A Vast Machine argues that all our knowledge about climate change comes from three kinds of computer models: data models, used to combine and adjust measurements from many different sources; simulation models of weather and climate; and reanalysis models, which recreate climate history from historical weather data. Meteorology creates knowledge through a sociotechnical infrastructure that covers the whole world — making global data. This infrastructure generates information so vast in quantity and so diverse in quality and form that it can be understood only through computer analysis — by making data global. A Vast Machine describes the science behind the scientific consensus on climate change, arguing that over the years data and models have converged to create a stable, reliable, and trustworthy basis for establishing the reality of global warming.