Predators often exert strong indirect effects in food webs. Trophic cascades occur when top-down effects of consumers are propagated through food web linkages to impact the abundance or biomass of non-adjacent trophic levels. Additionally, predators may affect primary producers via nutrient remineralization thereby influencing bottom-up processes. Predators with comparable direct effects on prey might be expected to elicit different indirect effects when they differ in foraging behavior and/or stoichiometry. Furthermore, environmental context, such as habitat heterogeneity and nutrient concentrations, should mediate the magnitude of both direct and indirect effects of predators. Using the classic largemouth bass < central stoneroller < algae food chain as a starting point, we compared effects of predator identity and background nutrient concentrations on top-down and bottom-up processes in heterogeneous experimental streams. Four consumer treatments (no fish, stoneroller only, stoneroller and bass, stoneroller and longnose gar) were replicated within two different nutrient regimes. We quantified stoneroller survivorship, periphyton biomass, benthic organic material, total nitrogen and total phosphorus in the water column, periphyton response on nutrient diffusing substrata (NDS), consumer excretion rates and stoichiometry.
Results/Conclusions
Both predators were different in top-down effects on the shared prey and nutrient remineralization. Stoneroller survivorship was reduced in the presence of bass vs. gar (p = 0.002). Stonerollers, bass and gar all differed in their excretion nitrogen and phosphorus concentrations and N:P ratios. Unexpectedly, gar excreted relatively more phosphorus than bass and had the lowest N:P ratio for excretion. Algal biomass on NDS indicated co-limitation of nitrogen and phosphorus under ambient nutrient levels. Significant time effects were observed for water column nitrogen and phosphorus concentrations (rmANOVA, both p < 0.002), with rapid declines in the nutrient enriched treatments to levels observed in ambient nutrient conditions. The decrease in water column nutrient concentrations was concomitant with an increase in periphyton density over time for all treatments under enriched conditions (rmANOVA, p < 0.001). This effect differed from treatments under ambient conditions that had a significant time effect and treatment x habitat x time interaction (rmANOVA, p < 0.001 and p = 0.047, respectively). Nutrient enrichment appears to preclude effects observed under ambient conditions. A strong trophic cascade was not observed for any of the food chains, which appears to have been overwhelmed by the magnitude of bottom-up forces in this nutrient-limited system.