THE CONCEPTS OF FEDERAL/STATE PARTNERSHIPS
THE CREATION AND MAINTENANCE OF SPATIAL DATA SETS
Historically, the federal spatial data infrastructure has been managed as a set of discrete mapping responsibilities within several federal agencies. The data management role of federal agencies has been that of data stewards for large homogeneous data sets (more often in paper map form rather than digital data sets). Examples of such data sets include the USGS 7.5 minute quadrangle maps, Soil Conservation Service soils maps, census geography maps from the U.S. Bureau of the Census, and the Fish and Wildlife Service national wetland inventory maps. As part of their mandates, federal agencies have collected and published data on maps that were then distributed to all levels of government and the private sector. Relationships with states have been largely limited to ad hoc or cooperative projects, with the states generally taking on the role of suppliers of data to the federal agencies.
This set of historical arrangements has many problems, both obvious and subtle. The activities of the federal government have been established largely by legislative mandate, and although these may be initially tied to need, it has been difficult to keep up with changing technologies and changing requirements in the user community. Mandates sometimes lead to redundancy, as legislators require different agencies to collect and maintain similar or even identical data. Costs are difficult to identify, quantify, or control, and the government finds it increasingly difficult to maintain established programs in the face of constantly increasing
pressures on the federal budget. Finally, programs of spatial data creation have often been slow and inadequately funded for data maintenance. As a result, data are often so out of date that their value is seriously compromised. These problems are increasingly evident when the state data programs are viewed in the same context and as part of the NSDI.
The role of state and local governments within the overall context of the NSDI is changing. The federal coordination of surveying, mapping, and related spatial data activities are defined in OMB Circular A-16. The most recent version of Circular A-16 (October 1990) added a major objective of developing a national digital spatial information resource with the involvement of federal, state, and local governments and the private sector. The mechanism for the involvement of state and local governments and the private sector has yet to be established. This problem was specifically recognized in Executive Order 12906 (signed by President Clinton on April 11, 1994):
The Secretary [of the Department of the Interior], under the auspices of the FGDC, and within 9 months of the date of this order, shall develop, to the extent permitted by law, strategies for maximizing cooperative participatory efforts with State, local, and tribal governments, the private sector, and other nonfederal organizations to share costs and improve efficiencies of acquiring geospatial data consistent with this order.
SPATIAL DATA STEWARDSHIP PRINCIPLES
Organizations that build and maintain spatial data have a vested interest in the quality of their data when the success of critical missions depends on the accuracy and availability of the data. This suggests that data stewardship roles may be served best by organizations that collect data for the purpose of meeting specific operational missions. These spatial data stewards have commitments to their own organization as well as obligations to meet the needs of partners throughout all levels of government. The mission of spatial data stewards could be expanded to meet data needs of multiple organizations as well as their own business needs. Currently, a major mission of the USGS is one of data collection. The stewardship concept introduced here is different from the data steward role of the USGS in that business needs other than a data collection mission also drive
the development and maintenance of data.
The challenge faced by all levels of government is to place data stewardship responsibility as close to the data originator as possible while maintaining an effective national infrastructure. Partnership agreements may result in data stewardship responsibilities shifting between federal, state, and local agencies. Federal agencies should take on new coordination roles to ensure that data are available to meet a growing national need. State and local government agencies with mandates and other needs for certain data should be responsible for building and maintaining these data. This is occurring on an informal basis with many states now but will take on entirely new significance in the future as more organizations depend on others for data.
One clear benefit of moving data responsibility closer to the source and in line with the business needs of organizations is that maintenance of the data is usually an integral part of their day-to-day activities. As an example, the Bureau of Land Management (BLM) has the federal responsibility for the Public Land Survey System (PLSS). But in many states, after years of requesting the necessary funds, BLM is still unable to meet its objectives for modernization of its Geographic Coordinate Data Base. However in Washington State, for example, the State Land Surveyor has the responsibility to maintain current records of surveys and to make them available to land surveyors on demand. The State Land Surveyor also has the responsibility to manage and make PLSS data available to GIS customers throughout the state. As new surveys are recorded, the PLSS section corners are added to the data base. A partnership opportunity clearly exists that would establish data stewardship responsibility close to the source of the data and provide opportunities to meet both state and federal objectives.
ECONOMICS OF THE NSDI
One of the strongest arguments for partnerships is their critical role in reducing the overall costs of the enterprise. After 30 years of experience in spatial data handling, there can be no remaining doubt about one fundamental truth—it is expensive. Spatial data bases are expensive to create and often more expensive to maintain.
Why is spatial data handling so expensive, and where does the money go? How can partnerships help to reduce costs, and what is needed to
ensure that economies are realized? This section looks briefly at these issues and at the ways in which federal/state partnerships in particular might help to minimize the costs of developing the NSDI.
The Costs of NSDI
Spatial data require specialized data collection systems, specialized hardware and software, and specialized training and education of the necessary support staff. All of these factors contribute to the overall cost of spatial data and the NSDI. Although GIS software development is a relatively small part of the overall electronic data processing industry, and few software companies claim a GIS business of more than $100 million annually, it is estimated that worldwide total annual expenditures on all aspects of spatial data handling are on the order of $10 billion.1 This figure should include every aspect of data base creation and use, from investments in mapping satellites, geodesy and positioning systems, through airborne photography, digitizing, data base design, software and hardware acquisition, and analysis, to specialized training. The cost of data input by digitizing and scanning (data conversion in the terminology of activities related to automated mapping and facilities management) is estimated by the Environmental Systems Research Institute to be $4.5 billion2 annually. The OMB recently surveyed federal spending on geographic data activities and found it to be about $4.4 billion (FY1994).3
These figures reflect current practice. As such, they miss much of the potential future role of spatial data in society, and the demands that are driving the interest in geographic information. We have only just begun to tap the potential that is illustrated by the range of consumer products based on digital spatial data bases that are now entering the market such as consumer GPS, visitor street guides to cities, kiosk vending of custom maps, and applications of IVHS.
As with any new field, there is no doubt inefficiency and waste in the current practices of spatial data handling. There is duplication of effort when the same map is digitized or scanned more than once, adding to the already high cost of converting spatial data to digital form. Duplication is hard to document, as it is rarely in anyone's interest to report when it occurs, and it would be impossible to come up with reliable estimates of the costs of duplication on an annual, nationwide basis. It occurs because there is little knowledge of what data sets exist or because the data might not be at appropriate detail or accuracy to be shared. The need to establish a clearinghouse has been recommended by the MSC4 and is being
developed as a prototype by the FGDC. Ironically, it can cost money to avoid duplication by developing data indexes, improving communication through conference attendance, and building data sets to meet other organizational requirements.
As the field matures, some problems will disappear as communication becomes easier and the level of general information on activities improves. But that will leave the more difficult problems. Duplication is obvious when two agencies each digitize the same map. It is much more difficult to persuade two agencies to collaborate in the collection of raw data, or in analysis, especially when collaboration requires some loss of autonomy or disruption of traditional professional boundaries. It is easy, also, to overestimate the occurrence of duplication by overlooking the subtler aspects of spatial data needs. Two agencies may both need wetlands data, but for different purposes, using different definitions and different levels of precision and accuracy. In such cases, reduction of duplication can be a lengthy and difficult business. Yet the rewards can be enormous. Although digitizing is expensive, and it is important to avoid obvious cases of duplication, the potential economies that can be achieved by collaboration and coordination in the total mapping effort are equally significant.
One of the major barriers to realizing potential economies from decreased duplication is the lack of agreed-upon standards, particularly data content standards, as previously discussed. The problems with different software systems should be allayed with adherence to the Spatial Data Transfer Standard (SDTS), which is now a Federal Information Processing Standard (FIPS-173). Specific content, accuracy, and metadata standards present much of the challenge facing the spatial data community. The various subcommittees of the FGDC are currently addressing many of these standard issues from the federal perspective. However, the state and local governments and the private sector need to be active participants in the standards development process.
Minimizing Costs Through Partnerships
It may appear that partnerships are exactly the wrong way to reduce costs in an expensive enterprise like the NSDI. It costs money to organize and facilitate partnerships. Partnerships can seem cumbersome when compared to a lean, efficient organization that carries out its mandate as inexpensively as possible. However, no matter how internally efficient an organization is, it is inefficient in the broader perspective if it duplicates the products of others. In practice, we believe that partnerships can reduce
the long-term cost of NSDI in three major ways.
First, partnerships are an effective way of achieving consensus. Instead of each agency acting independently, partnerships create a sense of shared responsibility for the product and its use. Partnerships broaden the basis of support for projects and help to ensure that they survive to meet the needs of society. Partnerships between federal, state, regional, and local governments act to dispel the perception that one level of government ignores the others or knows better. In economic terms, partnerships broaden the resource base by sharing costs while enhancing the benefits of spatial data.
Second, partnerships can encourage a clear division of responsibilities even when the data needs are shared. Historically, responsibility for spatial data has been divided between different levels of government in the United States on the basis of map scale and types of data. For example, the federal government has concentrated on producing maps at 1:24,000 scale and smaller while leaving larger-scale mapping to local and state governments and the private sector. States have mapped themes that match their areas of responsibility, such as transportation. These familiar divisions are becoming confused because of the radical changes resulting from the introduction of digital technology. For example, the federal government is supporting a digital orthophoto quarter-quad (DOQ) program with spatial resolution and positional accuracy higher than that of the 1:24,000 scale mapping programs. GIS is being used by all levels of government to take advantage of these new data types and to integrate data from a wide variety of sources. In a world of high-speed communications and distributed data bases, we may need entirely new concepts of ownership of data, or responsibility for its creation. This may take the form of a division of responsibility along entirely new lines, with the federal government responsible for data standards and quality control, and state and local governments responsible for data collection and maintenance.
Third, division of responsibilities within partnerships can promote investment so that we develop entirely new ways of reducing costs. Salaries account for by far the largest share of the costs of spatial data, whether they are paid to digitizer operators, programmer analysts, or field workers. The most effective ways of reducing those costs lie in better technology and better training. Spatial data handling and GIS have grown to the point where creative strategies are needed to promote new methods and better technologies and better development of human resources; however, no agency or level of government has assumed a leadership role in such developments. Partnerships could foster a sense of shared respon
sibility between all levels of government, the educational sector, and private industry. We need partnerships that foster more efficient data collection activities while at the same time fostering a more productive and responsive human resource sector.