“TOP–DOWN” TROPHIC INTERACTIONS IN LAKES: EFFECTS OF FISH ON NUTRIENT DYNAMICS

Abstract

We conducted enclosure experiments over two summers in Tuesday Lake, Michigan, to assess how a gradient of zooplanktivorous fish biomass affected the dynamics of nutrients (nitrogen, N, and phosphorus, P), and their partitioning among ecosystem compartments. In both years, fish (the cyprinid Phoxinus eos) reduced the abundance of large zooplankton species and increased the biomass of phytoplankton as predicted by the top–down control hypothesis. Fish had strong effects on the dynamics and fluxes of N and P. Total P concentrations in the water column declined over time in all enclosures, but fish slowed the rate of decline. Thus total water column P increased with increasing fish biomass. Total N increased less strongly with increasing fish biomass, and thus the total N:P ratio decreased with increasing fish biomass. The concentrations of particulate carbon, nitrogen, and phosphorus in the water column also increased with increased fish biomass. Particulate N:P ratio decreased with increased fish biomass, but effects were weaker compared to effects on total N:P ratios. Nutrient ratios of the zooplankton fraction (particles >63 μm) showed a response that was transient but consistent with observed trends in zooplankton species composition. In particular, when the large cladocerans Daphnia and Holopedium increased upon exclusion of fish, C:P and N:P ratios of the zooplankton fraction showed distinct declines, corresponding to the relatively high body P contents of these taxa. Phosphorus budgets revealed that fish were a net source of P to the water column, because they lost mass during the experiments, even at densities below those in the lake. However, loss of P from fish could not account for the higher total P concentration observed in enclosures with fish compared to fishless enclosures. The absolute amount of P sinking from the water column increased with increasing fish biomass but decreased when expressed as percentage of total P sinking, again suggesting that the presence of fish increases the relative retention of P the water column. The rate of decline in water column total P in the presence of fish was accurately predicted by sedimentation of P from the water column and other fluxes. Our results support the hypothesis that fish can exert major influences on the dynamics, distribution, and ratios of limiting nutrients.