Absolute homogeneity test of Kelantan catchment precipitation series

Abstract

Along the Kelantan River in north east of Malaysia Peninsular, there are several areas often damaged by flood during north-east monsoon season every year. It is vital to predict the expected behavior of precipitation and river runoff for reducing flood damages of the area under rapid urbanization and future planning. Nevertheless, the accuracy and reliability of any hydrological and climate studies vary based on the quality of the data used. The factors causing variations on these data are the method of gauging and data collection, stations environment, station relocation and the reliability of the measurement tool affect the homogenous precipitation records. Hence in this study, homogeneity of long precipitation data series is checked via the absolute homogeneity test consisting of four methods namely Pettitt test, standard normal homogeneity test (SNHT), Buishand range test and Von Neumann ratio test. For homogeneity test, the annual rainfall amount from the daily precipitation records at stations located in Kelantan operated by Department of Irrigation and Drainage Malaysia were considered in this study. The missing values were completed using the correlation and regression and inverse distance method. The data network consists of 103 precipitation gauging stations where 31 points are inactive, 6 gauging stations had missing precipitation values more than five years in a row and 16 stations have records less than twenty years. So total of 50 stations gauging stations were evaluated in this analysis. With the application of the mentioned methods and further graphical analysis, inhomogeneity was detected at 4 stations and 46 stations are found to be homogeneous.