Using Artificial Neural Network Models for Eutrophication Prediction

Artificial neural network (ANN), a data driven modeling approach, is proposed to predict the water quality indicators of Lake Fuxian, the deepest lake of southwest China. To determine the non-linear relationships between the water quality factors and the eutrophication indicators, several ANN models was chosen for the investigation. A commonly used back-propagation neural network model was used to relate the key factors that influence a number of water quality indicators such as dissolved oxygen (DO), total phosphorus (TP), chlorophyll-a (Chl-a), and secchi disk depth (SD) in Lake Fuxian. The measured data were fed to the input layer, representing forcing functions to control the in- lake bio-chemical processes. Eutrophication indicators such as DO, TN, Chl-a and SD were represented in the output layers. The results indicated that the back-propagation neural network model performs good in ten months prediction and the neural network is able to predict these indicators with reasonable accuracy. This study also suggested that the neural network is a valuable tool for lake management.