Runoff Analysis Study of Paddy Field Areas by Multiple Regression Model

Abstract

Shiraishis' multiple regression runoff model is effective for analyzing bad data. This model was calculated on the linear part of runoff separated by the condition of rainfall or discharge. The former is called the FMD method and the latter the FMR method. These two methods seemed to be equivalent. This model treats Y intercept b₀ as the base flow. The author clarified that these two points are wrong for paddy field areas. For separation of the base flow in paddy filed areas, a new method, a dummy variable method was proposed and used in the Naka River basin. This method can separate the base flow caused by rainfall from that caused by intake. In this basin, it was clarified that FMR method can separate the effect of intake from the effect of rainfall and intake. Additionally, a multiple regression intake model by rainfall was first developed in this basin in Japan.