Stochastic Modelling of Rainfall in Humid Region of North East India

Rainfall is one of the fundamental components of the hydrological cycle as its accurate estimation is necessary for planning, designing and operation of water resources development programmes. In the present study, monthly stochastic model was developed using rainfall data for Doimukh (Itanagar), Arunachal Pradesh, India. The rainfall series was assumed to be composed of deterministic and stochastic components. The trend component was found to be non-significant. Fourier series analysis was used to identify the periodic component. Three harmonics were found significant. The stochastic component was modeled by fitting into auto regressive model of order 6. The Mean and standard deviation of the generated series were found to be close to the historical values. The value of absolute error, relative error, correlation coefficient and Nash–Sutcliffe coefficient indicated a high degree of model fitness to the observed data series. Copyright Springer Science+Business Media B.V. 2008