Preliminary results of near-field modeling of the wastefield formed by the Sand Island, Honolulu, outfall are presented. Over one thousand simulations were run with the mathematical model RSB using as input data long time series of effluent flowrate and oceanographic observations over the whole water column including currents measured by Acoustic Doppler Current Profilers (ADCPs), and density stratification measured by thermistor strings. It was found that considerable variability in plume behavior occurs, whose extremes depend on particular combinations of flowrate, current, and density stratification. The wastefield is predicted to be submerged, usually deeply, about 90% of the time for the conditions simulated. It is demonstrated that the use of these recently developed instruments combined with appropriate mathematical models can lead to greatly improved predictions of the statistical characteristics of wastefield behavior in coastal waters than has been previously possible.