Performance, reliability and uncertainty of total phosphorus models for lakes—II. Stochastic analyses

Water quality problems in lakes are frequently analyzed with the assistance of models that simulate the dynamic behavior of total phosphorus. Part I of this series used deterministic methods to evaluate the performance and reliability of models with different formulations and mechanisms. Complex models are often amenable to accurate calibration but may be sensitive to uncertainties in the model coefficients and errors in measured lake data. This paper uses stochastic methods (Monto Carlo) to examine the supposition that the appropriate level of model complexity depends on the accuracy of measured data and the uncertainty associated with model coefficients. Three total phosphorus models are used to test the methodology for Shagawa Lake.