Probability Structure and Return Period of Multiday Monsoon Rainfall

The daily monsoon rainfall data recorded at Subang Airport, Malaysia, from 1960 to 2011 is examined in terms of probability structure for the estimation of extreme daily rainfall precipitation during the Northeast (NE) and Southwest (SW) Malaysian monsoons. The discrete autoregressive and moving average [DARMA(1,1)] model is preferable to the first-order Markov chain [DAR(1)] model. The conditional probabilities of t consecutive rainy days are time dependent. Nevertheless, a simple two-parameter gamma distribution appropriately fits the frequency distribution of multiday rainfall amounts. An algorithm is developed by combining the DARMA(1,1) and gamma models to estimate the return period of multiday rainfall. Extensive comparisons showed that the DARMA(1,1)-gamma model gives a reliable estimate of the return period of rainfall for both NE and SW monsoons at Subang Airport. Furthermore, values generated from the models enable the analysis of the frequency distribution of extreme rainfall events. (C) 2015 American Society of Civil Engineers.