The recently developed modeling and forecasting techniques such as quasilinerization and invariant imbedding are applied to the modeling of water quality systems. Due to the fast convergence rate and the sequential nature of the estimation scheme, a large amount of computer time and computer memory can be saved. Furthermore, these approaches form an effective forecasting technique for the prediction of pollutant concentration. The dispersion model with the consideration of axial mixing is used. By properly adjusting the axial diffusion coefficient, the dispersion model can represent streams with any degree of mixing. A water stream with intermediate reservoirs, and with waste discharges and water intakes along the stream, is represented by a system of second order differential equations. These equations are obtained by dividing the stream into different segments. An analytical solution is obtained for this system of equations to predict the BOD and DO concentrations. Based on this model, simulation study is also carried out to investigate the influence of the various parameters on the BOD and DO profiles. (Author)