National Academies Press: OpenBook
« Previous: 2 Current R&D Plans in NMD
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 25
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 26
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 27
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 28
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 29
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 30
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 31
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 32
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 33
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 34
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 35
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 36
Suggested Citation:"3 Research Opportunities." National Research Council. 1991. Research and Development in the National Mapping Division, USGS: Trends and Prospects. Washington, DC: The National Academies Press. doi: 10.17226/10986.
×
Page 37

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

RESEARCH OPPORTUNITIES The explosive growth of spatial data handing technology and applications has whetted the appetite of users for new and improved capabilities to analyze, model, and apply the data to meet their needs. The private sector seems to have satisfied some of these needs through the development of new hardware plat- forms and software to process the data. Still other needs or desires require innovative R&D to extend data handling and modelir g capabilities. Although some results are several years away from the marketplace, the private sector has demonstrated its rapidity to incorporate new spatial data handling advances into commercial products. To realize fully the potential benefits of new technology, researchers in government, the private sector, and the academic community should work synergistically. The research outlined in the following sections is not of exclusive interest to any particular sector; nor can any one sector accomplish what is needed. Precedents for cooperative activities include: (1) development of advanced mapping systems, wherein the USGS (Mark-II) and the DMA (Mark- 90) are developing such systems through both in-house activities and contracts with industry, (2) NSFs establishment of the National Center for Geographic Information and Analysis (NCGLA) to conduct fundamental research on a variety of issues (NCGIA is a consortium of the University of California at Santa 25

26 Barbara, the State University of New York at Buffalo, and the University of Maine at Orono), and (3) the support of academic research by industry. If NMD is to remain responsive to the needs of its current and future users, it must continue to be aware of emerging research problems. For example, some user-based professional associations and societies have attempted to define the research needs of their membership. The Urban and Regional Information Systems Association (URISA) has developed a statement of its research needs (see Table 8~. In addition, the NCGIA has generated a statement of its research objectives (see Table 9~. A useful analysis of how these two agendas construc- tively overlap is provided by Craig (1989~.3 He suggests a wide range of research needs from basic science to societal applications. From this analysis, it is appar- ent that the needs for research on spatial data handling and analysis are driven by applications and decisionmakers who use the resultant information. The remainder of this chapter attempts to identify an appropriate role for NMD in addressing the research agenda of the nation's spatial data user community. The committee strongly believes that the spatial data infrastructure of the nation would benefit greatly if NMD expanded its current research program by cooper- ating with and funding research with academia and the private sector. DAIA CAPTURE, REVISION, AD ACE To support its long-term mission as a primary spatial data supplier, NMD must be able to create and maintain efficiently and economically an accurate spatial representation of the United States. Innovative types of data and characteristics should be explored for incorpo- ration into NMD products, including, for example: (1) data at larger scales to support urban development, (2) varying scales and resolutions within a product based on the localized density of information, (3) linkages of NMD data to cadastral information, and (4) additional attributes to support address matching, coordination with TIGER products, and the Environmental Protection Agency's river reach files. In addition, research is needed to determine the appropriate forms of data capture to support automated systems. This need includes techniques for data aggregation and generalization, especially using GIS and image processing systems. Effective data update methods should also be developed, including techniques, manner of documentation, and acceptable frequency. Standards for unconventional NMD products (e.g., thematic and image maps) need to tee

27 TABLE 8 1987-1988 URISA Research Agenda (from Craig, 1989) , SOCIAL CONCERNS System Adoption · Codify concepts and terminology. Improve access to literature. Research real-world experiences to document actions and conditions that lead to success. · Develop educational programs for staff and other users. Social and Legal Impacts · Document the impacts on the host institution, management, staff, elected officials, and the public. · Determine criteria such as accessibility, objectivity, and equity that will ... . ~ . ~ ~ ~ . . encourage system ut~atlon by the broadest range of publics and policy- makers. Investigate legal imperatives for providing access to data, considering privacy laws, and determine conditions of legal liability for incomplete or inaccurate data. Management Issues · Develop effective strategies for day-to-day management, including how to bridge the gap between technicians and the user community. Assess the social, political, and behavioral conditions that inhibit data sharing and recommend means for improvement. . · Examine the problems and potential for a distributed corporate GIS, where each unit has its own unique domain of definitions, needs, and hardware. . Economic Factors Define a methodology and estimate costs and benefits. Measure cost effectiveness and productivity compared to manual systems. Explicitly measure the costs of data capture, conversion, and maintenance. Attempt to determine the value of "public information." · Explore the unique aspects of measuring costs, benefits, and decision structures of information systems as compared to other enterprises. To what extent is the uniqueness more profound in the public sector? TECHNICAL CONCERNS Data Base Development · Lower data capture costs through improvements in scanning technology and better utilization of remotely sensed data, especially through incorporation of artificial intelligence techniques. 1: , .

28 Develop data quality standards and methods for "stamping" or documenting the quality. Include both purity and spatial precision as measures of quality. Determine the need for access to a variety of human and physical geograph- ic data sets and define an approach for developing a "National Library" to meet that need. Develop tools to assist with data base design, procedures for updating, techniques to improve the data base over time, and methods for archiving. · Develop models for networked systems in governmental organizations where data are distributed to meet operational needs and analysis and problem solving must use the diverse data sets. Identify problem areas such as the need for data refining and the impact of independent upgrades at various nodes. User Interface and Empowerment Improve processing speeds so analysis can be done in "real time." This will require both improvements in hardware and data base structures as well as vastly improved processing algorithms. Add to the range of models available to the GIS analyst (e.g., transportation and ecological models). The possibilities are enormous, but a library of such functional modules would be very useful. GIS software might be modified to readily accept such modules; in many cases, the models have been developed, and their usefulness would be enhanced greatly with addi- tion of the graphical component that a GIS could provide. Gaming/simula- tion packages are another type of useful modules. Make GIS software accessible to users with different levels of technical expertise through the use of artificial intelligence, help screens, relational data bases, and software layering. Analyze the needs of planners and other public officials, see what potential applications they have for this technology, and develop specifications for products to meet those needs. Software Critique · Develop a comprehensive list of major software packages and the major application of each. · Develop a list of common and exceptional GIS functions. · Create benchmark tests that would fairly compare systems on features most important to users. Run these tests on the major software packages and report results. Determine and document the constraints imposed by selecting particular software, data scales and classification schemes, and data base structures. . .

29 TABLE 9 NCGIA Research Initiatives GROUP I 1. Accuracy of Spatial Data Bases: Develop methods for evaluating GIS data and products. 2. Languages of Spatial Relations: Develop a theory of spatial relations based on both cognitive/lingu~stic and mathematical/logical models. 3. Multiple Representations: Explore efficient data structures that accommo- date resolution-based change and determine how dissimilar representations of the same geographic feature can be related. 4. Use and Value of Geographic Information: Look at how decisionmakers use and value information in order to measure the benefits of GLA and GIS. 5. Architecture of Very Large GIS Data Bases: Develop and test prototypical systems with high performance. GROUP II 6. Spatial Decision Support Systems: Integrate operations research tools into the GIS environment. 7. Visualization of the Quality of Spatial Information: Develop methods for displaying the quality of data. 8. Expert System for Cartographic Design: Develop a system to design carto- graphic displays. 9. Institutions Sharing Spatial Information: Explore existing policies, prob- lems, and prospects for sharing data; develop models of political support for sharing. 10. Temporal Relations in GIS: Understand the modeling of time and assess the impacts on GIS design, with emphasis on logic and data base issues. 11. Space-Time Statistical Models in GIS: As above, with emphasis on statisti- cal models and remotely sensed data. 12. Remote Sensing and GIS: Improve the usefulness of remotely sensed data by applying GIS-based tools for data storage, processing, and classification. developed for transferring data to other users. Such research projects should be developed and tested in conjunction with either a program (e.g., global change and side-looking airborne radar) or a general category of digital data manipu- lation (e.g., image processing, thematic mapping, and data collection).

30 NMD's hardware needs in data capture include digital orthophotography generation; image scanning; and low-cost digital stereo photogrammetric digi- tizers for aerial photography, SPOT, and other remotely sensed data. Software needs are in the areas of digital orthophotography processing, optimal DLG hypsography and digital elevation model (DEM) algorithms, advanced algo- rithms for satellite image mapping, high-speed algorithms for raster/vector conversion that preserve the accuracy of the original data, and an advanced expert system for image analysis (e.g., improved feature extraction algorithms) for use with advanced satellite remotely sensed data including EOS data. As GIS and associated technologies become integral to the work of planning and resource management agencies, the demand for thematic data layers will surely increase. Timely, accurate, and current, detailed land-use and land-cover data as well as other thematic products are an immediate need of diverse govern- mental, scientific, and industrial users for both resource management and analysis of regional and global change. Existing thematic data products such as land-use and land-cover data sets produced by the USGS are at least 10 years old. To develop an intermediate-scale thematic mapping program based on current data, NMD will need to consider new data sources, innovative digital mapping techniques, state-of-the-art image processing techniques, automated cartography, advanced change detection algorithms, and automated procedures for assessing the accuracy of specific thematic data. In addition, considerable software research will be necessary to support both the digital data base and graphic thematic data products from the DLG-E data model. Additional research involving the application, structure, and accuracy of different thematic data products within a GIS framework would most appropri- ately be coordinated with other organizations. For example, the NCGL~ re- search initiative on the use and value of geographic information in decision- making is particularly relevant to NMD. The emphasis of this research is to: (1) identify primary and subsequent users of spatial information and explore methods for determining the value of such information, (2) identify problems associated with uncertainty and risk in decision-making from spatially referenced data, (3) develop and test models of decision-making regarding land use and land cover, focusing on the role of information, and (4) evaluate direct and indirect benefits of geographic information in an institution, agency, and/or industrial context. 'it::-,:

31 DAIA BASE ORGANIZATION Many recent successful applications of spatial data can be directly linked to technical advances in computer hardware and software that enable the user to store and retrieve large volumes of spatially referenced information efficiently. It is essential to NMD's future mission that it continue to be well aware of ongoing developments in several areas: (1) read/write mass storage devices, (2) improved data/information cataloging, indexing, and retrieval systems, (3) high data rate, high-capacity computing environments for cartographic and image processing, especially as related to remotely sensed data, (4) high-speed network technology for transmitting spatial data, and (5) interactive systems for processing, analyzing, and displaying digital data. The fundamental data base concern of NMD must be the basic format used to represent the digital versions of geographic features. Because NMD is committed to the DLG-E data model, the development of an unambiguous set of representation rules for all geographic features is important. Although the initial domain of features has been developed, several representations are possible for each instance of that feature. Work to derive consistent sets of feature-instance representation rules to complete the design of DLG-E should be given priority. Further work on feature classification needs to be conducted for the addition of new layers (e.g., geology and soils), for the use of classification hierarchies as an aid in generalization, and for the preparation of interactive reclassification tools. Extensions to the DLG-E data model must also be made. These extensions include modeling of 3-dimensional objects using attributes of relationships to improve flow information, expanding spatial models to describe accuracy, and storing and displaying temporal data in a spatial context. Research on implementing the DLG-E model in a variety of computer environments is important. Implementation of the model in the context of an extensible relational and/or object-oriented data base system should be tested. The goal is a data base system containing multiple representations of the DLG-E features. The global change research program, in particular, will require considerable research on temporal data models. This research includes: (1) documenting characteristic scales of spatial and temporal changes in various environments for basic social, natural, and applied sciences processes, (2) developing a taxonomy of space-time statistical models to help select appropriate data base structures for representing temporal variability of specific social and natural processes,

32 (3) implementing algorithms for efficient data update in systems with different characteristic frequencies and scales of temporal variation, and (4) developing and applying computationally efficient methods of multiple representation in the time domain. As noted in Chapter 2, NMD will establish an archive and data dissemina- tion facility for EOS land data. The EOS will be a major source of global change data that should become available beginning in the late 1990s. Other National Research Council committees believe that data should be archived at sites where there are active users. In light of this stated need, NMD should begin to expand its R&D program in data archive functions. The effort should include user needs assessments, test bed activities, advanced data and information design studies, improved algorithm development, and experimentation with new product generation. To meet the goals, NMD scientists at EDC will need to work closely with the EOS science user communities. SPATIAL DAM DISPLAY Historically, NMD has been concerned with producing quadrangle-based paper maps. Because future NMD data users would like to access seamless and scale-independent digital data bases, research on multiple representations is important. This research could focus on, for example: (1) developing models for digital description of cartographic features (object-oriented versus spatially addressed models, hierarchical models, conversion between models), (2) examin- ing the relations of the geometry of geographic features to the scale of repre- sentation (e.g., self-similarity versus scale dependence), (3) examining problems associated with scale changing, possibly leading to proposed solutions and algorithms based on pattern recognition of feature identification, inference across levels of resolution, and automation of feature simplification and selection, (4) characterizing effects of multiple presentation on error propagation, and (5) determining data base organizations capable of dealing with multiple repre- sentations of the same objects. Hardware and software that improve scientists' ability to visualize complex spatial relationships are also needed. Advanced research topics in visualization of spatial data include: (1) developing and implementing methods for displaying the quality (e.g., reliability and accuracy) of spatial information, (2) testing methods for multidimensional displays of spatial information, (3) evaluating the potential of time sequential (animation) and 2-dimensional (profile) and 3-

33 dimensional (depth) displays of complex spatial data, (4) developing and imple- menting methods for displaying hyperspectral multiparameter image data, and (5) developing methods for improving the display of multisource layers of spatial data in image and line form to further understanding of complex environmental processes. Artificial intelligence and expert systems research are also important in areas of cartographic display. For example, NMD should work closely with the Bureau of the Census on the expert system used for automatic name placement and inset selection for maps supporting the 1990 Census. APPLICATIONS As emphasized in the committee's previous report, if NMD is to improve its overall support for the spatial data user community, it needs more information on how those data are used. Spatial data are basic to geographic information systems, considered one of the fastest growing areas of data processing. In the context of modeling, research on the measurement of accuracy and the consequences of using inaccurate spatial data is essential. Other research is also required to: (1) assess statistical models of spatial data, (2) develop and evaluate techniques for interpolating estimates as a means to overcome problems of variable regions of data aggregation and missing values, (3) develop indices of uncertainty and confidence for spatial data products, and (4) determine the effects of aggregation on spatial modeling. Languages that enhance spatial modeling are also important research topics. A considerable amount of work is needed to: (1) identify formal cogni- tive/semantic models of spatial concepts/relationships in natural languages, (2) develop advanced methods for determining reference frames for spatial languag- es, (3) investigate formal mathematical/logical models of spatial concepts and relations based on topology and geometry, and (4) address the integration of relational based models into a general theory of spatial relations. Decision Support Systems Spatial information systems convert spatial data into information to facili- tate the decision-making process. In an optimal setting, a GIS would become a spatial decision support system (SDSS). Such a system would integrate analytical

34 models, data base management, graphical display, and report generation in an iterative, man-machine interactive setting. At a minimum, a set of GIS tools linked to an integrated data base should enable the user to conduct exploratory spatial analysis. In a more sophisticated environment, the GIS would enable the user to evaluate a set of alternatives, select the best alternative, implement the system or plan, and monitor the results. GIS technology introduces a further layer of complexity to the environment of decision support systems that is already complex. The task is challenging. Decision support systems have the following characteristics: They are designed to handle ill- or semistructured problems; They have an interface that is easy to use; They enable the user to have full access to the data base; · They are able to generate a number of alternative scenarios; · They support a range of decision-making styles; and · They support interactive and recursive decision-making processes. Given these challenges, a good SDSS needs to make GIS software accessible to users with different levels of technical expertise through the use of artificial intelligence, help screens, relational data bases, and software layering. In other words, the ultimate value of NMD's spatial data may depend on how well it can support an SDSS. The limiting factors of an SDSS relate most directly to data problems, inadequacy of the analytical functions, and inability to support the type of decision-making process that characterizes many private and governmental forms of decision-making. Over the past 20 years, the complex problems faced by resource managers and scientists have pushed system developers to create sophisticated spatial data handling tools that integrate and synthesize diverse forms of spatial information. The overriding challenge to GIS developers is to combine the appropriate information in a format that can be easily queried, analyzed, and displayed. The fundamental building block is the appropriateness of the data base (the principal concern of NMD). To be successful, future NMD data products must provide the appropriate data bases. In addition to maintaining and distributing data products, NMD also needs to keep abreast of changes in the analytical capabilities of automated systems. Commercially available spatial analysis tools today probably fall short of the needs (as expressed in the URISA research agendas; see Table 8) for combining spatial integration, optimal location allocation routines, and statistical analysis

35 in a common operating environment. Additional work should also include evaluation of: (1) the uncertainty and the risks associated with decision-making (e.g., the economic concept of utility applied to information, the role of information in uncertainty reduction, uncer- tainty reduction and absorption, and the limits to research for information), (23 the decision models (e.g., the decision-making process, the multiple role of information, information as a product, information as a public good, and the distinctions among data, information, and knowledge), (3) the demand for information (e.g., value as a demand-initiated concept, multilevel user identifica- tion techniques, the contrast between supply/push and demand/pull in the development of information systems, and public good aspects), (43 the benefits (e.g., direct and indirect benefits, uncertainty reduction, uncertainty absorption, expanded opportunity, and cost avoidance models), (5) the design of GIS data structures optimized to support decision systems, (6) the investigation of map/ image scanning technologies for effective capture of map linework and/or continuous tone images, (7) the development of methods for effectively structur- ing spatial search algorithms within a GIS framework, (8) the classification of spatial search problems and the identification of gaps in current models, and (9) the investigation of optimal data management procedures for effective use with a variety of spatial data types. Although the interface between GIS and analytical modeling systems needs improvement, such a linkage is emerging. One important aspect of this linkage is the technological ability to share information. Consider the following two points: the technology provides a means to decrease data collection costs, which can reduce institutional barriers, and organizations can maintain their institution- ally independent layer of data and their control over it. As individual agencies and organizations recognize the incentives to share their data with others, technology may break down traditional turf battles and lead to organizational innovation. By providing the base levels of spatial data contained in the NDCDB, the NMD may facilitate new levels of cooperation for divergent users. Global Change Research: An Example of Spatial Data Applications As discussed in Chapter 2, the USGS has begun major research initiatives in global change. These initiatives include the identification of suitable test areas for measuring and monitoring change. NMD has also begun to experiment with improved spatial analysis techniques. In addition, the USGS/NMD is developing

36 GLDIS at the EROS Data Center to archive, process, and distribute land-related data sets. The EROS Data Center has been identified as the distribution and archiving facility for land-related data collected by the EOS. Four elements are essential to a global change research program: (1) devel- oping improved spatial data analysis, management, and geographic information systems tools and methodologies, (2) identifying, delineating, and characterizing regions and the processes that create regional change, (3) developing methodol- ogies to integrate processes at local and regional scales with those occurring globally, and (4) supporting a program of land-use and land-cover monitoring that is required for analyzing and monitoring changes important to climate change. A variety of research is required in support of this theme: (1) analyzing the role of land-use and land-cover changes in the monitoring and modeling of climate change, (2) developing techniques and methodologies for dealing with very large spatial data bases, (3) developing quantitative techniques for linking hierarchical data sets at variable scales, and (4) exploring the full potential of remotely-sensed data from aircraft and satellites to support measurements, mapping, monitoring, and modeling of significant biophysical and socioeconomic processes. In addition, other significant research topics include: (1) developing tech- niques for near-real time indexing and browsing of map and image data of important Earth surface features (e.g., vegetation greenness, biomass, and albedo), (2) implementing advanced systems at regional and global scales to detect and identify significant temporal phenomena rapidly, (3) investigating procedures for producing regional, continental, and global-scale image maps from satellite-derived data, and (4) developing, testing, and evaluating prototype global data sets of value in the study of global change. RESEARCH OPPORTUNITIES: CONCLUSIONS A variety of significant research endeavors must be undertaken to improve understanding of our Earth. Just as the USGS has helped to improve knowledge of this vast nation and its potential for development, it should now provide the basic information upon which decisions can be made in our rapidly changing advanced technological society. To achieve sustainable (e.g., land-use decisions resulting in sustainable changes in land cover) development on a global scale, accurate, timely, and reliable information on the state of our national and global

37 environments is critical. A robust national mapping infrastructure is vital to achieving this goal. The establishment and sustenance of such an infrastructure will require an expanded NMD research commitment—one that includes both fundamental and applied research. Underlying this commitment must be coordi- nated research efforts by federal and state agencies as well as by academia and industry.

Next: 4 Conclusions and Recommendations »
Research and Development in the National Mapping Division, USGS: Trends and Prospects Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!