Relational Encoding of Simplex and Quadtree Data Models

Stephen C. Smyth

Many tools for spatial data retrieval and manipulation available to geologists are "blunt instruments." This is true primarily because the spatial structure of the stored objects is not modeled correctly. Spatial structure is very important because it embodies a strict set of relationships that must hold among individual spatial objects and their component elements. These relationships (topology and geometry) uniquely determine how physical objects can be assembled and located in space. Poor treatment of geometry and topological properties results in unnecessary complexity, inefficiency, and often disaster.

A complete (any physically realizable object can be modeled), consistent (any modeled object could be physically realized), and efficient (not much more time or space is required than is theoretically necessary) spatial data-base management system can be constructed using a simplex data structure with a quadtree (or octtree, in three dimensions) index. The simplex and quadtree models can be readily encoded in the data-definition and data manipulation languages (such as the ANSI standard language SQL) that accompany relational data-base management systems.

AAPG Search and Discovery Article #91038©1987 AAPG Annual Convention, Los Angeles, California, June 7-10, 1987.