Abstract |
The development of a multiple-input stochastic hydrologic system model is described for the analysis of hydrologic behavior of watersheds. Various components of the hydrologic system are expressed by time series, each containing a trend component, a periodic component and a stochastic residual component. For modeling the multiple-input system, a Markov-type mathematical formulation is proposed. For illustrative purposes, the model is applied to the analysis of monthly precipitations and streamflows of the upper Sangamon River basin in Illinois. The results indicate that the proposed model is feasible for the basin and thus can be used for filling missing streamflow data or generating stochastic streamflow sequences. (Author) |