[Establishment of the predictive model of source eutrophication using artificial neural network]

Area of Science:

• Environmental Science
• Water Resource Management
• Artificial Intelligence in Environmental Monitoring

Background:

• Eutrophication poses a significant threat to water quality in urban water sources.
• Monitoring key water quality parameters is crucial for assessing trophic states.

Purpose of the Study:

• To develop a predictive model for eutrophication in Zhengzhou City's main water sources.
• To evaluate the trophic state of Xiliu Lake and Huayuankou Pool.

Main Methods:

• Monitoring of water temperature, secchi depth, total phosphorus, total nitrogen, light illuminance, chemical oxygen demand, and chlorophyll-a.
• Application of grading points and comprehensive trophic state index methods for trophic state evaluation.
• Development of a eutrophication forecasting model using a backpropagation artificial neural network with the Levenberg-Marquardt algorithm.

Main Results:

• Both water sources were identified as being in a nutritional state, with Xiliu Lake showing a trend towards eutrophication.
• The artificial neural network model achieved a training error of 1e-11 and a coefficient of correlation of 0.871.
• A successful predictive model for eutrophication in Zhengzhou's main water resources was established.

Conclusions:

• Artificial neural networks are effective for establishing eutrophication forecasting models.
• The artificial neural network approach is well-suited to meet the demands of water resource management and eutrophication prediction.