This paper explores how the dominant scientific “way of knowing” weather shifted between 1920 and 1950. An empirical, geographical understanding of weather, characteristic of the US Weather Bureau in the 1920s, was largely supplanted by a quantitative and physical conception by the early 1950s. This intellectual change did not occur evenly, nor was it driven by the dominant meteorological organization, the Weather Bureau. Rather, a small group of people—initially peripheral members of the meteorological community—successfully utilized crises to mobilize powerful external groups to effect major changes in the way that meteorologists were produced, as well as expand the number and kind of institutions that supported meteorological activities.

The growth of aviation created clients who desired new kinds of meteorological products. It was trained, practicing meteorologists, however, who adapted American meteorology to the manufacture of knowledge for commercial and military aviation. As the Weather Bureau was slow to respond, especially in the 1920s, men who were rather peripheral to the American meteorological community took the lead. With one significant exception, these emerging leaders were advocates of the Bergen School. The key figures were Francis Reichelderfer and Carl-Gustaf Rossby, still minor figures in the 1920s, later assisted by an able but small group of American graduate students and illustrious Scandinavian expatriates. Rossby, in particular, became a master at using aviation’s needs to build support for research-driven atmospheric physics in the United States. While Rossby built two graduate programs and an influential airline weather service in the 1920s and 1930s, his key achievement was the wartime training program that eventually taught 6200 new meteorologists to understand the weather in the quantitative, physical concepts advanced by the Bergen School. These trainees swamped post-war meteorology, dominating new university, military, and commercial meteorological organizations for decades to come.

The exception was the charismatic Irving Krick, whose ideas regarding long-range forecasting, and especially his powerful connections at CalTech and the Army Air Force, threatened to upset the ascendancy of the Bergen School. The Bergen School’s successful attempts to isolate and discredit Krick, I argue, played an important role in shaping enduring scientific attitudes towards weather modification and long-range forecasting.