May 25, 2010 | In celebration of the 50th anniversary, we asked all current staff who were hired in the 1960s to comment on what has—or hasn’t—changed at UCAR/NCAR over the decades, and to share some memories and stories about the organization’s early days. Here are their recollections.
Ben Domenico (Unidata)
"My earliest image of NCAR was a view from the top of the Third Flatiron, looking down into the basement, which was being excavated at the time—June of 1965—just after I had graduated from college. From that distance, it wasn't obvious how large the structure would be. My climbing friend and I concluded that some rich Boulderite had found the ideal location for a new house. I didn't realize what I had seen until I came to HAO a year later as a graduate student assistant and learned about the new Mesa Lab construction. Little did I know how many years I'd be spending in an office in the basement I had first seen from high above.
I had just gotten my undergraduate degree and was headed for a summer institute at the Goddard Institute for Space Studies at Columbia and then to Yale for graduate school.
That era ranks among the best ever for students in the sciences. The government was pouring money into science, technology, and education to achieve President Kennedy's goal of "landing a man on the moon and returning him safely to the earth before this decade is out." That summer, we got tours of several other major science and technology centers under construction: the "Mission Control" space center in Houston, the Marshall center in Huntsville, where one could listen to one of the many German-language radio stations while watching the enormous sheet of flame shooting from a massive Saturn V engine test ignition, and Cape Canaveral, where they were installing huge fans in the top of the new Vertical Assembly Building to prevent precipitation from occurring inside the building.
Of course the ominous presence of the military draft was another other attractive aspect of being a science student at that time. In fact, the friend who accompanied ("dragged" is probably a more accurate term) me up the Flatiron that day had just finished his training as a "Green Beret" in the Special Forces.
Ben Domenico climbing the Third Flatiron on June 21, 1965.
After getting my M.S. at Yale the next year, I learned that Ludvig Oster, the professor I wanted to work with, was taking a position at the University of Colorado. I needed little convincing to make the move from New Haven to Boulder. Professor Oster put me in touch with Andy Skumanich, who set up a student assistant position for me that summer at HAO to help with his computer models for radiative transfer in stellar atmospheres. I enjoyed the numerical modeling computer work and ended up doing my thesis work with Andy as my advisor. In fact, two-thirds of my thesis committee are in a 40th anniversary photo of the 1971 HAO radiative transfer group. The other two were Jeff Linsky of JILA and John Taylor, who became locally famous as the original "Mr. Wizard" at CU.
The 1970s, however, were not such a good decade for finishing one's education in a scientific field. The space program was in the doldrums and there was a glut of young scientists entering (or attempting to enter) the research and education field at a time when government and academia were retrenching. As luck would have it, though, the computing field was beginning to blossom and NCAR was again in the forefront. So, after a post-doc at the Laboratory for Atmospheric and Space Physics (LASP), I took a programming position offered by Jeanne Adams in what was then the Computing Facility of the Atmospheric Technology Division. My only frustration was that my office in that basement of the Mesa Lab didn't look much bigger up close than it did from the top of the Flatirons.”
Vic Tisone (HAO)
“Before I started working at NCAR in 1968, I didn't know what the building being built on the mesa was going to be. I was born and raised in Boulder by what is now North Boulder Park, so I had a clear view of the construction. I was raised Catholic and went to Scared Heart Church, which is still at the same location on Mapleton. At this time there was a convent close to the church for the nuns. I remember that on clear days, I could hear bells or chimes that seemed to come from the mesa and the new structure there. I thought that the new building was going to be a monastery for monks, and bells or chimes were a signal for prayer. I later found out that the sounds were coming from Saint Gertrude Academy for Girls, which was a Catholic girls’ school close to the mesa. When I went for my job interview for NCAR, I learned that the building on the mesa was NCAR and not a monastery as I had once thought.
I was hired by HAO in 1968 as a research assistant for two HAO scientists. One was Jagdish Gupta and other was Sydney Chapman, a world famous geophysicist for whom the Chapman Room is named, who was to be turning 80 soon. Dr. Chapman was a very formal British man with a rigorous disciplined routine everyday. At noon everyday he would go swimming at the University and when he came back, you could hear very definitive firm strides of a man on a mission and ready for business. He was in good shape for his age. At that time, projects hired sets of research assistants because a lot of the work and calculations had to be done by hand and by Monroe desk calculators. Instead of computers or laptops, there was a room in the old HAO building on campus that had shelves of these calculators, which could be checked out for various projects. The calculators were mechanical wonders. I remember that when we got bored sometimes, we would divide a number by zero on one of the calculators and watch it go crazy trying to do the calculations. We would have to pull the plug before it vibrated itself off the desk.”
Dave Fulker (Unidata)
“Especially in my field, 50 years of technology advancements immediately come to mind as profoundly affecting how we work. But I’ll first comment on UCAR’s evolution from a modest (14 university) consortium, focused narrowly on fiduciary oversight of NCAR, to an organization that more broadly supports and represents a major segment of the world's atmospheric-science community. UCAR seems markedly more outward looking and less narrow in focus, both of which bode well for future relevance.
Dave Fulker, founding director of Unidata, performs in a jazz quartet as part of Unidata’s 25th anniversary in 2009.
Here are a few technology-driven changes I’ve seen:
Finally, while string-quartet recitals in NCAR’s lobby became fewer with Walt Roberts’ passing, it has been my recent good fortune to play jazz on special occasions such as the celebration of HIAPER’s arrival!”
Rol Madden (NESL/CGD)
“My NCAR adventure began in September 1967. I had only been on the job one week when I saw Robert White, director of the Environmental Science Services Administration (ESSA, which became NOAA in 1970), walking down the hall. I asked someone if that was White and was told that indeed it was. I was flabbergasted that Robert White could be in the building with no fanfare. A few short years earlier my Air Force colleagues and I would have been polishing the floors and cleaning the windows if a visiting colonel was on the way. If it had been the director of ESSA, I think we might have painted the walls.
What a place to work! In the summer of 1968, I was assigned to be Jule Charney’s “man Friday” as he spent several weeks at NCAR. Later, we would be eating and hiking with Ed Lorenz. I met every important meteorologist of the second half of the 20th century at NCAR and some who went further back: Bernhard Haurwitz, Sverre Peterssen, Eric Palmein, Tor Bergeron.... Many of the important meteorologists were on the NCAR staff. Their doors were always open. Their ideas were always shared.”
Akira Kasashara (NESL/CGD)
“In the fall of 1961, I visited Boulder at the invitation of Phil Thompson to join NCAR. After meeting with Walter Roberts, Phil, and Aksel Wiin-Nielsen, I was very impressed by the future plan of NCAR as described in the famous “Blue Book.” I thought that I would like to accept Phil’s attractive offer to work on any topic I wanted.
However, one drawback was that there were practically no research facilities and only a handful of scientists were present in the office located in the old CU Armory building. During my time working with George W. Platzman on hurricane prediction at the University of Chicago, I had to travel to Washington, D.C., many times to run jobs on electronic computers at the U.S. Weather Bureau and National Bureau of Standards. So, I didn’t want to move to an organization where no electronic computer was available on site.
I talked with Phil frankly about my concern. Phil assured me that NCAR would have the best computers maintained on site for NCAR scientists as well as for the atmospheric sciences community, and suggested that I take a leave of absence at the Courant Institute of Mathematical Sciences, New York University until NCAR installed its own computer.
Akira Kasahara and Warren Washington relied on the CDC 6600 supercomputer to run the NCAR General Circulation Model that they were developing in the 1960s and '70s. The output from the GCM runs was stored in the IBM 9-track magnetic tapes shown in this 1971 picture with Akira.
One and a half years later, my family drove to Boulder from New York City. When I showed up at NCAR there were almost one hundred staff members. The following year NCAR created the Computing Facility with a Control Data Corporation 3600. Soon afterward, I was acquainted with a bright, fresh Ph.D. graduate, Warren Washington, who expressed an interest in building a general circulation model (GCM) of the atmosphere. I was stunned, but applauded his foresight and courage, and enthusiastically promised my cooperation as I had been thinking the same thing.
Of course, Phil was delighted to hear our proposal and gave us his hearty endorsement. So, we started to build a GCM from scratch, working with Dave Houghton, Takashi Sasamori, Dave Williamson, Dave Baumhefner, Harry van Loon, Bob Dickinson, and many dedicated programmers and able support staff. The timing could not have been better as the Computing Facility continuously upgraded its supercomputers. Incidentally, these earlier supercomputers were designed by Seymour Cray, who devoted himself to building machines to materialize John von Neumann’s dream of advancing the problem of weather forecasting.
Even today, I occasionally stop by the CRAY 1-A exhibit and the wall display of Cray’s achievements located in the first basement of the Mesa Lab, recalling my appreciation for his dedicated undertaking. I always thought, as shared by a noted GCM modeler Joe Smagorinsky, that it is important to have the computers co-located with the model builders and to collaborate closely with computing facility staff, but the next-generation supercomputers finally have grown to the point that it is physically impossible to maintain this. A case in point is the recent feat of the formation of the NCAR-Wyoming Supercomputing Center. Looking ahead to the next 50 years, I can hardly imagine how the world will change, but I firmly believe that the spirit of UCAR/NCAR lives on.”
Charlies Knight (NESL/MMM)
“HAO was already well established in 1962, but the atmospheric part of NCAR was just starting when I was offered a job and Nancy and I came to Boulder. NCAR was situated in the old CU Armory on University Avenue. We had come to visit Jim Deardorff, and had talked with Jim Lodge, whose atmospheric chemistry program was gearing up. I applied for a job, the nature of which was and remained quite vague, so I had a lot of freedom.
Basic science was the watchword in the years right after Sputnik, and NCAR's dual purpose, as I understood it then, was supplying facilities for the university community and doing basic science, for its own sake and partly to keep the facilities "up to snuff." There was a perception then that atmospheric science in general needed a bigger component of interdisciplinary approaches, and a number of the early hires were outside the mainstream of meteorology.
We moved to 30th Street after about a year, and then to the Mesa Lab in about 1967. Discussions about successive designs of the Mesa Lab had been spirited. One of the objectives of the design then, which would be unheard of now, was to afford scientists the seclusion that was deemed necessary for individual, creative research.
Charlie Knight studying hail under a microscope in 1971.
Funding was not a major problem for several years. The physical services NCAR offered for research became magnificent rather quickly. For a number of years NCAR was small and centralized enough that everyone pretty much knew everyone else. Major, gradual changes have accompanied NCAR's growth and decentralization. Most significant for me has been the decline of research support, with nearly all support except computing now requiring outside funding. For instance, equipment and technical services (machine shop, glass-blowing, photography, and programming support) of the kind that used to be available, are, to a large extent, no longer supported.
One of the major events for us was the National Hail Research Experiment, roughly 1969–1975, which was controversial from start to finish. The first major controversy was whether NCAR should undertake it at all, and the motivation to do so was largely top-down. There was a lot of resistance to such an applied, weather modification project, but NCAR undertook the project’s management and was a major participant as well. To NCAR's credit, the experiment did produce worthwhile research results along with an evaluation of the hail suppression technique that had gotten such great acclaim in the Soviet Union. The experiment closed with one field season of research without cloud seeding after the decision to stop the actual test of hail suppression, two years short of its projected five.
It is interesting to me to reflect that the basic problems of cloud physics that were recognized and much discussed at the time we came to NCAR—the interconnected problems of getting a satisfactory, physical understanding of precipitation formation in clouds and of entrainment in cumulus clouds—are still basic problems today, in spite of considerable progress over the last 50 years. The biggest change in emphasis in this area is a sharp decline in experimental approaches, a more gradual decline in observational efforts, and a great increase in numerical modeling activity.”
Fred Clare (CISL)
“I have been in computing since first coming to NCAR in 1969. Initially, I was in the Computing Facility (CF) of the Atmospheric Technology Division (ATD), which became the Scientific Computing Division (SCD) in 1980. In 2004, SCD was integrated into the Computational and Information Systems Laboratory (CISL).
Of course the most dramatic change I have seen as a computer programmer is the mind-boggling increase in computing power. My personal Macintosh laptop today has about a thousand times the computing power of the CDC 6600 "supercomputer" I worked on when I first came to NCAR, and its memory is about two thousand times as large. This is not to mention its absurdly high internal disk space.
Things were a lot simpler in the early days. There was only one computer (the 6600), essentially one programming language (Fortran), and one graphic output device, a Data Display dd-80. It was a time when a programmer had a chance to understand the entire environment, and if something went wrong, there was every chance you could track it down. Now, when using one of the million-line applications that you got from some external source, it is almost hopeless to track down a bug yourself. Dealing with the increase in complexity of having many different computers, languages, compilers, graphic output devices, text editors, browsers, software development tools, and so forth, is a challenge.
The greatest change in the way that programmers work came with the advent of personal computing in the 1980s. It seems that every advance comes with a loss. In the early days, programmers submitted their programs for execution by way of punched cards. There was an area outside of the computer room where programmers congregated while waiting for their jobs to run. During the course of a couple of weeks, you would interact with almost every programmer in the division. This served as a way for programmers to exchange technical information, but perhaps more importantly it was a way for people to get to know each other on a personal basis. This created a sense of community that has been greatly diminished in an era when even those in neighboring offices communicate via e-mail."
Andrew Skumanich (HAO)
"THEN: In 1960, HAO started on an adventure in the fields of solar magnetism. Harold Zirin, during a sabbatical at the Crimean Astrophysical Observatory (CAO), discovered the utility of CAO's magnetograph in measuring line-of-sight (LOS) magnetic fields of solar prominences. Such structures were believed to be involved in interplanetary-terrestrial weather. Thus by 1964, an HAO-designed magnetograph was fielded at the HAO Climax Observatory.
The observations of prominence fields led to several papers and a CU Ph.D. thesis. With improvements in technology, a more versatile magnetograph was fielded in 1968 under the direction of Einar Tandberg-Hanssen. More publications and another CU Ph.D. thesis showed the success of the endeavor.
A more ambitious effort was initiated in 1969 by Jacques Beckers to measure the transverse as well as the LOS field. This completely new approach was brought to fruition by Lew House with the 1974 field testing of the STOKES-I polarimeter, named after G. Stokes, who showed that a complete description of polarized light required three parameters to describe the persistent polarization ellipse and a fourth to describe the random polarization of the light beam. Magnetic fields impose their signatures in the persistent component. Fairly involved matrix equations must be inverted to derive the vector field from these signatures. The Stokes I concept was a major success and was sited at the Sacramento Peak Observatory (SPO), where it became a national/international facility. A number of papers as well as a CU Ph.D. thesis resulted. But bandwidth limitations proved to be unacceptable and, with new technology, an updated Stokes II came on line in 1979.
In addition to the instrument development, an interpretive development was required to invert the matrix equations. This was provided concurrently by Lew House, Peter L. Auer, and James Heasley, but proved inadequate for Stokes II data. An ultimate breakthrough inversion method by Andrew Skumanich and Bruce Lites (1985) allowed Stokes II data to become useful, and with significantly improved technology, a new Advanced Stokes Polarimeter (ASP) was commissioned in 1992 at SPO. This has led to a very successful increase in our understanding of solar magnetic fields.
NOW: In 2010, a spacecraft version of the ASP has been successfully implemented by Lites on the Japanese Hinode mission and is the source of the highest resolution magnetic field observations so far. A major success story! Not cited above are the invaluable and decisive efforts of the HAO instrument staff and support scientists."