Short URL

AR-GAP Server

 

Other Geospatial Interoperability areas:

Loading, please wait...

The "GAP Server" project, conducted by Shane Covington with Bob Harris, between 2003 and 2004 developed an interoperable application to demonstrate the innate ability of Oracle Spatial to store, query, and retrieve large amounts of geospatial data and to develop a user interface allowing persons without SQL knowledge to query the Oracle Database. These applications were based on GAP layers which allowed users to query between species locations, habitats, ownership, and management areas.

The Arkansas GAPServer enabled results to be displayed as maps or reports based on GAP layer criteria defined by a user, as well as enabling users to download raw GAP data.

To achieve the goals of this project, all the GAP layers (species, ownership, management, and land cover in addition to counties and watersheds for the entire State of Arkansas) were imported into Oracle Spatial. Rather than other database architectures, Oracle has been designed to handle terabytes of data and millions of simultaneous transactions. Oracle also enabled geospatial data to be queried using SQL statements. These features offered a dramatic increase in processing ability over other database architectures that had been used to serve, query, and display GAP information. In addition to these features, Oracle’s GRID computing architecture could allow additional states to be added without reducing the performance of the application.

Once the geospatial data was imported and indexed in Oracle, a user interface was built to allow users to build queries based on the information stored in Oracle and to obtain results that matched a user’s criteria. This interface allowed users to obtain three types of information: map meeting a user’s criteria, reports meeting a user’s criteria, and the raw geospatial data layers. This client-side application could run as a standalone module or be integrated with a previously developed map viewer. When the application was integrated with a map viewer, the current map display was used as the bounding box for querying the database, and the resultant map was inserted back into the map display. Otherwise, a user could choose multiple counties or water basins for their area of interest.

 

Integration of AR-Gap Server with Interoperable Geo Observer (IGO)

 

To create queries, a natural query language builder was developed to enable users to build human understandable sentence fragments that defined criterion constraints. The natural language builder was implemented in HTML, XML, and JavaScript. Pull-down menus contained all the GAP layers and unique attribute values for those GAP layers, as well as, area of interest values such as water basins and counties for the State of Arkansas. These values could be added to create a sentence-based query using Boolean ANDs and ORs. Once a desired sentence was submitted, it was translated to SQL to perform processes.

Reports, maps, and downloads were processed by Java Servlets on the server-side. Reports could be created in HTML, text, two forms of XML, and a comma delimited text that could be imported into a common spread sheet. Map images of areas that met a user’s criteria were created using Oracle’s Map Viewer and displayed in either a previously opened map viewer or a new map display that also contained a base map for reference. The download of GAP layers were sent to GeoStor which processed the download request and sent the user an email of the location where the selected layer could be obtained.

 

 

All the initial goals of this project were met and exceeded. Using Oracle proved to be a wise architecture choice since large amounts of geospatial data could be stored, queried, and retrieved quickly. The natural language query builder has proven to be a very effective way to enable users without knowledge of SQL to build complex spatial queries. Furthermore, the reports, maps, and download ability have been very informative and useful to users.