This proposal is designed to evaluate how community interactions (predation, competition) produce changes in ecosystem dynamics during a series of whole-lake food web manipulations. The work proposed capitalizes on a program of fish stocking manipulations funded by an LTREB grant from the National Science Foundation, as well as on the long-term record of data collection at this site. Detailed study of resource-and predator-relations of the dominant components of the food web will allow the investigators the mechanisms responsible for changes in each tropic level and ultimately in the ecosystem itself. Fish populations will be monitored by acoustics and video observation, as well as by more conventional fisheries techniques, and changes in abundances, growth, diets, recruitment, and degree of plankivory evaluated as major changes in external fish stocking intensity are imposed. Changes in zooplankton dynamics (especially of Daphnia) will also be examines using techniques of zooplankton population biology (egg ratios, estimates of birth and death rates) as well as physiological measures of food status (lipid content). The PI's will also evaluate the degree to which trout planktivory accounts for observed Daphnia mortality. Phytoplankton-predator relationship will be assessed by measuring grazing rates by zooplankton dominants on the major algal taxa and determining whether changes in situ algal dynamics are attributable to grazing impacts. Phytoplankton-resource relations will be evaluated by monitoring algal nutritional conditions and using dilution bioassays to determine species-specific growth kinetics for major limiting nutrients. In addition, the roles of externally-supplied N vs. internally-supplied N (i.e. within the food web) in supporting algal productivity will be evaluated using 15 N isotope techniques. The roles of planktivory and nutrient/food supply in altering plankton dynamics will be assessed in controlled settings by experiments in large mesocosms. These experiments will determine plankton responses to: 1. a gradient of lanktivory pressure which will bracket the intensity of trout redation to be encompassed by the 4-year LTREB fish anipulations, and, 2. a factorial manipulation of trout lanktivory and NO3 loading designed to simulate the combination of high and low external nutrient loading and high and low fish stocking. These experiments also provide additional arenas within which to test the predications of simulations models of food web dynamics and phytoplankton community structure. A functional group model of the Castle Lake food web and a multi-species model of the Castle Lake food web and a multi- species model of the phytoplankton community (following Powell and Richerson 1985) will be developed as means of synthesizing the observations obtained during our mechanism-based observational and experimental program and to generalize the insights obtained. Parameters and simulation conditions will be obtained from specific experiments and from the long-term data set and used to further explore the interactions between community processes (predation, resource-based competition) and ecosystem properties (patterns of inter-and intra-annual productivity, the role of NO3 and NH4, etc.)
