RAINFALL ANALYSIS AND FORECASTING USING THE PROPHET METHOD ON TIME SERIES DATA

Ni Komang Sintya Dewi
Adie Wahyudi Oktavia Gama

Multidiciplinary Output Research for Actual and International Issue (Morfai Journal)
Vol. 6 No. 4 (2026): Multidiciplinary Output Research For Actual and International Issue

Climate change has increased rainfall variability, making it more difficult to predict rainfall patterns in terms of intensity, duration, and spatial distribution. This study aims to develop a daily rainfall forecasting model using the Prophet method, which is capable of handling seasonal patterns and long-term trends in time series data. The data used consist of daily rainfall records from 2015 to 2025 across nine regions in Bali Province, obtained from the NASA POWER platform. The research methodology includes data collection, data preprocessing, exploratory data analysis (EDA), Prophet model development with parameter optimization, cross-validation, and forecasting. Model performance is evaluated using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) metrics on test data. The results indicate that the Prophet method is capable of effectively modeling seasonal patterns and rainfall trends, producing stable predictions for future periods. This forecasting system is expected to serve as a decision-support tool in agriculture, water resource management, and hydrometeorological disaster mitigation.