Neural network modeling of salinity variation in Apalachicola River

Salinity is an important indicator for water quality and aquatic ecosystem in tidal rivers. The increase of salinity intrusion in a river may have an adverse effect on the aquatic environment system. This study presents an ap plication of the artificial neural network (ANN) to assess salinity variati on responding to the multiple Forcing functions of freshwater input, tide, and wind in Apalachicola River, Florida. Parameters in the neural network m odel were trained until the model predictions of salinity matched well with the observations. Then, the trained model was validated by applying the mo del to another independent data set. The results indicate that the ANN mode l is capable of correlating the non-linear time series of salinity to the m ultiple forcing signals of wind, tides, and freshwater input in the Apalach icola River. This study suggests that the ANN model is an easy-to-use model ing tool fbr engineers and water resource managers to obtain a quick prelim inary assessment of salinity variation in response to the engineering modif ications to the river system. (C) 2001 Elsevier Science Ltd. All rights res erved.