Search

Accurate quantification of dissolved oxygen (DO) is critically important for managing water resources and controlling pollution. Artificial intelligence (AI) models have been successfully applied for modeling DO content in aquatic ecosystems with limited data. However, the efficacy of these AI models in predicting DO levels in the hypoxic river systems having multiple pollution sources and complicated pollutants behaviors is unclear. Given this dilemma, we developed a promising AI model, known as support vector machine (SVM), to predict the DO concentration in a hypoxic river in southeastern China. Four different calibration models, specifically, multiple linear regression, back propagation neural network, general regression neural network, and SVM, were established, and their prediction accuracy was systemically investigated and compared. A total of 11 hydro-chemical variables were used as model inputs. These variables were measured bimonthly at eight sampling sites along the rural-suburban-urban portion of Wen-Rui Tang River from 2004 to 2008. The performances of the established models were assessed through the mean square error (MSE), determination coefficient (R 2 ), and Nash-Sutcliffe (NS) model efficiency. The results indicated that the SVM model was superior to other models in predicting DO concentration in Wen-Rui Tang River. For SVM, the MSE, R 2 , and NS values for the testing subset were 0.9416 mg/L, 0.8646, and 0.8763, respectively. Sensitivity analysis showed that ammonium-nitrogen was the most significant input variable of the proposal SVM model. Overall, these results demonstrated that the proposed SVM model can efficiently predict water quality, especially for highly impaired and hypoxic river systems.

In the context of water resource management and pollution control, the characterization of water quality impairments and identification of dominant pollutants are of critical importance. In this study, water quality impairment was assessed on the basis of 7 hydrochemical variables that were monitored bimonthly at 17 sites in 2010 along the rural-suburban-urban portion of the Wen-Rui Tang River in eastern China. Seven methods were used to assess water quality in the river system. These methods included single-factor assessment, water quality grading, comprehensive pollution index, the Nemerow pollution index, principle component analysis, fuzzy comprehensive evaluation, and comprehensive water quality identification index. Our analysis showed that the comprehensive water quality identification index was the best method for assessing water quality in the Wen-Rui Tang River due to its ability to effectively characterize highly polluted waters with multiple impairments. Furthermore, a guideline for the applications of these methods was presented based on their characteristics and efficacy. Results indicated that the dominant pollutant impairing water quality was total nitrogen comprised mainly of ammonium. The temporal variation of water quality was closely related to precipitation as a result of dilution. The spatial variation of water quality was associated with anthropogenic influences (urban, industrial, and agriculture activities) and water flow direction (downstream segments experiencing cumulative effects of upstream inputs). These findings provide valuable information and guidance for water pollution control and water resource management in highly polluted surface waters with multiple water quality impairments in areas with rapid industrial growth and urbanization.