Content area
Full Text
Attempts to explain spatial and temporal patterns of diversity in nature have been (e.g., Thoreau 1860, Clements 1916) and continue to be (May 1986, Walker 1989) a pervading theme of ecological research. Many factors have been linked with observed patterns of diversity (Krebs 1985, Begon et al. 1990), however, the interaction between disturbance and diversity has been one of the most debated issues (for reviews see Thiery 1982, Pickett and White 1985, Resh et al. 1988, Petraitis et al. 1989).
Physical disturbances (sensu White and Pickett 1985) that result in changes to population structure or resource availability, such as increases in discharge, are common in many streams (Power et al. 1988), and are postulated to have strong, possibly overriding influences, on community structure in rivers and streams (Lake and Barmuta 1986, Resh et al. 1988). A large number of observational (e.g., Fisher et al. 1982, McElravy et al. 1989, Scrimgeour and Winterbourn 1989) and experimental lotic studies (e.g., Robinson and Minshall 1986, Malmqvist and Otto 1987, Boulton et al. 1988, Doeg et al. 1989) indicate that disturbances reduce benthic invertebrate density and diversity, although both may recover quickly.
Nevertheless, although it is clear that severe physical disturbances reduce diversity, the effects of lesser disturbances are unclear. In reviewing the role of disturbance in lotic ecosystems, Resh et al. (1988) discussed three hypotheses relating disturbance regime to community structure: the equilibrium model based on Lotka-Volterra population dynamics (McIntosh 1985), the intermediate disturbance hypothesis (Grime 1973, Connell 1978) and the dynamic equilibrium hypothesis (Huston 1979). Equilibrium models invoke biotic interactions as the principal structuring force in communities, but only seem appropriate in very stable streams or during periods of stable flow (Minshall et al. 1983, 1985, Minshall and Peterson 1985).
Of the three hypotheses considered, only the intermediate disturbance hypothesis (Grime 1973, Connell 1978) has been widely applied in lotic studies (Ward and Stanford 1983, Resh et al. 1988). It predicts that diversity will be greatest at intermediate levels of disturbance, with competitive exclusion and physical elimination leading to species loss at either end of the disturbance continuum. The hypothesis has received support from both theoretical (Petraitis et al. 1989), and some empirical (Stanford and Ward 1983) studies. However, the results of experimental manipulations of substrate patches...