Author(s): Matthew O. Ward and Junwen Zheng, Computer Science Department, Worcester Polytechnic Institute, Worcester, MA 01609

Source: Proc. 1993 GIS/LIS Conference, November, 1993.

Abstract: Maps and other forms of geographic visualization rarely convey to users the quality or reliability of the information being presented. However, decisions which are made based on the analysis of spatio-temporal data may be seriously flawed if the quality of the data is not taken under consideration. This paper presents a technique for visualizing the quality of data which has both spatial and temporal attributes. We use a model of quality which incorporates both characteristics of the data gathering equipment as well as the model used to interpolate data at arbitrary locations in space and time. The resulting visualization system allows users to map five-dimensional data to five graphical attributes, where each attribute may be displayed in one of three modes: continuous, sampled, or constant. We show examples using US EPA data on dissolved inorganic nitrogen (DIN) concentrations in the Chesapeake Bay over a several year period.