Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 8
2
CURRENT R&D PLANS IN NMD
To date, the National Mapping Division has clearly concentrated on system-
development aspects of an R&D program. These development activities are
designed to facilitate its role as a major federal information supplier, both in
more efficient internal operation and in response to needs of the user commu-
nities. Although NMD has focused its developmental activities on product
generation, plans indicate an NMD desire for fundamental research activities
the spatial data models required to handle its data and a variety of data applica-
tions. However, resources directed toward these efforts are limited. Plans for
NMD's R&D activities and the FY 1990 level of effort (in full-time equivalents,
FTEs, which approximate person years) are summarized in Table 2. (See
Appendix A for more specific information about current activities.)
DEVELOPMENTAL PI^NS AND ACTIVITIES
The Mark-II System
NMD is developing a major advanced mapping system, Mark-II, to satisfy
National Mapping Program requirements through the year 2000. During the
acquisition phase, NMD is coordinating the system's development and procure-
8
OCR for page 9
9
TABLE 2 USGS Research and Development Plans and Activities
Level of Effort,
FTEs in FY 1990
DEVELOPMENTAL PLANS AND ACI IVITIES
Modernization Mark-II System: Activities directly related
to the modernization of the NMD map production system,
including the specification, procurement, testing, and instal-
lation of state-of-the-art hardware and software for the pro-
duction of cartographic data in graphic and digital form.
Part of this system is related to the Mark-90 system under
development by DMA.
Modernization—Product Generation: Development and
engineering activities required to maintain and enhance
current NMD production operations, focusing on the inte-
gration into NMD's production capabilities of proven off-
the-shelf advanced digital technology to support production
of standard and thematic graphic products.
Standards: Establishment of standards to support technical
aspects of NMD data production, data base, product gener-
ation, and product distribution operations.
Rules: Work related to the definition and documentation of
feature data specifications, including feature, attribute, and
attribute value definitions incorporated into the DLG-E
model; delineation, extraction, and representation rules;
product inclusion conditions, symbol specifications, and
generation rules.
Technology Transfer: Development and coordination activ-
ities to inform, instruct, and assist users of USGS and other
mapping organizations (OMO) data of data characteristics
and optimal utilization technologies, focusing on GIS, image
processing, and digital cartography technology.
DLG-E Development: Activities related to the implementa-
tion of the DLG-E model, including the phase II proto-
typing plan as the following (former) Mark-II modules: GIS
interface, text data collection, content conversion, and the
DLG-E processor.
57.2
6.7a
14.7
5.6
11.1
3.8
13.0
OCR for page 10
10
FTEs in FY 1990
RESEARCH PLANS AND ACTIVITIES
Theoretical Spatial Data Handling: Research on spatial
data base management, data structures, formats, spatial
data processing and manipulations, error propagation, AI
and GIS techniques, and advanced visualization methods.
Techniques Development: Activities related to the develop-
ment and testing of new and innovative techniques for data
manipulation, problem solving, and visual lion.
Remote Sensing and Image Processing: Research on sen-
sor and processing systems. data acquisition technologies
and data manipulation.
Thematic Mapping: Research related to the collection,
analysis, and display of thematic data sets. Includes devel-
opment of custom techniques and classifications for specific
categories of data and subsequent integration of these with
base category or other data.
Data Applications: Activities demonstrating the analysis
and application of spatial data from USGS and OMO
sources. Technologies used include GIS, image processing,
3-dimensional modeling, and customized thematic map
generation.
Global Change: Use of spatial data handling techniques for
research on global change issues.
Data Collection, Management, and Dissemination: Activ~-
ties related to the collection, management, and dissemina-
tion of USGS and OMO data holdings. This is principally
directed to data concerning global change.
, . . · .
34.2
0.5
7.0
12.1
0.9
0.3
10.4
3.0
a The Mark-II system affects almost all of the other development activities; thus
the total effort is actually greater than the 6.7 FTEs indicated.
OCR for page 11
11
ment with the DMA, which has taken the lead in developing an automated
mapping system (D3L9's Mark-90~. The Mark-90 system is being designed to
meet the Department of Defense's map production needs and digital data
requirements.
Mark-II development activities are directly related to the development of
the Mark-90 system, including specifications, procurement, testing, and installa-
tion of the hardware and software for production of cartographic data in graphic
and digital forms.
The current Mark-II development strategy calls for procurement of all
system capabilities through the DMA's System Center, with delivery scheduled
between 1991 and 1995.
Mark-II has been designed to collect, populate, and maintain the National
Digital Cartographic Data Base (NDCDB) and to generate products from it.
The NDCDB win contain digital representations of the 1:24,000-, 1:100,000-, and
1:2,000,000-scale map series. This topologically structured data base will support
the production of the 1:24,000-scale printed maps, and digital subsets will be
made available to the user community for use in GIS or other applications. The
digital data will be written in the Spatial Data Transfer Standard (SDTS) format
for exchange with the user community.
Because of national security concerns, many details of the Mark-90 and
Mark-II R&D are classified. The committee did not examine classified material
and therefore cannot critique the systems' components in detail. From the
unclassified information provided on the overall status and direction of the effort,
Mark-II appears, in concept, to meet NMD's data acquisition, revision, and
production requirements. It was designed essentially and has evolved to meet the
needs of maintaining an up-to-date data base to support the production of
existing NMD products. Because these products will likely evolve to meet
changing user requirements, it is vital that the technological capability of Mark-II
be flexible enough to continue fulfilling those needs.
Product Generation
Product development encompasses both high-volume, standard hard-copy
maps and thematic maps, often one of a kind. Development and engineering
activities focus on maintaining and enhancing current NMD hard-copy produc-
tion. A large part of the effort involves examination of the production process
for 1:24,000 maps in terms of digital production without sacrificing accuracy,
readability, or content. This activity is closely linked to the rules and standards
OCR for page 12
12
development activities.
These plans and activities focus on integrating proven GIS and image
processing techniques into NMD's production capabilities and may result in up-
grading current production prior to full implementation of Mark-II capabilities.
In addition, the effort plans to evaluate the utility of the Mark-II system, which
focuses largely on high-volume products, for production of many one-of-a-kind
products. The objective of the product generation effort is not to replace the
entire production line but to improve specific parts of it when off-the-shelf
technology can replace either aging equipment or improve the current (often
manual) techniques. As part of this effort, NMD ~11 develop a production
model using existing technology that would minimize additional development
costs.
Even though product generation for topographic maps and thematic maps
may appear quite different, they share a number of software, data bases, data
formats, and hardware needs. One of these commonalities includes the inte~a-
tion of the current GIS/commercial symbolization capability within the produc-
tion centers (thus requiring the creation and maintenance of a "digital symbol
library"~. NMD will develop a prototype of these digital standards and use them
to cooperate with vendors of existing commercial GIS software to develop
digitized symbol libraries.
Standards Development
As the organization with lead responsibility for creation and maintenance
of the NDCDB (and possibly an expanded geographic data base system), stan-
dards development is central to NMD's mission. Standards define NMD's
expectations for product content, including formats, topologic structure, treat-
ment of features on graphic products, degree of generalization, and symboliza-
tion specifications. They also include definition of products of uniform consis-
tency and content. Standards need to be established in the context of- well-
defined policy. Through configuration management and the technical instruc-
tions/documentation program, NMD is establishing a highly structured system
for maintaining and defining standards.
In 1983, NMD and the Federal Interagency Coordinating Committee on
Digital Cartography undertook an ambitious program to develop the SOTS. It
is a conscious effort to consider the needs of the entire spatial data community.
The standards mission of NMD includes: (1) awareness of technology,
applications, and programmatic changes and priorities that define new products
OCR for page 13
13
and system designs within NMD's operations and those of other agencies, (23 the
establishment of standards to ensure that standards precede production and that
they are well coordinated with existing production operations and products, (3)
the documentation and distribution of standards and related materials through
the technical instructions program, (4) maintenance of documentation to produce
notices of change, (5) provision of interpretation and assistance to the mapping
centers and others in applying the standards, (6) assessment of product confor-
mance to standards or quality assurance, including geometric accuracy, spot
checks of digital products, review of printed products, and response to user
concerns, and (7) initiation of corrective action when needed for the standards
or their use.
Core standards activities involve the technical instructions program, config-
uration management, and implementation of SDTS. Before 1981, standards
were expressed principally as topographic instructions. By then, the need to
include digital products had become clear, and a technical instructions program
was initiated. It is designed to exercise control of the content, distribution, and
maintenance of the standards. As part of the Mark-II development, NMD needs
to determine what degree of control or configuration management should be
exercised over these standards to make them acceptable.
Configuration management activities are an accounting mechanism for
maintaining a basic capability. The identification and ultimate alteration of
configured items is a highly structured process, particularly for Mark-II develop-
ment. It consists of identifying configured items, assigning responsibility for each
item to an office, and identifying discrepancies in hardware and software, docu-
mentation, and procedures. The configuration management activity also handles
and initiates changes that are warranted and approved by a configuration control
board. Verification and auditing procedures are used to monitor the implemen-
tation of baseline and authorized changes.
The SDTS should facilitate the exchange of spatial information among
agencies and between different hardware and software systems. Work with the
SDTS includes testing the standard and conducting workshops throughout the
user communities (federal, state, and other organizations). The SDTS, as pro-
posed, has four parts: definition and references; the data transfer specification
(ISO 8211~; statements relating to data quality, lineage, accuracy, consistency,
and completeness (user-supplied information); and the list of cartographic
features that supply the model definitions. Phase one testing was completed in
1988 and phase two testing was completed recently. In July 1990, NMD submit-
ted the SDTS to the National Institute of Standards and Technology for process-
OCR for page 14
14
sing as a Federal Information Processing Standard (FIPS). The objectives of
doing so are to: (1) promote the SDTS as a distribution mechanism for spatially
referenced data, (2) reduce redundant data collection efforts, (3) clarify federal
agency responsibilities, (4) coordinate standardized data content and quality
without requiring others to use a standardized data format (conversion should
be transparent to the users), (5) improve data user efficiency by minimizing
redundant data maintenance, and (6) broaden the access and creative use of data
in GIS.
Rules Development
Rules development, which began as a necessary part of product generation,
has grown into a broader activity. In addition to writing rules for generating
products, the effort includes defining the nature of features, their attributes, and
how and when data are collected. Current rules development centers on imple-
mentation of Mark-II: documenting the concept, specifications, principles, and
the rules for automated product generation. The digital line graph-enhanced
(DLG-E) data model is being used. It is a feature-based model used to capture
and store the relevant geographic objects that can be portrayed on a map. The
first step in using the model is to determine the features to be collected and to
define the domain of features- a major undertaking. As of November 1989, only
55 of the 203 identified geographic features had been considered.
NMD rules development is divided between data extraction and product
generation. The extraction specifications will be based on the level of informa-
tion required for specific data base construction at different scales (e.g., 1:24,000
or 1:100,000~; the generalization rules also depend on the scale of the data base.
If the data base is to contain more information than the graphics product needs,
a variety of other decisions ~11 need to be made in constructing the data base.
For example, the data integration applications of the GIS community will require
a robust set of attributes for point, line, and area features.
Technology Transfer
For technology transfer, NMD plans to identify the need for all the divi-
sion's research activities to increase and enhance their outreach efforts. They
also present an opportunity for identifying and making available evolving require-
ments for or innovative applications of NMD data and research results. Such
OCR for page 15
15
information is important in improving understanding of the changing user needs.
Two principal efforts are planned within this activity: one in technology
transfer and one in scientific exchange. Under technology transfer, proposed
projects include development of a communication network for users both within
and outside NMD and identification of NMD technological expertise and assign-
ment of specific responsibilities for technology transfer.
Under scientific exchange, possible projects include training and technical
assistance, a data clearinghouse, documentation, and publications. Training
courses could concentrate on applications of NMD data, and they could involve
participation in cooperative demonstration projects that apply NMD data. The
data clearinghouse could assemble a data base describing the content and avail-
ability of NMD data residing outside the NDCDB; it could also investigate
methods for the public to access such data. Documentation efforts could include
mechanisms to ensure full description of the techniques and results for all
demonstration projects. Publications efforts could include updated circulars
identifying technology transfer and exchange sources within NMD and develop-
ment of high-quality graphics production for publications and presentations.
DEG-E Development
Another part of NMD's ongoing development relates to moving the DLG-E
data structure from an experimental concept to an operational status. Begun as
a research effort a few years ago, DLG-E has been adopted as the data structure
for the Mark-II system. In addition to the explicit statement of rules and features
that NMD identified (see Rules Development above), it is important that NMD
can explicitly identify all elements of the file structures (see Table 3~.
Over the past 15 years, NMD has been instrumental in developing several
new spatial data models. The original impetus for this work was the Land Use
and Land Cover Mapping Program, which required a specific topological struc-
ture for handling land-use polygons. When NMD decided to address the broader
subject of topographic quadrangles, it developed the digital line graph (DLG)
data structure, which could accommodate point and line features in addition to
polygonal areas. The DLG data structure has been widely disseminated and is
often the de facto standard for interchange between different systems. In a
modern spatial data handling environment, there is a strong desire to interact
directly with geographical objects that usually consist of different geographical
elements. For example, a geographical feature such as an Indian reservation
actually consists of a variety of different point, line, and area symbols. A user
OCR for page 16
16
TABLE 3 Some DLG-E Definitions
-
Enti~: A real-world phenomenon that is not subdiv~ded into phenomena of the
same kind.
Feature: A set of phenomena with common attributes and relationships. The
concept of feature encompasses both Entity and Object.
Feature Instance: An occurrence of a feature defined by a unique set of attribute
and relationship values.
Object: A digital representation of all or part of an Entity.
Feature Object: A digital representation of an Entity to which only non-loca-
tional attributes and relationships are associated. A Feature Object may consist
of other Feature Objects and/or Spatial Objects. For the representation of an
Entity to be complete, however, a Feature Object must consist of one or more
Spatial Objects, either directly or through other Feature Objects.
would like to retrieve and display all these graphic symbols by simple reference
to the named feature. When a feature such as a road also forms the boundary
of another feature, the data structure must be able to recognize the fact that it
is shared by the two features. One such object-oriented data structure is now
part of the TIGER system developed by the Bureau of the Census, and the DMA
is developing another.
NMD RESEARCH INITIATIVES
In general, current research within NMD is directed toward developing
ways to capture, represent, store, display, and use spatially referenced infor-
mation to meet the mission of the USGS and the broader user community needs.
Most of these initiatives relate to creation of spatial data layers and the ways
these data are applied to problems, particularly within other USGS divisions. An
important area deals with applying spatial data handling capabilities to global
change; this new initiative expands NMD's mission beyond national boundaries
in response to a deified national need.
This section describes plans for research activities in NMD; only a small
portion of what is presented is ongoing (see Appendix A). The tables in this
OCR for page 17
17
section present possible activities within the research initiative condensed from
NMD planning documents.
Models of Digital Spatial Data
As one of the pioneering agencies in cartographic data structure develop-
ment, NMD has been at the forefront of research into spatial data models that
meet both cartographic production and GIS application needs. Most current
research focuses on possible extensions of the DLG-E data structure.
The DLG-E data model has been designated as the basis for NMD's future
digital spatial data holdings. Research needs to continue on testing the DLG-E
design concepts, extending the model, and implementing it in a variety of envi-
ronments. This research is particularly important in the long term because the
NDCDB provides the framework for a wide range of applications both within
and beyond the USGS.
The DLG-E data structure explicitly defines the logical organization of
components and component relationships of the DLG-E model. These data
structures can, in turn, be translated into file structures by rules specifying logical
implementations of the data structure within a given computing system environ-
ment. Many different data and file structures can be generated from a single
data model to improve the efficiency and general utility of DLG data products.
It is important that alternative implementations of the DLG-E data model be
prepared and tested. These initiatives need to be shared with the user communi-
ty to ensure compatibility.
NMD is aware that although DLG-E is a major step forward in the evolu-
tion of spatial data structures, it is not the end of the process. Current research
plans include various digital spatial data initiatives (see Table 4~. Initiatives
include logical extensions of the DLG-E data model to three dimensions and the
incorporation of temporal data.
Other research would address data handling needs, including manipulation
and analysis capabilities, accuracy assessment, generalization, artificial intelli-
gence, and visualization. The identification of these initiatives suggests NMD's
awareness of user needs. As shown in Table 2, NMD allocated only one-half an
FTE of effort to any of these activities during F Y 1990. Although the proposed
research could provide important steps in the evolution of its spatial data han-
dling capabilities, NMD is limited in both budget and the available expertise.
l
OCR for page 18
18
TABLE 4 Spatial Data Handling Issues
.
Spatial Data Bases
· Conceptual modeling of geographic phenomenon; extensions of the DLG-E
data model to 3-dimensions; handling of temporal data.
· Prototype implementation of DLG-E in extensible (object-oriented) data base
management systems.
· Evaluation and testing of spatial search and access methods.
· Multiple representations of geographic features.
· Use of active data bases for automatic updating across scales.
· Incorporation of spatial operators and computational geometry techniques
within an extensible data base management system.
Manipulation and Analysis
· Interfaces among data bases, GIS functions, and process modeling; modeling
of dynamic systems.
Data integration and analysis at multiple scales (delivering comparable
results).
· Location/allocation modeling.
· Methods for manipulating data from variable reporting zones.
.
Accuracy
· Development of accuracy measures and tests (for both locational and non-
locational attributes).
· Propagation of error through GIS processes.
· Indices of data uncertainty and confidence for GIS products.
Generalization
· Effects of aggregation on spatial modeling.
· Development of scale-to-scale statistics, both spatial and temporal.
· Cartographic generalization techniques.
Artificial Intelligence
· Spatial data handling approaches utilizing advanced data base architectures
that support AI.
· Knowledge-based decision support systems.
· Knowledge-based cartographic design tools.
Visualization
· Display of quality of spatial information.
· Display of time series, 3-dimensional data.
OCR for page 19
19
TABLE 5 Techniques Research Issues
l
Scientific visualization: Modeling and viewing of 3-dimensional data, including
new perspectives for the quality control and revision of DLG and DEM data.
Data capture and digital map production: Methods of pattern recognition to
automate attribute tagging and text placement.
Access to digital data: On-line status maps and text searching utilities coupled
with efficient data transfer.
Improved raster/vector integration: Procedures for viewing raster and vector
data together for analysis and updating of vector data.
Supercomputing platforms for GIS and image processing: Evaluation of the
costs and benefits of using supercomputers for the near-real time applications.
Interaction with GIS: Use of artificial intelligence and other methods to improve
user interfaces to complex geographic information systems.
Techniques Research
NMD's primary technical needs are improved procedures to capture, store,
and display geographical information efficiently. Although much of its R&D in
this area relates directly to the Mark-II modernization project, NMD needs to
keep up with the rapid changes in computer-oriented hardware and software
relating to the creation and use of its data. To meet this need, NMD maintains
a research program in new techniques (see Table 5) and has a variety of ongoing
technique-related activities to meet short-term goals (see Appendix A). In
particular, NMD is evaluating new sensing systems (e.g., radar) and existing
hardware and software products. They include several platforms, from micro-
computers to high-resolution workstations and supercomputers. NMD is also
experimenting with software systems that integrate raster and vector data.
Within NMD, plans include the development and testing of new and innova-
tive techniques for data manipulation, problem solving, and visualization. Most
of the plans and activities involve manipulating and using digital cartographic
data and remotely sensed imagery in natural resource investigations as well as
developing new techniques to facilitate future digital data use.
OCR for page 20
20
TABLE 6 Remote Sensing and Image Processing Issues
. . .
Preprocessing: Radiometric corrections, geometric corrections, merging of
multisource data sets.
Information extraction: Classification and pattern recognition, enhancement
and filtering, image analysis, feature extraction, DEM analysis and generation.
Calibration: Geometric corrections due to topographic effects, atmospheric
corrections, and absolute and relative radiometric corrections.
Interactive and batch processing development.
SPECIFIC TOPICS IMPORTANT TO THE INITL~TIVE
· Mapping (topographic and cartographic) includes optical and radar systems,
extraction of DEMs from stereo pairs, and cartographic correction and require-
ments.
· Change detection involves research on automated methods to detect and
identify changes from remotely sensed image data for both map updating and
global change mapping.
· Expert systems are of continual interest, and it is important for NMD to keep
up with developments and to develop in-house expertise for transferring expert
system research to an operational/production environment.
· Awareness of research involving high-resolution spatial and temporal data is
important for data handling, manipulation, and analysis.
· Calibration research for both geometric and radiometric effects needs to
continue.
Remote Sensing and Image Processing
In addition to techniques development, NMD has a specific initiative on
remote sensing and unage processing. Its scope is broad and usually depends on
the types of applications and interests of particular groups. Research in this area
is important for NMD to take full advantage of the multispectral-multitemporal
digital data acquired from advanced satellite sensor systems. Plans and activities
in remote sensing include sensor evaluation and instrumentation (see Table 6~.
Indeed, NMD may need to expand activities to take full advantage of the infor-
mation potential inherent in the data to be collected by the Earth Observing
System (EOS) planned for the late 1990s. In particular, NMD needs to experi-
OCR for page 21
21
ment with new data products used to monitor key land surface parameters, and
modeling procedures using EOS and other global data will be essential in
fulfilling the needs of global change research.
Through the Global Land Data and Information System (GLDIS) project,
NMD is likely to be the major EOS land remote-sensing archive. It is also
apparent that remotely sensed data will be a primary input to the mapping
process. Therefore, it is important that NMD remain in the forefront of new
sensing systems.
Thematic Mapping
Thematic maps derived from a wide range of activities within the USGS
(i.e., NMD, the Geologic Division, and the Water Resources Division) include
geologic and hydrologic maps, land-use and land-cover maps, national atlases,
and results of GIS analyses. NMD is involved with the thematic maps of the
other divisions primarily in the final production (printing) process. This initiative
emphasizes research related to the collection, analysis, and display of thematic
data sets. Plans and activities include development of custom techniques and
classifications of specific categories of thematic data and their subsequent
integration with base or other thematic data. Research topics of a more immedi-
ate nature deal with digital base maps, color electronic prepress preparation, line
work scanning, and plotting techniques. Plans for thematic activities for the
longer term include standards, data integration, data archiving, revision, deriva-
tive products, interactive map design, and visualization.
Data Applications Research
NMD's data applications initiative is designed to broaden and strengthen
the current research program to include cooperation with outside agencies.
Evaluation of NMD procedures and programs helps to ensure that current,
proposed, and future products and services meet the end-users' requirements.
A variety of cooperative research projects is needed for applications development
and product testing. These cooperative research efforts span the range of users
from other USGS divisions to other federal, state, and local agency users to
private concerns to academic institutions. Such efforts are often conducted on
a cost-sharing basis.
OCR for page 22
22
TABLE 7 Data Application Issues
.
Data Types: Evaluation of digital data needs of outside users
· Standards
· Formats
· Scale and resolution
· Linkage to large-scale data
· Expanded attribute tagging
Data Collection, Update, and Dissemination
· Aggregation and generalization
· Documentation and frequency
· Standards for non-NMD data
· Distribution of non-standard products (image maps, etc.)
Analysis Tools and Techniques
· New software systems
· New analytical functions
· New mapping products
· Socioeconomic applications
New User Communities
· Identification of new users
· Increased exposure of NMD digital products
· Innovative use of products
· Expanded user support system
This initiative comprises applications for GIS, image processing, 3-dimen-
sional modeling, and customized thematic map use in analyzing and displaying
spatial data.
Research tasks (see Table 7) relating to analysis tools and techniques
involve testing new software systems and internally developed analytical tech-
niques. These systems may produce new mapping products to support internal
and outside agency applications; these new products could be tested to determine
their effectiveness for information transfer and user needs. In addition, conven-
tional NMD data products need to be used with these new tools to see if the
products are effectively supported.
To maximize exposure of its products, NMD needs to identify, cultivate, and
educate new user communities. Experienced users of NMD data products can
OCR for page 23
23
often identify the strengths and weaknesses of current and proposed digital data
products and standards. Innovative use of the data will strengthen NMD's user
base and identify growth areas for future products.
Global Change Research
Scientists from many disciplines have realized the important interactions
among the components of the Earth's environmental system. There is now a
national focus on the need for improved understanding of the global environ-
ment. As scientists from many disciplines begin to define the range of complex
interactions among the lithosphere, biosphere, hydrosphere, cryosphere, and
atmosphere, the treatment of the spatial and temporal data bases for these
components offers several challenges. How can data collected on a local scale
be extrapolated to regional or global scales, and what types of generalizations can
be made in those extrapolations? How can spatial data analysis techniques be
applied to such global problems? How can the resulting data be managed and
disseminated? Such questions are interdisciplinary, requiring the attention of the
entire scientific community within government, academia, and the private sector.
NMD's research plan in global change has two parallel tracks: investigations
of specific (pilot) areas and development of improved spatial analysis techniques
applicable to global change issues. Plans include the following:
1. The long-range research objective is to develop geographic models of
large spatial extent, from continental to global scale. To meet this objective, site-
specific pilot areas will be used to develop and improve measurement, mapping,
monitoring, and spatial modeling procedures for land surface conditions, charac-
teristics, and behavior that will link directly to biogeochemical cycles, ecosystem
dynamics, and climate and the hydrologic system. Pilot study areas should be of
significant geographic extent, and they should be selected for relevance to global
change research objectives within the USGS.
2. Spatial analysis techniques will be developed, applied, and tested in the
pilot study areas. Examples of required techniques are spatial interpolation that
accounts for geographic features, observational strategies that account for scale
and temporal variability, procedures for linking information gathered through the
analysis of hierarchical data sets from field samples to satellite-acquired informa-
tion in multistage sampling techniques and/or other innovative spatial sampling
strategies, and visualization of model relationships and results.
OCR for page 24
24
A critical component of the U.S. Global Change Research Program is the
collection, management, and dissemination of earth science data sets that bear
on understanding the Earth system and its interactions. The EROS Data Center
(EDC) has been identified as the distribution and archiving facility for land-
related data collected by sensors of the EOS scheduled to be orbited by NASA
In the late 1990s. In addition, the USGS, through EDC, is currently negotiating
to become the North American node for the United Nations Environment
Programme/Global Resource Information Database (UNEP/GRID).
As recommended by NASA, NOAA, NSF, and the DOI, the USGS is
developing a Global Land Data and Information Management System (GLDIS)
at EDC to archive, process, and distribute land-related data sets. The goal of
GLDIS is to develop improved methods and systems for handling extremely large
data sets for global change research. These developments have four primary
components. Data collection includes the conversion of relevant earth science
data from raw form (e.g., aerial photography, radar imagery, and satellite data)
into usable digital or map formats. This activity will emphasize methods of
digitizing and evaluating the utility, accuracy, and quality of existing earth science
data. Information management Will provide an on-line global land data directory,
catalog, and inventory system, which includes on-line image and data browsing
capabilities along with network linkages to other data centers and directories.
Data set development will address the processing of raw or derived data sets to
create calibrated and derivative products or enhancements to existing derivative
products. Data management and dissemination Will address data storage, archiv-
ing, and dissemination. These activities could include: (a) research on durable,
high-density archive media, (b) conversion of existing Landsat data to these
media, (c) integration of Landsat, SPOT, and other satellite data with existing
earth science data, (d) development of graphic output products, and (e) research
on various standard formats for data distribution. The goals of GLDIS are
similar to those that should be encompassed with the NDCDB or a prototype of
a national spatial data base.
Representative terms from entire chapter:
data structure
