QUERYTOOL: DESIGN, IMPLEMENTATION AND APPLICATIONS

Abstract

This article describes the Querytool, a platform for querying multiple geographic data sets and associated thematic (legal) data available at the Cadastre in one integrated system. The system is nationwide and available for analyzing and performing consistency checks on the Cadastral data. It is designed to support ad hoc queries covering both the spatial and the thematic part of the data. The system is based on two important components: a relational DBMS and a (geo)graphic user interface for comfortable query formulation and showing the results. The usefulness of the tool, is shown by three quite different (implemented) cases from the real world, each illustrating different aspects of the system. These applications are based, among others, on the following generic concepts: spatial aggregates, historic spatial data, spatial join, integrated geometric and thematic data, SQL and shell scripts, using external spatial data, and spatial joins involving topologically structured data. As the relational DBMS can not execute the spatial join with topologically structured data, this is implemented in the (geo)graphic user interface. In this article the integration of multiple geographic data sets and associated thematic (legal) data in one database is described. The geographic data sets consists of large scale topographic data and the cadastral maps of the different provinces. Associated with the cadastral maps is thematic (legal) data, also organized per province. The relationship between the parcels on the cadastral map and the thematic data is through the nationwide unique parcel numbers. The cadastral maps are based on a topological structured model and manipulating area features in such a model requires navigation using the references to the boundaries. In addition, the topographic maps and the cadastral maps contain the full (update) history since the introduction in 1997. This is not (yet) the case for the thematic (legal) data. In the Dutch Cadastre such a database as described above has been set up. It is nationwide and available for analyzing and performing consistency checks on the Cadastral data. The goal being to improve the quality of the source data. The purpose is to create an environment with easy access to all the data. Therefore a user friendly interface is needed on top of this database. This was realized by using the standard GIS package GEO++ together with custom made add-on’s. GEO++ can best be described as a general Querytool for relational databases with spatial extensions, similar to the OpenGIS SFS/SQL (Buehler and McKee, 1998, Open GIS Consortium, Inc., 1998). The custom made add-on’s are used for: easier access to data: just one button, instead of querying 4 tables; analysis not possible in a relational DBMS:e.g. intersection of a topologically structured area feature with a polyline; and introduction of new interface concepts:e.g. the ’active set’. In summary the Querytool can be used for easy querying, analysis and visualization of data in one nation-wide database. Besides more or less complicated ad-hoc queries which can be dealt with in the (geo)graphic user interface, either the standard GEO++ part or one of the add-on parts, the Querytool database has been used in many applications. Section 2 gives a short overview of the Querytool system including a description of the computed spatial aggregates, sizes of the geometric and thematic data sets, and some remarks how good performance is obtained. The first application ’producing bills based on changes in the topographic map’ (Section 3) uses spatial aggregates (municipalities), historic data and the spatial join. The second application ’deriving zipcode map from cadastral data’ integrates geometric data (cadastral map) and thematic data (addresses) and uses a combination of SQL and shell scripts to derive the zipcode map; see Section 4. The third application, described in Section 5, ’quality improvement of the registration of legal notifications’ again integrates geometric and thematic data (legal notifications related to parcel), but also uses a spatial join between topologically structured parcels and linear pipe lines (data from third party). As the relational DBMS can not execute the spatial join with topologically structured data, this is implemented in the (geo)graphic user interface. The paper concludes with a short list of other implemented applications and some future developments in Section 6.