A simple model, based on tanks in series, for the estimation of mean annual loads and frequency distributions of loads from combined sewer systems is presented. The input data, dry weather flow, dry weather quality, and storm water quality are estimated from treatment plant influent data. Two similar methods for the estimation of flow-average storm water quality were tested by using treatment plant influent data generated by the model in comparison with the model input. Both methods are based on daily mass balances, but differ slightly with respect to the averaging procedures used. The performance of both methods is about the same. They show a small bias, but the variability introduced is small when compared with the variation occurring in real storm water quality data. Application of one of the methods on field data revealed no distinct relationships between the flow-averaged storm water quality concentration and the dry weather period or the total daily rain depth. By combination of continuous and Monte Carlo simulation techniques the model can be used to estimate mean annual loads and frequency distribution of loads from combined sewer overflows. For the extreme events a large 90 % confidence interval was found due to the large variations in storm water quality.
Rainfall, runoff, and water-quality data for the urban storm-water program in the Albuquerque, New Mexico, metropolitan area, water year 2001
