A hybrid stochastic model for multiseason streamflow simulation

A hybrid model is presented for stochastic simulation of multiseason stream flows. This involves partial prewhitening of the streamflows using a parsim onious linear periodic parametric model, followed by resampling the resulti ng residuals using moving block bootstrap to obtain innovations and subsequ ently postblackening these innovations to generate synthetic replicates. Th is model is simple and is efficient in reproducing both linear and nonlinea r dependence inherent in the observed streamflows. The first part of this p aper demonstrates the hybrid character of the model through stochastic simu lations performed using monthly streamflows of Weber River (Utah) that exhi bit a complex dependence structure. In the latter part of the paper the hyb rid model is shown to be efficient in simulating multiseason streamflows, t hrough an example of the San Juan River (New Mexico). This model ensures an nual-to-monthly consistency without the need for any adjustment procedures. Furthermore, the hybrid model is able to preserve both within-year and cro ss-year monthly serial correlations for multiple lags. Also, it is seen to be consistent in predicting the reservoir storage (validation) statistic at low as well as high demand levels.