Charney, Jule Gregory
CHARNEY, JULE GREGORY
(b. San Francisco, California, 1 January 1917;
d. Boston, Massachusetts, 16 June 1981), dynamic meteorology, numerical weather prediction
Charney was a leader of the Meteorology Project at the Institute for Advanced Study, Princeton, New Jersey, one of several centers focused on the development of numerical weather prediction in the mid-twentieth century. Numerical weather prediction, whereby future atmospheric conditions are forecast using computer-solvable models, was one of the most important scientific accomplishments in twentieth-century meteorology. Charney later played a critical role in international meteorological cooperation, spearheading the Global Atmospheric Research Program in the 1960s and 1970s. In other meteorological work, he made fundamental contributions through examinations of drought and the development of persistent high pressure areas known as blocks. Charney’s theoretical work extended to oceanography; his work on ocean currents was particularly important.
Early Years . Charney was born in San Francisco to Stella and Ely Charney—Yiddish-speaking, Russian Jewish émigrés who worked as laborers in the garment industry. As secular Jews, religion played no role in their family life. Intellectual pursuits, however, were highly prized, and the Charney home was full of lively discussions and music that would become integral parts of Charney’s life.
When Charney was five years old, his family settled in the Los Angeles area where he attended public schools. He excelled in mathematics, showing a tremendous ability to reduce complex problems to a number of smaller, simpler ones for which he would propose solutions—an ability critical for his later success in solving atmospheric dynamics problems. By the time Charney entered the University of California, Los Angeles (UCLA) in 1934, he had taught himself two to three years of university mathematics. Unfortunately, the teaching-intensive mathematics department had no mechanism for offering accelerated course work. Charney did exceedingly well in his studies with little effort, less faculty encouragement, and no direction. After completing his BS in mathematics in 1938, he remained for graduate studies and completed his MS in mathematics in 1940 with a thesis on curved spaces.
Meteorology Beckons . During this time, Charney made his first contact with meteorology when he attended a seminar given by the Norwegian Jörgen Holmboe, a former member of the Bergen School of Meteorology—the twentieth century’s premier meteorological research school. Charney was impressed. He knew nothing about meteorology, but it appeared to be a science worth pursuing. He also realized that atmospheric motions were subject to the laws of physics and represented by partial differential equations—a delightful pairing of the applied mathematics and physics theory beloved by Charney.
The first in a series of events that would profoundly affect Jule Charney’s professional life occurred in 1940. Norwegian Jacob Bjerknes, a Bergen School meteorologist of international reputation who had become exiled in the United States after Germany invaded Norway, established a meteorology program under the auspices of the UCLA Physics Department and brought Holmboe and Morris Neiburger in as faculty members. Their immediate purpose was to provide accelerated meteorology training to young military officers for the war effort. The United States needed several thousand weather forecasters and had only four hundred before the war started.
By 1941 the military draft offered Charney a choice: enlist in a military service or find a defense-related job. Holmboe presented Charney with the option of joining the new meteorology program as a teaching assistant (exempting him from military service) and taking graduate meteorology courses at the same time. Unsure about making a commitment, Charney consulted California Institute of Technology physicist Theodor von Kárm án and asked if he should shift to meteorology or take a position in the aeronautics industry. Von Kármán recommended meteorology—he felt aeronautics had veered too far into engineering. The combination of the draft, Holm-boe’s invitation, and von Kármán’s recommendation nudged Charney into meteorology, a switch he made that July.
Graduate School . Charney’s initial heavy teaching load allowed little time for research, yet the curriculum must have grounded him in the fundamentals of atmospheric dynamics. Once the demand for new meteorologists cooled in 1944, Charney returned to a more regular graduate student schedule.
He also identified his research topic: the problems of genesis and development of extratropical cyclones, that is, low-pressure circulations in the middle latitudes, an extremely important area of meteorological research. Charney was convinced that he could develop a basic model that would produce cyclones if he could carry out some very complicated mathematics. In his model, the westerlies (air moving from west to east in the zone between 30° and 60° north latitude) would be in near geostrophic balance (the horizontal pressure gradient balancing the Coriolis force) and show an increase in speed with height until they reached a maximum at the tropopause (approximately 10 kilometers above Earth’s surface). Charney’s chosen topic was extremely difficult, but it was perfect for his abilities and psychological need for a challenge. He was convinced that rewards in absence of hard work were meaningless—the greater the challenge, the happier he was.
Charney attacked his dissertation research, concentrating on the dynamics of the lower 10 kilometers of the atmosphere. To simplify the problem, he assumed an approximate balance between the pressure gradient and Coriolis forces (quasi-geostrophic approximation).
This assumption reduced the problem to one that could be solved by a second-order differential equation. However, no tabulated solution existed for this equation, so Charney had to solve it on a mechanical calculator using finite difference methods. In 1946, Charney finally found the solution and completed his dissertation.
He also married his first wife, Elinor Kesting Frye. Their twenty-year marriage produced two children— Nora and Peter—and Elinor’s son Nick from her first marriage took Charney’s name. Charney would marry again in 1967. His nine-year marriage to painter and color theorist Lois Swirnoff ended in divorce.
As evidence of the outstanding quality of his work, the Journal of Meteorology published his dissertation as a single article (“The Dynamics of Long Waves in a Baro-clinic Westerly Current”) entirely filling one of its 1947 issues. Meteorologists immediately recognized its importance; the model and results made meteorological sense even if many of them could not follow the mathematics he had used in the solution.
En Route to Norway . His PhD complete, Charney wanted to further his studies with the leading meteorological theorists of the time, the Norwegians. Awarded a prestigious National Research Fellowship, he left Los Angeles to study with Halvor Solberg at the University of Oslo. While en route, Charney stopped at the University of Chicago to visit with Swedish American meteorologist Carl-Gustaf Rossby—one of the most eminent meteorologists in the United States—who was the discoverer of large-scale waves in the atmosphere.
Rossby’s dynamic view of meteorology meshed nicely with Charney's, making it a much better fit than UCLA’s more geometrical approach. While chatting with Rossby, Charney found out that Solberg would not be in Norway when he arrived because he was due shortly in Chicago. Therefore, Rossby encouraged Charney to postpone his fellowship and stay with Chicago’s Meteorology Department for a few months. Charney accepted, and later considered the time with Rossby to be the defining experience of his professional life. Not only did Rossby provide tremendous intellectual stimulation, but he was available at almost any time for discussions ranging over a wide variety of issues. For the first time in his life, Charney felt he had discovered true intellectual rapport with another person. Their close working relationship continued until Rossby’s death in 1957.
After a year at Chicago, Charney continued on to the University of Oslo. Working with meteorologists Arnt Eliassen and Ragnar Fjørtoft, he began seeing a way to reduce the number of equations required to describe the atmosphere. Charney’s primary focus was to eliminate all equations that were not meteorologically important and to find a numerical solution for those remaining.
This was not a new idea. Vilhelm Bjerknes and his Bergen School had been working on a rational physics-and mathematics-based meteorology since before World War I. Lewis Fry Richardson had also picked up the challenge, publishing the result of his flawed attempt to solve the hydrodynamic equations defining atmospheric motion by iterative methods in his 1922 book, Weather Prediction by Numerical Process. Now Charney wanted to put forecasting on a firmer mathematical basis.
By assuming a geostrophic and hydrostatic atmosphere, Charney simplified the problem down to a single equation involving only air pressure, from which sound and gravity wave solutions were excluded. His work would be particularly important several years later because a special case of the quasi-geostrophic approximation model—the equivalent barotropic model—was used in the first numerical forecast in 1950. Charney’s 1948 article “On the Scale of Atmospheric Motions” set the stage for advancements in numerical weather prediction.
Numerical Weather Prediction . With Charney’s year in Norway almost up, he received an invitation from Hungarian-born mathematician John von Neumann to join the Meteorology Group at the Institute for Advanced Study in Princeton. Von Neumann had met Charney in August 1946 during a meeting to which Rossby had been invited. Rossby had insisted that Charney come along. At the meeting, von Neumann tried to convince meteorologists that the atmospheric “problem” was ideal for solving on his new electronic digital computer. Charney had been intrigued by von Neumann’s proposal, and it had driven part of his work in Norway. Hearing of Charney’s research, von Neumann invited him to join the Princeton team.
Charney took over the rudderless Meteorology Project in spring 1948, immediately offering the project a better sense of direction. He insisted that the team start with a very well-defined, simple model that could be assembled with workable algorithms that would operate in small steps. Once the simplified model worked, they would add more factors one at a time until the model became more realistic and complex. Charney had already prepared his quasi-geostrophic prediction equations, but they needed to be adapted to the computer. The result was a 500 millibar (approximately 5,500 meters above sea level) barotropic model, that is, a single-parameter, single-level model based solely on the horizontal motion of the initial circulation field.
While the model was ready, von Neumann’s computer was not, so project members gave their model a trial run on Aberdeen (Maryland) Proving Ground’s ENIAC computer in April 1950. The results of this first “expedition” produced forecast maps that looked meteorological, but needed refinements. Over the next four years, project members continued to improve their atmospheric models, which were finally run operationally in May 1955 at the Joint Numerical Weather Prediction Unit, a combined U.S. Weather Bureau–Navy–Air Force organization that was an outgrowth of the Meteorology Project.
The creation of numerical weather prediction was a tremendous achievement whose success was due to Charney’s attention to the details of hydrodynamic factors in the atmosphere and von Neumann’s computational techniques. Charney’s quasi-geostrophic model had been critical to getting the project started and he had been absolutely determined to see the model progress from research project to daily operations. He also wanted to ensure that the theoretical models were not only internally consistent mathematically, but represented observational evidence.
Charney was forced to make assumptions during the development of his model. He first removed all vertical motion in the atmosphere from consideration and then reduced movement to one dimension. He added in friction and topography while trying to make the model more “realistic,” but found those factors had little effect for twenty-four-hour forecasts and removed them. The numerical weather prediction models run by the Meteorology Project proved it was possible to predict cyclogenesis (the birth and development of a low-pressure circulation system), which is the precursor to the formation of weather-producing fronts. The models also proved that a geostrophic system could successfully create an elementary simulation of the atmosphere’s general circulation pattern. Charney also showed that model complexity was constrained by computing power. As computing power increased, the models included more variables and became more sophisticated.
The results of Charney’s research were widely felt throughout the meteorology community. Although he had had to make simplifying assumptions, Charney had not done so in the absence of supporting theory. His assumptions enabled dynamic meteorologists to understand the atmosphere’s very complex physical processes and analyze the motion of large-scale systems. Therefore, Charney’s assumptions not only reduced the number of years required to produce operational computer-created forecasts, but their basis in systematic quantitative arguments was equally important for meteorological theory.
The MIT Years . The work of the Meteorology Project completed, Charney moved on to the Massachusetts Institute of Technology (MIT) faculty in 1956. While at MIT, he continued his research on problems in dynamic meteorology including the vertical propagation of planetary waves (very long waves of energy that encircle the globe), the stability of jet streams, the generation of hurricanes, geostrophic turbulence, and climate dynamics in desert areas.
In addition to his theoretical research, Charney was also active in organizational work. He was a leading member of the Committee on Atmospheric Sciences at the National Academy of Sciences and was involved in the decision to form the National Center for Atmospheric Research in Boulder, Colorado, which became one of the premier sites for meteorological research in the United States in the early 1960s.
In 1968, Charney accepted the chair of the committee directing the Global Atmospheric Research Program (GARP)—a huge international undertaking that proposed to observe “the entire atmosphere of the Earth and the sea surface... in detail for the first time.” Charney had been mulling over the possibility of such a project since the early 1960s. He had realized that meteorology was a global science, but it was decidedly lacking in the global observations that would allow atmospheric scientists to cooperate across national boundaries. Even a short-term observational experiment would give meteorologists a tremendous data set with which to work.
Although several smaller experiments took place during the early 1970s, notably the GARP Atlantic Tropical Experiment (or GATE), the main yearlong event—the Global Weather Experiment—started in December 1978. This five-hundred-million-dollar project involved four polar orbiting and five geosynchronous satellites, more than three hundred ocean buoys, and more than three hundred constant-level balloons that would drift on air currents approximately 14 kilometers above Earth’s surface while measuring temperature and upper level winds. The experiment was a huge success. These data—so critical for meteorological research—were almost immediately used to improve operational weather forecasting.
With his GARP responsibilities behind him in 1978, Charney turned to what would be his last major research project, which examined the creation of blocks (masses of high-pressure air that prohibit the normal flow of atmospheric systems). Although blocking highs had been readily identifiable features for many years, no accepted theory for their life cycle existed until Charney began to examine them in the context of a seminar he was teaching at UCLA. Charney and one of the students, John DeVore, developed and published a dynamical theory of blocking in 1979. As had been the case throughout Charney’s life, this scientific study encouraged further investigations, particularly among the ranks of fluid dynamicists.
The supervisor of many graduate students during his MIT years, Charney left behind many academic progeny. They continued the various strands of his research agenda after his death at age sixty-four from lung cancer.
WORKS BY CHARNEY
“The Dynamics of Long Waves in a Baroclinic Westerly Current.” Journal of Meteorology 4 (1947): 135–162.
“On the Scale of Atmospheric Motions.” Geofysiske Publikasjoner 17, no. 2 (1948): 1–17.
“On a Physical Basis for Numerical Predictions of Large-Scale Motions in the Atmosphere.” Journal of Meteorology 6 (1949): 371–385.
With Ragnar Fjørtoft and John von Neumann. “Numerical Integration of the Barotropic Vorticity Equation.” Tellus 2 (1950): 248–257.
With Robert G. Fleagle, Vincent E. Lally, Herbert Riehl, et al. “The Feasibility of a Global Observation and Analysis Experiment.” Bulletin of the American Meteorological Society 47 (1966): 200–220.
With John G. DeVore. “Multiple Flow Equilibria in the Atmosphere and Blocking.” Journal of the Atmospheric Sciences 36 (1979): 1205–1216.
Harper, Kristine C. “Research from the Boundary Layer: Civilian Leadership, Military Funding and the Development of Numerical Weather Prediction (1946–1955).” Social Studies of Science 33 (October 2003): 667–696.
Lindzen, Richard S., Edward N. Lorenz, and George W.Platzman, eds. The Atmosphere—A Challenge: The Science of Jule Gregory Charney. Boston: American Meteorological Society, 1990. This volume contains Charney’s curriculum vitae, an oral history interview conducted approximately one year before his death, articles about his work and life, and reprints of several of Charney’s most significant articles.
Nebeker, Frederik. Calculating the Weather: Meteorology in the20th Century. San Diego, CA: Academic Press, 1995. See especially chapters 10 and 11.
Phillips, Norman A. “Jule Charney’s Influence on Meteorology.” Bulletin of the American Meteorological Society 63 (1982): 492–497.
——. “Jule Gregory Charney, January 1, 1917–June 16,1981.” Biographical Memoirs, vol. 66. Washington, DC: National Academy of Sciences, 1995.
Richardson, Lewis Fry. Weather Prediction by Numerical Process.Cambridge, U.K.: Cambridge University Press, 1922. Reprinted by Dover Publications, New York, 1965, and by Cambridge University Press, 2006.
Kristine C. Harper