Modeled on rainfall-runoff process by artificial neural network coupled with simple linear model

The rainfall-runoff process is modeled by coupling a simple linear(black box) model with the a artificial neural network(ANN).In most of the previous studies,the previous runoff is used as one of the inputs to the ANN,while in this paper,the previous runoff is substituted by the runoff estimated from a linear model.It provides an approach for modeling the rainfall-runoff relationship in catchments with sparse data.It demonstrates that the coupled model adopted herein produces the reasonably satisfactory results.