NUMERICAL METHODS FOR ASSESSING WATER QUALITY IN LAKES AND RESERVOIRS (MODELING ECOSYSTEMS).

摘要

Water quality models are used as tools for predicting both short-term and long-term trends in water quality. They are generally classified into two groups based on the degree of empiricism. The two groups consist of the purely empirical types known as "black-box" models and the theoretical types called "ecosystem" models. The dissertation deals with both types of water quality models. The first part deals with empirical phosphorus models. The theory behind this class of models is discussed, leading to the development of an empirical phosphorus model using data from 79 western United States lakes. A new approach to trophic state classification is introduced. The data used for the model was obtained from the Environmental Protection Agency National Eutrophication Study (EPA-NES) of western United States lakes.; The second portion of the dissertation discusses the development of an ecosystem model for culturally eutrophic Liberty Lake situated in eastern Washington State. The model is capable of simulating chlorophyll-a, phosphorus and nitrogen levels in the lake, on a weekly basis. For computing sediment release rates of phosphorus and nitrogen, equations based on laboratory bench-top studies using sediment samples from Liberty Lake are used. The model is used to simulate certain hypothetical nutrient control techniques such as phosphorus flushing, precipitation, and diversion.