Few authors in this volume attempted to explicitly address the question of whether the climate event or episode and the ecosystem response had identifiable upper and lower limits. This is because it is important to study climate events that cause severe ecosystem change so the limits of the ecosystem response are clearly bounded. Not all of the subject matter of our chapters meets this criterion. In retrospect, the reason for the lack of consideration of limits is that the subject is partially timescale-related. If an individual investigator is not familiar with all the time-scales at which their ecosystem operates, he or she will not have all the information needed to answer this question. The interdisciplinary approach of LTER research is often helpful in extending an individual's knowledge of an ecosystem, so we can expect more answers to this question to appear as the LTER program further matures.
Although the answer to this question of limits is partially related to the timescale for which we have information, it is also related to the physics or biophysics of the climate event and ecological response in question. This is well demonstrated for white spruce in Interior Alaska (chapter 12), where the gateways in the suggested model are specifically related to certain limiting values of climate variability (e.g., summer temperature and its relation to drought) and ecosystem response or preexisting condition (e.g., growth reserves). In another case, Schaefer (chapter 8, p. 154) states "although there is no fixed upper limit to the amount of rain that can fall within a 24-hour period, there are no records that . . . it has exceeded 600 mm in Puerto Rico." A longer record might produce a higher 24-hour record. Particular wind velocities are used to define the strength category of a hurricane, but the highest category, 5, is open ended at a wind velocity exceeding 69 m/sec (155 mi/hr). It should be possible to use physical principles and information on sea surface and air temperature extremes, as well as maximum and minimum storm wind velocities, to make a fairly good estimate of the maximum possible pre-cipitable water for the location. Alternatively, the theory of extreme statistics could be applied. Neither of these approaches has yet been used much at LTER sites. One sense of a limit to ecosystem response to a hurricane is implicit in chapter 2, where Boose describes a range of responses from partial defoliation to complete blow-down of a mature forest. The latter might be taken as the upper limit of ecosystem response in the hurricane context.
The Palmer LTER site (chapter 9) shows some situations where climate variability and ecosystem response display limits and other cases where it does not.
The limits are clearly shown in the ENSO-dominated timescale, but they are completely unknown in relation to the multidecadal warming trend because it is not possible to say where and when this warming trend will end. In an analogous fashion, the limits of climate (precipitation) variability seem to be well established at the quasi-quintennial scale and, to some extent, at the century scale at the Sevilleta site. But we know less about them at the millennial scale. Additionally, where the ecosystem is subject to alteration by human activity, the response to a particular climate event may be quite different from one event to another. This was shown by the emerging dominance of shrubland over grassland in the southwestern LTER sites partly related to cattle grazing in the 1890s.
The millennial timescale, which here is taken to be the Holocene but which also can involve the Pleistocene and some of its preceding geological epochs, does not play such an important role for present-day ecosystem managers. But it does help to be aware of changes at this timescale for two reasons. First, changes at the millennial scale can give information concerning the extremes to which the system can move and/or give some feel for its degree of homeostasis. The millennial-scale changes set the "limiting values" on the natural system changes in a practical and hierarchical sense. It is conceivable that human influence can help exceed these limits, but it is useful to have some idea of where the limits are or have been in the past. Second, it is also important to recognize that many of the floral and faunal species presently found in the ecosystem, or close relations of current species, have survived throughout all these extremes. This helps us to understand the degree of resilience of the ecosystem to natural changes. Third, it helps to recognize that many atmospheric phenomena that are important today have been present for a long time. For example, radiolarian records from the Santa Barbara basin indicate that El Niños have been occurring for at least 5.5 million years (Casey et al. 1989).
There have been warmer and cooler periods throughout the Pleistocene. Yet the extreme climates from the Last Glacial Maximum (LGM) 20,000 years ago to the warmer climates of the Holocene treated in chapter 18 may, in many ways, be regarded as representing, or approaching, limits of values of climate variables that modern Rocky Mountain and semiarid southwestern ecosystems may have to withstand in the absence of human influences. One suite of ecosystem responses to these changes is the varying assemblages of beetles. Elias (p. 370) argues that "ecological changes take place at many timescales, but perhaps none is more significant than the truly long-term scale of centuries and millennia, for it is at these timescales that ecosystems form, break apart, and reform in new configurations." Vegetation response in the Colorado Front Range took the form of a change from alpine tundra to subalpine forest and a decrease of the tree-line elevation of 500 m during the colder times of the mid-Pinedale glaciation. Elias' statement is also applicable to the ecosystem changes described by Monger for southern New Mexico.
Was this article helpful?