of Environmental Hydrology
Journal of the International Association for Environmental Hydrology
JEH Volume 11 (2003), Paper 8, July 2003 Posted July 31, 2003
INVERSE MODELING TO IDENTIFY NONPOINT SOURCE POLLUTION USING A NEURAL NETWORK, TAIHU LAKE WATERSHED, CHINA
College of Water Resources and Environment, Hohai University, Nanjing, China
Various studies have been carried out for the evaluation of non-point source pollution using physically-based distributed hydrological and water quality models. A number of modeling interactions have been developed using remote sensing, geographical information systems, best management practices, decision support systems, and water quality modeling tools, for the identification and quantification of non-point sources of pollution in the watersheds of lakes and rivers. In the most recent artificial neural network applications, an intelligence system is commonly used for the management of the dynamic and complex nature of watersheds. In this paper, the author has proposed an inverse modeling approach, using artificial intelligence for the identification of non-point source pollution based on pollution indicators in storm water and agriculture runoff. The study is carried out in the Xishan county sub basin of the Taihu Lake watershed. A back propagation neural network model has assisted the development of an inverse modeling system to identify pollution sources beyond the presence of pollution indicators.
Reference: Zaheer, I. and G. Cui; Inverse Modeling to Identify Nonpoint Source Pollution Using a Neural Network, Taihu Lake Watershed, China, Journal of Environmental Hydrology, Vol. 11, Paper 8, July 2003.
College of Water Resources and Environment
China E-mail: email@example.com
|Return to HydroWeb Homepage|