Statistical Modeling of Space-Time Data

by Charles T. Kufs, Roy F. Weston Inc, United States,

Document Type: Proceeding Paper

Part of: Critical Water Issues and Computer Applications


A common practice in hydrogeology is to draw contour lines to depict the occurrence of spatially distributed (i.e., regionalized) variables such as groundwater elevations, aquifer properties (e.g., transmissivity) or contaminant concentrations. Where the occurrence of such variables must be evaluated quantitatively, the statistical method known as trend-surface modeling can be used. Trend-surface modeling is a simple extension of multiple regression in which a spatially distributed variable is represented by an equation consisting of a constant, or error term, and algebraic functions of sample location coordinates. Optimized and innovative trend-surface modeling, as well as statistical dimension modeling is discussed.

Subject Headings: Optimization models | Statistics | Water pollution | Water resources | Errors (statistics) | Hydrologic models | Spatial distribution | Aquifers

Services: Buy this book/Buy this article


Return to search