Having explored the most obvious of the possible explanations of the observed cyclic and quasi-cyclic variations in the climate, we must now draw some conclusions. But before doing so there is one final fly in the ointment: addressing the ultimate consequence of non-linearity. Wherever possible the aim of the book has been to attempt to provide manageable explanations of the observed fluctuations by using linear models. In Chapter 7 this approach was extended to take in certain aspects of non-linearity to model observed behaviour, notably in the case of the progression of the ice ages (see Section 7.5). In so doing, it has been necessary to note continually that this approach flirted with the unpalatable fact that what we are observing is effectively unpredictable. Frequently, the adjective ‘chaotic’ has been used in this context and now we need to put this expression on a firmer footing. To do this we need to look at what has become a fashionable areas of scientific thinking since the 1980s – chaos theory.

Chaos theory

Chaos theory has attracted much public attention for two principal reasons. First, it seeks to bring some understanding to the fascinating boundary between order and disorder in physical systems. Second, it presents the concepts with an intoxicating mixture of imagery. But at a more basic level it provides some particularly important insights into the search for meteorological cycles.