Search

The availability of a long and comprehensive rainfall record is crucial for the successful completion of a hydrological study. Real observed rainfall data, however, are frequently insufficient or absent and typically contain numerous missing data. In order to generate high-quality data, an appropriate method is needed. Hence, this study presents a comparative analysis of two methods, namely spatial interpolation and probabilistic methods for the estimation of missing rainfall data. The aim is to determine the best estimation method that can replace missing rainfall data in the study region. The methods are illustrated using 10 rainfall stations in Pahang, from year 2011 to 2020. This research employs different available methods, namely normal ratio, arithmetic average, inverse distance weighting and coefficient of correlation weighting (spatial interpolation method), and gamma distribution (probabilistic method). Results are to be compared and evaluated based on mean square error, root means square error, similarity index, and correlation coefficient values. The study shows that for the spatial interpolation method, the inverse distance weighting gives good predictions. When comparing a gamma distribution (probabilistic method) and the inverse distance weighting (spatial interpolation method), the inverse distance weighting (spatial interpolation method) gives a good performance in estimating missing rainfall data. Studies on missing rainfall are important to produce high-quality data for hydrological and water management purposes. [ABSTRACT FROM AUTHOR]