A Vast Machine: Computer Models, Climate Data, and the Politics of Global Warming by Paul N. Edwards (MIT Press, 2010).
This is a dense, scholarly book which took me a while to read, but it was worth the effort. When I left the world of climatology (the source of my summer jobs while I was in college) back in the early 1980s, computer models ran at processor speeds people probably have on their cell phones now, and even the scientists weren’t talking about global warming yet. Thirty years later, the evidence of climate change is all around us, so the much better models we have now don’t have to be perfect for people to believe in global warming. Ironically, all those years that Republican administrations told federal agencies to “do more research” before making any policy decisions have resulted in some pretty strong research evidence for anthropomorphic climate change.