Determining Probability Distribution For Streamflow Regions Using Partial-Moments
An attempt has been made to model the annual maximum streamflow, utilizing the guidelines in the regional
flood frequency analysis. The Partial L-moments (PL-moments) at several censoring levels are employed to estimate the
regional parameters of three extreme value distributions, namely; generalized extreme value (GEV), generalized logistic
(GLO) and generalized Pareto (GPA) distributions. A total number of 18 streamflow stations located throughout the eastern
region of Peninsular Malaysia were used as a case study. Firstly, the data is screening out for data verification and quality
control. Next, identification of homogeneous regions is made using homogeneity test based on PL-moments. The PL -
diagram is then constructed and GEV and GLO distributions appeared to be the acceptable distributions for representing the
regional data. However, it is relatively difficult to identify a particular distribution that most fitted the regional data.Thus,
goodness-of-fit test (Z-test) is used and the result showed that the most appropriate distribution for modeling maximum
streamflow in the East Coast of Peninsular Malaysia, based on PL-moments is the GLO distribution.
Keywords— Partial L-Moments, Frequency Analysis, Parameter Estimation, Censoring Level.