Introduction

The project is funded by a grant from U.S. National Geospatial-Intelligence Agency(NGA) NURI(NGA University Research Initiatives) program. The research addresses the knowledge discovery and dissemination subtopic of NGA NURI. Work on this project started in 2004.

With the advancement of sensor and platform technologies, the capability for collecting geospatial data has significantly increased in recent years. In the U.S., both military and civilian agencies have collected a huge amount of geospatial data via remote sensing. While those data are potentially valuable for our nation’s security and economic growth, before they can be useful, they have to be converted to geospatial information and knowledge. The challenge is how to discover useful knowledge embedded in the mountains of the geospatial data in an effective and timely way. The traditional methods of analyzing data by expert analysts fall far short of today’s increased demands for geospatial knowledge. As a result, much data may never been analyzed even once after collection. Therefore, technologies for automated geospatial knowledge discovery and dissemination are urgently needed for both military and civilian geospatial applications. This research will develop such technologies.

Three significant features distinguish geospatial knowledge discovery (GKD) from other scientific endeavors: 1) the research is multidisciplinary; 2) the research is data, information, and computation intensive; and 3) the research regions may vary from micro to global scale. Nor-mally, the processes of geospatial knowledge discovery involve three consecutive steps:

Geo-query: locating and obtaining data from data repositories.
Geo-assembly: assembling the data and information from data centers based on the needs of geocomputation.
Geocomputation: analyzing and modeling the complex geospatial phenomena by using data and information from the geoquery.

Because of the multidisciplinary nature of GKD, data from data centers are diverse. Often, the temporal and spatial coverage, resolution, origin, format, and map projections are incompatible. As a result, even when the analysis is very simple, considerable time is required in the geo-query and geo-assembly to obtain and assemble the data and information into a form ready for analysis. If datasets the analysis requests are not readily available at data centers, the data and information system (DIS) at the data centers cannot provide the datasets on demand even if the process to make them is very simple. Therefore, analysts have to spend a considerable amount of time in ordering and processing the raw data to produce the data they need in the analysis. Although DIS are the core support for GKD, current systems, which generally facilitate only data search and ordering, are inadequate.