The concept of ecological resilience has come to serve as an integrative framework for the environmental sciences (including ecology, environmental sociology, and environmental economics). This integration has been achieved largely by defining common concepts and processes that are applicable to both ecological and human systems. One of the next steps in the maturation of this endeavor is to develop a research paradigm that is also integrative. In this talk, I describe the memory theory of resilience based on five logical propositions: (1) Resilience of historic states of a system depends on the sources of memory available to that system. (2) System memory comprises stocks, attributes, and processes and their collective effects are substantially greater than the effects of any one source by itself. (3) Degrading forces that reduce resilience will often be new (nonhistoric) conditions or agents. Degrading forces can act to both reduce memory and alter responses. The ways that degrading forces interact with (amplify or dampen) disturbance or change is also critically important to their impact on resilience. (4) Resilience varies strongly with background context. It is not a general attribute. (5) The relationship between rates of productivity and the resilience of systems will be complex. There is a choice to be made between managing for productivity and managing for resilience.
Results/Conclusions
In this talk I use data from a study of the responses of grasslands to prescribed fire to illustrate the applicability of the memory theory. An instantiation of the theory is developed to study the factors that determine whether a system will remain in or recover to its historic state following a disturbance event. The categories of controlling factors examined include historic influences, degrading agents, sources of memory, and responses to fire. Structural equation modeling is used to specify a statistical model to evaluate the theoretical model. Results from this analysis show that resilience was enhanced by a high gamma diversity, particular species, high community biomass, high fire severity, and summer burning. Resilience was impaired by the increasing amounts of a degrading agent (Chinese tallow trees). Resilience was found to be highly conditional based on location along an environmental (microelevational) gradient. I conclude that resilience depends on many elements of memory and increased exploration of these elements is badly needed. Structural equation modeling provides a useful framework for testing models of the factors controlling resilience.
