EFFECTIVENESS OF NUMERICAL WEATHER PREDICTION USING RAPID UPDATE CYCLE (RUC) AND IMPLEMENTATION IN DATA ASSIMILATION SYSTEM

• Main Author: Putu Ferry Wistika, I
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/84283
• Institution: Institut Teknologi Bandung

This study evaluates the effectiveness of the Rapid Update Cycle (RUC) method with a 1-hour update cycle (Cycle 1) in improving weather prediction accuracy in Bali Province. The results indicate that Cycle 1 has the lowest Root Mean Square Error (RMSE) for temperature and rainfall compared to other cycles (3, 6, 12 hours) and the model without data assimilation (Non DA). This method has proven to capture temperature and rainfall variability more accurately, closely aligning with actual observational data. Data assimilation from Automatic Weather Stations (AWS) significantly enhances the reliability of the prediction model, allowing for more accurate adjustments to actual atmospheric conditions. Further analysis shows that Cycle 1 provides more precise and accurate predictions regarding wind speed bias and spatial reflectivity patterns. The model successfully reproduces more detailed and accurate rainfall and wind speed distribution patterns, as indicated by BMKG radar data and moisture flux transport. The improved reliability of predictions enhanced by the RUC Cycle 1 method offers significant benefits in mitigating the impacts of extreme weather in Bali, especially in the tourism sector. This study confirms that more frequent data assimilation through the RUC method provides significant advantages in producing more accurate and consistent short-term weather predictions.