Field Experiments on Responses of a Freshwater, Benthic Macroinvertebrate Community to Vertebrate Predators

We examined the seasonal importance of vertebrate predators in potentially regulating the abundance and diversity of the benthic macroinvertebrates in the littoral zone of a soft-bottom reservoir that receives thermal effluent from a nuclear production reactor. Thirty-six predator (fish and turtle) exclusion cages (4 m-') were placed in shallow water at six locations along a thermal gradient in Par Pond, a 1100-ha cooling reservoir on the Savannah River Plant near Aiken, South Carolina, USA. An additional 36 control plots (4 m-) were also set up. Cages were in place during three, 3-mo test periods beginning in September 1977. Estimates of benthic density, taxon richness, and distribution within functional groups (defined by feeding mechanism) were calculated for each test period. Effects of temperature on predator-prey relationships were also determined. Experimental results of this study suggest that vertebrate predation was not the fundamental parameter organizing the benthic macroinvertebrate community in the littoral zone of this reservoir. Neither taxon richness nor density of total macroinvertebrates was conclusively related to predator treatment. Relationships between predator treatment and community response (changes in density and taxon richness) were generally unaffected by either plot locality, temperature fluctuations from thermal effluent, or seasonal changes. When data from caged and control plots were pooled, however, both location and water temperature individually had direct impacts on the benthic community. From our results and other field studies we hypothesize that individual species of "keystone" benthic predators (Paine 1969b) do not occur in the littoral zone of freshwater lentic environments with soft bottoms. This hypothesis is based on the following four arguments. First, environmental heterogeneity should reduce predator efficiency relative to that found in rocky intertidal systems where keystone predation may be present. Second, relative food web complexity should reduce the community importance of an individual predator species. Third, the apparent relatively greater resource partitioning of space and food in the freshwater littoral zone hinders monopolization of ratelimiting environmental resources by a single dominant prey species. Fourth, the effects of predation may not be extensive enough in time or area to provide sufficient resource space (and thus food) for exploitation by fugitive species. Possible regulation by a guild of predators and cases where the general hypothesis might be falsified are discussed