Knowledge Infrastructures for a Changing Planet: the past, present, and future of relations between climate science and nuclear weapons research

Abstract: Nuclear war and climate change present the two most serious threats to global security since World War II. This talk shows that nuclear weapons research and climate science were historically connected in deep, sometimes intimate ways. Each developed its own knowledge infrastructure, including people, technical systems, and organizations, with surprising parallels and frequent exchanges across the classified/civilian divide. From the 1940s on, nuclear weapons research and climate science both relied heavily on computer models, used related physics and numerical methods, and shared human as well as technical resources. Radiocarbon from nuclear weapons tests contributed to understanding of the global carbon cycle, while fallout monitoring networks produced critical knowledge about the stratosphere. In the 1980s, the potential for “nuclear winter” — a war-induced climatic catastrophe — became a major political issue, but the groundwork for this concern had been laid long before.

This interplay not only continued, but became even more significant after the Cold War’s end, when the weapons labs’ expertise, equipment, and observing systems were partially repurposed. Several US national laboratories now play essential roles in climate and Earth system science. Among these roles are the Program on Climate Model Diagnosis and Intercomparison, based at Livermore and responsible for the important Coupled Model Intercomparison Project (CMIP), a major unifying force in climate modeling for the Intergovernmental Panel on Climate Change assessments. The cyberinfrastructure underlying CMIP and similar projects must address mounting challenges related to data access controls, software support, and the security of huge data collections, while their institutional and human bases depend on ongoing national support. Crafting effective climate policy, I argue, will require understanding and rethinking the dynamics of these knowledge infrastructures for the present, rapidly evolving context.

About the Speaker: Paul Edwards is a Professor in the School of Information (SI) and the Dept. of History at the University of Michigan. SI is an interdisciplinary professional school focused on bringing people, information, and technology together in more valuable ways.

His research explores the history, politics, and cultural aspects of computers, information infrastructures, and global climate science. His current research focuses on knowledge infrastructures for the Anthropocene.

Dr. Edwards is co-editor (with Geoffrey C. Bowker) of the Infrastructures book series (MIT Press), and he serves on the editorial boards of Big Data & Society: Critical Interdisciplinary Inquiries and Information & Culture: A Journal of History. His most recent book is A Vast Machine: Computer Models, Climate Data, and the Politics of Global Warming (MIT Press, 2010).