Alternative stable state theory was first proposed by Richard Lewontin (1969), but other early key authors include Holling (1973), Sutherland (1974), May (1977), and Scheffer et al. (2001).
This definition is an extension of stability analysis of populations (e.g., Lewontin 1969; Sutherland 1973) and communities (e.g., Drake 1991; Law and Morton 1993).
It may be possible for the system to exist under different community structure regimes depending on initial conditions (e.g., population densities or spatial arrangement of individuals) (Kerr et al. 2002).
The community perspective requires the existence of alternative stable states (i.e., more than one valley) before the perturbation, since the landscape is not changing.
In the ball-and-cup model, this would be the energy required to push the ball up and over a hill, where it would fall downhill into a different valley.
Although the mechanisms of community and ecosystem perspectives are different, the empirical evidence required for documentation of alternative stable states is the same.
In addition, state shifts are often a combination of internal processes and external forces (Scheffer et al. 2001).
As a result, benthic vegetation cannot receive light and decline, increasing nutrient availability and allowing phytoplankton to dominate.
In this ecological context, hysteresis refers to the existence of different stable states under the same variables or parameters.
Others (e.g., Beisner et al. 2003) claim that this is not so; although shifts often involve hysteresis, a system can show alternative stable states yet have equal paths for "A → B" and "B → A".
The 1983 crash of sea urchin populations in Caribbean reef systems released algae from top-down (herbivory) control, allowing them to overgrow corals and resulting in a shift to a degraded state.
When urchins rebounded, the high (pre-crash) coral cover levels did not return, indicating hysteresis (Mumby et al. 2007).
There are at least three ways in which anthropogenic forces reduce resilience (Folke et al. 2004): (1) Decreasing diversity and functional groups, often by top-down effects (e.g., overfishing); (2) altering the physico-chemical environment (e.g., climate change, pollution, fertilization); or (3) modifying disturbance regimes to which organisms are adapted (e.g., bottom trawling, coral mining, etc.).
Shifts to less-desirable states often entail a loss of ecosystem service and function, and have been documented in an array of terrestrial, marine, and freshwater environments (reviewed in Folke et al. 2004).
Most work on alternative stable states has been theoretical, using mathematical models and simulations to test ecological hypotheses.
Verifying the existence of alternative stable states carries profound implications for ecosystem management.
Understanding the nature of these thresholds will help inform the design of monitoring programs, ecosystem restoration, and other management decisions.
Empirical evidence for the existence of alternative stable states is vital to advancing the idea beyond theory.
Schröder et al. (2005) reviewed the current ecological literature for alternative stable states and found 35 direct experiments, of which only 21 were deemed valid.
However, the Schröder et al. (2005) analysis required evidence of hysteresis, which is not necessarily a prerequisite for alternative stable states.
Although alternative stable state theory is still in its infancy, empirical evidence has been collected from a variety of biomes: