<![CDATA[Rainfall constitutes a fundamental climatic factor with a crucial influence on hydrological processes, agricultural production, and the mitigation of hydrometeorological hazards. In West Java Province, rainfall exhibits substantial variability and a strong annual seasonal cycle, largely driven by intricate monsoon systems and diverse topographical conditions. These characteristics necessitate the application of a reliable modeling framework capable of accurately representing seasonal behavior. This study seeks to analyze and forecast monthly rainfall in West Java Province using the Seasonal Autoregressive Integrated Moving Average (SARIMA) approach, with particular emphasis on assessing its performance in a region marked by pronounced spatial and temporal rainfall variability. Monthly rainfall observations obtained from the Indonesian Agency for Meteorology, Climatology, and Geophysics (BMKG) for the period 1995–2023 were employed in the analysis. Data preprocessing involved the application of a Box–Cox transformation to stabilize variance, followed by seasonal differencing with a 12-month lag to address seasonal non-stationarity. Model identification was conducted through examination of the Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF), while model selection relied on the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). The findings reveal that the selected SARIMA model successfully represents the dominant annual monsoonal rainfall pattern in West Java, as supported by residual diagnostic tests indicating satisfactory statistical adequacy. Forecasting results further show that the predicted rainfall values closely track historical dynamics, with forecast uncertainty increasing as the prediction horizon extends. Overall, this study underscores the ongoing relevance of SARIMA as a baseline forecasting model in monsoon-influenced regions and offers practical insights for seasonal water management, flood and drought risk reduction, and the enhancement of rainfall-based early warning systems in West Java Province]]>
Copyrights © 2026
