simulation.”
Here’s how things will go. In Section 5.1, we’ll begin to examine some of the more difficult points of climate science, with special attention to features of the global climate system that contribute to its high dynamical complexity. In particular, we’ll focus on two aspects of the global climate which, while neither necessary nor sufficient for high dynamical complexity in themselves, are characteristic of complex systems: the presence of non-linear feedback mechanisms, and the presence of chaotic behavior. We’ll think about what it means for a system to be chaotic, and how the presence of feedback mechanisms (which are represented as non-linearities in the mathematics describing the system’s behavior) can contribute to chaos. I shall argue that careful attention to these two factors can shed a tremendous amount of light on some of the vagaries of climatology. We will see that the kind of model we constructed in 4.1 is incapable of handling these issues, and will survey some more robust models which attempt to come to terms with them.
After describing some of the problems endemic to the study of the Earth’s climate (and the models designed to solve them), we shall consider how climate scientists meet the methodological challenges they face in actually using more sophisticated models. In Section 5.2, we will discuss one of the defining tools in the climatologist’s tool-kit: computer simulation. The construction of simulations—computer-solved models designed to be run repeatedly—is a methodological innovation common to many complex system sciences; we’ll think about why this is the case, and consider the relationship between the challenges presented by non-linearity and chaos, and the unprecedented methodological opportunities presented by modern supercomputers. I will argue that while “science by simulation” is an absolutely indispensable
148
