Strategic Enhancement of NGA's Geographic Information Science Infrastructure
Executive Summary
This proposal describes a
three-year program of research into two related topics in geographic
information science (GIScience). Users of geospatial technologies are
increasingly faced with the problem of multiple data sources and the
incompatibilities between them. Although geographic information systems
(GIS) are often presented as the solution to problems of integrating
data from multiple sources, in practice a large proportion of the resources
of any project are devoted to overcoming initial problems of format,
syntax, semantics, quality, and spatial support. The first two of these
problems have been addressed over the past ten years through the development
of a number of common standards and specifications, but the others
are proving more difficult. This project will bring focused research
effort to bear on them, and develop and implement solutions.
The project is divided into
two phases, addressing first spatial webs and then data integration,
since work on the second phase will depend in part on completion of
the first. We define a spatial web as an information community actively
sharing geospatial data, and propose to work to remove the impediments
that currently exist to such sharing, in the face of issues of semantics,
quality, and spatial support. We propose a four-task research agenda
to address these issues over the first two years of the project.
The second phase will address
data integration, which we define as the steps undertaken to combine
data once the problems of the spatial web have been addressed. Such
combination can be for a range of purposes, from fusion to concatenation
and averaging, and we will develop methods to address all of these
purposes. Integrated data inherits many of the properties of the data
being integrated, but new properties are often created in the process;
data quality, for example, may be better for data that have been averaged.
We will research and implement appropriate methods for the automatic
creation of metadata for integrated data sets, as part of a comprehensive
four-part research program to address data integration.
The project will leverage
the facilities and research expertise of the three institutions that
are partners in the National Center for Geographic Information and
Analysis (NCGIA), a consortium founded in 1988 with a major grant from
the National Science Foundation. The bulk of the effort will be at
the University of California, Santa Barbara, but subcontracts are proposed
with the University at Buffalo and the University of Maine. Buffalo
researchers will contribute to the effort on spatial webs, and Maine
researchers to the effort on data integration.
We will also provide opportunities for students in the Masters program at the University of Redlands to participate in the research, to conduct research projects of their own, and to interact with researchers and graduate students at NCGIA.