<![CDATA[This study aims to predict rainfall in Bengkulu Province for January 2024 using the Robust Kriging method, an advanced geostatistical approach designed to handle outliers and non-ideal spatial characteristics. The novelty of this study lies in integrating Robust Kriging with a Gaussian semivariogram for short-term rainfall prediction in Bengkulu Province. This combination has not been explored in previous hydrometeorological studies. Rainfall data were obtained from the Meteorology, Climatology, and Geophysics Agency (BMKG) and analysed to identify spatial dependency and variation. The analysis began with descriptive statistics, assumption testing, and outlier detection, followed by the construction of robust empirical and theoretical semivariogram models. Three semivariogram models, Spherical, Exponential, and Gaussian, were compared to determine the most suitable model based on Mean Squared Error (MSE) and Mean Absolute Percentage Error (MAPE) values. The results indicate that the Gaussian model produced the smallest MSE and MAPE values, showing the best fit to the empirical semivariogram. The Robust Kriging interpolation generated spatial predictions of rainfall intensity across Bengkulu, showing higher rainfall in the north and lower rainfall in the south. The findings demonstrate that Robust Kriging effectively improves prediction accuracy by minimizing the influence of outliers and optimizing spatial weighting. These results provide valuable insights for water resource management, agricultural planning, and hydrometeorological disaster mitigation in Bengkulu Province.]]>
Copyrights © 2025
