Smagorinsky, Joseph

views updated


(b. New York, New York, 29 January 1924; d. Skillman, New Jersey, 21 September 2005), dynamic meteorology, numerical modeling.

Smagorinsky was the founder and director of the Geophysical Fluid Dynamics Laboratory, Princeton, New Jersey—an outgrowth of work on theoretical and operational numerical weather prediction in the late 1940s and early 1950s. Under his leadership the laboratory considered problems related to the general circulation of the atmosphere, pioneering collaborative research and expanding its reach to include numerical modeling of the oceans and a variety of atmospheric phenomena.

Early Years Smagorinsky was born in New York City to Byelorussian immigrants Nathan and Dinah (Azaroff) Smagorinsky, the youngest of five sons. Intrigued by meteorology at an early age, during the thirties he would visit the top of the New York Daily News building to look at the weather instruments and see what was happening at their weather observatory as forecasters prepared the latest information for the newspaper. Extremely interested and talented in science and mathematics, Smagorinsky took and passed the entrance examination for one of New York’s three science high schools, Stuyvesant High. As a student there, he received an enriched education in mathematics and science, and further indulged his interest in meteorology.

However, his true passion was naval architecture. Smagorinsky designed and built model boats that he would race on the lake in Central Park. When he was unable to pursue naval architecture in college due to financial considerations, he turned to his second love: meteorology.

Meteorological Education Smagorinsky entered New York University (NYU) in 1941 as a student in its recently formed Meteorology Department. Within a few months, the United States entered World War II, and by 1943 he found himself in the Army Air Corps.

The United States had been woefully short on meteorologists at the beginning of the war. The military services needed thousands of forecasters to support aviation, land, and sea missions, but there were only four hundred professional meteorologists in the country. A training program, hastily assembled by Swedish American meteorologist Carl-Gustaf Rossby of the University of Chicago, had started training young college graduates with degrees in mathematics and physics to be meteorologists, and then turned to students who did not have degrees in hand. Those handpicked students were sent to the “B” course, offered at universities around the country, and then sent on to the final “A” course at one of the “Big Five” meteorology programs (University of Chicago; New York University; Massachusetts Institute of Technology (MIT); University of California, Los Angeles; or California Institute of Technology).

Smagorinsky, selected to attend the “B” course at Brown University, was one of the very few of these young men and women who had ever entertained the idea of being a meteorologist before the war. Successfully completing the course at Brown, he finished up the “A” course at MIT and was commissioned in 1944 the day before the D-day invasion. As an Army Air Corps weather officer, Smagorinsky provided flight forecasts for B-29 squadrons in Nebraska, and then concluded his military service as a weather reconnaissance officer in the North Atlantic.

Returning to NYU at the end of the war, Smagorinsky finished his BS in 1947, and his MS in 1948. While finishing his undergraduate degree, he met U.S. Weather Bureau statistician Margaret Knoepfel. One of the few women who had ever worked for the Weather Bureau, Knoepfel had been sent to NYU for graduate meteorology training. She had to return to headquarters in Washington, D.C., in 1947. A year later Smagorinsky moved south to join her and they were married. (The Smagorinskys would later have five children.) He had no intention of working for the Weather Bureau, but his other job options did not work out, and he soon found himself working in the scientific services division—the bureau’s research arm.

Numerical Weather Prediction When he first entered meteorology, Smagorinsky was under the mistaken impression that it was a rational, scientific discipline whose forecasts were based on numerical calculations and the laws of physics. Inquiring as to why that was not the case, he was told by one of his professors that it had been suggested by Bergen School of Meteorology founder Vil-helm Bjerknes in 1904, and unsuccessfully tried during World War I by British meteorologist Lewis Fry Richardson. Calculating the weather was still a possibility; it just had not happened yet.

While at NYU, Smagorinsky heard dynamic meteorologist Jule Charney speak on the scale properties of the equations of motion—the early work that would lead to numerical weather prediction. Smagorinsky was fascinated by the possibilities of this rational approach to weather forecasting. He had the opportunity to hear Charney again when the latter spoke at the Weather Bureau in Washington, D.C. On that occasion Smagorinsky had several questions ready for Charney, who responded by inviting him to join the Meteorology Project at the Institute for Advanced Study in Princeton, New Jersey, where his team was developing numerical weather prediction models. Gaining a release from the Weather Bureau, Smagorinsky moved to Princeton to help with the one-dimensional linear barotropic model calculations while he simultaneously worked on his PhD at NYU. Margaret Smagorinsky joined the project also, working to program John von Neumann’s new computer that would solve the atmospheric models being prepared by Charney’s team.

Having completed his PhD in 1953, Smagorinsky was unsure of his next move. That was settled for him when Weather Bureau chief Francis W. Reichelderfer asked him to return to Washington, D.C., to head up the new numerical weather prediction unit. Reichelderfer, anticipating that numerical weather prediction would become an operational reality within a short period of time, realized the bureau needed to train people and procure equipment. The job went to Smagorinsky, who was given plenty of guidance, and a long list of things to do, by Charney. The Joint Numerical Weather Prediction Unit, an effort funded and staffed by the Weather Bureau, navy, and air force, opened its doors in 1954 and issued its first operational forecast in 1955. Smagorinsky had been responsible for evaluating and obtaining the unit’s computer, determining efficient ways to handle data, and ensuring that Weather Bureau employees were trained in numerical techniques. Once the unit was operational, he turned his attention from operational modeling (prediction) to theoretical modeling.

General Circulation Research The new joint unit was concerned with operational numerical weather prediction—the issuance of short-term forecasts. Von Neumann and Charney encouraged Reichelderfer to establish a separate group that would continue research on the general circulation of the atmosphere, work that had been started by meteorologist Norman A. Phillips. With funding from the navy and air force, Reichelderfer agreed and the Weather Bureau established the General Circulation Research Section in 1955 to conduct basic research on the general circulation of the atmosphere and on climate. Smagorinsky became the director in October 1955, and through three name changes for the section remained its director until his retirement in 1983.

At first, the small group (five people at the end of 1955) worked on general atmospheric circulation (i.e., the overall global movement of air), a complex problem for which there would be no quick solution. Smagorinsky realized, however, that if they were to extend their atmospheric circulation modeling to climate modeling they would need to incorporate ocean modeling, which they did in 1960. Until that time, the ocean had been held static. Indeed, early models did not include surface interactions because they were too difficult to model. Smagorinsky also realized the importance of modeling individual meteorological processes and taking the lessons learned from those virtual experiments and applying them to weather forecasting and climate models. His work with G. O. Collins in 1955 on numerically predicting precipitation was a forerunner of all later efforts on specific processes. Smagorinsky pioneered interdisciplinary interactions in meteorological modeling, and his team gradually included radiation, condensation, boundary layer processes, and ocean surface interactions into their increasingly complicated models.

By 1963, Smagorinsky’s group had changed its name along with its mission, becoming known as the Geophysical Fluid Dynamics Laboratory (GFDL). Five years later the laboratory moved from Washington, D.C., to Princeton, New Jersey, to become associated with, but remain independent of, Princeton University. Princeton had a strong interest in fluid dynamics and the arrival of the laboratory provided them with a ready-made graduate program in the subject.

However, the laboratory was not tied to a specific department. It had formal ties to the civil engineering, aerospace and mechanical sciences, and geology and geophysical sciences departments, and unofficial ties to astrophysics, chemical engineering, and statistics. These ties were important to Smagorinsky, who looked upon the modeling work being done by the laboratory as being at the intersection of a variety of disciplines. By the mid-1970s, however, the laboratory became administratively part of the Geology and Geophysical Sciences Department even though it maintained a separate identity as a national weather service-funded activity.

The laboratory tested circulation models by using real initial condition data and then considering it as a forecast run, that is, as if the model were predicting the weather instead of providing insight into the atmosphere’s behavior, to check the results. In 1963, Smagorinsky published his primitive equation model of the general circulation. The so-called primitive equations of the atmosphere— equations of motions, continuity equation, equation of state, and the first law of thermodynamics—had been proposed as the way to calculate the weather by Norwegian Vilhelm Bjerknes in 1904, and had been taken up by Lewis Fry Richardson in his failed World War I–era attempt to produce a numerical forecast. However, the early numerical weather prediction modeling led by Jule Charney in the late forties and early fifties had opted for simpler models that would run on extant computers. By the 1960s, the primitive equations became viable modeling options due to advances in computing power. Smagorinsky’s model was a significant breakthrough in model sophistication. Over time, the GFDL team increased the model run times, and turned its attention to modeling wind- and heat-driven ocean circulation, convection, hurricanes, mesoscale features, and climate. The weather and climate models developed by the laboratory since its inception have significantly increased meteorologists’ understanding of atmospheric circulation, including heat and moisture transport, and human impact on Earth’s changing climate.

Observational Experiments The GFDL consistently needed large amounts of data to feed into its models, so Smagorinsky was always on the lookout for additional observation sources. He was extremely enthusiastic when Charney—mulling over ways to increase international cooperation for this most global of sciences—proposed a major observational data-gathering experiment. Later known as the Global Atmospheric Research Program (GARP), Charney’s grand scheme for pulling together massive amounts of observational data from every conceivable direct and remote sensing platform got started in the late 1960s. Smagorinsky was asked by the National Academy of Sciences to be a member of the U.S. National Committee for GARP, and later served from 1967 to 1975 as one of twelve members of the Joint GARP Organizing Committee assembled under the auspices of the World Meteorological Organization and the International Council of Scientific Unions. He was chair of the organizing committee from 1976 to 1979, playing an important role in coordinating GARP’s main event: the five-hundred-million-dollar, year long Global Weather Experiment of 1978 that involved scientists from 147 nations gathering data from satellites, buoys, ships, and constant-level balloons.

Awards Smagorinsky received many awards during his distinguished career. He was elected as an American Meteorological Society (AMS) fellow in 1967 and selected as an honorary member—the highest distinction bestowed by the society—in 1992. Smagorinsky received several other awards from the AMS, including the Clarence Leroy Meisinger Award given for distinguished work by a scientists under age forty when nominated (1967), the Carl-Gustaf Rossby Research Medal for outstanding research leading to understanding of atmospheric behavior or structure (1972), the Cleveland Abbe Award for distinguished service to the atmospheric sciences by an individual (1980), and the Charles Franklin Brooks Award for outstanding services to the society (1991).

Smagorinsky was also recognized internationally. He received the Royal Netherlands Academy of Sciences’ Buys Ballot Medal (1973), the World Meteorological Organization’s IMO Prize—the most important international meteorological award—in 1974, and the Royal Meteorological Society’s Symons Memorial Medal in 1981.

In 2003, Smagorinsky and longtime colleague and general circulation model creator Norman A. Phillips received the prestigious Benjamin Franklin Award in Earth Science given by the Franklin Institute of Philadelphia, for their work in computer modeling and Smagorinsky’s role in establishing the global observational network.

Retirement Smagorinsky retired from federal service in 1983 after leading the GFDL for twenty-eight years. Having survived a heart attack some ten years before, he had decided that it was time to leave the stress of running the laboratory to someone else. Although he had told the World Meteorological Organization’s Hessam Taba that he had no specific plans, he did anticipate spending a considerable amount of time of on non-meteorology-related projects. However, he remained professionally active into the 1990s.

In 1986 he served as president of the American Meteorological Society, and remained on the executive committee until 1988. He also was chair of the National Research Council’s Climate Research Committee from 1981 until 1987, and was an officer on the National Research Council’s Board on Atmospheric Sciences and Climate. Smagorinsky continued his academic affiliations as a visiting senior fellow with Princeton University’s Department of Geological and Geophysical Sciences, and served as the Brittingham Visiting Professor at the University of Wisconsin in 1986. Although Smagorinsky had spent almost his entire career with GFDL, through his ties with Princeton he came to influence many graduate students in meteorology, encouraging them to pursue a broad range of interests within the field instead of becoming too specialized.

Smagorinsky died in 2005 after a long struggle with Parkinson’s disease.



“The Dynamical Influence of Large-Scale Heat Sources and Sinks on the Quasi-Stationary Mean Motions of the Atmosphere.” Quarterly Journal of the Royal Meteorological Society 79 (1953): 342–366.

With G. O. Collins. “On the Numerical Prediction of Precipitation.” Monthly Weather Review 83 (1955): 53–68.

“General Circulation Experiments with Primitive Equations: I. The Basic Experiment.” Monthly Weather Review 91 (1963): 99–164.

Interview by Richard R. Mertz. Tape recording. 19 May 1971. National Museum of American History, Smithsonian Institution, Washington, DC.

“GARP—An Interim Retrospection.” WMO Bulletin 29 (1980): 92–100.

Interview by John Young. Tape recording. 16 May 1986 and 21 May 1986. National Center for Atmospheric Research, Boulder, Colorado.


Smagorinsky, Margaret. Interview by Kristine C. Harper, Ronald E. Doel, and Terry Smagorinsky Thompson. Tape recording. 2 January 2006. National Center for Atmospheric Research, Boulder, Colorado.

Taba, Hessam. “Professor J. Smagorinsky.” In The “Bulletin” Interviews: Reprints of the 32 Interviews with Eminent Personalities in Meteorology and Hydrology That Appeared in the WMO Bulletin between 1981 and 1988. Geneva: World Meteorological Organization, 1988.

Kristine C. Harper