Development Of An Integrated Gis Based System For Surface Water Quality Assessment And Management Gis Swqam

Download or read book Development of An Integrated GIS-Based System for Surface Water Quality Assessment and Management (GIS-SWQAM) written by Lu Fang and published by . This book was released on 2016 with total page 190 pages. Available in PDF, EPUB and Kindle. Book excerpt: It is an fact that surface water receives a large volume of pollutants from industrial, agricultural, and municipal sources. The adverse health and environmental effects of surface water pollution have been a major concern in environmental management. Water quality models are useful tools to simulate the complex transport and fate of pollutants in a water body and predict the short-term and long-term effects on water quality variation. The emergence of spatial information technologies, such as Geographic Information System (GIS) make it possible to assess and predict surface water quality with more details with respect to spatial information. The focuses of this thesis is to develop a comprehensive system named as GIS-SWQAM, which includes: (1) the development of a GIS-based water quality assessment system to assess the water quality and provide spatial distribution of water quality variables; (2) the development of an artificial neural network model to predict the change of water quality variables; (3) the development of a user interface that integrates the above models and functions; furthermore, a comparative analysis of the modeling approach developed in the GIS-SWQAM and the commercial model MIKE 21 was performed through field case studies. The GIS-based water quality and ecological risk assessment models (MWQ module for marine water quality assessment and LWQ module for lake water quality assessment) are developed by integrating a fuzzy risk assessment model, a eutrophication risk assessment model, a heavy metal risk assessment model, a dynamic database, the ArcGIS Engine, and a graphical user interface (GUI). The assessment results are both spatially and visually presented in the form of contour maps and color-coded maps that indicate risk levels. A large amount of data with both spatial and temporal distributions is managed by the developed system and analyzed by the assessment modules. The developed MWQ and LWQ modules are respectively applied in the Liaodong Bay of China and Lake Champlain. The MWQ and LWQ produce risk maps that depict the spatial distribution of integrated water quality index values, eutrophication risk levels and heavy metal risk levels in the study area. The maps generated can provide a better understanding of the distribution of the water quality and ecological risk levels. The primary factors that affect the water quality are subsequently examined using the visualized results. An artificial neural network model with the back-propagation algorithm (BPANN) is first developed using Matlab to predict the chlorophyll-a concentration in Lake Champlain. Then, the algorithm of the BPANN model is built using the C# programing language and integrated with GIS and the database to build the ANN module, which is applied to predict the total phosphorus concentration in Lake Champlain. The best performing model is determined among the results of models built with different combination of input variables, which are preliminarily selected by linear correlation analysis and domain knowledge. Subsequently, the performances of the BPANN models are validated by a new set of field data. Similar to the MWQ and LWQ modules, the ANN module also produces the spatial distribution maps of the predicted concentrations; errors made during the prediction are presented in the user interface. The results indicate that the developed BPANN models can provide acceptable prediction results and can be used to provide a quick modeling assessment of water quality variation for managers. In this thesis, the MIKE 21 FM software is also used to establish a hydrodynamic model coupled with a transport model to simulate the total phosphorus concentration in Lake Champlain. A comparative analysis is performed between the results of the MIKE 21 model and the BPANN model. The results of the MIKE 21 model are acceptable, but not as good as that of the BPANN model. This further verifies that the developed BPANN model is a reliable tool to assess the lake eutrophication and to help managing lake water quality. The developed system can be also applied to surface water management in other area.