Apoala Project

.

AN INTEGRATED APPROACH FOR REPRESENTATION
AND ANALYSIS OF SPACE/TIME ENVIRONMENTAL DATA

The growing complexity of environmental analysis and policy decisions demands not only introduction of high performance computing technologies, but innovative approaches to their application. Environmental risk assessments and decision support for ecosystem management requires the integration and analysis of large volumes of data derived from both models and direct measurement of the environment. These data exist at multiple scales and vary both spatially and temporally.

Past research directed toward developing approaches to dealing with the volume and complexity of environmental data has often focused on isolated issues associated with data structures, visual display, query protocols, computer architecture, and others. What is needed, in contrast to such a fragmented approach, is an integrated vision for data analysis and problem solving environment that can cope with spatiotemporal data and multiscale analysis.

The objective of the research being initiated is to design, implement, and assess a prototype that meets these needs. Specifically, we plan to develop a prototype Temporal Geographic Information system with integrated multivariate spatiotemporal visualization capabilities. The prototype will include:

(1a) a fully-integrated multi-representational, multi-scale database model which integrates object-based, location-based and time-based representations of environmental data in a tightly coupled manner,
(1b) implemented in a flexible and highly efficient manner that employs parallelism for storage, retrieval and manipulation to maximum advantage,

(2) including a flexible database query protocol linked to a graphical user interface that facilitates multidimensional and multi scale query and analysis, and

(3) a multi-dimensional, multi-scale visualization capability which facilitates exploratory and traditional analysis of time sequences of three-dimensional data.

The system proposed will be implemented in a UNIX workstation environment and use real-world data in assessing system functionality. Although all four components of the proposed research will have independent utility, the principle contribution of the research will be the basis it establishes for building integrated Temporal Geographic Information and Visualization systems (TGIS/VIS) that will support a full range of environmental science and decision support applications. Merging TGIS and geovisualization capabilities not only links computational problem solving environments with data access and analysis methods, but will lead to more robust assessment of conceptual approaches to spatiotemporal representation.

Copyright © 1998-1999 Pennsylvania State University


.
	AN INTEGRATED APPROACH FOR REPRESENTATION AND ANALYSIS OF SPACE/TIME ENVIRONMENTAL DATA The growing complexity of environmental analysis and policy decisions demands not only introduction of high performance computing technologies, but innovative approaches to their application. Environmental risk assessments and decision support for ecosystem management requires the integration and analysis of large volumes of data derived from both models and direct measurement of the environment. These data exist at multiple scales and vary both spatially and temporally. Past research directed toward developing approaches to dealing with the volume and complexity of environmental data has often focused on isolated issues associated with data structures, visual display, query protocols, computer architecture, and others. What is needed, in contrast to such a fragmented approach, is an integrated vision for data analysis and problem solving environment that can cope with spatiotemporal data and multiscale analysis. The objective of the research being initiated is to design, implement, and assess a prototype that meets these needs. Specifically, we plan to develop a prototype Temporal Geographic Information system with integrated multivariate spatiotemporal visualization capabilities. The prototype will include: (1a) a fully-integrated multi-representational, multi-scale database model which integrates object-based, location-based and time-based representations of environmental data in a tightly coupled manner, (1b) implemented in a flexible and highly efficient manner that employs parallelism for storage, retrieval and manipulation to maximum advantage, (2) including a flexible database query protocol linked to a graphical user interface that facilitates multidimensional and multi scale query and analysis, and (3) a multi-dimensional, multi-scale visualization capability which facilitates exploratory and traditional analysis of time sequences of three-dimensional data. The system proposed will be implemented in a UNIX workstation environment and use real-world data in assessing system functionality. Although all four components of the proposed research will have independent utility, the principle contribution of the research will be the basis it establishes for building integrated Temporal Geographic Information and Visualization systems (TGIS/VIS) that will support a full range of environmental science and decision support applications. Merging TGIS and geovisualization capabilities not only links computational problem solving environments with data access and analysis methods, but will lead to more robust assessment of conceptual approaches to spatiotemporal representation.

Copyright © 1998-1999 Pennsylvania State University