Past Awardee

The Development of Point-to-Zone Pattern Learning (P2Z) for Groundwater Recharge

Yu-Feng Lin
Yu-Feng Lin

College: Illinois State Water Survey
Award year: 2006-2007

Recharge and discharge (R/D) rates define the relationships between groundwater, precipitation, and surface water that constrain management options for water supply. R/D rates and patterns result from of a set of complex, uncertain processes that generally are difficult to study. There is currently no single method capable of estimating R/D rates and patterns for all practical applications. Therefore, cross analyzing results from various estimation methods might be a more adequate approach than using only a single estimation method. The Illinois State Water Survey (ISWS) has been leading recent research in R/D estimation by pattern recognition using numerical methods and image processing algorithms. In the collaboration with federal and multi-state research institutes, we have created a set of software tools that can help hydrogeologists to estimate groundwater R/D in a more efficient way than conventional methods. Feedbacks from the Beta-test users of this new system indicate a great desire for decision support systems in addition to the original subjective pattern recognition procedure. We propose a joint research project to develop the software that provides quantified pattern learning based on the algorithms developed by the Image Spatial Data Analysis Group at NCSA. The quantifiable reliability indices that result from pattern learning algorithms will bridge the gap between the traditional subjective zonation R/D estimations and advanced stochastic and uncertainty analysis. The software will be tested against a field test site in Wisconsin which has been intensively studied by many research institutes and universities. The final product developed in the proposed research will be applied immediately to several current active groundwater studies in northeastern Illinois. The results of the proposed research will be accessible and beneficial across the hydrologic sciences, with potential uses including water resources planning and management, the evaluation of groundwater development alternatives, and the modeling of the fate and transport of environmental contaminants. The proposed research could be the seed for significant advances in hydrology, geology, environmental science, and geography because it will create a tool that will enable hydrogeologists to recognize R/D patterns objectively and compare R/D estimations from various methods. This project will disseminate and promote the use of NCSA technologies for image spatial data analysis and user-friendly decision-making systems, and build the technological capabilities of the ISWS. This proposed project will be developed by the joint efforts of graduate students and researchers at the Department of Computer Sciences, NCSA, and the ISWS.