A variety of approaches are employed by federal agencies and SHPOs for determining the need for an archaeological survey. Many SHPOs rely on the professional judgment of staff archaeologists. Many employ a "short-list" of defined criteria based on the environmental characteristics of a project area and/or other relevant factors. A few have begun to develop predictive models that use a larger set of environmental characteristics shown to be reliable indicators of site sensitivity.
For many years, the New York SHPO has relied on the presence or absence of known archaeological sites in proximity to a project area as the sole criterion for determining the need for survey. If the Area of Potential Effect for a federal, state, or local undertaking falls within one-half mile of a known archaeological site on the New York SHPO's inventory map, the SHPO will recommend that a Stage 1 survey be conducted. This is based on the premise that the project area is in an environmental zone that may contain archaeological sites, given its proximity to other known sites. The SHPO relies on its own statewide inventory, which includes over 11,000 archaeological sites, and the inventory of the New York State Museum, which contains over 9,000 sites, to make survey determinations.
This "inventory map" system was developed at the time that the New York State Historic Preservation Act of 1980 was being promulgated. During the 1970s, the only archaeological surveys that were being undertaken in New York State (with rare exceptions) were surveys for Department of Transportation (DOT) and Environmental Protection Agency (EPA) projects--the two agencies identified as the principal "land-disturbing" agencies in New York. Through the efforts of the archaeological community and the SHPO, agreements were reached with these agencies, and archaeological surveys became a routine component of DOT and EPA projects. However, the SHPO was not successful in its efforts to get other federal or state agencies to conduct archaeological surveys unless there was a known archaeological site within the Area of Potential Effect.
With the passage of the state historic preservation act, the SHPO proposed the use of environmental criteria as a way to determine the need for archaeological surveys. This approach was opposed by the development community, however, and did not survive the political climate of that era. The inventory map system was subsequently developed as a more politically palatable method to encourage archaeological survey activity in New York State.
In the intervening years since the adoption of this system, the inventory map approach has supported a considerable amount of survey activity in New York. The wide distribution of known sites, mapped with one-half mile buffers, creates a "blanket" of sensitivity zones over most of the major drainages of the state, and approximately 50 percent of the projects reviewed by the SHPO fall within one of these archaeologically sensitive areas. These zones are continually expanding as new sites are reported and added to the statewide inventory. The results have been impressive. Currently, the New York SHPO's library of archaeological surveys includes over 4,000 reports. Each year since 1989, for example, we have been adding between 250 and 400 new survey reports to the library, and between 150 and 500 archaeological sites to the statewide inventory, most discovered as a result of Stage 1 surveys (Fig. 1). During the same seven-year period from 1989 through 1995, these surveys covered an average 12,057 acres of New York State each year.
Nevertheless, using an inventory map as the basis for determining the need for archaeological surveys has inherent limitations. The statewide inventory of known sites in New York represents a biased record of archaeological activities that cannot be said to reflect a scientifically meaningful distribution of sites across the state. There are whole towns and major portions of some counties where there are no sites reported, probably because there has been so little archaeological activity there over the years. Relying on the site inventory to determine the need for survey guarantees that there will be little in the future as well, since these areas fall outside the sensitivity zones.
The inventory map reflects only locational information on archaeological sites, and as such is not truly a predictive tool. It cannot and does not attempt to use known site locations, and the behavioral and environmental factors that they embody, to predict the locations of sites where they are not known. For example, despite the fact that many of the sites in the inventory are floodplain sites, all floodplain locales in the state are not identified as sensitive. Only those floodplain locales that are within one-half mile of a known archaeological site fall within a sensitivity zone. While this criterion draws in most of the major floodplains across the state, it also excludes some which are undoubtedly highly sensitive for sites.
These limitations have been emphasized by numerous archaeologists and professional archaeological organizations over the years, especially the New York Archaeological Council (NYAC). Indeed, the community of professional archaeologists in New York almost universally rejects the approach currently employed by the SHPO to trigger surveys. The SHPO's State Board for Historic Preservation appointed a committee to study the issue and the 1990 report of that committee recommended replacement of the inventory map approach with a predictive model. That same year NYAC also passed a resolution denouncing the use of the inventory map approach and supporting the implementation of the committee's recommendations. In addition, NYAC recommended that the SHPO discontinue using the inventory map approach in lieu of professional evaluation and judgment, until an appropriate predictive model could be developed. In 1991 the SHPO acknowledged the need to develop a better approach for determining survey and began to consider means by which a predictive model could be created.
There is a consensus among archaeologists that predictive modeling is one of the most effective techniques for determining archaeological sensitivity. Sophisticated models are quantitative, observation-based predictive frameworks, the results of which are scientifically reproducible, verifiable, and directly applicable to the decision-making process. In application, most complex models have been shown to be accurate predictors of site sensitivity. Therefore, they are extremely valuable tools for determining the need for an archaeological survey.
Unfortunately, an informal survey conducted by the New York SHPO indicated that most federal agencies and SHPOs were not widely using predictive models. Most have not had the time or the resources to develop, apply, or test appropriate models. Even where predictive models are available for use, their application typically involves the time-consuming quantification of a large number of environmental characteristics for a project area and entering these attributes into a computer format suitable to run a model. Undertaking this level of effort is often impossible with the existing staff limitations and project time frames that hamper most state and federal offices.
While this discovery was discouraging, the rapid development of Geographic Information System (GIS) software suggested an obvious solution. The development of a standardized, broadly applicable, GIS-based predictive model would ensure that application of the model on a project-by-project basis would not be a time-consuming or labor intensive task. Geographic Information Systems such as ArcInfo can automatically calculate the environmental measurements necessary for the modeling process and, equipped with an appropriate predictive algorithm, can compute site sensitivity from the data gathered. Furthermore, GIS output can be produced cartographically, allowing the results to be interpreted and disseminated quickly and easily in map form. The easy application of a GIS-based model could make it practical for SHPO functions, which often require evaluation of hundreds of projects within a short time frame.
The potential development of a GIS-based predictive model for New York began to come into focus during 1992. The SHPO, which is a part of the New York State Office of Parks, Recreation and Historic Preservation (OPRHP), established a relationship with OPRHP's State Parks Research Institute in that year. The Research Institute maintained state-of-the-art GIS technology and expertise, and since 1989 had been assisting the SHPO to create a GIS database of state- and National Register-listed properties in New York. Although the institute was largely supported by external grants and income-producing projects, it pledged its support of the predictive model project, as time allowed.
That same year the SHPO entered into a cooperative relationship with the New York State Department of Environmental Conservation (DEC) to create a GIS database from the SHPO's statewide inventory of archaeological sites. The SHPO database of known archaeological site locations, combined with GIS overlays for floodplain, surface hydrology, soils, and other environmental overlays available from the DEC, could then be used to develop a predictive model to be used in the state. Over the subsequent year-and-a-half, the statewide inventory of over 11,000 known archaeological sites was methodically entered into GIS by a single DEC staff person. Since then, SHPO staff have been verifying every single entry, as well as adding the 150-500 new sites that are reported to our office each year. This effort is ongoing.
While progressing steadily, the SHPO also began exploring other options, knowing that the development of a modeling tool would take time. In 1994 the SHPO began relying on professional staff evaluations (rather than the inventory map) on a trial basis for all state parks (OPRHP) projects. Taking advantage of the close relationship between the SHPO and the OPRHP, this approach is currently being employed for all projects on the quarter-million acres of park lands owned by New York State. Nevertheless, the experience suggests that decisions based solely on professional judgment, without the use of well-defined and objective decision-making tools, are often difficult to defend. More recently, the SHPO has been working closely with the DEC on the possibility of developing a checklist approach that would incorporate environmental factors into the archaeological sensitivity assessment and place decisions within the context of project planning and environmental review.
As the DEC data-entry portion of the GIS database project neared completion in 1994 the SHPO began to explore the additional needs of the predictive model project, including the construction of appropriate attribute fields for the GIS inventory of sites; acquisition of GIS overlays not currently available for critical environmental factors; and researching existing predictive models that might be appropriately applied. A number of predictive models developed in New York, including one of statewide application, were identified which could likely contribute to the SHPO effort. In addition, on a more national basis a number of planning, environmental, and cultural resource management firms have developed GIS-predictive models that may be adaptable to New York. With the assistance of the state's DOT, the SHPO has begun to evaluate models recently developed by other sources within the industry.
Unfortunately, the reductions in resources and staffing that all state and federal agencies have suffered in recent years have conspired to slow the New York SHPO's progress. After suffering a staff reduction in 1994, the State Parks Research Institute was eliminated in 1995. Although the SHPO obtained some of the GIS hardware and software maintained by the institute, additional major purchases had to be made, and the loss of much of the Institute's technical expertise has not yet been replaced. At the same time, our own SHPO staff has continued to suffer a series of reductions. Staffing levels have been reduced by 35 percent since 1990. Whereas the SHPO once had eight archaeologists on staff, we now have three (Fig. 2). Only one of those three is dedicated to archaeology full-time. In an effort to speed the pace of the project the SHPO applied for a National Center for Preservation Technology and Training grant in 1995, but was unsuccessful.
Despite these setbacks, the New York SHPO is optimistic that the goal of developing a GIS-based predictive model is attainable. The most encouraging factor in this regard is the fact that many other institutions both within New York State and nationally are developing GIS for similar purposes. We are not in this endeavor alone, and the advances that others are making will in many instances be adaptable to our needs and likely to the needs of other SHPOs as well. Despite the reduction in staffing, the development of a GIS-based predictive model that can be used to replace existing tools remains a priority within the office, and staff continue to pursue the effort, as time permits. While the effort has been progressing slowly, we are much closer to the end than to the beginning.
The New York SHPO is very interested in benefiting from the work of others in both academia and the cultural resources management field who are developing GIS-based predictive modeling tools. Undoubtedly, the expertise in this area would be of tremendous benefit to our efforts. The New York SHPO contact person for this project is Robert D. Kuhn, Office of Parks, Recreation and Historic Preservation, Peebles Island, P.O. Box 189, Waterford, N.Y. 12188, (518) 237-8643 ext. 255, fax (518) 233-9049, email firstname.lastname@example.org.
Robert D. Kuhn is historic preservation program coordinator for the New York SHPO