To answer the question we must assess ecosystem resilience, resistance, and stability. Is the system easily modified? Does it readily recover from perturbation? Are there thresholds that result in major or irreversible changes in processes, ecosystem conditions, or population numbers?
Theory and field studies have shown that some ecological systems change abruptly from one relatively stable state to another. In these situations, simply removing the factor or factors that caused change may not return the system to its previous state. For example, sustained, heavy livestock grazing in arid grasslands of the western United States, in the absence of fire, has led to invasion and establishment of shrubs and trees (Archer 1994). Once trees gained sufficient stature to capture much of the moisture supply, elimination of grazing did not result in reestablishment of grassland (Glendening 1952). Such a process is consistent with “state and transition” models and with the existence of multiple stable states (Allen-Diaz and Bartolome 1998). These conceptual models help us to appreciate the complexity of ecosystem relationships and processes and should be used to evaluate management of the northern range. How do these concepts help us to evaluate changes in the GYE? Many aspects of the northern range have been intensively studied, but it has not been experimentally shown, for example, how large a reduction in the consumption of aspen by ungulates would be required to permit their “recovery.” Consequently we do not know whether changes in plant communities during the 1900s indicate that a new state, characterized by fewer communities dominated by willows and aspen, is likely to persist. Research outside the park, however, does not support the hypothesis that a new state has become established (Kay 1990). To evaluate whether the northern range is approaching a threshold, beyond which willow and aspen communities will be unable to reestablish themselves, we must have some idea of the range of natural variation (Landres et al. 1999). Are changes on the northern range within limits to be expected since Europeans arrived? How important are rare events? The “natural” interval between large fires is thought to be on the order of 200 to 300 years—can we realistically expect to manage such events? Despite claims to the contrary, we found no evidence that the northern range is approaching a threshold after which we would observe irreversible changes, such as loss of local reproductive potential of key plant species (e.g., sagebrush or aspen), that would not have occurred if the park actively controlled ungulates. This finding results, in part, because much of the evidence of dramatic changes comes from communities that are successional or the result of disturbance (e.g., aspen and riparian communities) (Houston 1982). However, changes in