man action at one level of aggregation may depend on events at another level. Theory about these relationships is relatively weak, but the problem is of active interest to social scientists in several disciplines. If excellent data sets are compiled, the problem of connecting levels of analysis may attract leading disciplinary researchers to the topic of global change to build theory that would aid in understanding it while advancing their own fields.

Linking time scales is also critical to the global change agenda. The question is this: Which social changes, occurring on the time scale of months to years, are likely to persist or be amplified over time, to the extent that they will be significant to the global environment on a scale of decades to centuries? Obversely, which short-term changes are likely to disappear over time? Physical scientists know which halocarbons are long-lived catalysts for the destruction of stratospheric ozone and which ones are quickly destroyed; social scientists do not yet know much about which social changes catalyze other changes or about which ones are relatively irreversible. Historical cases, such as the CFC case, suggest some interesting hypotheses; over the near-term, efforts to catalogue and compare such hypotheses would be a useful first step toward a theory of the long-term effects of social change. The general problem has received very little attention from social scientists. Improved understanding of the human analogues of long-lived catalysts may contribute to increased interest in long-term phenomena in social science.



Some species, such as rosewood, are selectively eliminated from the forest for economic reasons. It is reasonable to expect that in an ecosystem characterized by many smaller species, such as insects dependent on a single species for food, that the selective cutting of one tree species will cause multiple extinctions.


The mechanism is rather complex. Evapotranspiration in the Amazon forest appears to cause a regional climatic increase in precipitation. In such a regime, large-scale clearing, which reduces evapotranspiration per land area even if trees are replaced by other vegetation, will decrease rainfall downwind. Because species diversity in Amazonia is directly related to levels of rainfall, lower rainfall in any region can be expected to reduce the number of species in that region.


Species with large area requirements are disproportionately affected when forest clearing is fragmented, as it typically is in Amazonia. Under those conditions, an individual or functional group of individuals with a large area requirement is less likely to find adequate forest resources

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement