Rainfall analysis in the Indian Ocean by using 6-States Markov Chain Model

In this study, we used Markov chain approach to analyze rainfall dataset from one buoy (4N90E) in the eastern Equatorial Indian Ocean. This study aims to determine the opportunity for transition (displacement) of daily rainfall intensity, where there are six states or conditions of rainfall intensity, i.e. no rain, very weak, weak rain, moderate rain, heavy rain, and very heavy rain. The Markov Chain method used is the Chapman-Kolmogorov equation and the steady state equation. The investigation of the 6-states in Markov chain model show that dynamic probability of transition state for rainfall data is reflexive properties majority. By using the model, it is concluded that the transition rate matrix of the largest transition probability in the area of 4N90E occurs at the transition from state-1 to state-1 as much is 0.72 and state-2 to state-2 is 0.60. The transition probability value becomes 0.5184 and 0.36 for the same state dominant of two periods P-2. The use of the 1st order Markov chain is better than 2nd order.