Assessment of frost events on the tropical highlands: Analyzing meteorological conditions and wrf-arw model performance on the Dieng plateau, Indonesia

Abstract

This study investigates frost formation on the Dieng Plateau and evaluates the Weather Research and Forecasting model with Advanced Research WRF core (WRF-ARW) in simulating these events. The primary goal is to identify atmospheric conditions leading to frost and assess the performance of two parameterization schemes within the WRF-ARW model: the Tropical Physics Suite (TPS) and a scheme tailored for tropical frost prediction, referred to as the Bussoni scheme. Meteorological data from the Batur Automatic Weather Station were analysed for eight frost events between August and September 2023. Key parameters such as surface air temperature, dew point, relative humidity, and wind patterns were examined. The study identified critical thresholds for frost occurrence during night-time, including temperatures near or below 4°C, dew points close to air temperature, relative humidity above 95%, and wind speeds below 2 m/s. Comparative analysis revealed that the Bussoni model marginally outperforms the TPS in simulating temperature and dew point, indicated by lower Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values, and a higher Index of Agreement. The TPS, however, showed a slight edge in predicting relative humidity and wind speed, with lower RMSE and higher correlation coefficients. Both models faced challenges in accurately predicting wind direction, highlighted by high RMSE and low correlation, which underscores the complexity of modelling wind patterns in mountainous terrain.