacked elsewhere. This unique arrangement came about through the complex politics of science typical of Big Science facilities. Complicating relations between the Arecibo Observatory and its parent organization, as well as relations with its funding agency, were turf battles between competing Big Science fields (radio astronomy and ionospherics) and personality conflicts.


[88] The Arecibo Ionospheric Observatory


Cornell University's 1,000-ft (305-meter) Arecibo dish began as a UHF radar managed by a civilian institution, Cornell University, but funded by the military. The (Air Force) Rome Air Development Center largely funded Cornell ionospheric research, and the original purpose of the Arecibo telescope was to conduct ionospheric research. The Arecibo facility started in the mind of William E. Gordon, Cornell professor of electrical engineering, who had devised a new incoherent scatter technique for studying electrons in the upper levels of the ionosphere by bombarding them with powerful radar waves. He worked on the technique with Cornell electrical engineering colleagues Henry Booker, his dissertation advisor, and Ben Nichols, both of whom shared Gordon's interest in ionospheric research. A Cambridge graduate, Booker had worked in the radio section of the Cavendish Laboratory, and during World War II he led the theoretical division of the radar effort at the Telecommunications Research Establishment.2

In order to apply his technique, Gordon realized he needed a powerful radar, which he proposed to build with state-of-the-art components. Gordon also realized that the instrument would be costly, too costly to have a single purpose. He proposed, therefore, that it also do radar astronomy experiments. Henry Booker added radio astronomy, a field that interested him.

Funding for the initial radar design studies, completed by Gordon, Booker, and Nichols in December 1958, came from the military: the Office of Naval Research, the Rome Air Development Center, and the Aerial Reconnaissance Laboratory, Wright Air Development Center, Wright-Patterson Air Force Base, Ohio. The studies outlined the radar parameters: a pulse radar with 2.5 megawatts of peak power and 150 kilowatts average power (essentially the Millstone radar transmitter), a low noise temperature, and an operating frequency around 400 MHz (430 MHz; 70 cm in the final design). The availability of components and antenna technical limits largely determined the operating frequency. The antenna itself was to be a parabolic dish 305 meters (1000 ft) in diameter fixed in a zenith-pointing position and fed from a horn on a 152-meter (500-ft) tower.3

Concurrent with these design studies, Bill Gordon sought funding. The budget of the NSF, the agency of choice for basic research, was not large enough for the project, and NASA was interested in building spacecraft. The National Bureau of Standards already had built ionospheric radars and was building a dipole array radar in Jicamarca, outside Lima, Peru, that incorporated Gordon's incoherent scatter technique. Its director, Ken Bowles, a Cornell graduate, had demonstrated the feasibility of Gordon's technique with a Bureau of Standards meteor radar in Illinois.

Gordon first presented his project to ARPA in the summer of 1958. ARPA was an entirely new agency. Although Gordon was not aware of it at the time, ARPA's Defender Program was an effort to research, develop, and build a state-of-the-art defense against [89] Soviet missiles. Though some ARPA scientists saw the scientific value of Arecibo, ARPA's main interest in the project was as part of the Defender Program to track the ion trails created by missile exhaust.4

Gordon campaigned in Washington for two years. The Sugar Grove dish was a barrier to gaining approval; reviewers wanted to know why he needed to build the 305-meter (1000-ft) dish, when the Navy had a fully-steerable antenna under construction. Finally, Gordon met Ward Low of the Institute for Defense Analysis and an ARPA adviser, and ARPA agreed to finance the engineering and construction of the dish. The Air Force Office of Scientific Research (AFOSR), through the Electronics Research directorate, Air Force Cambridge Research Laboratories (AFCRL), Bedford, Massachusetts, monitored the contract. The AFCRL now influenced the design of the telescope. Low introduced Bill Gordon to the AFCRL antenna group, which had been studying spherical reflectors for over a decade. They redesigned the fixed, zenith-looking paraboloid into a spherical reflector with a movable antenna feed mounted on a suspended platform.5

The antenna was larger than any other attempted for radar or radio astronomy, larger even than the Sugar Grove dish. The size required an unprecedented support structure. Cornell civil engineering professors proposed placing the dish in a natural bowl in the earth. The proposal was practical from an engineering perspective and cut costs, according to preliminary studies by William McGuire and George Winter, Cornell School of Civil Engineering.

Topographical, political, and scientific factors influenced the choice of a site. In the tropics, the planets would pass nearly overhead and into the antenna's cone of view. After considering sites in Hawaii, central Mexico, Cuba, Puerto Rico, and some smaller Caribbean islands, the search narrowed to the Island of Kauai, the Matanzas area of Cuba, and northern Puerto Rico. Political and import problems eliminated Cuba; Hawaii was too far and too remote. Puerto Rico had a favorable location, political stability, and minimum distance, as well as a karst topography full of sinkholes in which to locate the giant reflector. After looking at locations in Puerto Rico, Cornell chose a natural bowl in the mountains south of the city of Arecibo.6

With feasibility and location established, ARPA and Cornell signed a contract on 6 November 1959 in which the University agreed to perform three tasks: 1) conduct design studies on a vertically-directed ionospheric radar probe; 2) consider ionospheric and other scientific uses for the instrument, then propose a priority list of the first experiments; and 3) lay out structures and buildings needed for the initial facility.7

Meanwhile, also in 1959, Henry Booker launched the Center for Radiophysics and Space Research (CRSR), an umbrella organization for mainly astronomy and electrical engineering faculty research, as well as management of the Arecibo facility. Booker shared its administration with fellow Cambridge graduate Thomas Gold. Like Booker, Gold had [90] worked on radar during World War II, but at the Admiralty Research Establishment. After the war, Cambridge, the Cavendish Laboratory, and the Royal Observatory in Greenwich, Gold arrived in the United States in 1957 and taught astronomy at Harvard. Booker thought Gold ideal for running the CRSR.8

The CRSR staff, professors from the astronomy, electrical engineering, and physics departments, drew up a research program for the Arecibo telescope. Following ARPA guidance, they listed 20 experiments arranged in order of priority. The first three explored the ionosphere. Then came proposals for planetary, lunar, solar, and other radar work, followed by three more ionospheric experiments. The last 10 were all radio astronomy experiments. The first 10, the CRSR staff concluded, were "clearly within the scope of the ARPA missions," but the "relation of experiments 11 through 20 [in radio astronomy] to the ARPA mission is not so clear." ARPA did not appear interested in radio astronomy. Well before the telescope's inauguration, however, radio astronomy had been assigned a major role in its scientific mission.9

Cornell next began building the Department of Defense Ionospheric Research Facility, as the telescope was named originally. Construction of the structure, antenna, concrete towers, and electronics were let out to over a half dozen commercial subcontractors, while the Army Corps of Engineers supervised the construction and civil engineering. The raising of the 300-ton feed platform from the bottom of the bowl, where it had been assembled, to its approximate final position 152 meters (500 ft) overhead, was an awe-inspiring sight. As Bob Price recalled, the raising of the pylons was also "Very impressive....They had all these very strong Puerto Ricans pulling at cables. It was like some 1930s Mexican mural painting. Labor at its best. All coordinated pulling at these cables, and pouring cement at the same time, and getting the right tension on everything."10



Figure 13. Aerial view of the Arecibo Observatory showing its location in a natural sinkhole in the hills of north central Puerto Rico.

Figure 13. Aerial view of the Arecibo Observatory showing its location in a natural sinkhole in the hills of north central Puerto Rico. The antenna is so large that its can only be seen in its entirety from above. (Courtesy of National Astronomy and Ionosphere Center, which is operated by Cornell University under contract with the National Science Foundation.)


After its inauguration on 1 November 1963, the Arecibo Ionospheric Observatory (AIO) was not just a Cornell-ARPA facility; it also became part of an international agreement for the exchange of faculty and graduate students between Cornell and the University of Sydney, signed in September 1964. The University of Sydney was a major, worldwide center for radio astronomy. The agreement gave Americans access to some of the most advanced radio astronomy instruments in the world, as well as some of the most renowned researchers.11

Bill Gordon directed the observatory at Arecibo. After meeting Gordon Pettengill at Millstone, Thomas Gold "twisted his arm" to get Pettengill to take the position of associate director. At Lincoln Laboratory, Pettengill had carried out radar astronomy experiments, but more as a hobby. When he arrived at Arecibo in July 1963, "A totally new world opened up down there. This was a university-operated facility....And there was no direct military work!" Pettengill devoted his entire time to planetary radar and achieved recognition in the field.12

What made the Arecibo world so different, apart from the lack of "military work" that was the bread and butter of Lincoln Laboratory, was the fact that planetary radar astronomy was an integral part of the scientific agenda. Arecibo's university connection would supply graduate student researchers. Moreover, as associate director, Pettengill could hire people to do planetary radar. Thus, the earliest Arecibo planetary radar [92] astronomer was not trained in the traditional way, as a graduate student in an academic setting, but was hired to do planetary radar. The first such hire was Rolf B. Dyce.

Pettengill first met Dyce years earlier, when Dyce was with the Rome Air Development Center, Griffiss Air Force Base, in Rome, New York. Dyce had a B.A. in Physics and a Ph.D. from Cornell, where he did radar studies of auroras. Dyce eventually landed a job with the Stanford Research Institute (SRI) at Menlo Park, California, where he worked on classified ionospheric and radar research, including auroral, meteor, and lunar studies. Dyce and Pettengill also toured Europe together and visited key radar research centers, including Jodrell Bank, the Dutch facility at Dwingeloo, the Chalmers Institute in Gothenborg, Sweden, and the Norwegian Defense Research Establishment. Pettengill hired Dyce in January 1964, just weeks after the Arecibo dedication in November 1963.13

Arecibo was different from Lincoln Laboratory and Haystack in many other ways, too, because of the relationships between Arecibo and Cornell and between Arecibo and Lincoln Laboratory. While MIT did not train radar astronomers to work at Lincoln Laboratory, Cornell sent graduate students to Arecibo to work on doctoral dissertations in radar astronomy. MIT students also carried out radar astronomy dissertation research at Arecibo. As a result, Arecibo became a training ground for future radar astronomers.

Some of the earliest graduate student radar research was done on the Sun and Moon, not the planets. Vahi Petrosian, a Cornell graduate student working on a masters thesis, attempted some solar radar work in July and August 1964. After later attempts by two other graduate students, solar echo experiments were abandoned; the results were neither as good nor as productive as those achieved by the El Campo solar radar.14

On the other hand, starting in 1965, Arecibo undertook a far more vigorous and productive program of lunar radar research with supplementary funding from NASA, which hoped to use the results to help select Apollo landing sites.15 Carrying out the lunar work in collaboration with Dyce and, occasionally, Pettengill was Cornell graduate student Thomas W. Thompson. The research formed the basis of his 1966 doctoral dissertation. Thompson worked briefly at Haystack, then again at Arecibo, before he found a position at JPL. He returned to Arecibo occasionally to make lunar radar observations. 16

[93] The next graduate student was Raymond F. Jurgens, whose 1969 dissertation used Arecibo radar data to form some of the first range-Doppler images of Venus.17 Then came Donald B. Campbell, originally from Australia. Using the radar interferometric method developed at Haystack, and working under both Dyce and Arecibo director Frank Drake, Campbell began a lifelong career devoted to the radar imaging of Venus. Both he and Jurgens later were key figures in planetary radar astronomy.

While its relationship with Cornell turned Arecibo into a breeding ground of radar astronomers, its relationship with Lincoln Laboratory and Haystack, forged through the presence at Arecibo of Gordon Pettengill, provided entree to the software, techniques, and ephemerides developed by Lincoln Laboratory. Pettengill was a vital factor not only as associate director from 1963 to 1965, but also as Arecibo director from 1968 to 1970.

At the heart of that relationship was the business of creating radar ephemerides. The standard planetary ephemerides issued by the U.S. Naval Observatory were simply not accurate enough for radar work, so special ephemerides computer programs had to be developed. In order for them to be as accurate as possible, these radar ephemerides had to draw on a data base of radar observations. At Lincoln Laboratory, Irwin Shapiro started such a radar ephemerides computer program. Haystack provided a large amount of the ephemerides data, and so did Arecibo at the instigation of Gordon Pettengill, with a modest grant from NASA. Pettengill recalled the speed with which radar observational data arrived at Lincoln Laboratory: "I remember we used to send it back by special delivery mail. We would mail it by six in the evening at Arecibo, and it would be delivered in Lexington, Massachusetts, at nine the next morning. Very efficient. Then it would be put into the Lincoln Laboratory ephemeris program."18 In addition to the ephemerides, Lincoln Laboratory supplied Arecibo with software and techniques. As mentioned earlier, Don Campbell adopted the Haystack radar interferometry technique at Arecibo, and the special fast Fourier transform software created for the Haystack interferometer also migrated to Arecibo.19

When Pettengill left Arecibo in 1970, he returned not to Lincoln Laboratory, but to MIT, where he became professor of planetary physics in the Department of Earth and Planetary Sciences. The change from Lincoln Laboratory to MIT was as stimulating to Pettengill as the original move to Arecibo. He continued planetary radar research, using both Haystack and Arecibo. He was not alone; both Tommy Thompson and Don Campbell used both telescopes.20 Moreover, Pettengill, who already had guided the radar astronomy dissertations researched at Arecibo, began offering a course in radar astronomy at MIT and sending MIT graduate students to Arecibo to do their doctoral research.

The fruit of this cross-fertilization between Arecibo and MIT and Lincoln Laboratory was that Arecibo evolved into a common research facility for both Cornell and MIT, so that by the time planetary radar astronomy research ended at Haystack, Arecibo already was in position to continue the research programs underway at Haystack. That did not mean, though, that the Arecibo telescope provided the same amount of observing time as Haystack.

At Haystack, planetary radar astronomy accounted for a greater percentage of observing time than at Arecibo. Although planetary, lunar, and solar radar experiments occupied roughly 9 percent of Arecibo antenna time for the period December 1965 through September 1969, only 2.4 percent of total observing time was given over to radar [94] astronomy in 1970, while radar accounted for about a third of Haystack antenna time in the same year.21 Moreover, as radar astronomy use of Haystack declined from 17 percent in 1971 to 12 percent in 1973, radar use of the Arecibo telescope increased, but not proportionally, and peaked in 1972 at 9.5 percent, somewhat lower than the lowest use at Haystack. The combined absolute number of total observing hours on the two telescopes suggests that planetary radar astronomy activity in the early 1970s was not increasing or even remaining stable, but was declining. It was Little Science becoming smaller.


From ARPA to the NSF


In November 1974, eleven years after the dedication of the Arecibo Ionospheric Observatory (AIO), a second dedication ceremony took place to denote the instrument's upgrading to S-band. The upgrade was not achieved by simply adding higher-frequency equipment. The reflector surface had to be refinished, the suspended platform accommodated to the new equipment, a new power supply provided, and the S-band transmitter and maser receiver designed, built, and installed. Each component of the instrument had to be adapted in order that the whole might function in the higher frequency range. For planetary radar astronomy, the upgrade essentially created a new instrument with entirely different and expanded capabilities. Nonetheless, however critical the upgrade was for radar astronomy, both radio astronomy and ionospheric research benefited significantly from the resurfacing and equipment improvements, too.

The conversion of the AIO into an S-band radar telescope was a long, indirect, and difficult process, even if considered only as a technological feat. The conversion paralleled and was inextricably enmeshed in the transformation of the AIO into a National Science Foundation National Research Center. That transformation was set in motion by cutbacks in the ARPA budget, not the Mansfield Amendment.

The realization that the S-band upgrade was possible is said to have been born in August 1966, during Hurricane Inez. The 100-kilometer-per-hour (62-mile-per-hour) winds moved the telescope less than a half inch (1.27 cm), instead of the foot (30 cm) it was feared. A subsequent study of the telescope structure showed that it was sufficiently stable to operate at wavelengths of the order of 10 cm (3,000 MHz). Optimistically, Frank Drake, successor to Bill Gordon as observatory director, thought that the dish could be resurfaced in less than two years for under $3 million.22

But funds were not readily available. Moreover, the annual budget allotted by ARPA started to shrink, from over $2 million initially to $1.8 million in the period 1965 through 1969. Although ARPA was cutting back all research in order to support the Vietnam War,23 the Arecibo budget suffered because ARPA felt that the telescope performed below expectations.

The antenna feed operated at only 21 percent efficiency; the dish received less than half the power it should have received. That was a huge dollar loss, too; the cost of building a dish half the area would have been much less. Nonetheless, it was still an extremely sensitive telescope. The inefficiency of the antenna feed became a source of friction between Thomas Gold and Bill Gordon, who insisted that the feed could be improved, and between AIO management and ARPA.24


Figure 14. Linear antenna feeds attached to the suspended platform of the Arecibo Observatory.

Figure 14. Linear antenna feeds attached to the suspended platform of the Arecibo Observatory. (Courtesy of National Astronomy and Ionosphere Center, which is operated by Cornell University under contract with the National Science Foundation.)


[96] The line feeds were an ongoing serious problem. After a three-day visit to Arecibo in October 1967, Bart J. Bok, director of the Steward Observatory, Tucson, observed that the line feed problem "seems to be the most critical one facing the Arecibo-Cornell group." A number of Ithaca researchers attempted to improve the feeds. One Cornell graduate student considered the use of Gregorian optics, an option also studied by the AFCRL's Antenna Laboratory. However, not until 1988 was the first Gregorian feed tested and installed at Arecibo.

Arecibo had three feed research programs going on at the same time. Only one, for a high-powered, 430-MHz radar feed operating at both circular polarizations, was vital to its radar functions. Of two competing radar feed designs, the AIO selected that of Alan Love of the Autonetics Corporation, a subsidiary of North American Rockwell. Love worked with Cornell's L. Merle Lalonde to construct an appropriate feed, which was installed on the antenna in early 1972. The new radar feed was a success.25

ARPA's funding of the AIO dropped to a great extent because of the inefficient feed. Too, radio astronomy at the AIO was expanding rapidly in the wake of the discovery of pulsars (the AIO had tremendous advantages for investigating them), and ARPA felt more and more that it should support just the facility's ionospheric work, which was the only research relevant to Department of Defense interests. The AIO, though, hoped that ARPA would pay for the resurfacing and a new radar feed.

Although the ARPA contract did not allow the AIO to seek funding from other agencies, ARPA was now receptive to the idea of sharing the AIO budget with the NSF. So with ARPA's blessing, Thomas Gold and Frank Drake approached the National Science Foundation about civilian operational money for the AIO. The AIO also submitted a proposal to the NSF in early 1967 for detailed engineering studies and a cost estimate to resurface the reflector.26 The search for both resurfacing and operational funds thus proceeded concurrently and was boosted by the report of the Dicke Panel.

Thomas Gold, Frank Drake, and Rolf Dyce pitched the Arecibo resurfacing project before the Dicke Panel. The Panel gave the project highest priority. As a result, Cornell obtained an NSF grant for a study and cost estimate of the reflector resurfacing. The AIO selected the Rohr Corporation, which also built JPL's Mars Station, to conduct the study. Rohr planned to install light aluminum panels for the reflector surface at a total cost of $3.5 million.27

The NSF, however, did not ask Congress to underwrite the resurfacing of the Arecibo reflector. The feed problem stood in the way. At its meeting of 16-17 October 1967, the NSF Astronomy Advisory Panel resolved:28

The NSF Advisory Panel will be hesitant to favor the improvements of the surface of the Arecibo dish and/or the undertaking of substantial operating expenses for Arecibo until a successful radio astronomy feed has been constructed and made operational at frequencies low enough that the surface is not critical.

[97] In short, if an adequate feed design were not feasible, investing in an expensive resurfacing of the reflector for operation at higher frequencies made no sense. The feed problem held up the resurfacing and by implication the entire S-band upgrade. Consequently, Cornell undertook an in-house effort to design a 327-MHz feed at its own expense.

Although the reflector resurfacing project came to a temporary halt, the drive to secure NSF operational support succeeded in the wake of the Dicke Panel report. In July 1967, as the Dicke Panel was meeting in Washington, Cornell Vice President for Research and Advanced Studies Franklin A. Long asked Leland Haworth, director of the NSF, for a meeting about the possibility of jointly funding the operation of the AIO with ARPA. The NSF and ARPA soon entered into discussions and, by late August 1967, the NSF was agreeable to replacing the AFOSR as the government agency monitoring the Arecibo contract.29 This was the first step in converting the AIO into a civilian observatory.

ARPA was prepared to underwrite the full AIO budget to the end of September 1968. Beginning 1 October 1968, for fiscal years 1969 through 1972, ARPA would pay for a third of the AIO budget, representing the portion of telescope time spent on ionospheric work. "It is very much hoped," the ARPA negotiator expressed, "that the entire facility will be identified as a National Science Foundation Observatory with ARPA as one of several users."30

In December 1967, well before passage of the Mansfield Amendment, Cornell and ARPA came to an agreement on the AIO contract. Cornell, NSF, and ARPA would negotiate a one-year contract for AIO operation from 1 October 1968 through 30 September 1969. The ARPA-NSF Memorandum of Understanding, signed in late April 1969, left the AIO under ARPA and the AFOSR until 1 October 1969, when the NSF took over, thereby anticipating the effect of the Mansfield Amendment. For the fiscal year starting 1 October 1968, each agency agreed to pay half the facility's annual budget. For the two years beginning 1 October 1969, ARPA agreed to transfer to NSF a third of the annual budget to support just ionospheric research. ARPA did not commit any funding after 1 October 1971, but left the door open to the possibility.

The Memorandum of Understanding defined ARPA's step-by-step divestment of Arecibo. Although ARPA initially had funded Arecibo for Project Defender, the telescope was never engaged in classified military research. Moreover, one clause in the Memorandum of Understanding specifically forbade the participation of the AIO in secret work: "The Observatory shall not be used to make measurements which are themselves classified nor be used as a repository for classified information."31 The AIO was on the rocky road to civilian supervision and funding.


What's In a Name?


The transformation of the AIO into an NSF National Research Center involved two interconnected issues, the observatory's management structure and the status of ionospheric research, both of which were complicated by personality conflicts and turf fights between Big Science fields. Implicit in being a National Research Center was free access to the telescope for all qualified scientists. The AIO always maintained that it operated as [98] a national center, and the Cornell-Sydney agreement opened the observatory to foreign scientists. The real problem was that radio astronomy use of the telescope had skyrocketed, especially in contrast to ionospheric research. From December 1965 through September 1969, for example, ionospheric research accounted for 30 percent, while radio astronomy took up 50 percent of antenna time.32

Ionospheric research had been the reason for creating the AIO in the first place, and it was more interesting to the electrical engineering than to the astronomy department. The name of the facility changed to the Arecibo Observatory, discarding the "ionospheric" of the original name. To some individuals, the name change did not reflect the facility's multiple research agenda, which was the intent of the change, but instead signified lack of interest in ionospheric work. As Gordon Pettengill explained: "We settled on that name early, because it encompassed the radio astronomy, radar astronomy, and ionospheric research. There was quite a group that wanted to call it the Arecibo Ionospheric Observatory, which was the original name under Bill Gordon."33 Many accused Thomas Gold, who had fostered the expansion of radio astronomy, of thwarting ionospheric work, but Gold insisted that no ionospheric researchers ever were turned down.

Perceptions outside Arecibo and Cornell confused the presumed reduction of ionospheric studies with the rift between astronomy and electrical engineering within the CRSR, and colored everything with the friction between Bill Gordon and Thomas Gold. Gold found Gordon "a little difficult, because he really wanted to cut himself off from Cornell, from everything completely, and I realized that if he did so, then the telescope would never be used for radio astronomy and radar, and it would become merely an ionospheric instrument, and that I was very opposed to, being nominally in charge of building such a huge wonderful instrument and then finding it's not used for what it's capable of."34 Bill Gordon, for his part, stated, "If you ask me, I was mad at the time, and whatever I tell you has some personal bias built in." In short, he explained, "I thought I was removed from a job that I deserved to have."35

Frank Drake, radio astronomer and one-time Arecibo director, explained the conflict rather precisely. "I had picked up enough innuendo in Gold's tone and Gordon's words to realize that the two of them were engaged in a bitter battle for the Arecibo turf," he wrote. Gold "wanted the Arecibo telescope freed to do more research in radio astronomy. He was lobbying the university administration to put it under his jurisdiction." Gordon "could not bear to relinquish control of it." He left, however, after Gold pointed out to the university administration that Gordon had been off-campus far longer than the university bylaws allowed. "It was a fact people might have been willing to overlook, but once Gold seized on it, Gordon was forced to make a choice."36

Feelings about the friction between Gold and Gordon, as well as the perceived neglect of ionospheric work, also shaped how the NSF handled the AIO. The chief personality at the NSF was Tom Jones, director of the Division of Environmental Sciences. He explained the situation to the NSF director in 1968:37

[99] The operation of AIO has been tainted by a great deal of political infighting on the Cornell campus. Results of these confrontations included the departure from Cornell of Drs. W. Gordon and H. Booker, both aeronomers, who were the originators of the backscatter concept for probing the ionosphere and who saw the Arecibo venture through from the proposal stage right on up to its final construction and initial operation. There are indications that, aside from accepting opportunities for professional growth, they left Cornell because the administrative control of AIO was removed from the director of the Observatory and placed in the hands of another individual on the Cornell campus. We do know, from conversations with aeronomers, that they do not want to give up the use of the Arecibo instrument.

Jones maintained a vigil on the AIO case, as he moved from the Division of Environmental Sciences to the Office of National Centers, which directly handled the Arecibo Observatory. Thomas Gold found that Jones "kept expressing a sort of paranoia about ionospheric work, but constantly. I mean, I couldn't talk to him without getting a lecture that far too little ionospheric work was being done, and he couldn't support any funding for Arecibo if this were done, even though at the time it was doing very good work in radio and radar astronomy, but not enough ionosphericists wanted to go there. I couldn't help it!"

According to Gold, Jones told him that he could not support funding for Arecibo if the reduction of ionospheric research continued. As for his relations with Bill Gordon, Thomas Gold insisted that it had nothing to do with ionospheric research. He and Gordon disagreed over the management of the observatory. According to Gold, Gordon wanted to operate it "in a way independent of Cornell," and he did not want to return to Cornell. Bill Gordon "wanted to make all the decisions as to who gets what time and all that," and Gold objected.38

Control of the observatory was the key issue dividing Gordon and Gold. The issue of where management of the AIO should rest, at Arecibo or at Ithaca, was precisely the concern of the NSF, too. The issue was clouded by both personality conflicts and the status of ionospheric research. On 27-28 February 1968, the NSF Advisory Panel for Atmospheric Sciences, which included Bill Gordon, issued a formal statement on the future of ionospheric research at the AIO: "As the NSF assumes increasing operational responsibility, the Panel strongly recommends that any management changes be made in such a way as to insure the availability of the AIO for experimental research in aeronomy and solar-terrestrial physics." Moreover, "The Panel considers it important to establish a management structure for the AIO whereby scientists from institutions throughout the United States may use the Observatory. To accomplish this, it is suggested that the scheduling and operating policy be established by the scientific community and implemented by the resident director. An appropriate way to assure representation of the scientific community would be to place the management of the AIO in the hands of a consortium of interested universities."39

The Advisory Panel was not alone in suggesting management by a university consortium along the lines of NEROC or the NRAO.40 However, Cornell and Gold wanted to retain control of the Arecibo Observatory (AO). Harry Messel, head of the University of Sydney School of Physics and joint director, with Gold, of the Cornell-Sydney University [100] Astronomy Center, protested to Donald F. Hornig, the special presidential assistant for science and technology, that any change in the AO management structure would affect the Cornell-Sydney arrangement, too. Despite Hornig's assurances to the contrary, the evolving AO management structure led to the termination of the Cornell-Sydney agreement.41

The crux of the management structure question, however, all personality and turf conflicts aside, was separation of observatory administration from all academic departments, like the CRSR. The NSF did not want to fund National Research Centers that were prisoners of an astronomy department or of any other academic unit. It was clear, though, that if the AO were to become a National Research Center, with a secured budget from the NSF, Cornell would have to draft a new management structure; otherwise, a university consortium might take over Cornell's managerial role.

In March 1969, as the NSF looked toward assuming full responsibility for the AO on 1 October 1969, the Foundation asked Cornell to prepare a proposal for the operation of the AO for the two-year period beginning 1 October 1969. The proposal was to discuss the AO management structure, "bearing in mind our opinion that a director of a National Center should report to a level of management significantly above that of a department or similar unit."42 The April 1969 proposal outlined a management structure drafted the previous summer. The director of the AO reported to a policy committee, which consisted of only the university provost, the director of the CRSR (Gold), and the vice president for research and advanced studies.43

A special National Science Foundation AIO Group reviewed the proposal. Their major objection was the management plan: "It does not show much change from the existing management structure at Cornell and does not appear to be suitable for a National Center. No member of the AIO group finds it acceptable." Specifically, the problem was the three-man policy committee. "This Committee seems clearly intended by Cornell to be the group which runs the show. It is proposed that it be made up exclusively of Cornell employees resident in Ithaca. The suggestion that such a group should be considered 'national management' has reduced the undersigned [Fregeau] to a conviction that his education in the art of strong language is grossly inadequate."

The AIO Group felt that a more appropriate structure would have the observatory director report directly to the vice president of research, a single individual, and not a committee; otherwise, "the implication [is] that the committee is the AIO director's boss." In the judgement of the AIO Group, "The Cornell proposal is not, in its present form, suitable for review by the scientific community. If it were to be sent out in this form, the community reaction would probably poison the beginnings of what we expect to be a fruitful venture for NSF."44

On 1 October 1969, when monitorship of the Arecibo contract passed to the NSF, Cornell reorganized the AO's management structure to conform more closely to the Foundation's guidance. The observatory was removed from CRSR supervision and placed under an Arecibo Project Office headed by Assistant Vice President for Research (Arecibo Affairs) Thomas Gold.45

[101] In the following months, Cornell and the NSF continued to consider the observatory's management structure. The result was a new organizational structure effective 1 July 1971 that brought it more in line with other National Research Centers, and a new name, the "National Astronomy and Ionospheric Center" (NAIC). The name and acronym were intended to emulate the NRAO as a model and gave assurances of the importance of ionospheric research.

In the new management structure, the title of Assistant Vice President of Research (Arecibo Affairs) was discontinued. Gold had quit. Those duties were given to the observatory director, who was responsible to Cornell, through the vice president for research, for the overall management and operation of Arecibo. He prepared the annual budget, annual program plan, and long-range plans for the AO. The observatory director was to be located primarily in Ithaca and was also the director of the NAIC. The director of observatory operations, who answered to the director, had responsibility for the operation, maintenance, administration, and improvement of the facility, oversaw personnel and time allocations, and helped prepare the budget. He was required to be located in Arecibo.46

For the new director of observatory operations, the NAIC hired Tor Hagfors in 1971. His selection reassured those who worried about the status of ionospheric research. Hagfors had an impressive background in ionospheric (and radar) research and administration at the Norwegian Defense Research Establishment, Stanford University, the Jicamarca Radio Observatory (where he was director, 1967-1969), and Lincoln Laboratory's Millstone Hill radar.47


The NSF and NASA Agreement


As the new management structure emerged, and as the National Science Foundation took over the Arecibo contract, the search to fund the S-band upgrade continued. In May 1969, the Subcommittee on Science, Research, and Development of the Committee on Science and Astronautics, headed by Emilio Q. Daddario (D-Conn.), recommended deferring the NSF request for resurfacing money. Gordon Pettengill, speaking as Arecibo director, pointed out that "many throughout the radio astronomy community were seriously disappointed at the failure of the Congress to authorize funds for the resurfacing of the AO reflector." When the Dicke Panel reconvened in June 1969 and reaffirmed the need for the resurfacing, they too expressed disappointment that the new reflector surface had not yet been started.48

A major breakthrough occurred when NASA took an interest in the project. Throughout the 1960s, NASA had funded only mission-oriented radar research, but not radar telescope construction. In January 1969, Harry H. Hess, chair of the Space Science Board, wrote to John Naugle, associate administrator of Space Science and Applications at NASA, urging NASA to fund the Arecibo radar upgrade. The cost, estimated to be $5 million for the resurfacing plus $2 or $3 million more for the radar equipment, was "small in comparison with the construction of a new radar facility but would make it possible to [102] map the surface of Venus with a resolution of a few kilometers. Such a map would obviously be a tremendous step forward in our knowledge of the planet. The NASA contribution to the total cost of improving the Arecibo facility would be very small compared to the cost of obtaining the same information from some future orbiter."49

Hess's argument closely resembled that of Werner von Braun in an anecdote related by Don Campbell: "I don't know if it's apocryphal or not, but there is the story that Werner von Braun said, if you can get a two kilometer resolution on Venus for $3 million, which is roughly what we were talking about, that it was an immense bargain, and that NASA should take it straight away."50

Indeed, NASA became interested in funding the upgrade for one major reason. The S-band equipment could make radar maps of the Venusian surface with a resolution of two to five kilometers. The space agency was interested in a one-megawatt radar operating at 10 cm (3,000 MHz). And as NASA chief of planetary astronomy William Brunk came to realize, the total cost of the upgrade was a fraction of the initial cost of the facility.51

The country was discovering that it could not afford both guns and butter, the Vietnam War and the Great Society. NASA and NSF were under serious pressure to cut their budgets, and in December 1969, the new Republican President shut down the NASA Electronics Research Center in Cambridge. Budgetary austerity perhaps led NASA to support the Arecibo S-band upgrade, at a cost of a few million dollars, over the NEROC telescope, with an estimated price tag of $30 million dollars. Furthermore, given the superior transmitter power and receiver sensitivity of the upgraded Arecibo S-band radar over the proposed NEROC radar, NASA would be getting a better investment for its dollars.

Budgetary belt tightening also induced NASA to realize that every planetary mission had to do something that could not be done from the ground, and missions would have to rely on ground-based results more than ever. The radar images obtainable from the upgraded radar would be invaluable to the exploration of the planets. Thus, the mission-oriented logic of NASA, combined with budgetary restraint, led to its adopting the Arecibo upgrade project.52 NASA now approached the NSF.

When NASA and NSF representatives met on 2 December 1969, the NASA budget for fiscal 1971 was among the topics of discussion. The space agency was going to ask for an extra $1 million to build a major planetary research facility as part of its Planetary Astronomy Program. Three candidate projects were under consideration: a 60-inch (1.5-meter) planetary telescope at Cerro Tololo, Chile; a large-aperture infrared telescope; and the Arecibo upgrade. The final choice pivoted on the NSF budget submission to the Bureau of the Budget and Congress.

The NASA strategy was to pay for the resurfacing, if the NSF failed to win funds from Congress, and to worry about the rest of the S-band upgrade later. Brunk knew that NASA had to be prepared to pay for the radar equipment. Because radar equipment was "not a high priority item for general radio astronomy," he reasoned, "the development of a high power radar transmitter at a wavelength of 10 centimeters will be a low priority for NSF funding and must therefore be included in the NASA Planetary Astronomy budget."53

Soon after the NASA-NSF meeting, in February 1970, Cornell submitted a funding proposal for the S-band upgrade to both NASA and the NSF. The proposal asked for $3 million over three years, with work to begin February 1971. Both the NSF and NASA fiscal 1971 budget requests contained money for Arecibo. The NSF proposed to underwrite [103] the reflector resurfacing, while NASA budgeted for the radar equipment and its installation. Congress approved both the NASA and NSF Arecibo S-band expenditures. The NSF funds were frozen, however, until a new cost estimate became available. The estimate, completed in November 1970, was $5.6 million.54

The upgrade brought together the NSF and NASA into a special relationship that started with joint discussions in December 1970 between William Brunk, chief of the NASA Planetary Astronomy Program, and Daniel Hunt, head of the NSF Office of National Centers and Facilities Operations. As discussions progressed, on 6 March 1971, NASA formally expressed its intent to enter into an agreement with the NSF for the addition of the S-band equipment. The two agencies entered into negotiations and, on 24 June 1971, signed a Memorandum of Agreement, which went into effect 1 July 1971.55

Under the agreement, the NSF funded the resurfacing and NASA the addition of a one megawatt S-band radar transmitter, receivers, and associated changes to the antenna to provide radar capability at a wavelength of ten centimeters (3,000 MHz). The project was to be managed under the existing NSF-Cornell contract. The NSF would serve as the monitoring agency, and NASA would transfer its portion of the funds to the NSF. The agreement deferred the issue of S-band operational costs until later, although the two agencies intended to share those costs proportionally.56

In this way, NASA came to fund radar instrument construction and committed itself to supporting the research performed with the instrument. This deviation from earlier policy was motivated by an interest in mission-oriented research, namely, to obtain radar images in support of space missions to the planets, particularly to Venus. The consequence was a permanent institutional and funding arrangement for planetary radar astronomy at Arecibo, as well as a unique instrument.


Arecibo Joins the S-Band


The NASA-NSF agreement, backed by Congressionally-approved funds, provided the legal, financial, and managerial framework for the actual upgrade work to take place. The upgrade began with a search for a contractor to undertake the reflector resurfacing. Two firms bid, the Rohr Corporation and LTV Electrosystems of Dallas, and in November 1971 Cornell awarded the contract to LTV Electrosystems, which shortly afterward changed its name to E-Systems. The original spherical reflector consisted of 1/2-inch (1.3-cm) steel wire mesh (chicken wire) supported by heavy steel cables. Over 38,000 thin aluminum panels, fabricated on-site by E-Systems, replaced the chicken wire. From the beginning, inefficiencies and equipment failure plagued panel installation, but they were overcome, and the last panel was installed in November 1973.57

[104] Concurrently, the NAIC oversaw the design and construction of the S-band radar transmitter, receivers, and associated equipment; the necessary modifications to the suspended feed platform; and construction of a new carriage house to hold the S-band equipment. Ammann & Whitney, a well-known structural engineering consulting firm, reported on the suspended structure and reflector cable anchorages as well as on the feasibility of upgrading the suspended structure. They found no basic deficiencies in the structure that would make upgrading impractical or inadvisable.58

In order to develop specific transmitter characteristics that met scientific goals, yet represented realistic state-of-the-art feasibility, NAIC staff discussed its design with experienced radar astronomers and with experts from Varian Associates, Raytheon, and Continental Electronics. The operating frequency, 2380 MHz (12.6 cm), appeared to be the optimum choice for both radar and radio astronomy and was close to the JPL planetary radar frequency (2388 MHz; 12.6 cm).59

Originally, the transmitter was to produce 800 kilowatts using two klystrons. The NAIC based the decision on the experience of JPL Goldstone, where a single klystron produced 400 kilowatts of average continuous-wave power at 2388 MHz. Although Varian was developing a one megawatt continuous-wave klystron, the advantages of proven reliability and ready availability of spares militated against using a single, experimental one-megawatt klystron. Finally, after extensive discussion with representatives of NASA and the NSF, the NAIC reduced the transmitter power requirement to 450 kilowatts, thereby lowering costs "without impacting upon scientific goals of the program." In contrast, the original UHF radar transmitter produced only 150 kilowatts of average power.60

The experience of JPL in operating at S-band proved invaluable to the Arecibo radar upgrade. In addition to providing expert advice to the NAIC staff, a former JPL employee reviewed technical matters for the NASA technical monitor. The maser receivers, moreover, were excess Deep Space Network equipment. The agreement between the NAIC and JPL for the transfer of the masers noted that JPL was "a pioneer in the development of maser systems," and that "no commercial firms have the required capability, experience and expertise to produce an S-band maser system that would be operational at 2.38 GHz."61

The renovated reflector was dedicated on 15-16 November 1974. After delivery of the keynote speech, Rep. John W. Davis (D-Ga.) gave the signal for the transmission of The Arecibo Message, 1974, an attempt to communicate with extraterrestrial civilizations. The radar upgrade, however, was not yet completed and would not be entirely ready until the following year.62

Shortly after the resurfacing dedication, however, Gordon Pettengill (Arecibo) and Richard Goldstein (JPL) used the S-band transmitter to bounce signals off the rings of Saturn. Because the Arecibo maser receiver was not yet installed, Arecibo sent and JPL's Mars Station received. The bistatic experiment worked, despite line feed and turbine generator problems at Arecibo.

[105] As Don Campbell recalled: "The initial feeds that we used with the transmitter needed cooling; we were having trouble attaching the cooling lines. People would line up late at night in front of the control room during this experiment. We would turn on the transmitter, and there would be this sort of flash of light, as things burned up up there and everybody went 'Ah!' It was a bit like fireworks. When the problem finally got solved, I think everyone was rather disappointed that there wasn't any flash of light up there!"63

The Arecibo S-band upgrade literally created a new instrument with which to do planetary radar astronomy, a field whose outer limits of capability still leaned strongly on the availability of new hardware. Although the Arecibo telescope made S-band radar observations of Mars beginning August 1975 for NASA's Viking mission, the arrangement with NASA freed Arecibo also to do radar research that was not mission related. So, late the following month, on 28 September 1975, the radar detected Callisto, followed two nights later by Ganymede, the first detections of Jupiter's Galilean moons.64

The NASA agreement guaranteed planetary radar astronomy an instrument and a research budget. Nowhere else did planetary radar astronomy operate with such extensive institutional and financial support. These unique advantages, combined with its relations with Cornell and MIT, have sustained Arecibo as the focal center of planetary radar astronomy to the present day.


The JPL Mars Station


In sharp contrast to Arecibo, JPL did not formally recognize planetary radar astronomy as a scientific activity. Planetary radar had neither a budget line nor a program at JPL; it was invisible. Its role was to test the performance of the Deep Space Network (DSN). Eb Rechtin, the architect of the DSN, deliberately avoided creating a radar astronomy program. He saw no reason, other than for science, why NASA ought to fund it. Instead, planetary radar became, in the words of JPL radar astronomer Richard Goldstein, the "cow-catcher on the DSN locomotive," financed at the "budgetary margin" of the DSN. In Goldstein's own words, "I was the cow-catcher and still am."65

Radio astronomy, on the other hand, held a more privileged position. Nick Renzetti, the DSN manager responsible for links between the Network and its users (NASA space missions), forged an agreement with NASA Headquarters that permitted qualified radio astronomers to perform experiments on Goldstone antennas at no cost, provided the experiments did not conflict with the antennas' prime mission, spacecraft communications and data acquisition.66

Not only was JPL planetary radar astronomy invisible, but relations between JPL and its oversight institution, the California Institute of Technology, were about as distant as those between Lincoln Laboratory and MIT. JPL employees, like their peers at Lincoln Laboratory, could not have graduate students, unless they held a joint appointment at Caltech. Although Dick Goldstein taught a radar astronomy course at Caltech during the 1960s, his students did not become radar astronomers, but went into other fields. An unusual case was Lawrence A. Soderblom, who took Goldstein's course in the fall of 1967. Soderblom later joined the U.S. Geological Survey, where he interpreted planetary radar data.67

[106] Unlike Lincoln Laboratory or Arecibo, JPL did not hire people to do radar astronomy, because JPL officially did not have a radar astronomy program. Goldstein and Dewey Muhleman had participated in the 1961 Venus radar experiment, not because they were Caltech graduate students, but because they were JPL employees in Walt Victor's group. Roland Carpenter, another JPL planetary radar astronomer of the 1960s, also worked under Walt Victor.68 Once Roland Carpenter and Dewey Muhlemen left JPL, Dick Goldstein remained the sole JPL radar astronomer for several years.

Planetary radar astronomy subsisted at JPL during the 1960s on money earmarked for various space missions and on the budget of the DSN. The NASA budget then was more generous. NASA paid the cost of operating and maintaining the Goldstone radar as part of the DSN, so that the costs of the radar instrument were paid. When Goldstein needed a piece of hardware designed and built, he assigned the job to one of the employees he supervised as manager of Section 331. The Advanced Systems Development budget of the DSN paid for hardware design and construction.69

In order to obtain time on the Goldstone radar, Goldstein went from mission to mission and explained why the mission ought to support his radar experiments. With approval from a mission, Goldstein could then request antenna time from the committee in charge of allocating antenna use. Mariner missions supported many of the radar observations, while Viking and Voyager supported experiments on Mars, Saturn's rings, and the Galilean satellites of Jupiter. Officially, the experiments were done for neither radar improvement nor the science, but for "better communications" with spacecraft.70

If the NASA Headquarters Planetary Science Program, then headed by William Brunk, approved of a particular set of radar experiments, getting antenna time was much easier. As Goldstein explained, Brunk "would support me a little and I took that as a positive thing, and I guess later on he turned that off, but it wasn't very big in the first place....It was a kind of a way to get legitimacy. If he funds you a little, that means it's important. If he doesn't fund you at all, that means it's not important....I would go to great lengths to get antenna time and a little funding from Brunk was helpful."71

Planetary radar astronomy at JPL thus came into existence and continued to function because of the Laboratory's Big Science space missions and Deep Space Network activities. In particular, it was the idea, put forth by the DSN's chief architect Eb Rechtin, that radar astronomy would have the dual function of testing the DSN's ability to support interplanetary missions and developing new hardware for the DSN (that is, the justification for having Advanced Systems Development underwrite radar astronomy hardware). It was specifically for developing and testing new DSN hardware that, shortly after the 1961 Venus radar experiment, Rechtin arranged with NASA to set aside a Goldstone radar for that purpose. On that instrument, Goldstein and Carpenter made Venus radar observations during the 1962 and 1964 conjunctions. Planetary radar research benefitted from the developmental work, which increased the continuous-wave radar's average power output from 10 to 13 kilowatts in 1962 and then to 100 kilowatts for the 1964 Venus experiment.72

When Goldstein made observations during the 1967 Venus conjunction, however, he used a new, more powerful 64-meter-diameter (210-ft-diameter) S-band antenna, the Mars Station. The need to handle missions at ever increasing distances from Earth furnished [107] the raison d'être for JPL's entry into the Big Dish arena, and incidentally supplied its future radar astronomers with an ideal instrument for imaging and other planetary radar work. With the Mars Station, Goldstein and his colleagues discovered three rugged sections of Venus; the largest received the name Beta. The needs of NASA space missions, not radar astronomy, dictated the design of the Mars Station.

The Mars Station represented the DSN's commitment to the S-band and its need for large antennas capable of communicating with probes at great distances from Earth. Starting in 1964, all new space missions were to use the higher S-band. Despite the commitment to S-band, NASA still had active missions operating at lower frequencies. The switch to S-band throughout the Deep Space Network therefore required a hybrid technology capable of handling missions operating at either the higher or lower frequency bands. A JPL design team devised the equipment, which was installed throughout the DSN.

The hybrid equipment, however, was only a transitional phase before the construction of more powerful and more sensitive antennas specifically intended to handle unmanned missions to the planets. In order to determine the essential characteristics and optimal size for those antennas, JPL initiated a series of studies in 1959 that culminated in the Advanced Antenna System. NASA's Office of Tracking and Data Acquisition, which oversaw the DSN, sponsored a pioneering conference on large antennas on 6 November 1959. Speakers reported on three kinds of antennas: steerable parabolic dishes, fixed antennas with movable feeds (e.g., Arecibo), and arrays, the same antenna types considered later by NSF panels.

NASA and JPL decided to stay with the proven design of steerable dishes. The next decision was antenna size. JPL engineering studies showed that antenna diameters between 55 and 75 meters (165 to 225 ft) were near optimal and the most cost-effective. The final choice, 64-meters (210 ft), was the same size as the recently-completed Australian radio telescope at Parkes. This was no coincidence. JPL engineers had received a lot of help from the Australian designers. Their studies of the Parkes telescope provided JPL engineers with a wealth of data and ideas to use in the design of their 64-meter (210-ft) dish.

JPL also commissioned private firms to carry out feasibility and preliminary design studies for the Advanced Antenna System beginning in September 1960, before awarding a construction contract to the Rohr Corporation in June 1963. Construction proceeded after JPL analyzed and approved the Rohr design in January 1964. Rohr completed the antenna in May 1966, following the formal dedication on 29 April 1966.



Figure 15. JPL Goldstone Mars Station (DSS-14) upon completion in 1966.

Figure 15. JPL Goldstone Mars Station (DSS-14) upon completion in 1966. (Courtesy of Jet Propulsion Laboratory, photo no. 333-5967BC.)


The dish was dubbed the Mars Station, because its mission was to support Mariner on its journey to Mars in 1964, long before the antenna was operational. Nonetheless, on 16 March 1966, the big dish received its first signals from Mariner 4 and provided operational support for Pioneer 7, launched in August 1966. The Mars Station subsequently supported several other missions, including the first Surveyor flights, and made possible live Apollo television pictures from the Moon, not to mention planetary radar images and topographical maps. In order to systematize its growing number of antennas around the world, the DSN instituted a numbering system, so that each Deep Space Station (DSS) would bear a unique number. The original Echo antenna became DSS-12, while the antenna used in the Venus radar experiments became DSS-13. The Mars Station was DSS-14.73

The Mars Station, as part of the Deep Space Network, underwent a major upgrade in the 1970s in order to accommodate the needs of the Viking and Mariner Jupiter-Saturn spacecraft (later known as Voyager). For the Viking mission, each DSN station would have to handle six simultaneous data streams from the two Viking Orbiters and the one Lander. [109] Viking, in fact, was a dual-frequency craft; it used both S-band and X-band frequencies. For the Mariner flight to Jupiter and Saturn, the telemetry rates were the same as those for Mariner 10, but the data were coded and transmitted at X-band from distances up to nine astronomical units. Operating in the higher X-band range gave the increased sensitivity needed to remain in contact with Mariner, as it flew by Jupiter and Saturn. Construction of the 400-kilowatt, X-band (8495 MHz; 3.5 cm) transmitter for the Mars Station was completed by Advanced Systems Development, and the DSS-14 began operating at X-band in 1975.74

During the 1970s, the population of JPL planetary radar astronomers grew. Jurgens had an undergraduate and graduate degree in electrical engineering from Ohio University and had taught electrical engineering at Clarkson College (Ohio), before pursuing a doctoral degree at Cornell. Sometime after he finished researching his dissertation, a study of the radar scattering properties of Venus, at Arecibo, JPL hired Jurgens in 1972 to serve on the technical staff of the Telecommunications Research Section, not to do planetary radar astronomy.75

Also working in Goldstein's section was George Downs, who had studied radio astronomy at Stanford University under Ronald Bracewell. Goldstein had Downs analyze Mars radar data and make observations at Goldstone to assist in the selection of the Viking landing site, a project funded by the Viking Project Office. The planetary radar work, however, was in addition to his regular JPL duties, which involved studying newly discovered radio sources as potential timing sources for the Deep Space Network.76

During the heyday of the Viking Mars radar observations, Goldstein called upon other JPL employees, such as Howard C. Rumsey, Jr., who had a strong background in physics and mathematics, and the hardware experts George A. Morris and Richard R. Green. Jurgens described the atmosphere at JPL: "We all knew each other's talents. It was very efficient. Nobody ever felt like we were working terribly hard. It was just like a big playpen. Everybody came here, and we sort of did our thing and thought about what we wanted to do. We'd talk to each other, and we'd go out to lunch. It was the period of the long lunches sometimes. We had the Gourmet Society. The Gourmet Society was really headed by Howard Rumsey, who really liked good food. He would read the Sunday gourmet page and the Thursday gourmet page in the L. A. Times, and pick out interesting restaurants. At least one day a week, we went trudging off-lab to eat decent food at some interesting place that Howard had selected. These things often involved bicycle trips as far as Long Beach."77

Once Viking project funding ended in 1976, JPL radar astronomy hit hard times. Getting time on the DSN become more difficult. It was easy to get time in the early and middle 1960s, when the DSN was tracking few spacecraft. As Dick Goldstein explained: "Back in the sixties I thought of myself as director of the Goldstone Observatory. I got to choose what we could do, if I could get support for it."78 During the 1960s, the JPL radar experiments conducted on Venus involved hundreds of hours of runs; for example, the 1961 Venus experiment involved 238 hours of data collected over two months. But by the end of the decade, the amount of time available had declined. The JPL 1969 Venus observations were not made daily for a period of months during inferior conjunctions, but only "on 17 days spaced from 11 March to 16 May 1969."79

[110] The reduction in available antenna time was in direct proportion to the increasing number of spacecraft with which the Deep Space Network communicated. By 1977, the DSN was in communication with a record 14 spacecraft. In addition to the three Viking craft (two orbiters and one lander), the DSN communicated with Helios 1 and 2, Pioneer 11 (Saturn), Pioneer 10 (which was leaving the solar system), Voyagers 1 and 2, and Pioneers 6, 7, 8, and 9. That number grew to 19, a new record, the following year, when the DSN also handled communications with Pioneer Venus, which was an orbiter and four probes.80

Then the Deep Space Network stopped funding radar astronomy hardware. The ability to carry out radar astronomy without official recognition was maintained thanks to the presence at high levels of JPL management of Eb Rechtin and Walt Victor, who watched over planetary radar activities. But Rechtin left JPL, and Victor transferred in December 1978 from the DSN to the Office of Planning and Review.81 Without their guardianship, JPL radar astronomy was vulnerable.

As Goldstein explained: "From a chauvinistic point of view, it was a disaster, because the rest of the world passed us by....We went from being a couple years ahead to being a couple years behind."82 Without funding for hardware, the radar system was at risk. Moreover, the Goldstone Mars Station was in desperate need of repairs, and the equipment was becoming harder and harder to maintain. In 1976, the antenna already was ten years old, and the electronic equipment transferred to the Mars Station from the Venus Station (DSS-13) was even older.83

The termination of Deep Space Network funding of planetary radar astronomy grew out of two concerns, one within JPL and the other within the Deep Space Network. One of Bruce Murray's chief concerns after taking over as Laboratory director was the state and status of science and scientists at JPL. The basic criticism was that JPL lacked a commitment to scientists. But the problem had a cultural side; technologically-centered team-work dominated laboratory culture. Also, many of those doing science were like Dick Goldstein and Ray Jurgens; trained and hired as electrical engineers, they carried out radar astronomy science experiments. Murray made the problem the topic of mini-retreats, meetings, and seminars and, as a first step in elevating the status of science at JPL, appointed in October 1977 the first JPL chief scientist, Caltech physics professor Rochus E. Vogt, who had authored a report on relations between Caltech and JPL, another topic of great concern.84

Despite, or rather because of, Murray's concerns for science at JPL, planetary radar astronomy did not fair well under his reign as laboratory director. Goldstein was transferred out of the section where he had guided and supported the JPL planetary radar effort. JPL management decided that it was not proper to do science under the guise of improving the DSN. Radar astronomy should compete with other JPL science activities, and the Office of Space Science and Applications (OSSA; now the OSSI, Office of Space Science and Instruments) at NASA Headquarters should fund it, they ruled.85

At the same time, the DSN budget was suffering from monetary and manpower limitations.86 To make matters worse, a routine review of the Deep Space Network, chaired [111] by Eb Rechtin, declared that radar astronomy was no longer the "cow-catcher" of the DSN, meaning that the role of radar astronomy in creating new hardware to help drive forward the Deep Space Network had come to an end. It was, therefore, time to pull the plug on planetary radar astronomy, after previous reviews had lauded it.

From time to time, acting under instruction from NASA Headquarters, the Deep Space Network called into existence the TDA (Tracking and Data Acquisition) Advisory Panel to review DSN long-term plans. Planetary radar was held high as an integral part of DSN development activities by Ed Posner, a DSN manager. Among the hardware contributions of radar astronomy he listed were microwave components, signal processing techniques, and station control concepts, all of which were tested in a "realistic environment." Planetary radar fell from that favorable position during the 1978 review. The head of the review panel, now called the DSN Advisory Group, was none other than Eb Rechtin, the architect of the Deep Space Network and the one responsible for making planetary radar a testbed of DSN technology. DSN management asked the panel to consider, among many other questions, radar astronomy. In the opinion of the Advisory Group, which Eb Rechtin wrote, "Another DSN technology which may have had its day as a foundation for DSN technology is DSN radar astronomy. Radar astronomy served the DSN very well for many years. The Advisory Group wonders what the next area might be."87 Radar astronomy no longer produced the cutting edge hardware that justified the support of Advanced Systems Development. As a result, planetary radar astronomy at Goldstone went begging for money.

A good part of the problem was the perception of planetary radar as just a testbed for DSN technology. The value of the science was simply not recognized by either DSN management or NASA Headquarters. After all, in accordance with Eb Rechtin's plan, planetary radar was not to occupy a budget line nor to have program status; it was simply a DSN activity to assist in the development and testing of new technology.

The lack of money to even maintain the Goldstone radar, whose age and one-of-a-kind design engineering made it all that much harder to maintain, began to frustrate the performance of experiments. By 1980, the Goldstone radar was in such bad shape that planetary radar astronomy experiments were no longer carried out on a regular basis. The radar was resurrected for attempts at asteroid 4 Vesta on 28 May 1982 and comets IRAS-Araki-Alcock and Sugano-Saigusa-Fujikawa in 1983, but only Comet IRAS-Araki-Alcock was detected successfully. As Ray Jurgens reflected on the situation: "Basically, it looked like it was the end of the radar."88




In discussing radar systems available for planetary research, an instrument that one must not overlook is the bistatic Goldstack radar, which used Haystack as the transmitting antenna and the JPL Goldstone DSS-14 radar as the receiving antenna. In the past, planetary radar astronomers seldom used bistatic radars, let alone radars requiring the coordination of two unrelated institutions. Bistatic radars require a daunting amount of coordination on both the technical and institutional level. Nonetheless, transmitting power and antenna receiver sensitivity can combine to create a radar capable of doing more than either facility operating monostatically. In theory, Goldstack could outperform either Haystack or DSS-14 separately and achieve a nearly tenfold increase in overall radar performance.

[112] When radar astronomers Irwin Shapiro and Gordon Pettengill pitched Goldstack to NASA in 1968, they outlined an ambitious program of research: 1) observations of the Galilean satellites Ganymede and Callisto; 2) maps of the surfaces of Mercury and Venus; 3) a Moon-Earth-Moon triple-bounce experiment to study the Earth's radar-reflecting properties; 4) topographical studies of Mars at a resolution of 150 meters; and 5) a radar test of general relativity.89 Haystack and JPL engineers worked out the technical details of those experiments, and by May 1970 JPL had installed an X-band maser tunable to the Haystack frequency. The demands of the space program on the Mars Station, however, forced postponement of the experiments. As Shapiro recalled: "DSN always had scheduling problems. Scheduling was the biggest pain in the neck. From the point of view of science, I never felt the best things were done with scheduling; the engineering and mission pressures were too enormous. It always seemed to be as impossible as possible to schedule ground-based science experiments, but good science, in fact, was done." Goldstack eventually searched for Ganymede and Callisto in late May and early June 1970.90


Jodrell Bank


Several years before planetary radar astronomy ended at Haystack and declined at JPL, radar research at Jodrell Bank came to an end, too. In contrast to JPL, Jodrell Bank officially recognized and funded its radar astronomy program, and Sir Bernard Lovell proudly and, in the face of adversity, stubbornly maintained radar research. The Jodrell Bank facility was an example of British Big Science; private and civilian governmental funding underwrote the building of the large dish. While the U.S. military funded some meteor radar research, Jodrell Bank radar astronomy was not, in any sense, an extension of American Big Science. The demise of planetary radar astronomy at Jodrell Bank was a lesson in the dangers inherent in Little Science, not Big Science.

Thanks to NASA and the American military, Jodrell Bank did not lack for radar equipment. The still secret agreement between Lovell and an unidentified Air Force officer had as its immediate objective the sending of commands to the Pioneer 5 spacecraft. The U.S. Air Force funded Space Technology Laboratories (STL), a Los Angeles-based wholly-owned subsidiary of Ramo-Wooldrige (later TRW), to install a continuous-wave 410.25-MHz (73-cm) radar transmitter and other equipment on the Jodrell Bank telescope in order to track lunar rocket launches. Although the STL transmitter had only a few kilowatts of power, it was stable, reliable, and free of the problems that plagued the pulse radar apparatus pieced together by John Evans. Ownership of the STL transmitter passed to NASA, which provided operational funds between 1959 and 1964 to track rocket launches, not to perform radar experiments. NASA left the equipment on the Jodrell Bank antenna "on an indefinite loan basis," so that the University of Manchester might use it for scientific research.91

[113] In 1962, the Jodrell Bank radar group consisted of only John Thomson and his graduate student John E. B. Ponsonby. Evans had sought his fortune at Lincoln Laboratory. As Ponsonby characterized the radar group, "We were always two men and a boy [K. S. Imrie]." When Ponsonby arrived at Jodrell Bank in 1960, he was shocked to discover that he was the only one in the radar group with a flare for electronics; Thomson, according to Ponsonby, was happy doing computations. Jodrell Bank radar astronomy was small not only in terms of staff, but also in observing time, which varied between 1 and 10 percent.92

Thomson and Ponsonby abandoned much of the equipment Evans had been using; they used a simpler approach with less technical risk. The old apparatus used vacuum tubes; the new was all solid-state digital electronics. A grant from the DSIR underwrote the cost of these modifications, as well as the purchase of a parametric amplifier and spare klystron tubes. The 1962 and 1964 Jodrell Bank Venus radar experiments were carried out with this digital continuous-wave equipment.93

The focus of Jodrell Bank's Venus radar research after 1962 was a bistatic experiment with the Soviet Long-Distance Space Communication Center located near Yevpatoriya in the Crimea. The experiment was possible only because Lovell had succeeded in thawing Cold War relations. The opportunity came in March 1961, when Soviet space trackers lost contact with a Venus probe launched the previous month. The Soviet Academy of Sciences approached Lovell to use the Jodrell Bank telescope to search for signals. As the months passed, and Jodrell Bank attempted to make contact with the probe, communications between the British and Soviets increased. The collaboration led to the establishment of a telex link between Jodrell Bank and the Yevpatoriya radar station, as well as an invitation for Lovell to visit the Soviet Union two years later.

The idea of doing the bistatic experiment came to Lovell during his visit to Yevpatoriya, when he discovered the extremely powerful Soviet transmitter. Vladimir Kotelnikov, who headed the Soviet planetary radar effort, joined Lovell as the other moving spirit behind the bistatic project. An Iron Curtain of secrecy hindered the project, however. In order to set up the bistatic radar, Jodrell Bank had to know the frequency and precise coordinates of the Yevpatoriya radar. The Soviets were loathe to disclose their frequency, transmitter size, location, or even antenna dimensions, but the British established those parameters step by step. Nonetheless, the experiment did not work initially, because Jodrell Bank lacked the correct Doppler shift. After testing the bistatic arrangement on the Moon, the Yevpatoriya facility began to transmit radar signals to Venus, and Jodrell Bank received them from January through March 1966. Data tapes were delivered to Kotelnikov by way of the British Embassy. As a long-distance bistatic radar experiment, the effort was a first. However, it was an opportunity lost.94

Ponsonby set forth a cogent analysis of the bistatic Venus experiment in his dissertation: "Planetary radar has proved to be a field in which new results are only obtained by the groups which have the most sensitive systems and the data processing capacity to make the best use of the data acquired. In both respects the group at Jodrell Bank has never been in a leading position."95

Lovell disagreed; the bistatic experiment was "just too late." JPL and Lincoln Laboratory already had determined the rate and direction of rotation of Venus. "But if [114] only my 1963 conversations and agreement with the Soviet Union could have been facilitated without trouble at this end and without trouble at the transmitter," Lovell argued, "we would have been first on that."96

Lovell's analysis, as well as that of Ponsonby, raised the vital question of the ability of the Jodrell Bank radar group to effectively compete against American radar astronomers. The STL transmitter operated in the UHF band (410.25 MHz; 73 cm). Although the Arecibo Observatory operated in the same band, the trend in planetary radar astronomy was toward higher frequency ranges, the S-band at JPL (and later at Arecibo) and the X-band at Haystack. The higher frequencies allowed the radar to do much more radar astronomy science than was possible at UHF.

Ponsonby raised another point in his dissertation: "If a true state-of-the-art transmitter were acquired it would cost an appreciable fraction of the cost of the telescope, and clearly to justify investment on that scale it would have to be used much more extensively than would be compatible with the predominantly passive radio astronomical programs at Jodrell Bank. Passive radio-astronomy may appropriately be done as a secondary line of research at a primarily radar installation, but experience has shown that the two activities do not combine well the other way. Appreciating this, the research reported in this thesis is not, at least for the present, being pursued further." Indeed, Ponsonby continued, "The limited computing facilities available in the University and the lack at the time of on-line computers at Jodrell Bank in effect prevented a thorough analysis of the data that was acquired, and this took away much of the value of the observations."97

The acquisition of new radar and computer equipment certainly would have constituted a significant expenditure, but Lovell probably could have raised the necessary money. Could Jodrell Bank have kept up with the development of planetary range-Doppler mapping in the United States? Thomson was working on an aperture synthesis technique for making lunar radar maps. The mathematical process for constructing the image was analogous to that now used for tomographic brain scanners and differed entirely from that used in the United States to construct range-Doppler maps. The technique was not very practical, however; it required computer capacity not then available at Jodrell Bank and ultimately could not be generalized to the planets.98

In the end, the small scale of planetary radar astronomy at Jodrell Bank did it in. Thomson and Ponsonby grew tired of the Soviet bistatic Venus experiment. They carried the main load of the work at irregular hours of the day and night. Finally, on 18 March 1966, Thomson and Ponsonby could take no more. They handed Lovell a list of ten good reasons for ending the experiment. Kotelnikov agreed to "an interval" in the observations, which never resumed.99 Ponsonby remained rather cynical about the venture, which he has characterized as a political exercise. "The signals were recorded on magnetic tape and sent off to Russia, and I never heard from them again!" 100

Ponsonby already was tired of the bistatic experiments, when the death of John Thomson from an inoperable brain tumor in August 1969 devastated the Jodrell Bank planetary radar program. Without Thomson, and certainly without Ponsonby's interest, Jodrell Bank had no radar group. Through sheer stubbornness, however, Lovell tried to keep the radar program going. In October 1969, he and his Jodrell Bank colleagues drew up a scientific program for a proposed 122-meter (400-ft) telescope, the Mark V. The program included a series of planetary radar experiments outlined by Ponsonby. Was this the [115] telescope that could have revived Jodrell Bank radar research? Like its American cousin the NEROC telescope, the Mark V was never built. In retrospect, Lovell realized that "It was now out of the question for us to continue....I saw the passing of radar as inevitable, but with regret." 101

The sixties was the era of the Big Dish; large antenna projects came and went, and so did planetary radars. In 1965, four antennas supported planetary radar experiments: Arecibo, Haystack, Jodrell Bank, and the Goldstone Mars Station. A fifth dish, the NEROC telescope, was on the drawing board. But ten years later, the NEROC telescope had not been built; Haystack and Jodrell Bank no longer performed planetary radar experiments. By 1980, Goldstone had joined their number. Only Arecibo remained. Planetary radar astronomy appeared to be a collapsing field.

At Arecibo, nonetheless, radar astronomy had found a patron in NASA. Planetary radar there also had a recognized and guaranteed budget, as well as a world-class research instrument, and both Cornell and MIT fed graduate students to the Arecibo facility. Given the financial, institutional, technological, and other resources available at Arecibo for planetary radar astronomy, one would have expected the field to have occupied an increasing amount of antenna time from 1974, when Haystack ceased radar astronomy, to 1980, when JPL activity virtually ended. Instead, antenna use remained relatively stable, averaging about six percent between 1971 and 1980 and passing seven percent concurrently with the inferior conjunctions of Venus.102

In terms of personnel, one could count the field of planetary radar astronomy as consisting of nine individuals. At MIT was Gordon Pettengill; at JPL, Dick Goldstein, Ray Jurgens, and George Downs. The Arecibo Observatory supported four radar practitioners: Don Campbell, associate director at the Arecibo Observatory since 1979; John Harmon, AO research associate since 1978; Steven J. Ostro, Cornell assistant professor of astronomy since 1979; and Barbara Ann Burns, a graduate student of Don Campbell.

In 1980, planetary radar astronomy was indeed a small field in terms of available instrumentation and active practitioners. It was an example of Little Science, but one which depended on Big Science for its very existence. Moreover, although that Big Science had been as diverse as military, space, ionospheric, and radio astronomy research at the emergence of radar astronomy, by 1980 Big Science had come to mean one thing: NASA. The financial and institutional arrangements with NASA influenced the kind of science done. In order to understand how that science was influenced, we must first look at the evolution of planetary radar astronomy as a science.



1. Emilio Q. Daddario, "Needs for a National Policy," Physics Today 22 (1969): 33-38; James E. Hewes, Jr., From Root to McNamara: Army Organization and Administration, 1900-1963 (Washington: U.S. Army Center of Military History, 1975), pp. 299-315.

2. Gordon 28/11/94; Benjamin Nichols, telephone conversation, 14 December 1993; "Cornell University Center for Radiophysics and Space Research," typed manuscript, 12 August 1959, Office of the Administrative Director, NAIC; Gordon, "Incoherent Scattering of Radio Waves by Free Electrons with Applications to Space Exploration by Radar," Proceedings of the IRE 46 (1958): 1824-1829; George Peter, Evolution of Receivers and Feed Systems for the Arecibo Observatory (Ithaca: NAIC, 1993), pp. 4-5; SCEL Journal Vol. S-1, no. 32 (6 August 1953): 2, "Signal Corps Engineering Laboratory Journal/R&D Summary," HAUSACEC; Gillmor, "Federal Funding," p. 126.

3. Gordon 28/11/94; Benjamin Nichols, telephone conversation, 14 December 1993; Gordon, Booker, and Nichols, Design Study of a Radar to Explore the Earth's Ionosphere and Surrounding Space, Research Report EE 395 (Ithaca: Cornell School of Electrical Engineering, 1 December 1958), pp. 1 & 10-11; Gordon, Antenna Beam Swinging and the Spherical Reflector, Research Report EE 435 (Ithaca: Cornell School of Electrical Engineering, 1 August 1959), pp. 1 & 8; CRSR, Construction of the Department of Defense Ionospheric Research Facility - Final, Research Report RS 55 (Ithaca: CRSR, 30 November 1963), p. 2; Gillmor, "Federal Funding," p. 127.

4. Gordon 28/11/94; Nichols, telephone conversation, 14 December 1993; Jack P. Ruina, "Arecibo," Electronics 7 April 1961, n.p., article in publicity folder, Office of the Administrative Director, NAIC; CRSR, Ionospheric Research Facility, p. 2; Herbert F. York, Making Weapons, Talking Peace: A Physicist's Odyssey from Hiroshima to Geneva (New York: Basic Books, 1987), pp. 142-143; Gillmor, "Federal Funding," p. 126.

5. Gordon 28/11/94; Philip Blacksmith, "DODIRF 1000-foot Spherical Reflector Antenna," and Alan F. Kay, A Line Source Feed for a Spherical Reflector, Technical Report 529 (Hanscom AFB: AFCRL, 29 May 1961), Phillips Laboratory; Roy C. Spencer, Carlyle J. Sletten, and John E. Walsh, "Correction of Spherical Aberration by a Phased Line Source," Proceedings of the National Electronics Conference 5 (1949): 320-333; Gillmor, "Federal Funding," p. 127.

6. Gordon 28/11/94; Gordon, "Arecibo Ionospheric Observatory," Science 146 (2 October 1964): 26; Gordon, "Arecibo Ionospheric Observatory," p. 26; Gordon, Booker, and Nichols, pp. 12-13; Donald J. Belcher, "Site Locations for a Proposed Radio Telescope," Appendix C in ibidem; R. E. Mason and W. McGuire, "The Fixed Antenna for a Large Radio Telescope: Feasibility Study and Preliminary Cost Estimate," Appendix B in ibidem.

7. CRSR, Design Studies for the Arecibo Radio Observatory, Research Report RS 9 (Ithaca: CRSR, 30 June 1960), NAIC, p. 1.

8. Gold 14/12/93; Nichols, telephone conversation, 14 December 1993; "Center for Radiophysics;" Annual Summary Report, Center for Radiophysics and Space Research, July 1, 1965-June 30, 1966, 30 June 1966, p. 10; Arecibo Observatory Program Plan, October 1, 1970-September 30, 1971, May 1971, pp. 62-63, AOL.

9. CRSR, Scientific Experiments for the Arecibo Radio Observatory, Research Report RS 5 (Ithaca: CRSR, 31 March 1960), pp. vii & 31-33; AIO, Research in Ionospheric Physics, Research Report RS 41 (Ithaca: CRSR, 30 June 1962), p. 7.

10. Price 27/9/93; CRSR, Construction of the Department of Defense Ionospheric Research Facility, Research Report RS 22 (Ithaca: CRSR, 30 June 1961), pp. 1-2; ibid., Research Report RS 34 (Ithaca: CRSR, 31 December 1961); ibid., Research Report RS 40 (Ithaca: CRSR, 30 June 1962), pp. 12-15; ibid., Research Report RS 45 (Ithaca: CRSR, 31 December 1962), pp. 1 & 11-12; various items in publicity binder, Office of the Administrative Director, NAIC; Thomas C. Kavanagh and David H. H. Tung, "Arecibo Radar-Radio Telescope Design and Construction," Journal of the Construction Division, Proceedings of the American Society of Civil Engineers 91 (May 1965): 69-98.

11. CRSR Summary Report, July 1, 1964-June 30, 1965, 1 July 1965, CRSR, p. 5; Cornell-Sydney University Astronomy Center, 1965, p. 4, AOL; Gold and Harry Messel, "A New Joint American-Australian Astronomy Center," Nature 204 (1964): 18-20.

12. Pettengill 28/9/93; Gold 14/12/93.

13. Dyce 22/11/94; Pettengill 28/9/93.

14. AIO, Research in Ionospheric Physics, Research Report RS 61 (Ithaca: CRSR, 31 December 1964), pp. 46-48; ibid., Research Report RS 72 (Ithaca: CRSR, 31 January 1968), p. 127; Vahi Petrosian, Two Possible Methods of Detecting UHF Echoes from the Sun, Research Report RS 54 (Ithaca: CRSR, 30 September 1963), which was his masters thesis. His doctoral thesis, completed in June 1967, however, was on "Photoneutrino and Other Neutrino Processes in Astrophysics." Petrosian later went to Stanford. CRSR, "Proposal to National Science Foundation for Research Ionospheric Physics, Radar-Radio Astronomy, October 1, 1969 through September 30, 1971," April 1969, pp. 138-140, Office of the Administrative Director, NAIC.

Donald B. Campbell obtained solar continuous-wave echoes at 40 MHz (7.5 meters) during the summer of 1966. AIO, Research in Ionospheric Physics, Research Report RS 70 (Ithaca: CRSR, 31 January 1967), p. 75. Alan D. Parrish, a NASA Trainee, and Campbell made more solar observations in 1967. ibid., Research Report RS 71 (Ithaca: CRSR, 31 July 1967), pp. 78-79; Campbell 8/12/93.

15. Thompson and Dyce, "Mapping of Lunar Radar Reflectivity at 70 Cm," Journal of Geophysical Research 71 (1966): 4843-4853; Thompson, "Radar Studies of the Lunar Surface Emphasizing Factors Related to Selection of Landing Sites," Research Report RS 73 (Ithaca: CRSR, April 1968); Gold, CRSR Summary Report, July 1, 1964-June 30, 1965, 1 July 1965, CRSR, p. 4; Annual Summary Report, Center for Radiophysics and Space Research, July 1, 1966-June 30, 1967, 30 June 1967, p. 10; Annual Summary Report, Center for Radiophysics and Space Research, July 1, 1968-June 30, 1969, 30 June 1969, p. 4; AIO, Research in Ionospheric Physics, Research Report RS 61 (Ithaca: CRSR, 31 December 1964), pp. 39-41.

16. Thompson 29/11/94; NAIC QR Q2/1970, n.p.; Thompson, "The Study of Radar-Scattering Behavior of Lunar Craters at 70 Cm," Ph.D. diss., Cornell, February 1966; Thompson, "Radar Studies of the Lunar Surface Emphasizing Factors Related to Selection of Landing Sites," Research Report RS 73 (Ithaca: CRSR, April 1968). The lunar radar measurements were made at 40 MHz (7.5 meters) and 430 MHz (70 cm) at the AIO.

17. CRSR, "Proposal to National Science Foundation for Research Ionospheric Physics, Radar-Radio Astronomy, October 1, 1969 through September 30, 1971," April 1969, Office of the Administrative Director, NAIC, pp. 138-140; Jurgens, "A Study of the Average and Anomalous Radar Scattering from the Surface of Venus at 70 Cm Wavelength," Ph.D. diss., Cornell, June 1968.

18. Pettengill 28/9/93.

19. Rogers 5/5/94; Hine 12/3/93. For a discussion of radar interferometry at Haystack and Arecibo, see Chapter Five.

20. Pettengill 28/9/93.

21. Sebring to Hurlburt, 27 March 1970, 18/2/AC 135, MITA; AIO, Research in Ionospheric Physics, Research Report RS 69 (Ithaca: CRSR, 30 June 1966), p. 87; ibid., Research Report RS 70 (Ithaca: CRSR, 31 January 1967), pp. 124-125; ibid., Research Report RS 71 (Ithaca: CRSR, 31 July 1967), pp. 113-124; ibid., Research Report RS 72 (Ithaca: CRSR, 31 January 1968), pp. 125-134; ibid., Research Report RS 74 (Ithaca: CRSR, 31 July 1968), pp. 137-145; ibid., Research Report RS 75 (Ithaca: CRSR, 31 March 1969), p. 51; ibid., Research Report RS 76 (Ithaca: CRSR, 30 September 1969), p. 44; NAIC QR Q1-Q4/1970, passim. The Arecibo Observatory quarterly reports for the years 1971 to 1975 indicate the fraction of radar astronomy use of the antenna: 2.9 percent in 1971; 9.5 percent in 1972; 6.9 percent in 1973; 1.9 percent in 1974; and 7.2 percent in 1975. At Haystack, in March 1970, for example, of the 290 hours scheduled, 90 (31 percent) were spent on radar observations. See Chapter 3 for Haystack radar use.

22. Peter, p. 12; AIO, Research in Ionospheric Physics, Research Report RS 70 (Ithaca: CRSR, 31 January 1967), p. 1.

23. John Lannan, "An Example of Scientific Research under Scrutiny," The Sunday [Washington] Star, 30 March 1969, p. F-3. For the AIO budget, see CRSR Summary Report, July 1, 1964-June 30, 1965, 1 July 1965, CRSR, p. 1; Annual Summary Report, Center for Radiophysics and Space Research, July 1, 1965-June 30, 1966, 30 June 1966, pp. 2 & 7; ibid., July 1, 1966-June 30, 1967, 30 June 1967, pp. 8, 10 & 12; ibid., July 1, 1967-June 30, 1968, 30 June 1968, pp. 1, 11 & 13; ibid., July 1, 1968-June 30, 1969, 30 June 1969, pp. 1 & 15. AFOSR contract F44-620-67-C0066 allocated $5,210,200 for the term 1 February 1967 through 30 September 1969.

24. Gordon 28/11/94; Gold 14/12/93; Campbell 7/12/93; L. Merle Lalonde and Daniel E. Harris, "A High-Performance Line Source Feed for the AIO Spherical Reflector," IEEE Transactions on Antennas and Propagation AP-18 (January 1970): 41.

25. Bok to George B. Field, "Arecibo," NSFHF; Kay, A Line Source Feed; J. Pierluissi, A Theoretical Study of Gregorian Radio Telescopes with Applications to the Arecibo Ionospheric Observatory, Research Report RS 57 (Ithaca: CRSR, 1 April 1964), NAIC; Peter, p. 18; Campbell 7/12/93.

26. Diary note, Hurlburt, 15 December 1967, and Long to Haworth, 27 July 1967, "Arecibo," NSFHF; Annual Summary Report, Center for Radiophysics and Space Research, July 1, 1965-June 30, 1966, 30 June 1966, pp. 12 & 18; Ibid., July 1, 1966-June 30, 1967, 30 June 1967, p. 8; AIO, Research in Ionospheric Physics, Research Report RS 71 (Ithaca: CRSR, 31 July 1967), p. 1; Gold 14/12/93.

27. National Science Board, Approved Minutes of the Open Sessions, pp. 113:14-113:15, National Science Board; "Report of the Ad Hoc Advisory Panel for Large Radio Astronomy Facilities," 14 August 1967, typed manuscript, pp. 2-3, NSFL; Lalonde and Harris, p. 42; AIO, Research in Ionospheric Physics, Research Report RS 72 (Ithaca: CRSR, 31 January 1968), p. 4; AIO, Ibid., Research Report RS 74 (Ithaca: CRSR, 31 July 1968), pp. 8-9.

28. Haworth to John Foster, 9 November 1967; Memorandum, Gerard Mulders to Haworth, Randal M. Robertson, and William E. Wright, 25 August 1967; and Memorandum, Mulders to Robertson, 3 January 1968, "Arecibo," NSFHF.

29. Hurlburt diary note; Long to Peter Franken, 23 August 1967, and Long to Leland Haworth, 27 July 1967, "Arecibo," NSFHF.

30. Franken to William Wright, 23 August 1967, "Arecibo," NSFHF.

31. S. J. Lukasik to Long, 12 December 1967; Memorandum of Understanding, AIO, attached to letter, Haworth to John Foster, 30 April 1969; and Memorandum of Understanding, AIO, attached to letter, S. E. Clements to Haworth, 12 May 1969, signed by Haworth and Foster, "Arecibo," NSFHF.

32. Maintenance and equipment improvements were 11 percent and radar astronomy 9 percent of antenna time. AIO, Research in Ionospheric Physics, Research Report RS 69 (Ithaca: CRSR, 30 June 1966), p. 87; Ibid., Research Report RS 70 (Ithaca: CRSR, 31 January 1967), pp. 124-125; Ibid., Research Report RS 71 (Ithaca: CRSR, 31 July 1967), pp. 113-124; Ibid., Research Report RS 72 (Ithaca: CRSR, 31 January 1968), pp. 125-134; Ibid., Research Report RS 74 (Ithaca: CRSR, 31 July 1968), pp. 137-145; Ibid., Research Report RS 75 (Ithaca: CRSR, 31 March 1969), p. 51; and Ibid., Research Report RS 76 (Ithaca: CRSR, 30 September 1969), p. 44.

33. Pettengill 28/9/93.

34. Gold 14/12/93.

35. Gordon 28/11/94.

36. Frank Drake and Dava Sobel, Is Anyone Out There? (New York: Delacorte Press, 1992), pp. 77 & 79.

37. Jones to Haworth, 8 February 1968, "Arecibo," NSFHF.

38. Gold 14/12/93. Bill Gordon declined comment on the whole affair. Gordon 28/11/94.

39. Statement of the National Science Foundation Advisory Panel for Atmospheric Sciences to the Director of the National Science Foundation, 21 March 1968, "Arecibo," NSFHF.

40. See, for instance, Haworth to Long, 23 January 1968, "Arecibo," NSFHF.

41. Messel to Donald Hornig, 12 June 1968, and Hornig to Messel, 9 July 1968, "Arecibo," NSFHF; Gold 14/12/93.

42. Randal N. Robertson to Long, 17 March 1969, "Arecibo," NSFHF.

43. CRSR, "Proposal to National Science Foundation for Research Ionospheric Physics, Radar-Radio Astronomy, October 1, 1969 through September 30, 1971," April 1969, Office of the Administrative Director, NAIC; advanced draft, "The Management of the AIO as a National center," July, 1968, "Arecibo," NSFHF.

44. Memorandum, J. H. Fregeau to Associate Director (Research), NSF, 28 April 1969, "Arecibo," NSFHF.

45. Annual Summary Report, Center for Radiophysics and Space Research, July 1, 1969-June 30, 1970, 30 June 1970, p. 9.

46. Campbell 9/12/93; Arecibo Observatory Program Plan, October 1, 1970-September 30, 1971, May 1971, pp. 35-39, Office of the Administrative Director, NAIC.

47. Pettengill 28/9/93; Campbell 7/12/93; Arecibo Observatory Program Plan, October 1, 1971-September 30, 1972, January 1972, NAIC, pp. 25-31; Arecibo Observatory Program Plan, October 1, 1970-September 30, 1971, May 1971, p. 63, AOL; NAIC QR Q3/1971, 9.

48. AIO, Research in Ionospheric Physics, Research Report RS 76 (Ithaca: CRSR, 30 September 1969), p. 1; "Report of the Second Meeting of the Ad Hoc Advisory Panel for Large Radio Astronomy Facilities," 15 August 1969, typed manuscript, p. 3, NSFL.

49. Hess to Naugle, 27 January 1969, NHOB.

50. Campbell 7/12/93.

51. Brunk, Planetary Astronomy New Starts, FY 1971, n.d., NHOB.

52. Tatarewicz, p. 98. For the creation and demise of the NASA Electronics Research Center, see Ken Hechler, Toward the Endless Frontier: History of the Committee on Science and Technology, 1959-1979 (Washington: USGO, 1980), pp. 219-231.

53. Henry J. Smith, Memo to the files, 11 December 1969, NHOB.

54. AIO, Proposal to NSF and NASA for Major Additions and Modifications to the Suspended Antenna Structure and Equipment of the Arecibo Observatory, February 1971 through February 1973, February 1970, and Daniel Hunt to Brunk, 11 December 1970, NHOB; National Science Board, Minutes of the Open Meetings, 132:6 and 133:7-8, National Science Board; NAIC QR Q3/1971, p. 13.

55. Hunt to Brunk, 11 December 1970; NASA Deputy Associate Administrator for Space Science and Applications to Assistant Administrator, Office of DOD and Interagency Affairs, 6 March 1971; Memorandum, Director of Planetary Programs, Office of Space Science and Applications, NASA, to Associate Administrator for Office of Tracking and Data Acquisition, 20 May 1971; and Memorandum of Agreement between NASA and the NSF for the Addition of a High-Power S-Band Radar Capability and Associated Additions and Modifications to the Suspended Antenna Structure of the NAIC at Arecibo, 24 June 1971, NHOB.

56. Memorandum of Agreement between NASA and the NSF for the Addition of a High-Power S-Band Radar Capability and Associated Additions and Modifications to the Suspended Antenna Structure of the NAIC at Arecibo, 24 June 1971, NHOB; "High Power Transmitter to Boost Arecibo Radar Capability," NSF press release, 17 August 1971, "Radar Astronomy," NHO.

57. NAIC QR Q1/1971, p. 5; Q2/1971, p. 6; Q3/1971, p. 6; Q4/1971, p. 7; Q1/1972, p. 10; Q2/1972, p. 13; Q3/1972, p. 11; and Q1/1973, p. 13; National Science Board, Minutes of the Open Meetings, 144:4-5, National Science Board.

58. AIO, Proposal to NSF and NASA for Major Additions and Modifications to the Suspended Antenna Structure and Equipment of the Arecibo Observatory, February 1971 through February 1973, February 1970, pp. 7-8, 11-15, NHOB; NAIC QR Q3/1972, pp. 11-12.

59. NAIC QR Q4/1971, p. 8; and Q2/1972, p. 13.

60. Campbell 8/12/93; NAIC QR Q2/1970, p. 4, Q4/1970, p. 9, Q1/1972, p. 12, Q3/1972, p. 12, and Q1/1973, p. 14; AIO, Proposal to NSF and NASA for Major Additions and Modifications to the Suspended Antenna Structure and Equipment of the Arecibo Observatory, February 1971 through February 1973, February 1970, pp. 14-15, NHOB; AIO dedication brochure, no page numbers, Cornell, 1974, NHOB.

61. Brunk to Claude Kellett, 18 April 1973, NHOB; Jack W. Lowe to W. E. Porter, 29 March 1977, Office of the Administrative Director, NAIC; NAIC QR Q1/1972, p. 11; Peter, p. 13. Documents relating to the transfer can be found in the Office of the Administrative Director, NAIC.

62. Drake and Sobel, pp. 180-185; Campbell 8/12/93; Dedication publication, Cornell University, 1974, NHOB; National Science Board, Minutes of the Open Meetings, 168:2, National Science Board; NAIC QR Q3/1974, p. 10, Q2/1975, p. 4, and Q3/1975, p. 4.

63. Campbell 8/12/93; Campbell 7/12/93; NAIC QR Q1/1975, p. 4.

64. NAIC QR Q3/1976, pp. 4-5.

65. Goldstein 14/9/93; Goldstein 7/4/93; Rechtin, telephone conversation, 13 September 1993.

66. Renzetti 16/4/92; Renzetti 17/4/92.

67. Soderblom 27/6/94.

68. Carpenter, telephone conversation, 14 September 1993.

69. Jurgens 23/5/94; Goldstein 14/9/93; Downs 4/10/94.

70. Memorandum, Carl W. Johnson to Murray, 31 October 1977, 62/3/89-13, JPLA; Goldstein 14/9/93.

71. Goldstein 14/9/93.

72. Victor, "General System Description," p. 3 in Goldstein, Stevens, and Victor, eds., Goldstone Observatory Report for October-December 1962, Technical Report 32-396 (Pasadena: JPL, 1 March 1965); Waff, ch. 6, pp. 17 & 19; Goldstein and Carpenter, "Rotation of Venus," pp. 910-911; Carpenter, "Study of Venus by CW Radar," p. 142.

73. Corliss, Deep Space Network, pp. 37-38, 50, 60-61, 82, 84, 87, 129 & 131; Renzetti, A History, pp. 25-26, 32, 52 & 54; Robertson, pp. 255-261; The NASA/JPL 64-Meter-Diameter Antenna at Goldstone, California: Project Report, Technical Memorandum 33-671 (Pasadena: JPL, 15 July 1974), pp. 7-17; Rechtin, Bruce Rule, and Stevens, Large Ground Antennas, Technical Report 32-213 (Pasadena: JPL, 20 March 1962), pp. 7-10.

74. Rob Hartop and Dan A. Bathker, "The High-Power X-Band Planetary Radar at Goldstone: Design, Development, and Early Results," IEEE Transactions on Microwave Theory and Techniques MIT-24 (December 1976): 958-963; JPL Annual Report, 1974-1975, p. 22, JPLA.

75. Jurgens 23/5/94.

76. Downs 4/10/94.

77. Jurgens 23/5/94.

78. Goldstein 14/9/93.

79. Golomb, "Introduction," in Victor, Stevens, and Golomb, p. 4; Goldstein and Howard C. Rumsey, Jr., "A Radar Snapshot of Venus," Science 169 (1969): 975.

80. JPL Annual Report, 1976-1977, p. 22, and ibid., 1978, p. 20, JPLA.

81. Murray to Allen M. Lovelace, 30 November 1978, 75/5/89-13, JPLA.

82. Goldstein 14/9/93.

83. Jurgens 23/5/94.

84. Agenda, Director's Mini-Retreat, "How Does Science Fit In at JPL?," 22 March 1977, 55/3/89-13; Director's Letter, no. 22, 30 September 1977, 61/3/89-13; Roger Noll to Murray, 23 November 1977, 63/3/89-13; and typed manuscript, First Annual "State of the Lab" Talk by Murray to Management Personnel, 1 April 1977, 55/3/89-13, JPLA.

85. Goldstein 14/9/93.

86. Notes from a discussion of TDA problems discussed during a mini-retreat held 8 November 1977, 63/3/89-13, JPLA; Jurgens 23/5/94; and Stevens 14/9/93.

87. "TDA Advisory Panel, 1971-1972," and "TDA Advisory Council, 1978-1981," JPLPLC.

88. Jurgens 23/5/94; Jurgens, "Comet Iras," pp. 222 & 224.

89. Memorandum, NEROC Project Office to Wiesner, 19 September 1968, regarding "Proposed Contact with Newell Regarding Possible Partial Support of Haystack by NASA," 8/2/AC 135, MITA; NEROC, Proposal to the National Science Foundation for Programs in Radio and Radar Astronomy at the Haystack Observatory, 8 May 1970, pp. III.8-III.10, LLLA; Brunk to Distribution List, 4 October 1968, NHOB.

90. Shapiro 4/5/94; Shapiro 1/10/93; "Funding Proposal, 'Plan for NEROC Operation of the Haystack Research Facility as a National Radio/Radar Observatory,' NSF, 7/1/71-6/30/73," 26/2/AC 135, and Sebring to Hurlburt, 27 March 1970, 18/2/AC 135, MITA; NEROC, Proposal to the National Science Foundation for Programs in Radio and Radar Astronomy at the Haystack Observatory, 8 May 1970, pp. III.8-III.10, LLLA; JPL 1970 Annual Report, p. 14, JPLA.

91. Lovell, 11/1/94; Evans 9/9/93; Ponsonby 11/1/94; Lovell, "Astronomer by Chance," pp. 322-325 & 328-329; Edmond Buckley to R. G. Lascelles, 8 November 1961, and related documents in 2/53, Accounts; Able, Thor, and Pioneer 5 materials in 4/16, Jodrell Bank Miscellaneous; materials in 1/4, Correspondence Series 2; 2/53, 2/52, 2/55, 7/55, 8/55, 1/59, and 3/59, Accounts; and 4/16, Jodrell Bank Miscellaneous, JBA.

92. Ponsonby 11/1/94; summaries of telescope use in 1/2, Correspondence Series, JBA.

93. Ponsonby 11/1/94; 2/51, Accounts, JBA; Ponsonby, Thomson, and Imrie, "Radar Observations of Venus," pp. 1-17; Ponsonby, Thomson, and Imrie, "Rotation Rate of Venus Measured by Radar Observations, 1964," Nature 204 (1964): 63-64.

94. Lovell, 11/1/94; Ponsonby 11/1/94; Lovell, "Astronomer by Chance," pp. 370-372; Lovell, Out of the Zenith, pp. 186-188 & 201-204; Ponsonby and Thomson, "U.S.S.R.-U.K. Planetary Radar Experiment," pp. 661-671 in R. W. Beatty, J. Herbstreit, G. M. Brown, and F. Horner, eds., Progress in Radio Science, 1963-1966 (Berkeley: URSI, 1967); Ponsonby, "Planetary Radar," pp. 6.11-6.22.

95. Ponsonby, "Planetary Radar," p. 6.21.

96. Lovell, 11/1/94.

97. Ponsonby, "Planetary Radar," pp. 6.21-6.22.

98. Ponsonby 11/1/94; Lovell, "Astronomer by Chance," pp. 373-375; various documents in 2/51, Accounts, JBA.

99. Lovell, Out of the Zenith, pp. 207-208.

100. Ponsonby 11/1/94.

101. Lovell, 11/1/94; Ponsonby 11/1/94; Lovell, Out of the Zenith, p. 203; Lovell, The Jodrell Bank Telescope, Chapters 5-6 and 9-10, especially pp. 55-56 & 257. In analyzing the demise of radar astronomy at Jodrell Bank, though the smallness of the active radar astronomy staff, technical and technological factors, and the American lead had a more determinant role, to be sure, one must not overlook the lure of radio astronomy.

102. These figures are based on the NAIC quarterly reports for the years 1971-1980. The percentage of radar use annually was 2.9 percent in 1971; 9.5 percent in 1972; 6.9 percent in 1973; 1.9 percent in 1974; 7.2 percent in 1975; 5.8 percent in 1976; 7.3 percent in 1977; 4.7 percent in 1978; 5.0 percent in 1979; and 7.8 percent in 1980. The average percentage for the period 1971-1980 was 5.9, while the average for 1971-1975 was 5.68 percent and for 1976-1980 6.12 percent.