A Nonlinear Time-lag Differential Equation Model for Predicting Monthly Precipitation

This paper investigates the nonlinear prediction of monthly rainfall time series which consists of phase space con-tinuation of one-dimensional sequence, followed by least-square determination of the coefficients for the terms of the time-lag differential equation model and then fitting of the prognostic expression is made to 1951-1980 monthly rainfall datasets from Changsha station Results show that the model is likely to describe the nonlinearity of the an-nual cycle of precipitation on a monthly basis and to provide a basis for flood prevention and drought combating for the wet season.