Wednesday 03 Dec 2014: Science and Simulation: Whatever happened to Conjecture and Refutation?
Prof. Geoff Vallis - University of Exeter
Harrison 004 17:15-19:30
In the good old days the scientific method was simple. A hypothesis was put forward, refined into a theory, and tested with an experiment (or so it goes). If the theory was consistent with the experiment all well and good, but if not the theory was discarded or at least modified, and so back to the drawing board. But these days we try to understand complex systems with complicated models, and if the model doesn't agree with observation we don't discard it, we tweak it! Is this really the same thing? Should we and can we have a theory of a complex system - is that what a model is? Just what is the relationship of models, theories and understanding?
I'll discuss (but won't really answer) these questions from the perspective of a practicing scientist and in the context of practical problems in the climate sciences, where a complex nature and complicated models come together as never before.