Weather by the Numbers: The Genesis of Modern Meteorology

by Kristine C. Harper
The MIT Press, Cambridge, Massachusetts, 2008
308 pp., illus. 20 b/w. Trade, $40.00
ISBN: 978-0-262-08378-2.

Reviewed by Jennifer Ferng
Department of Architecture
Massachusetts Institute of Technology

jferng@mit.edu

Farmers sometimes relied upon the intuitive signs of wind direction, the natural instinct of animals, imminent storm clouds, and heavy rainfall to predict how the weather would influence the prosperity of their forthcoming harvest. The U.S. Weather Bureau during the early 1920s, partially comprised of several thousand unpaid volunteers who collected and transmitted local observations for regional forecasts that affected crop and road services, was, at the time, an ineffective government bureaucracy that envisaged how extreme heat temperatures would impact the transport of livestock and published empirical data on fruit frost for tobacco and alfalfa seed districts. Placing the Weather Bureau as one of the many strategic organizations at the heart of meteorology’s expansion into a theoretically sophisticated science, Kristine Harper traces the tribulations and successes of American and European scientists from World War II into the 1960s who introduced numerical computing techniques to the art of forecasting the weather as part of the Meteorology Project based at the Institute for Advanced Study in Princeton, New Jersey. She shifts the spotlight away from the better-known contributions of mathematical prodigy John von Neumann and, instead, accentuates the daily decisions made by American meteorologists collaborating with external Norwegian experts who possessed more theoretical training and time-tested experience. Harper challenges some historians’ interpretations of von Neumann’s central role in developing numerical weather prediction, represented by such works as Frederik Nebeker’s Calculating the Weather: Meteorology in the 20th Century (1995) and William Asprey’s John von Neumann and the Origins of Modern Computing (1990). Computational modeling during World War II – though widely prevalent in chemistry and physics – became defined in meteorology, in Harper’s view, through the pursuit of a mathematics-based theory of general circulation in the earth’s atmosphere. By pursuing the “control of nature,” American scientists attempted to master a new technology that would demonstrate a different style of scientific practice from that of their Norwegian counterparts.

Harper’s claim stresses that von Neumann’s development of the “electronic brain,” which would eventually take only 90 minutes to make a 24 hour forecast, was influenced by a cadre of mathematicians and physicists who were responsible for the majority of the tedious advancements as part of an international effort that transformed the computer into a feasible instrument for weather prediction. This book is shaped as an institutional history since Harper closely focuses on the interactions of the “military industrial academic complex” that developed between the Air Force, the Navy, the Weather Bureau, and the Big Five schools which were among the first that instigated academic programs in meteorology, including Caltech, MIT, NYU, University of California at Los Angeles, and the University of Chicago. This type of approach in studies of the sciences during World War II has also been predominant among scholars such as Paul Edwards, who defines the relationship between computer models, data, and reality in his seminal The Closed World: Computers and the Politics of Discourse in Cold War America (1997); Stuart Leslie, who writes about the research and development contracts acquired by MIT and Stanford in The Cold War and American Science (1994); and Jessica Wang, who analyzes scientists’ reactions to Cold War anticommunism in American Science in an Age of Anxiety (1998), to name a few.

In chapters five through seven, the reader is quickly immersed into the staffing shortages, funding discrepancies, and ideological conflicts that surface between the strong personalities of von Neumann, Jule Charney, Philip Thompson, and Carl-Gustav Rossby, who along with Arnt Eliassen were known as the part of the “Scandinavian Tag Team.” In the face of dwindling financial funding for advanced research and a dearth of academically-qualified practitioners, Rossby remained an essential mediator in disseminating concepts between various team members who closely mentored many of his students while founding the Meteorological Institute in Stockholm, Sweden in later years. During the initial exercises for the electronic digital computer, “canned” data was employed rather than live data in order to correlate historic patterns to what had been calculated, and Charney, in becoming a “servant to the machine,” became conversely fluent with the prototypical models entered into the computer. In 1952, Thompson’s underhanded attempt to secure control of the direction of the Meteorological Project ironically accelerated the gradual progression towards operational forecasting by forcing those from the Weather Bureau to consolidate their manpower and information with those from the von Neumann-Charney research group.

The fact that meteorology as a type of geophysical science became more theoretical in nature and less reliant upon analogue techniques seems self-evident, perhaps too much so, but since many scientific disciplines are a combination of human skills, subjective evaluation, and technological computation, to say that meteorology moved from being an art to a full-fledged “science” with the sole assistance of numerical weather prediction seems slightly drastic. Providing more examples of the forecasters’ skills or “tacit knowledge” (non-quantifiable or non-transferable aspects of research assessment) that were given over to computational techniques could have alleviated this overstatement on the part of the author. To her credit, Harper does profile many other researchers recruited for the IAS Meteorology Project, divided between theorists and those who had a sense of the “physical factors” that influenced the weather, and it is clear that the “sanity check” provided by forecasters’ knowledge of a model’s biases and strengths informed the accuracy and temporal range of weather predictions for operational meteorology. Despite these minor drawbacks, the full cast of personalities that Harper portrays in this book, nevertheless, defies the conventional expectation that a groundbreaking scientific development belongs only to the labors of an individual genius.