Rainfall is the most critical and key variable both in atmospheric and hydrological cycle. Its patterns usually have spatial and temporal variability. Its variability is assumed to be the main cause for the frequently occurring climate extreme events such as drought and flood. In this study, spectrum analysis, cross-spectral analysis as well as seasonal auto-regressive integrated moving average (SARIMA) model were used in intending to predict the pattern and its extreme event frequency of rainfall in Dire Dawa Region based on data obtained from Dire Dawa and adjacent stations: Dengego and Haramaya. The result indicated that the amount of rainfall at Dengego and Haramaya are more or less the same on average in all seasons and much higher than that of Dire Dawa during last 30 year study period.