Food security in India is tightly linked to rainfall variability. Trends in Indian rainfall records have been extensively studied but the subject remains complicated by the high spatio-temporal variability of rainfall arising from complex atmospheric dynamics. In the present study, the rainfall data of 35 years (1978-2012) for Hatti has been analyzed. The rainfall characteristics were analyzed on annual, seasonal, monthly and weekly basis. The onset, withdrawal and length of rainy season were determined. The dry and wet spell probabilities on weekly basis for rainy season period were determined by Markov-chain process which will be helpful in crop planning, adoption of in-situ moisture conservation practices, knowing the need of supplementary irrigation and predicting potential weeks for water harvesting. The minimum assured weekly rainfall was derived by using incomplete gamma distribution. The one day and upto five consecutive days maximum rainfall for various return periods were obtained using different probability functions which will be helpful in designing soil and water conservation and hydraulic structures.