Predatory fish impact on competition between stream insect grazers: a consideration of behaviorally- and density-mediated effects on an apparent coexistence pattern

Abstract

A manipulative field experiment and theoretical analyses of a simple competition model were used to show how exploitative competition between a caddisfly (Glossosoma sp.) and three mayfly grazers (Ameletus sp., Baetisthermicus and Cinygmula sp.) was mediated by a predatory fish, freshwater sculpin (Cottus nozawae). The field experiment followed a two-factorial design, with Glossosoma densities (natural vs reduced) and sculpin presence (present vs absent) as treatments. Diet analysis revealed that all four prey species were eaten under the natural condition and the sculpin preferred mayfly grazers to Glossosoma. Our experiment showed that although mayfly densities in the presence of either sculpin or Glossosoma were lower than in the no-sculpin plus reduced-Glossosoma treatment, no difference in mayfly densities was found between the following three treatments: sculpin plus natural-Glossosoma, no-sculpin plus natural-Glossosoma, and sculpin plus reduced-Glossosoma. These results indicated that fish predator produced no effects additive to the competitive operation on the mayfly density, and also that competitive operation of Glossosoma on the mayfly densities produced no additional effects to fish predators. In addition, although the competitive effect of the mayflies on Glossosoma could not be manipulated in isolation, the density of Glossosoma in the presence of sculpin was greater than that in the absence of sculpin in the natural-Glossosoma treatment. Thus, the densities of these competing grazers either stayed the same or increased in the presence of the fish predator relative to predator-free treatments. A theoretical model, developed to explain the experimental results here, predicted that the densities of the two competing prey under predation pressure could be simultaneously greater than those under predator absent conditions when the behaviorally-mediated effects of the predator were strongly operative. Although we were unable to distinguish experimentally the two different effects of predator on the prey competition, the behaviorally- and density-mediated effects, the reality of the behaviorally-mediated effects in the experiment was discussed.