Modeling Multistep Ahead Dissolved Oxygen Concentration Using Improved Support Vector Machines by a Hybrid Metaheuristic Algorithm

Dissolved oxygen (DO) concentration is an important water-quality parameter, and its estimation very for aquatic ecosystems, drinking water resources, agro-industrial activities. In the presented study, a new support vector machine (SVM) method, which improved by hybrid firefly algorithm–particle swarm optimization (FFAPSO), proposed accurate of DO. Daily pH, temperature (T), electrical conductivity (EC), river discharge (Q) DO data from Fountain Creek near Fountain, United States, were used model development. Various combinations T, EC, Q as inputs to models estimate The outcomes SVM–FFAPSO compared with SVM–PSO, SVM–FFA, standalone SVM respect root mean square errors (RMSE), absolute error (MAE), Nash–Sutcliffe efficiency (NSE), determination coefficient (R2), graphical methods, such scatterplots, Taylor violin charts. showed superior performance other methods in best each method was also assessed multistep-ahead (from 1- 7-day ahead) DO, superiority observed comparison. general recommend use modeling, this can be useful decision-makers urban planning management.