In this talk we will be discussing attractors of dynamical systems, which are the formal way a "steady state" is defined. Finding attractors used to be a messy and ambiguous process, but a new method utilizing fundamental dynamical systems theory is able to find attractors effortlessly, provided one has equations for a dynamical system. In the real world, we typically measure dynamical systems and don't have equations. There we want to be able to quantify the attractors in a way that is not affected by our measurement process. Statistical quantifiers of the dynamics of the system, like mean or variance, get altered by the measurement process. But the fractal dimension is a dynamic invariant that characterizes the true system, and not its measurement, while providing some additional knowledge inaccessible from statistics. Unfortunately, its computation comes with many pitfalls and difficulties, which I will try to highlight using illustrative examples from paleoclimate.
19.04.2022
15:15 Uhr