Click for next page ( 24

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 23
24 opportunity to take on long-term curation of collections Mn/Model also includes three-dimensional, GIS-format recovered from DOT property. The agreement includes a set geomorphic maps for a number of the state's major river of criteria that the repository must adhere to before receiving valleys, in addition to a bog region and several small upland the collection. locations. These maps are used to define Landscape Suit- ability Ratings. These ratings indicate the potential for and possible depth of buried sites, in addition to whether or not INNOVATIVE STATE HISTORIC PRESERVATION the surface of an area is disturbed. By consulting the Land- OFFICE GUIDELINES scape Suitability Rating maps, the DOT determines whether a subsurface survey is necessary and if a proposed project A search of all of the SHPO websites identified only one state might disturb buried sites. (Vermont) that had moved beyond the standard approaches to archaeological investigations. It should be noted that not The North Carolina DOT, with funding from FHWA, has all SHPOs have posted their state's guidelines on their web- completed the first stage of a GIS-based Archaeological Pre- sites. What makes the Vermont guidelines unique is inclu- dictive Model (APM) for seven counties in the state. This sion of explicit guidance on establishing the significance of first phase of work involved collecting digital environmental prehistoric sites, listing the characteristics that make a site data, digitizing and georeferencing data from archaeological eligible for listing in the National Register. The guidelines sites and historical maps, and creating a Microsoft Access also indicate the categories of data that sites must contain to database. The next task will be to create a predictive model address the important research topics for the state. These that will indicate whether any given 30-m-square area within questions are included in a historic context on "Vermont's a project corridor has a high, medium, or low probability for Prehistoric Cultural Heritage." The Vermont guidelines also containing archaeological sites. The model is currently include extensive discussions on integrating public outreach focused on prehistoric sites. This predictive model will allow and education into archaeological investigations and sets the DOT to choose highway alternatives that have the least standards for this outreach. The guidelines also include a list- costs for complying with NEPA and Section 106. Eventually, ing of recommended projects and programs for public out- the APM, which is housed in the Office of State Archaeol- reach. Finally, the guidelines stress many times that one of ogy (located within the state's HPO), will be developed for the goals for Vermont archaeology is to increase creativity the entire state of North Carolina (27). The North Carolina and flexibility in the conduct of archaeological studies (25). DOT noted that once in place, the models will be continu- ously tested and refined through future transportation proj- INFORMATION TECHNOLOGY AND ects within the areas covered by the APM. INFORMATION MANAGEMENT SYSTEMS The use of archaeological predictive models is relatively State DOTs are using various IT tools to manage archaeo- new to the transportation project development process. Sev- logical resource data. The most common tools are GIS and eral military installations across the country developed such database programs such as Microsoft Access (13). These IT models in the 1980s and 1990s and have used these models tools are used to maintain site inventories and develop com- with mixed success. In 20022003, Statistical Research, Inc., puterized archaeological predictive models. and the SRI Foundation carried out a project, funded by a Department of Defense (DoD) Legacy Grant, to evaluate the use of archaeological predictive models on military installa- Archaeological Predictive Modeling tions (28). The project was designed to answer four questions: One of the most comprehensive and longest operating, Do predictive models that have been created for mili- transportation-related archaeological GIS databases and tary installations work? predictive models is Minnesota DOT's Mn/Model. From Can they be refined to work better? 1995 to 1998, using funds from FHWA, the Minnesota DOT Are they sufficiently accurate so that land managers and developed Mn/Model to improve the ability of archaeolo- SHPOs can use them in evaluating management deci- gists to assess the likelihood of finding archaeological sites sions about installation of archaeological resources? throughout the state. Creation of the model involved the Can a predictive model be integrated into a more dynamic collection and mapping of both environmental and archaeo- operational model that would be useful across the DoD to logical site data for the entire state. Mn/Model is a statisti- increase cost-efficiency of cultural resource management cally based predictive model that displays the probability of at large installations? finding a pre-1821 archaeological site at any given location. Probability categories are based on statistical correlations To address these questions the project team first tried to between known archaeological site locations and environ- determine the pervasiveness of predictive modeling in the mil- mental attributes. The model maps low, medium, high, and itary through a questionnaire sent to installations representing unknown probabilities of finding an archaeological site in a all branches of the service. Although not intended to be a com- particular area (13,26). plete canvassing of the military use of predictive models, the

OCR for page 23
25 objective of the questionnaire was to achieve a reasonable These are questions that predictive models can assist in sample from which inferences could be drawn. The second answering. Determinations of eligibility require archaeolo- gists to state why a site is significant, and what can be learned step was to choose models from four of the responding instal- from the site. Models could be used to highlight why a lations for an in-depth evaluation of their technical quality, particular site's location is unusual or typical of a class of accuracy, and general utility as a cultural resource manage- past behaviors. Data recovery plans could incorporate model ment tool. The four installations included Fort Bliss in predictions about the type of site and the resources available Texas/New Mexico, Fort Drum in New York, Eglin Air Force to its residents (28). Base in Florida, and Fort Stewart in Georgia (28). The study produced several important findings about the Archaeological Resource Inventories and Portals use of predictive archaeological modeling at military installa- tions. These findings, which follow, are presented in this syn- Several states have computerized archaeological resource thesis report because they provide transportation planners and inventories (13). For example, the Ohio DOT, working with historic preservation specialists working with state DOTs the Ohio SHPO, developed a GIS program based on the with valuable guidance and lessons about archaeological pre- National Park Service's MAPIT (Mapping and Preservation dictive models. Information Technology) software. The Ohio database includes more than 120,000 cultural resources, including Despite all the interest in predictive modeling in the mili- archaeological sites. The DOT funded the development of tary, there is no centralized instruction on how to create, use, this GIS. The database allows DOT staff to electronically and maintain these models. Each installation is left to sur- plot archaeological site locations and make early evaluations mount the difficulties associated with site recording, GIS of potential impacts from proposed transportation project development, and predictive modeling by itself. This alignments. The GIS has become a valuable project planning approach has encouraged innovation and led to the develop- ment of a wide variety of models, but the potential of many tool and is used for the analysis of alternatives within the of these models is restricted because of decisions made early NEPA process (4). in the process. It was clear that installations could have profited greatly from one another's miscalculations and Florida DOT's Efficient Transportation Decision-Making successes. (ETDM) process Most models are rudimentary in nature. In many respects, predictive modeling has witnessed a loss of sophistication in links land use, transportation, and environmental resource the models developed in recent years. Most models are planning and facilitates early interactive involvement to pro- simple intersection models or simple correlation models. duce better and more consensual environmental outcomes. Few models are based on multivariate statistical techniques Through electronic data sharing and comment entry, maps or theoretically based constructs about past human behavior. can be viewed and comments filed and read by others on-line Because of the simplistic nature of the models, some instal- at various stages in the process (4). lations have added judgmental criteria into their models to increase their accuracy; even though by doing so they reduce ETDM includes an interactive, user-friendly GIS database their systematic and objective character. that is accessed through the Internet. The ETDM database contains all of the state's computerized archaeological Models tend to be restricted to predicting surface manifes- resource data, in addition to a wide range of environmental tations. Despite the importance and predictability of buried sites, geomorphology is not a component of most modeling and land use data. Through the use of ETDM, the Florida efforts, and neither are remote sensing techniques. The lack DOT can identify potential impacts to archaeological of satellite imagery is particularly noticeable. Such imagery resources and develop management strategies during the can be a useful proxy for ground cover and land surfaces. transportation planning process and during the early phases The imagery exists in digital form that can easily be of project development. All management decisions are included as a separate theme in an installation's GIS. Impor- tantly, much of this imagery is available to the military at lit- documented electronically and can be accessed by trans- tle or no cost. portation and resource agencies involved in the Section 106 process, facilitating concurrent agency reviews and In most cases, models are not integral to the historic preser- approvals. Agencies also use the system to exchange infor- vation compliance process. In part this may be a result of mation (4,13). The Florida DOT noted that the Seminole models going out of date. While much effort has gone into Tribe of Florida is currently participating in the ETDM creating models, little effort has been expended in refining them. Models are treated as final products rather than being process. The DOT is working with several other tribes about viewed as a process that involves continual modification and participating in the ETDM process. improvement. But even for models that have been refined and kept current, decisions regarding level of inventory, One area where IT tools have become very important to determinations of National Register eligibility, and resolu- DOTs is in the recordation and management of archaeologi- tion of adverse effects rarely include model predictions. Yet, this does not have to be the case. How many acres should be cal sites located in DOT ROWs. The Oregon DOT has a GIS surveyed? Where should they be conducted? How should mapping program that locates archaeological sites within its sites be identified (e.g., shovel testing or pedestrian survey)? ROWs. The GIS also indicates areas of high archaeological