ABSTRACT

The aim of this project is to construct a mathematical model for the rainfall data of Onne which could be used to predict reliable and dependable future rainfall values. Splecifically, it was intended to estimate and isolate the components of time series data namely: Trend, seasonal variation, cyclical variation and irregular component present in the data using the descriptive· time series approach. This estimation and isolation was done based on the original data. The study covered the period of 1989 to 1997 while 1998 data was removed so that it can be compared with the predicted values. The data was collected from IITA Onne in Rivers State. The results of the analysis showed amongst others that: (1) All the components of time series were present in the data. (ii) On the average, the rainfall of Onne within-the period under consideration is 203.14mm with a standard deviation of 154.87mm. Based oh the results of the analysis, it was suggested that a probability model be filled to the data in other to obtain a more sound result. Also, observations of more than 20 years should be analysed so that the effect oc cycles will be greatly felt.