Lecture slides for UCLA LS 30B, Spring 2020
License: GPL3
Image: ubuntu2004
Learning goals:
Know the four defining properties of chaos in a dynamical system.
Be able to explain what each one of them means.
Example: The three-species food chain model (the Hastings model)
Example: The discrete-time logistic model
Definition: A mathematical model has chaotic behavior if it has the following four properties:
The model is deterministic.
The behavior is bounded.
The behavior is irregular (or aperiodic).
The behavior has sensitive dependence on initial conditions.
Properties of chaos:
The model is deterministic.
Recall that this means that (at least in theory) the current state determines all future states. In short, there is no randomness in the model.
Note that this is a property of the mathematical model, not a property of the behavior.
All models that we have studied and will ever study in LS 30 are deterministic. Models that are not deterministic are called “stochastic”, a fancy word for random.
Properties of chaos:
The behavior is bounded.
Or, in other words, there is some bounding box that can be drawn in the state space, so that a trajectory that starts inside that box will remain inside it.
Example: The amount of grass is bounded.
Example: The sheep population is bounded.
Example: The wolf population is bounded.
Example: The whole trajectory is bounded.
Example: The behavior of the discrete-time logistic model is bounded.
Properties of chaos:
The behavior is irregular (or aperiodic).
This means that a typical trajectory never repeats itself, nor does it asymptotically approach a periodic cycle.
Note that, together with deterministic, this means that a trajectory will never reach the same exact state twice. Or, in other words, it does not form a closed loop in the state space.
Example: The populations in the three-species food chain are irregular (aperiodic).
Example: The discrete-time logistic model is irregular.
Properties of chaos:
The behavior has sensitive dependence on initial conditions.
This means that making a tiny change in the initial state will result in much bigger changes in the long run.
Colloquially, this is often referred to as the butterfly effect: the tiny air currents produced by a butterfly flapping its wings today in Sri Lanka can affect entire weather systems months from now in Kansas, halfway around the world from Sri Lanka.
Example: Sensitive dependence on initial conditions in the grass
Example: Sensitive dependence on initial conditions in the sheep
Example: Sensitive dependence on initial conditions in the wolves
Example: Sensitive dependence on initial conditions in the grass
Example: Sensitive dependence on initial conditions in the sheep
Example: Sensitive dependence on initial conditions in the wolves
Example: Sensitive dependence on initial conditions in the discrete logistic
Conclusions: The defining properties of chaotic behavior
The model is deterministic: the current state determines all future states; i.e. there is no randomness in the model.
The behavior is bounded: none of the state variables go to ∞.
The behavior is irregular (or aperiodic): it never repeats itself, nor does it asymptotically approach a periodic cycle.
The behavior has sensitive dependence on initial conditions: a tiny difference in two initial states will grow into a much bigger difference eventually.