Although analysis is an integral to GIS by definition, it has not been fully integrated into functioning systems to any significant degree for several reasons. First, early research and development efforts were put toward data input and storage techniques. This put analysis issues in the background in terms of research and development priorities. A second reason why analysis is still not an integral component of GIS is that there is no perceived need for analysis within a large portion of the GIS user community. Many GIS users benefit simply from the storage capabilities of GIS for the inventory of spatial data. These applications, as they have currently defined them, do not require the use of analytical functions. Other users who are adopting GIS are accustomed to obtaining analysis capabilities from other sources, such as environmental models and spatial statistical analysis (e.g., hydrologic, transport, and groundwater models). For them, GIS represents a means of data management and efficient access rather than an integral part of the analytical process. The problem of no perceived need for analysis by the users is exacerbated because the software industry is market driven. The software industry is working toward improving the access to spatial data through “user friendly” and “easily accessible” tools and yet increase “power” and “flexibility” of the overall software. The concept of an “easy” yet “powerful” system is potentially contradictory.
One of the overriding trends to solving the problem of incorporating spatial analysis in GIS is based on developing links between software packages. The benefits of such a linkage are clear. The result of linked software would be an integrated environment where research effort could be applied to solving a particular substantive problem rather than managing the technology. This approach to improving spatial analysis in GIS offers the advantages that the techniques exist, are known to the users, and the algorithms and code exist. However, this approach as the basis for a research agenda to improve the analytical capabilities of GIS is limited. Developing an integrated environment is simply a technical one and does not improve our ability to solve problems beyond the pre-computer methods to address the complexity of real-world spatial issues.
In general, I believe spatial analysts and GIS specialists who wish to integrate spatial analysis techniques into GIS are taking a research focus that is too narrow. My goal as a researcher to expand the analytical capabilities of GIS by broadening the perception of spatial analysis in GIS to open up new research opportunities. The avenues by which to accomplish this expansion are several.
First, there remains reluctance to identify modeling as a component to spatial analysis. Modeling, although often specific to one application context, represents the culmination of many spatial analysis techniques. Limiting research to analytical processes just prior to this step is shortsighted and does not serve to broaden the capabilities of GIS. Research examining the related aspects of spatial statistics and modeling should be examined to define a comprehensive framework for spatial data analysis.
Secondly, spatial analysts rarely look at what GIS can offer spatial analysis beyond basic data storage and manipulation. There is the common perception that GIS can gain significantly by integrating more analysis (again, a view that I believe is a technical issue). Rarely do we examine the ways in which research in spatial analysis can benefit from the strengths in GIS. Needed is a turn toward examining how GIS can benefit spatial analysis, such as definitions of geometric properties, topology, and visualization.
Thirdly, the computer is not yet viewed as an alternative medium to the paper map for representing spatial data. This view should be reflected in the assumptions associated with defining spatial data and subsequently the form of the analysis that are performed on these new data types. With the computer as an alternative medium for representing spatial data, we are not limited to describing spatial phenomena with points, lines, and areas. Definition by objects, for example, allows us to build spatial representations that include both the data and the processes. Needed is research toward defining objects that can describe complex phenomena such as qualitative data and spatio-temporal data. Questions regarding how alternative forms of data representation influence and change the types of analysis performed can then be examined.
Finally, spatial analyst are not examining what is happening with regards the increasing rate of data processing, which is resulting in the definition of near real-time problems. The rate of information processing is increasing due to the expansion of the Internet, direct data capture through remote sensing and GPS, and improved hardware. There is growing need for developing concepts to go along with the data processing such as real-time GIS and virtual laboratory. The role of role of spatial analysis in these environments needs to be evaluated. Applications are being built to use the Internet and available data in contexts such as:
- use of real-time GIS for emergency evacuation routes
- electric companies assessing the cause of power outages and dispatching
repair crew
- palm-top GPS tracking with GIS data handling capabilities
- real-time GIS for storm tracking and analysis
- simulation laboratories for pollution modeling
These types of applications are growing and the role of spatial analysis in them is virtually undefined.
In summary, geographic problems are being defined to include qualitative
data, spatio-temporal relationships, and the expansion of these data into
multiple scales. Research in spatial analysis and GIS must recognize the
kinds of problems that are being defined and build tools to help answer
them. Spatial analysts need to untie themselves from the traditional techniques,
and move GIS and spatial analysis beyond replicating the paper map.
Master of Arts, Geography August 1989
The Ohio State University, Columbus, Ohio
Bachelor of Science with honors, Mathematics June 1987
The Ohio State University, Columbus, Ohio
Teaching/Research Assistant and Instructor January 1993 – August 1997
Department of Geography, The Pennsylvania State University
Associate in Research July 1991 - December 1992
Organization for Tropical Studies, Duke University
Research Associate III September 1989 - June 1991
Department of Natural Resources Science, The University of Rhode Island
Research Assistant September 1988 - September 1989
Department of Geography, The Ohio State University
Brewer, Cynthia A. and Elizabeth A. Wentz 1997. A simplified GIS interface for learning introductory geomorphology. Geocal, 16: 6-10.
Wentz, Elizabeth A. 1997. Shape Analysis and GIS. Auto-Carto 13, 204-213.
Wentz, Elizabeth A. 1996. Book review of Innovations in GIS Michael F. Worboys (ed.) 1994. In Annals of the Association of American Geographers 86:376-377.
Wentz, Elizabeth A. and Joseph A. Bishop 1995. Geographic Information System and Database Management at a Tropical Rainforest Research Station, Biology International 30:10-19. Republished in GIS Methodologies for Developing Conservation Strategies, 1998, Basil G. Savitsky and Thomas E. Lacher Jr. (eds), Columbia University Press, New York, 83-95.
Peuquet, Donna J. and Elizabeth A. Wentz 1994. An Approach for Time-Based Analysis of Spatiotemporal Data, Advances in GIS Research 1:489-504.
Campbell, Aimee F. and Elizabeth A. Wentz 1993. Monkeying Around with ARC/INFO: GIS Methods in the Study of Primate Ecology and Conservation, ARCNews Redlands, CA: Environmental Systems Research Institute, Inc. Reprinted in Liana Durham, NC: Organization for Tropical Studies.
Wentz, Elizabeth A. and Marco Vinicio Castro Campos 1993. Management of Research Areas at a Biological Field Station, in Proceedings of the Thirteenth Annual ESRI User Conference, Redlands, CA: Environmental Systems Research Institute, Inc., 303-314.
Central Arizona – Phoenix, Long Term Ecological Research (CAP – LTER)
Core Scientist, Center for Environmental Studies, Arizona State University
August 1997 - present