FISH PREDATION AND BENTHIC COMMUNITY STRUCTURE - THE ROLE OF OMNIVORY AND HABITAT COMPLEXITY

In population models, omnivorous predation (i.e., predation on > 1 trophic level) generally has a destabilizing influence, whereas habitat heterogeneity tends to stabilize both predatory and competitive interactions. I experimentally evaluated the impact of a native omnivorous predator and of habitat structural complexity on the structure of a freshwater littoral macroinvertebrate community. Perch (Perca fluviatilis) were stocked at three densities (no fish, and low and high natural densities) into pond enclosures with dense submerged vegetation present or absent. Macroinvertebrate abundance, biomass, size structure, and diversity as well as gut contents of perch were monitored over a 2-mo summer period. Perch consumed both predatory and nonpredatory invertebrates except molluscs. In vegetation, perch consumed macroinvertebrates almost exclusively during the whole experiment, whereas in the absence of vegetation, perch successively increased their consumption of microcrustaceans. The biomass of predatory invertebrates and the abundance of the dominant predatory invertebrate Sialis lutaria (Insecta: Megaloptera) decreased with increasing perch density in both vegetation treatments. In contrast, perch density had no effect on the biomass of nonmolluscan herbivores and detritivores. Vegetation increased the biomass of all macroinvertebrates and tended to delay the negative effect of perch on predatory invertebrates. Perch density had a negative effect on the mean size of total nonmolluscan macroinvertebrates, but not of predatory invertebrates and nonmolluscan herbivores/detritivores considered separately. While perch density had an effect on the diversity (Shannon evenness and Shannon diversity) of only nonmolluscan macroinvertebrates (i.e., potential prey), vegetation had a positive effect on species richness (but not on evenness) of all macroinvertebrates. The results suggest that perch disproportionally preyed on predatory invertebrates, but that characteristics other than size per se were responsible for their higher vulnerability to perch. Perch may have affected nonmolluscan herbivores/detritivores simultaneously through direct consumption and through indirect release from invertebrate predation, which could explain the lack of a perch effect on nonmolluscan herbivores/detritivores. Complex interactions between the omnivorous predator, intermediate consumers, and resource species (macroinvertebrate herbivores and detritivores) may explain the relationship between perch density and macroinvertebrate diversity. The reduction in foraging efficiency of perch in submerged vegetation, and the observed diet shift of perch towards microcrustaceans when predatory invertebrates were still fairly abundant, suggest that relatively large predatory invertebrates were still fairly abundant, suggest that relatively large predatory invertebrates can coexist with fish in the littoral of lakes to a higher extent than in the pelagic zone. Predatory invertebrates may thus play a more important role in food web dynamics of littoral than of pelagic communities.