Improving geospatial data discovery by enhancing public metadata catalog search services

Abstract

Most public geospatial data is served via standard compliant Web Mapping Servers and Web Feature Servers available globally. However, it is quite complicated to search and locate the relevant data if the exact data source name and other parameters are unknown. Catalog servers, which store metadata and provide search protocols provide a standard way to handle this problem, but only afew Catalog servers exist which limits the usefulness of such servers for adhoc geospatial data discovery. This paper looks at the advantages of making available metadata of geospatial data through Catalog Servers for adhoc geospatial data discovery. It looks at the limitations of searching geospatial data using Web Mapping Servers and Web Feature servers and search engines such as Google, and presents a case for having increased Catalog Servers. It then looks at a novel way of building metadata from existing Web Mapping Servers using common data base techniques, which could then lead to more metadata in Catalog Servers, which would in turn lead the way to better and more efficient Geospatial Data Discovery. It then looks at a specialized client that is used to search catalog servers over the web, providing domain experts with a powerful tool to accurately search for data. 1. Introduction I The rapid increase of public spatial data on the Internet, coupled with the need for using spatial data in meaningful ways, is shifting the focus towards the discovery and integration of data. During the past decade there has been tremendous growth in the number of spatial data sources available as public web services, and these sources are being used in domains as diverse as finance, education, citizen services, digital media and emergency response, to name a few. The majority of this data is made available to the general public as raster maps in Web Mapping Service (WMS) Servers [I] and vector data on Web Feature Service (WFS) Servers [2]. The Geospatial Information Database (GlOB) [3] project lists 1400 WMS servers serving over 330,000 layers, indicating the scale of data that needs to be searched to identify data. With such large amounts of data available for public use, the task is left to the end user to search, identify and fetch the relevant spatial data he/she is looking for. The ability to discover and access geographical data for planning, visualization and decision making is a requirement to support communities and activities at local, regional, national and international level. This is summarized well in this statement from the Spatial Data Infrastructure [4] cookbook in its chapter on geospatial M. Careem and D. Karunarathne are with the