Web geoprocessing services on GML with a fast XML database

Abstract

Nowadays there exist quite a lot of Spatial Database Infrastructures (SDI) that facilitate the Geographic Information Systems (GIS) user community in getting access to distributed spatial data through web technology. However, sometimes the users first have to process available spatial data to obtain the needed information. The geoprocessing services can be provided over the web using the conventional databases (such as the relational databases or object-oriented databases) as backend, though this causes a concrete problem of overhead of data conversion. In this research, we proposed an approach to provide the web geoprocessing services, using an XML (eXtensible Markup Language) database system as back-end and the Geography Markup Language (GML) as data encoding standard. Currently there is not yet a formal standard query language for GML, we demonstrate that the XML Query (XQuery) language can be extended with spatial semantics to carry out spatial computation upon GML data. A scenario was chosen, namely the assessment of soil erosion caused by rainfall, to apply the proposed approaches. After describing and analysing the requirements of an assessment of soil erosion caused by rainfall, we proposed a suitable system prototype design combining the Model View Controller (MVC) architectural pattern with Service-Oriented Architecture (SOA) principles. To add robustness and flexibility to the system, the implemented web geoprocesses were provided through Web service orchestration. Such system prototype can support the decision-making activities, such as planning the land use for environmental conservation purposes.

Keywords: Geoprocessing, XQuery, XML database, spatial queries, GML, Web service orchestration.