A Regional Frequency Analysis of Annual Maximum Rainfall in Terengganu using the Index Rainfall and L-moments Approaches

In data scarce regions, regionalization provides a means for estimating hydrological extremes. In this study, regional pooling of 1-, 3-, 5- and 10-day annual maximum rainfall from 1948 to 2013 at 32 stations across Terengganu, Malaysia is used in a standard regional frequency analysis using the well-known L-moments approach. The objectives of this study are (1) to estimate extreme rainfall in Terengganu, Malaysia using regional pooling of frequency analysis, (2) to establish a well-designed and accurate procedures of analysis of rainfall extremes in Terengganu using the regional frequency approach and (3) to improve the understanding of spatial rainfall characteristics as a main cause for design and planning practices in flood control. Cluster analysis is performed to determine the homogeneity of rainfall regions based on at-site characteristics. The cluster analysis has identified four rainfall regions based on homogeneity and heterogeneity tests using Monte Carlo simulations with regional average L-moment ratios fitted to the Kappa distribution. On the basis of the accuracy of the derived index rainfall quantiles, the regional rainfall frequency approach is found to be acceptable and had smaller uncertainty as compared to at-site estimates