3
Priorities for Maintaining and Enhancing Spatial Data Management

For successful science, geographic research depends on specific tools and methods for acquiring, managing, processing, and analyzing spatial data. Since its inception the USGS has held responsibility for the basic geographic mission of mapping the surface and sub-surface features of the United States. Mapping methods have evolved from field-based surveys to include interpretations of hard-copy aerial photographs, and manipulation and analysis of remotely-sensed digital data. In response to changing customer needs the USGS’s mission and vision statements articulate the need to accelerate the transition from traditional products (typically paper maps and reports) to new digital geospatial products. This chapter reviews the origin and types of existing geographic tools and methods at the Survey for data management and identifies priorities for their maintenance and enhancement. The following topics are primary and secondary priorities in spatial data management that coincide with the scientific mission and capabilities of the USGS:

Primary

  • Collection and handling;

  • Representation; and

  • Integration.

Secondary

  • Data mining;

  • Historical data;

  • Managing the security of data; and

  • Spatial data reserves for decision making



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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 47
Research Opportunities in Geography at the U.S. Geological Survey 3 Priorities for Maintaining and Enhancing Spatial Data Management For successful science, geographic research depends on specific tools and methods for acquiring, managing, processing, and analyzing spatial data. Since its inception the USGS has held responsibility for the basic geographic mission of mapping the surface and sub-surface features of the United States. Mapping methods have evolved from field-based surveys to include interpretations of hard-copy aerial photographs, and manipulation and analysis of remotely-sensed digital data. In response to changing customer needs the USGS’s mission and vision statements articulate the need to accelerate the transition from traditional products (typically paper maps and reports) to new digital geospatial products. This chapter reviews the origin and types of existing geographic tools and methods at the Survey for data management and identifies priorities for their maintenance and enhancement. The following topics are primary and secondary priorities in spatial data management that coincide with the scientific mission and capabilities of the USGS: Primary Collection and handling; Representation; and Integration. Secondary Data mining; Historical data; Managing the security of data; and Spatial data reserves for decision making

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey This chapter focuses on the data that support USGS research efforts in pursuit of its mission as one of the nation’s leading producers of natural science information. PRIMARY PRIORITIES Spatial Data Collection In addition to aerial and ground survey methods, imagery is the USGS’s most common source of spatial data. The Survey began using aerial photographs for mapping in the 1930s (USGS, 2001a; Figure 3.1); by the 1970s satellite imagery became the primary source for capturing broad-scale geographic information (Figure 3.2). FIGURE 3.1 Early aerial photography, a precursor to modern remote sensing, was an adventure. This pilot (W.Sidney Park, on the right) and photographer are about to begin a 1922 photography flight with the most modern photographic equipment then available, the small camera on the ground under the plane. SOURCE: USGS Field Records and Photography Library, Denver, Colorado.

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey FIGURE 3.2 Landsat 7 image of the United States. SOURCE: U.S. Geological Survey EROS Data Center.

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey FIGURE 3.3 Digital Elevation Model, North American Continent. SOURCE: U.S. Geological Survey EROS Data Center. The most important link between imagery and the final map is the Digital Elevation Model (DEM; Figure 3.3). Through the Geography Discipline (previously National Mapping Program) the USGS led the collection and dissemination of digital elevation data for the United States from the 1970s through the 1990s. In September 1999, the USGS completed national coverage (except Alaska) of 7.5-minute DEMs with data points at 30-m intervals. Toward the end of the mapping effort, most of the elevation models were produced by private firms, rather than the USGS. As data sources change and improve, the trend in the USGS is to involve a greater number of participants in map and geographic information production and use. The National Digital Elevation Program (NDEP), for

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey example, encourages partnerships with federal, state, local, and private sector organizations for building and maintaining digital elevation data. The USGS is positioned to provide the framework for geospatial information and should serve as the depository and portal for the Department of the Interior and other clients, and to furnish other information and derivative products in support of effective research and decision making (NRC, 2001b). One long-standing USGS data collection project is in Antarctica, where the Survey coordinates mapping and geodetic activities with the international scientific community under the terms of the Antarctic Treaty. The USGS maintains the Antarctica Geographic Names Data Base and the Atlas of Antarctic Research. Agency scientists, including geographers, collect Antarctic satellite imagery, GIS, gravity and glaciological data, and establish a geodetic control database for Antarctica (Draeger et al., 1997). Successful data handling relies on standardization, especially when many users are involved. At the USGS, data transfer (i.e., the movement of data from one software or hardware system to another) follows spatial data transfer standard (SDTS) formats. In addition, the USGS is the lead agency for national elevation data, one of the seven framework layers of the NSDI. Finally, the Geography Discipline of the USGS produces a wide variety of digital spatial products for distribution, all using a common software platform. Metadata are important components of spatial datasets that document the datasets and include the chronology of the data processing steps, extraction guidelines, desired and actual positional accuracy, and the temporal characteristics of the data. Metadata are an essential component of database management. Spatial Data Representation Digital data from aerial photography and satellites are represented as models, usually in the form of maps, for meaningful interpretation and archiving. The USGS uses two primary types of data models, raster and vector, for converting raw data into finished map and image products. In raster models the region is divided into rectangular blocks, or cells, with each cell containing a single measured value. Thus, a raster model for terrain elevation consists of a region of cells containing the land surface elevation within each cell. In vector models points, lines, and areas form the basis of representation for the originating data. In a vector model points are represented by coordinates, lines have starting and stopping points, and areas have lines as boundaries. Objects in a vector model therefore have characteristics or measurements that determine their appearance. An outline

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey FIGURE 3.4 Color-infrared color composite mosaic of the United States prepared from 16 images from the Advanced Very High Resolution Radiometer (AVHRR) sensors on the meteorological satellites NOAA-8 and NOAA-9. The images were acquired between May 24, 1984 and May 14, 1986. SOURCE: NOAA and USGS.

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey of states and counties of the United States, for example, can be represented in a vector model consisting of points and lines that describe the boundaries of each political division in the overall map. At the present time the USGS provides paper and digital maps that subdivide the entire nation into topographic map quadrangles or special maps such as those defined by state boundaries. To examine data from large areas users may need to assemble several quadrangles or other maps; however, many of these maps do not fit together seamlessly. USGS researchers are designing new seamless databases that can be accessed without subdividing the information into topographic quadrangles. A seamless geographic database is more efficient to search, and researchers can download only the data they need instead of receiving all the data for a particular quadrangle. This sort of advancement in geospatial technology represents the next generation of maps, and the USGS should engage in fundamental research to support the approach in order to make it a reality. The seamless characteristic faces problems in matching features from one section of the map to adjacent sections, as well as connecting sections with different levels of detail. Landscape models are now core research tools in the Biology Discipline and Water Discipline (Figure 3.4). Aquatic habitat patterns studied at the John Day Reservoir in Oregon combine a spatially explicit hydrological model with GIS data. Landscape models form the core of the Geography Discipline’s urban dynamics research. Simulations based on algorithms derived from past patterns of growth can be used to predict future urban expansion. Links between these models and models of hydrological response or habitat quality could improve predictive capability. Urban growth models can also incorporate feedback from changes in the biophysical environ-ment. The committee urges the Geography Discipline to increase and streng-then its ties with the other disciplines in order to establish powerful integrative models with productive linkages. This objective is difficult because of the disparate vocabularies, scales, and methods of data collection used by various natural and social sciences. Computational and content limitations in current GIS data models impede the generation of realistic landscape models of temporal geographic processes, such as stream discharge, chemical runoff through agricultural sites, and headwall erosion. Digital representations of geographic information lack appropriate data structures to handle elements with temporal variation in geometry or content. In addition, network models of spatial routing and capacity analysis are not integrated with visualization tools and mapping functions. For example, visualization tools could help transportation engineers monitor drift in iterative models and breakdowns in complex

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey networks such as regional or national power grids, telecommunications networks, or sewage or water pipelines beneath city streets. Emerging information technologies, such as animation and multimedia, introduce new requirements for cartographic design which remain underdeveloped and untested. Web interfaces could provide public access to data archives. Users would benefit from simple, easy-to-use interfaces for cartographic data made available on the Web. Without this access to public domain data for citizens and other users outside of the USGS, the nation’s investment in improved data representation and analysis will not reach its full potential. Integration of Spatial Data In addition to collection and representation, another critical aspect of maintaining and enhancing spatial data is the integration of a variety of data types in a single product. For many years the production of 1:24,000 topographic maps was the framework for the nation’s spatial data. The coverage of the continental United States by topographic maps was completed by 1991, but this paper-based method is now obsolete. It is impossible to adequately update the 55,000 topographic quadrangles in a timely fashion. Government agencies, companies, and private citizens require digital products. In response, USGS is developing The National Map (see Chapter 4). (The USGS refers to the programs and people in the Geography Discipline as “the National Map” but the committee believes it would be more appropriate to label only the product, “The National Map.”) The National Map demands a seamless database with thematic layers that will require substantial research and time for development. The first phase of the project is the creation of the National Elevation Dataset (NED), a seamless raster product for the entire nation that was assembled for the continental United States based on 7.5-minute Digital Elevation Model (DEM) source data (10-m and 30-m spacings). The committee recognizes that a major challenge is integrating source data other than standard USGS DEMs into the NED. For example, joining digital raster graphic (DRG) quads into seamless images is not a simple procedure. DRGs are reprojected to Universal Transverse Mercator (UTM) coordinates setting up a conflict between the data and the map frame. Attributes in a national seamless database must be rectified and standardized, creating another challenging research problem. For example, many resource classifications do not have clear, mutually exclusive definitions for each class of feature, and there is little standardization among the many state and private generators

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey of such data. The effective management of these so-called fuzzy classes, construction of within- and between-class connections, exploration of relationships between classes, and evaluation of the results of these actions are formidable tasks. They require integrated database management, statistical graphics, and statistical analyses that are not yet fully developed (Fosnight, 1992). SECONDARY PRIORITIES Spatial Data Mining1 USGS clients require increasing amounts of data to solve research and management problems. Research on data mining can inform users about exploring and retrieving information from large data archives, but efficient and effective data mining remains an important unsolved problem (Ester et al., 1997). Many large databases currently being constructed contain spatial and temporal attributes, offering the possibility of discovering or confirming geographic ideas relevant to natural science and the USGS mission (Miller and Han, 2001). Decision makers can discover new data and improve the quality of their environmental policies by mining these existing archives (Yuan et al., 2001). Relevant areas of data mining research include algorithms to browse the extensive archives of satellite data for predicting droughts, for evidence of pest infestations in forested areas, and for similar content-based pattern matching. Satellite information is being collected at such a rate that human inspection of individual data frames is no longer feasible. The USGS should develop automated data mining methods to provide timely natural science and hazards-related information. Historical Spatial Data The USGS has a long-established role of maintaining and archiving historical data. Its archives have become a valued resource for geographic research involving changes through time, and they are critical to numerous research applications ranging from hazard prediction and land use planning 1   Data mining is defined as an information extraction activity whose goal is to discover hidden facts contained in databases. NASA Astronomical Data Center (http://adc.gsfc.nasa.gov/adc/adcdatamining.html).

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey to assessments of environmental change. Survey archiving and distribution of spatial data are guided by the Land Remote Sensing Policy Act of 1992 (15 U.S.C. § 5601 et seq.), and by rules established by the Executive Branch as the 1996 National Space Policy. Each policy requires the USGS to maintain a national archive (Draeger et al., 1997). USGS archives contain aerial photographs used for topographic mapping with black and white photos dating back to the 1930s. The USGS’s Earth Resources Observation Systems (EROS) Data Center stores negatives of more than 8 million aerial photographs from many federal agencies (USGS, 2001a). Also archived at USGS-EROS Data Center is the vertical color and color-infrared aerial photography generated by the National High Altitude Photography (NHAP) and National Aerial Photography Program (NAPP). In addition to the aerial photography, USGS-EROS archives Landsat Multispectral Scanner (MSS) and Thematic Mapper (TM) imagery from NASA, Advanced Very High Resolution Radiometer (AVHRR) data from NOAA, and some declassified military imagery. The Center’s holdings are expanding to include enhanced TM data from Landsat 7, digital terrain data from the recent Shuttle Radar Topography Mission, and data from new multispectral sensors aboard NASA’s recently launched TERRA satellite. The USGS is now working with the SPOT Image Corporation to acquire and archive 10m-resolution images, primarily of North America (Kelmelis, 2001). The USGS has generated most of the historical environmental data in the United States, ranging from stream discharge data collected by the Water Discipline to plant and animal data collected by the Biology Discipline. Methods of collection for land data varied from land survey records made a century and a half ago to the most recent satellite images. These data provide an indispensable context for present processes in the Critical Zone, the place of interaction between nature and society. The federal government and others continually collect new information for monitoring the landscape. With the automation of data collection primarily through remote sensing, the rate of data collection has increased, with a concurrent increase in the amount of historical data. The rapid increase in historical data will be accelerated further with the advent of The National Map. All high-quality spatial data collected should be archived in such a way that it can be retrieved easily and used with modern GIS (Jensen et al., 1996). Metadata should also be archived and accessible. As technology continues to advance, storage media will change, data formats will be modified, and data from new domains will be collected. The committee urges that the USGS take a national leadership role in research on how to preserve the national geospatial data archives, how to manage access and

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey update of this national treasure, and how to integrate new datasets and data domains as they become relevant. Furthermore, the committee notes that research is needed on how best to incorporate historical geographic data into the current digital formats. Historical geospatial data include data collected or created in digital form (remotely-sensed data and digital cartographic datasets), as well as geospatial data in hardcopy (paper maps) that will need to be converted to digital form. The present demand and access needs are large [e.g., in FY2000, the Land Processes Distributed Active Archive Center (LP DAAC) distributed over 9,200 scenes (views of particular areas of the surface of the earth), consisting of 160,000 files or 1.1 terabytes of data to customers via FTP or the web] and this usage will continue to grow. Historical spatial data priorities for the USGS lie in the following areas: digital conversions of hardcopy spatial data, research using historical data to assess human impacts on the natural environment, understanding slow environmental change, applications in ecosystem management and restoration, development of decision support systems, and finding solutions to methodological problems. Some of these priorities address research needs, others address data processing needs. Digital Conversion of Hardcopy Spatial Data Converting hardcopy spatial data into a digital format is a way to link the past and the future. The paper maps produced by USGS for over a century are a unique and important resource of topographic and land use data. The USGS can and should play a major role in the maintenance, digital conversion, and provision of U.S. historical environmental data from the pre-digital era. For example, in the restoration of the Everglades ecosystem of south Florida, USGS reconstructions of the original flows of water and plant communities provide a guide for engineering efforts to recreate a more natural hydrologic and biologic system in the region. As the Clean Water Act, Endangered Species Act, and other policies make environmental restoration an increaseingly common activity in the United States, there will be greater demands for digital reconstructions of former environments. One of the most important aspects of this task is digital conversion and georeferencing of the Survey’s historical topographic maps. Digital con-version is the most efficient way to preserve these historic maps and make them accessible to users. Other agencies will continue to archive their own datasets, but the USGS, through the FGDC, should provide leadership for data archiving, digital conversion, and access. The FGDC has established a Historical Data

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey Working Group, although it has not been active recently. The USGS should take a leadership role in this group and use it as an advisory body. Documenting Human Impacts on the Environment Historical spatial data play a significant role in the documentation of human impacts on the environment. These data can be linked with other numerical data to assess these impacts. For example, the USGS has historically collected data in tabular and text form on mineral production, stream flow, surface and ground water quality, water use, and other topics in the course of preparing early reports on resources and settlement potential. Other federal agencies that have been significant collectors of historical environmental data include the Bureau of Land Management (original land survey records for the central and western United States), the U.S. Forest Service (forest maps inventory from the 1930s), Agricultural Stabilization and Conservation Service (aerial photographs), U.S. Bureau of the Census, and the National Weather Service In addition, newspapers, company documents, and personal diaries and journals provide important historical data. Historical data have been used for environmental research for decades, but their use is now rapidly expanding. Examples of human impacts that are traceable through historical spatial data are modification of land cover and vegetation, land drainage, wetland loss, changes in river channels, and filling along coastlines. Historical research often reveals unexpected, past land configurations or land uses that are not evident at present and are unknown to the present occupants. Subjects for future research include topics that have received little attention in the past. For example, there is little research done in the United States on whether there are impacts from historical land cover change on water quality and quantity. Relatively little work has been done on the interplay of human-caused change and natural dynamics in geomorphic processes, such as coastal erosion and river channel change. Historical and ongoing land use change has major implications for the global carbon cycle, for energy and water transfer between the land surface and the atmosphere, and therefore, for climate change. Understanding Infrequent or Slow Environmental Processes Understanding of infrequent or slow environmental processes is enhanced by historical spatial data. For many environmental hazard processes, such as severe weather, floods, droughts, wildfire, and landslides, the primary source

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey of information is direct instrumental monitoring or assessment of remnant geologic evidence. However, these processes typically have recurrence intervals close to or longer than the length of the instrumental record, and geologic evidence may be obscured by subsequent events. Therefore, historical records are critical for identifying and estimating recurrence intervals and possible magnitudes of the events. In addition, some environmental processes do not create immediate acute hazards, such as river channel change and shoreline retreat, and they may show relatively little change within the span of instrumental record. Historical time series are important for understanding processes that operate at rates not adequately captured by recent instrumental records. As long as data are appropriately archived, continued instrumented measurements will create a valuable historical record. Historical records may also be useful in documenting human responses. For example, historical maps may be used to show the growth of flood protection measures (e.g., levees) that may influence the magnitude of subsequent floods. Historical data are important in assessing the character of nature-society interactions as well. Longitudinal data on the extent and value of built structures and other infrastructure are important for comparing the economic impact of different natural disasters in the same area. Historical data can contribute to the development of models displaying slow or infrequent environmental processes. Ecosystem Management and Ecosystem Restoration Access to historical data is often important for research on ecosystem management and restoration. Egan and Howell (2001, p.1) observed, “A fundamental aspect of ecosystem restoration is learning how to rediscover the past and bring it forward into the present—to determine what needs to be restored, why it was lost, and how best to make it live again.” This perspective also applies to the broader challenge of bioregional assessment for ecosystem management. A particularly important challenge in ecosystem management and ecosystem restoration is to determine the historical range of variation (HRV) for the ecosystem in question. Ecosystem status responds to natural disturbances that operate on time scales of years to centuries. The HRV represents the range of conditions and disturbance regimes under which the ecosystem is self-sustaining; if pushed beyond the HRV, the ecosystem will move into a state of disequilibrium, threatened species may be reduced or eliminated, and environmental quality may decline significantly. Historical spatial data are crucial to establishing the range of natural variation (Egan and Howell,

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey 2001) and, in some cases, may be the only or best way of establishing a reference model for restoration or management. Recently, historical studies have been initiated as part of regional ecosystem management planning. The Interior Columbia Basin Ecosystem Management Plan provides an instructive example: historical spatial data on forest management practices, including livestock grazing, timber harvest, mining, and road building from diverse sources provided an understanding of the causes of present forest conditions (Oliver et al., 1994). American Indian nations are involved increasingly in managing native lands, often with a goal of restoring to some extent the pre-European conditions and processes, and/or of reviving traditional ways of managing their landscapes (Egan and Howell, 2001, p. 14). Spatial data provide information on the extent and locations of traditional land uses, providing a guide to the process of cultural and environmental restoration. Historical Data for Decision Makers Often, the USGS utilizes historical spatial data to calibrate and test decision support systems. Historical data also are often used as input in simulations, which are critical to developing the scientific understandings that underlie predictive tools. Historical data explain how natural and social systems have responded to trends, stresses, and perturbations. In fact, analysis of historical data is essential for research on processes and behavior of social systems. The USGS’s research on urban dynamics, based on historical data, illustrates the relative roles of natural and human-created features in shaping the spatial pattern of urban expansion. Such findings can enhance models of urban growth that may be used in decision support systems. Also of great value are the USGS’s holdings of historical data on aquatic systems and wetlands, minerals production, energy resources, and other topics. The USGS should continue to work to make these historical data reserves useable in decision support systems. Additional and primary methodological challenges in using historical spatial data are related to data availability and quality. Availability, level of detail, and accuracy decline as the age of data increases. The scale at which the original data were collected may not be the same scale as needed or as appropriate for the current problem, and data cannot necessarily be transferred across scales. Egan and Howell (2001, p. 13) argued that the best approach is to “combine appropriate techniques in a way that is multiscale and cross-referential in order to build convincing, corroborative lines of evidence.” Because the USGS is the custodian for large amounts of historical spatial data, part of the Survey’s mission encompasses basic

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey research into a variety of questions that are historical in nature. In order to meet these responsibilities, the USGS should develop projects within the Survey and through cooperation with external researchers to address basic geographic research questions related to the accuracy, availability, quality, and scale issues for historical spatial data. Managing the Security of Data The policy of many government data centers, including the USGS-EROS Data Center, is to make their databases publicly available. However, in response to the terrorist attacks of 2001, federal agencies are reassessing public access to potentially sensitive data (DOI, 2001). Data that have previously been published may not be updated, and, in some cases, existing data have been withdrawn from the public domain. For example, National Imagery and Mapping Agency (NIMA) maps of certain areas in the United States are not released for general use. The USGS also must be sensitive in the arena of geospatial data that might be used for targeting purposes. However, the restriction of general geographic data, including locational data and digital elevation models, can be detrimental to the basic mission of the USGS. From an economic perspective, the publicly available geospatial data provide important support for government management and private economic activity; sequestering too much data may result in significant disruption. The success of the American economy relies in part on available, low cost, public data. The archiving of valuable data, and the ready availability of those data for legitimate purposes, must not be abrogated unnecessarily. In many cases, it will be impossible to eliminate all public sources of geospatial information on the location of potential targets, so that the elimination of a single USGS source is not likely to result in improved security. The USGS’s general guidelines for assigning restrictions to data distribution (DOI, 2001) define a range of restrictions that might be assigned to documents and data (see Appendix D); these categories include data that are not restricted, partly restricted, or placed off-limits completely for public use. Decisions on restriction of distribution rest with the associate directors. While the current fundamental philosophy is to restrict distribution, distributing data is the driving philosophy of public agencies such as the USGS. In addition, since there are no substantive guidelines, the four associate directors’ restriction categorizations have the potential to be inconsistent. The inherent conflict between security and the need for accessible information remains relatively unresolved because the policy articulated by the DOI is drawn from a legal perspective rather than a scientific one (DOI, 2001).

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey A uniform security policy for spatial data should be developed, and the associate directors should serve as advisors to a single USGS decision maker. To make as much data available as possible, the policy should clearly outline how the mission of the Survey and the security of the nation should be balanced in making decisions for data management. Spatial Data Reserves for Decision Making Good policy decisions that help maintain a robust economy and healthy environment are dependent on data that are as complete, accurate, and freely available as possible. Among U.S. government agencies, the USGS has been one of the major generators and providers of natural science and social science data and information for policy making. The USGS has extensive holdings of past data that will continue to be essential for contributing to informed policy making, and its new mission includes the continued collection of data vital for informed public policy. Before the advent of digital data collection and storage technologies, the USGS and other agencies had already collected a large amount of data, and these pre-digital are still extremely useful. Early topographic maps produced by the USGS include information on settlement, transportation infrastructure, river channels, and, to some extent, land cover and land use. In addition to first-generation topographic maps that may date from a century or more ago, revised editions of maps provide data at subsequent points in time that document changes from the first generation’s measurements. Knowledge about the topography, land use/ land cover, transportation network, hydrography, political boundaries, and place names of the American landscape over time is very important to a vast array of biogeographic, geologic, urban planning, and environmental spatial modeling applications. The integrated archive of this information has historically been the USGS topographic map series, which is currently archived in digital form as Digital Line Graphs (DLG) data in several series. These data form a record of changes to federally owned lands, terrain morphology, and transportation networks dating back over 100 years and provide a unique background for planning and decision making for all levels of government, private companies, and citizens. Complementing these historical data will be an increasing quantity of new data. In the future, a considerable amount of spatial data is expected to emerge from federal agencies, state and local government agencies, commercial firms, and some private not-for-profit sources. A federal agency is the optimal steward for the archiving and preserving of these data for the nation. Although many federal functions have devolved to state and local authorities

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey over the past two decades, management of the nation’s spatial data infrastructure is a federal responsibility because placement in a federal-level agency is the only way to insure national uniformity in content, organization, and accessibility. Although the provision of geospatial data was a government function in the past, the public and private sectors share this responsibility today. However, despite the increasing role of the private sector, local governments, and non-profit groups in geospatial data provision, the USGS should continue to be a major data provider in the future, in part because it is directed to do so by Congress as part of its mission, but also because it administers the nation’s primary imagery and spatial data base archives. In addition, the USGS plays a critical role in developing standards through the FGDC. SUMMARY The USGS has a primary responsibility in the creation, archiving, distributing, and management of the nation’s spatial data related to natural science. The role of the Survey in maintaining and enhancing these data includes the need for basic supporting geographic research, for special attention to historical spatial data and the development of security rules for their use and distribution, and for a continuation of the USGS leadership role with respect to spatial data and its standards. Data processing is not enough. Research into methods for data processing is the only way by which the USGS will improve its provision of natural science data and information that are relevant to society’s ever-changing needs.

OCR for page 47
Research Opportunities in Geography at the U.S. Geological Survey This page in the original is blank.