Determination of return period for flood frequency analysis using normal and related distributions

Abstract

Flood frequency analyses are used to predict design floods for sites along a river. The technique involves using the observed annual maximum rainfall depths data to calculate statistical information such as mean values, standard deviations, skewness, and recurrence intervals. This research is to determine the return period for annual maximum rainfall depths data for two sites in Kemaman, which Jabor and Air Putih. All the data at two sites of study area are yearly data and the data of rainfall depths are taken starting from 1985 until 2015. The distributions used in this study are Normal, two-parameter Lognormal (LN(2)) and three-parameter Lognormal (LN(3)) distributions. The first step is to find the parameter estimates of rainfall depths in these two sites for each distribution. The parameters are estimated using the Method of Moments (MOM) and Probability Weighted Moments (PWM). Then, the best methods for each distribution in each site are obtained by using the performance measurements. Based on the best method for each distribution, the comparisons for the best method for each site are made. The results show that the best distribution that fits the characteristics of performance indicators for Jabor and Air Putih are two-parameter Lognormal distribution and three-parameter Lognormal distribution, respectively. Then, the last step is to determine the return period for the best distribution for 2, 5, 10, 50 and 100 years. Lastly, for the return period of 2, 5, 10, 50, 100 years, it can be concluded that Jabor has the higher annual maximum rainfall depths than Air Putih.