The existence of alternative stable states has been reported in a variety of ecosystems. Bistable systems with Hopf bifurcations are prone to abrupt and highly irreversible shifts from one attractor to another. Gradual changes in environmental conditions or disturbance regime may lead to a dramatic change in the state of the system. These state transitions become particularly alarming when they occur in the direction of a more “degraded” (or undesired) state, in which important ecosystem functions and services are lost. In some cases the processes determining bistability are well understood, the drivers of transitions between states are known, and well-tested conceptual frameworks relating the state of the system to its drivers exist. However, in many applications this knowledge cannot be effectively used because to date there are no well-tested ecological theories that can be applied to assess the proximity of an ecosystem to its threshold conditions for regime shift.

Results/Conclusions

We capitalize on phase transition theories from statistical mechanics to develop some criteria for the detection of the imminence of state change in ecosystems. These methods rely on existing models of bistable ecosystem dynamics to determine the scaling relations between state variables and environmental parameters in the proximity of the critical point. The scaling exponents are then used as precursors of regime shifts  to recognize the symptoms of imminent state change. The effectiveness of this method is evaluated on some examples of  bistable ecosystem dynamics with a focus on the case of arid grasslands.