First-difference fluctuations and the complexity of simple population models exhibiting chaos

A new method for distinguishing deterministic chaos from random noise in very short time series is presented. It is based on the scaling properties of the fluctuations present in the first-difference series generated by simple non-linear models widely used in population dynamics