Effects of stoneflies on local prey populations: mechanisms of impact across prey density

Ecology, July, 1996 by Kim W. Kratz

INTRODUCTION

Predators can have important effects on the abundance of prey species and the composition of prey communities (Paine 1980, Kerfoot and Sih 1987). Experimental manipulations of predators have shown that the magnitude of predator effects may vary across communities and among taxa within communities (Sih et al. 1985). Examinations of the effects of predators on the density of benthic prey populations in streams have produced equivocal results. Experiments using predatory fish have produced predator impacts ranging from no effects to profound effects (Allan 1982, Gilliam et al. 1989, Flecker 1992, Power 1992, Sih et al. 1992, see review in Wooster 1994), whereas invertebrate predators generally reduce populations of common prey in local areas (Peckarsky and Dodson 1980a, Walde and Davies 1984a, Peckarsky 1985, 1991a, Malmqvist and Sjostrom 1987, Lancaster 1990, Lancaster et al. 1990). This suggests that it may not be possible to generalize about the effects of predators on local prey densities in streams. These previous studies, however, did not address specifically the influence of density-dependent processes on predator effects. Because predator - prey interactions are sensitive to prey density (Walton 1980, Hildrew and Townsend 1982, Murdoch and Stewart-Oaten 1989, Williams et al. 1993) and streams are characterized by patchy distributions of benthic invertebrates (Pringle et al. 1988), differences among studies in the magnitude of predator impact may arise if predator impact is sensitive to prey density. It is of interest, then, to examine the effects of prey density on predator impacts on prey populations.

The degree that predators can alter densities of local prey populations depends, in part, on the interplay among behavioral responses of predators to prey (i.e., predator numerical and functional responses) and prey behavioral responses to predators (i.e., refuge use and dispersal) (Hassell 1978, Lima and Dill 1990, Huang and Sih 1991). Strong predator numerical (aggregative) responses, where individual predators spend more time in patches with more prey, influence the impacts of predators on prey populations (Hassell and May 1974) and can enhance the stability of predator - prey dynamics in models (Murdoch and Stewart-Oaten 1989). Even weak responses of individual predators to areas of high prey density may cumulatively produce strong aggregation of predator populations in areas of high prey density (Ives et al. 1993). The strength of a predator's impact on prey populations is additionally influenced by the form of the predator's functional response, with density-dependent foraging generally conferring stability to predator - prey dynamics (Hassell and May 1985). The form of the functional response may reflect behavioral responses of prey. Predator presence has been associated with increased use of refugia by prey (Cooper 1984, Sih 1987, Huang and Sih 1991, Garvey et al. 1994), which can produce density-dependent functional responses (Murdoch and Oaten 1975). Predators may alter per capita rates of prey dispersal and further influence the magnitude of observed predator impacts (Lima and Dill 1990, Douglas et al. 1994). Thus predator consumption and prey anti-predator behaviors contribute to observed predator impacts on local prey populations.

In open stream systems, predator effects on local prey density can arise from two mechanisms, direct consumption and predator-induced emigration (Cooper 1988, Lancaster 1990, Forrester 1994a, Sih and Wooster 1994). Although both mechanisms may result in declines in local prey density, consumption represents a net loss of prey to the global population whereas emigration results in the redistribution of prey among local patches. In addition to effects on local prey populations, predators may affect adjacent patches by inducing or suppressing prey emigration rates (Kareiva 1986, Cooper et al. 1990, Lancaster et al. 1991, Sih and Wooster 1994, Wooster and Sih 1995). Dispersal influences not only the degree of coupling among patches in heterogeneous environments, but also the stability of predator - prey interactions (Kareiva 1986, Taylor 1990, Hanski 1991). Therefore, models of predator - prey dynamics must consider the relative importance of each of these loss terms, because they have significantly different implications for local patches and metapopulations. Despite widespread observations of predator effects on prey density in streams, attempts to assess the relative importance of predation and emigration to prey loss from patches in streams are rare. The few studies that have been done have reported different results (Peckarsky and Dodson 1980a, Peckarsky 1985, Feltmate and Williams 1989, Lancaster 1990, Sih et al. 1992, Forrester 1994a).

Here I assess experimentally the effects of natural densities of predaceous stonefly larvae (Doroneuria baumanni Stark and Gaufin: Perlidae) on the behavior and density of baetid mayfly nymphs (primarily Baetis tricaudatus Dodds: Baetidae) across a gradient of prey density. The major goals of this experiment were to: (1) determine the functional response of predatory stonefly nymphs and the effect of prey density on predator-induced per capita prey emigration from patches; (2) determine predator impact, and partition predator impact into direct predator consumption vs. predator-induced drift; and (3) determine the effects of predatory stoneflies and prey density on the proportion of baetid nymphs found on substratum surfaces over a diel cycle. I also addressed the relationship between prey density and predator dispersal from patches.