Research Article An Architecture for the Management of GIS Data Files Gregory Vert Molly Stock Department of Computer Science Department of Forest Resources University of Idaho University of Idaho Piotr Jankowski Paul Gessler Department of Geography Department of Forest Resources University of Idaho University of Idaho Abstract Management of GIS data presents unique problems. These include the need to manage a wide range of differing applications' data formats, group files into meaningful organizations that are driven by geospatial concerns, provide lineage and version support, provide documentation of the nature of the data and other descriptive information, and keep track of relationships among data files and groups of data files. Current approaches to data management can provide partial solutions to these problems, but none is complete. We have developed a software architecture that addresses all levels of this problem. This architecture centers around a data model that is populated with metadata and extended with fuzzy set theory notation. Fuzzy set theory allowed us to solve data ambiguity problems and to develop an object-oriented framework that served as the basis for a proof-of- concept prototype. In addition, we developed file system organizations, computer network communications strategies and the mechanics of basic system operation so that the architecture provides a blueprint for a new type of software. 1 Introduction The University of Idaho College of Natural Resources manages an area called the Experimental Forest. This forest encompasses 7300 acres of land in multiple non- contiguous tracts. The land for the forest was acquired in a series of purchases and donations starting in 1932. It provides an ideal environment for research on various Transactions in GIS, 2002, 6(3): 259±275 ß 2002 Blackwell Publishers Ltd, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA. Address for correspondence: Molly Stock, Department of Forest Resources, University of Idaho, Moscow, ID 83844-1133. E-mail: mstock@uidaho.edu