Abstract

Diverse complex systems can reach tipping points where they abruptly switch to a state with contrasting features. Such

transitions are also called regime shifts or critical transitions. In this talk, I will present how fairly simple mathematical

models of ecosystems, e.g. extensions of logistic population models, can help us to devise early warning signals of critical

transitions. I will then focus on our work on analysing large-scale remotely-sensed data of Savanna ecosystems and a

dryland ecosystem which provide evidence for these theories. We show while there is indeed empirical evidence for these

theories, stochasticity in real ecosystems can confound some interpretations. I will talk about our work on how spatial self-

organisation can offer insights on understanding critical points and regime shifts in ecosystems.