Document Type : Original Article

Authors

• Zahra Sheikholeslami ¹
• Majid Ehteshami ²
• Zeinab Ghasemi ³

¹ PhD of Civil and Environmental Engineering KN Toosi University of Technology

² Associate Professor of Department of Civil Engineering, K.N. Toosi University of Technology

³ PhD Candidate of Environmental Engineering, Department of Civil and Environmental Engineering University of Auckland, Auckland, New Zealand

Abstract

The monthly flow rates of the Chalus River in Mazandaran Province, Iran are predicted using the Seasonal Autoregressive Integrated Moving Average (SARIMA) model in this research. The SARIMA model was created and verified with MiniTab software by analyzing historical data spanning from 2006 to 2023. The modeling process involved checking data stationarity with the Augmented Dickey-Fuller (ADF) test, normalizing data using the Johnson Transformation, and determining the best SARIMA parameters by analyzing Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) plots. The SARIMA model with parameters (2,0,0)(0,1,1)12 was determined to be the most precise in predicting future outcomes, exhibiting a strong R² value and reliable forecasting capabilities. Despite effectively modeling the seasonal changes of the Chalus River, the model proved to be inadequate in predicting extreme flow rates. The findings indicate that utilizing the SARIMA model proves to be a dependable instrument for overseeing water resources in the area, with potential for further investigation into integrating SARIMA with alternative approaches to improve forecasting of exceptional occurrences.

Keywords

• SARIMA
• Prediction
• River Flow Rate
• ACF
• PACF
• Time Series