Productive reservoirs typically contain abundant gizzard shad Dorosoma cepedianum. As juveniles (25 mm total length, TL), gizzard shad potentially compromise success of larval planktivores by virtually eliminating zooplankton in early summer. Hence, as competitors and, perhaps prey, gizzard shad can drive food web dynamics. I explored how variable abiotic and biotic factors influence the role of gizzard shad as a strong interactor in reservoirs.

During years in which larval gizzard shad and its congenor threadfin shad Dorosoma petenense (henceforth, shad) appear early relative to bluegill, they may reduce bluegill success. In experiments (1-m^3 bags) with varying larval appearance times, bluegill growth indeed varied with their appearance relative to gizzard shad. By quantifying larval densities in reservoirs as a function of their relative appearance, I discovered that, in systems with > 10 shad/m^3, both zooplankton and bluegill densities were consistently low, despite varying larval appearance times.

Low bluegill densities may compromise growth and overwinter survival of juvenile largemouth bass Micropterus salmoides (henceforth, bass). Hence, I conducted experiments with bass and varying bluegill densities in cages, pools, and aquaria. In reservoirs, I quantified bass growth with abundant bluegill (N = 1) and abundant shad (N = 2). Though experiments demonstrated that bass should grow more rapidly in systems with bluegill than with shad, bass often grew moderately or rapidly in shad-dominated reservoirs. However, bass growth varied more with shad, rendering predictions for their success difficult. In a second reservoir survey and set of multi-scale experiments, I explored how variable first summer growth, fall size, food availability, and predation influenced overwinter growth and survival of bass. Poor growth of bass only should compromise their survival in reservoirs where predation and energy depletion interact. Hence, variable first summer growth of bass only should translate to variable recruitment success in systems characterized both by abundant predators and low food during winter.

By increasing our understanding of reservoir food webs through surveys and multi-scale experiments, I developed a predictive model for managers facing these complex systems. In turn, I improved our insight into how variable predation and competition mediate food web interactions in midwestern and southeastern reservoirs.