Linear Regression-based Time-series Prediction of Total Suspended Solids in the Day River Basin, Vietnam

Forecasting the concentration of Total Suspended Solids (TSS) plays a critical role in the monitoring and management of surface water quality, particularly in regions exposed to agricultural, domestic, and industrial activities. In this study, observed TSS data from January to October at Tan Lang floating bridge, located in the Day River Basin, were employed to construct and evaluate a predictive model for TSS concentrations in November and December. Based on the time series collected from January to October, a linear regression model was established, with time (expressed as the number of days from the initial measurement) as the independent variable and the corresponding observed TSS values as the dependent variable. After calibration, the model achieved a high coefficient of determination (R² ≈ 0.94), indicating a strong linear relationship between time and TSS variation. This result demonstrates the potential of the model to provide relatively reliable forecasts for subsequent time points. The two unmonitored dates, November 15 and December 8, were predicted by the model with TSS values of 13.92 mg/L and 12.29 mg/L, respectively. Keywords: Prediction of total suspended solids, Linear Regression, Day River Basin, Vietnam.