Predicting BOD under Various Hydrological Conditions in the Dongjin River Basin Using Physics-Based and Data-Driven Models

The water quality of the Dongjin River deteriorates during irrigation period because supply river maintenance to main is cut off by mass intake agricultural weirs located in midstream regions. A physics-based model and a data-driven were used predict under various hydrological conditions. Hydrological Simulation Program–Fortran (HSPF), which model, was constructed simulate biological oxygen demand (BOD) Basin. Gamma Test derive optimal combinations observed variables, including external inflow, intake, rainfall, flow rate, for non-irrigation periods. adaptive neuro-fuzzy inference system (ANFIS) then built using these results. ANFIS this study capable predicting BOD from data periods, without running model. predicted results have high confidence levels when compared with data. Thus, proposed method can be reliable rapid prediction only monitoring as input.