Rainfall is one of the most complex and difficult elements of the hydrology cycle to understand and to model due to the tremendous range of variation over a wide range of scales both in space and time. The complexity of the atmospheric processes that generate rainfall makes quantitative forecasting of rainfall an extremely difficult task. Thus, accurate rainfall forecasting is one of the greatest challenges in operational hydrology, despite many advances in weather forecasting in recent decades. In this article, a data set containing the monthly rainfall of Tamilnadu for a period of 136 years (1871 - 2006) was analysed and the prediction of annual rainfall of Tamilnadu was made through Box-Jenkins ARIMAmodel, traditional statistical time series forecasting model and some of the learning algorithms of artificial neural networks. The results are compared using the error measure MAPE. It is found that the Radial Basis Function learning algorithm gives better prediction when compared to the other models discussed in this paper.

Keywords: ARIMA, LeamingAlgorithm, MAPE.