Due to the complex nature of aquatic food webs, the interaction between abiotic and biotic factors that govern ecosystem dynamics is often elusive. Recent advancements in both the collection of reliable field data and the development of ecological models have enabled researchers to gain insights into these more complex interactions. In this study the relationship between physical and ecological processes has been explored by applying a process based coupled physical and ecological model (DYRESM-CAEDYM) to the data sets of three aquatic ecosystems. In the first, the role of zooplankton in the nutrient cycles of Lake Kinneret, Israel was quantified. The model was parameterized and calibrated using an extensive field data set. It was found that the excretion of dissolved nutrients by zooplankton accounted for up to 58% of phytoplankton demand and that this value varied seasonally in response to patterns of stratification and mixing. In the second ecosystem, Mono Lake, USA, results from model simulations were studied to determine the significance of the transport of nutrient rich hypolimnetic water via the benthic boundary layer (BBL) on lake productivity. Model results indicated that although on average the impact of BBL transport on Mono Lake ecology was not large, significant nutrient fluxes were simulated during periods when BBL transport was most active. The timing of these fluxes in the context of seasonal changes were found to be critical to specific aspects of food web dynamics. In the final application, the ecological gradients of the primary salt ponds of Shark Bay, Australia were studied with specific focus on the role of zooplankton as a determinant of ecosystem dynamics. Model results indicated that zooplankton grazing was responsible for reduced water column particulate organic matter and increased light available for the development of microbial mats. However, no direct 8 link between zooplankton grazing and observed changes in planktonic algal species composition or nutrient limitation across the salinity gradient of the ponds was found. Results from this study demonstrate the potential of a lake ecosystem model to extract useful process information to complement field data collection and address questions related to the relationship between physical and ecological processes in aquatic ecosystems.