Zusammenfassung: Studies of predator-prey interactions have been limited to laboratories or relatively small scale field experiments because of logistic difficulties in large scale field experiments. Much evidence has suggested that predictions based on small scale experiments on predator-prey interactions may not be simply extended to large scales, e.g. whole system. This study was designed to examine predation effects on a prey fish assemblage at the whole lake scale. After a year of premanipulation study, northern pike (Esox lucius) were introduced into a small piscivore-free lake inhabited by 14 species of fishes. In the third and fourth years, the lake was divided by a large mesh fence and northern pike were introduced into one half of the lake. Predation effects on individual prey population were estimated for both direct effects (i.e., consumption of prey species) and indirect effects, including habitat changes associated with predator avoidance behavior, increase of emigration rates and changes of population size structure. The relative importance of direct versus indirect effects was examined using randomized intervention analysis along with a model that quantifies the indirect effects. Decreases of prey fish biomass due to increased emigration was at least as great as that due to the direct effect of consumption when prey biomass was high (>40 kg/ha) but not when prey biomass was low (<30 kg/ha). Significant indirect effects included dramatically decreased abundance of the dominant species (northern redbelly dace, Phoxinus eos), increase of some rare species (sunfish, Lepomis spp. and creek chub, Semotilus atromaculatus), and decreased mean sizes of redbelly dace. Indirect effects occurred rapidly and were most apparent immediately following the introduction of northern pike. Community responses to northern pike predation included changes of species composition, species diversity and community stabilities. The fish communities were significantly different among the pre-pike period, pike period and post-pike period. The greatest variance was related to abundances of redbelly dace and sunfish. However, predation effects on the prey fish community depended on the observational scales used to determine community structure. (DBO)

Zusammenfassung: This technical report describes a FORTRAN program (HISEA.FOR) to be used in the study of mixed stock fishery composition estimation. Program HISEA is a versatile research tool capable of performing analyses. The simplest use of HISEA is to perform a composition analysis. The program can then be run in bootstrap mode to provide a non-parametric estimate of the reliability of the estimated compositions. In simulation mode, HISEA can quantify the effect of changing sample sizes; of adding or removing variables; and of adding or removing stocks. Simulation also identifies potential problems (such as high bias and variance in the composition estimators) that may arise due to different stocks being too alike.