Decision Support System for Non−Point Source Pollution Control

Overview This paper presents the development of a Graphical User Interface (GUI) to a Decision Support System (DSS) for NonPoint Source (NPS) pollution control on watershed scales. The GUI and all components of the DSS are entirely embedded in ERDAS IMAGINE and the system is launched from the main toolbar. The system uses IMAGINE's GIS functions for data storage and pre−processing, SML for hydrologic simulation (finite differences) and the Knowledge Engineer for spatial decision−making processes. The GUI, written in EML, ties these components together and permits interactive selection of study watershed, pollutant of concern and model, interactive input of design meteorological parameters and integrated support for data pre− and post− processing commonly used in NPS pollution analyses. The integrated system is expected to be of tremendous instructional value for the graduate engineering course ENBE 633 "NonPoint Source Pollution Control" and also represents the next generation of tools needed by land managers and planners to evaluate the effects of past and future land uses on water quality. The demand for this type of tool is expected to significantly increase during upcoming years because of growing public concern over the environment. Non−Point Source (NPS) pollution, that which emanates intermittently from spatially distributed sources as a result of land use and meteorological events, is a major contributor to the degradation of surface water quality in the US and worldwide (Novotny and Olem, 1994). This type of pollution can be controlled by the adoption of improved land surface management methods (Best Management Practices, BMPs) which essentially prevent potential pollutants from leaving the land. The process of identifying appropriate BMPs typically involves 6 steps: 1) select an area of interest (typically a watershed); 2) gather all pertinent spatial data about this area; 3) identify pollutant(s) of concern in this zone; 4) identify critical sources (hot spots) of the pollutant(s) in the study area; 5) select appropriate BMPs for each hot spot and 6) evaluate the reduction in pollution resulting from application of the selected BMPs. Performing these steps has historically been rather difficult because of the wide land base over which NPS pollution is generated and the complex transport processes that it undergoes prior to reaching surface water bodies. In recent years, researchers have started to combine Geographic Information Systems (GIS), distributed parameter hydrologic models and Expert Systems (ES, or other forms of Artificial Intelligence (AI)) to aid in the analysis of NPS pollution and …