The application of geographic information systems to the modern world is well known, but their capabilities also make them ideal for analyses of ancient civilizations and they are becoming common tools for archaeologists around the world. The country of Jordan, with important archaeological sites, such as Petra and Jerash, is rich in ancient monuments and artifact, and as attention in the region turns from conflict to cooperation, the discovery and management of these ancient resources is becoming increasingly important. The Madaba Plains Project has been involved with the archaeology of Jordan since 1968, discovering hundreds of archaeological sites during surveys of the Madaba region and conducting excavations at the sites of Tell Hesban, Tell el-Umeiri, and Tell Jalul. Since 1991, the Madaba Plains Project, in cooperation with the Advanced Resource Technology Group at the University of Arizona, has incorporated an ArcInfo based GIS as an integral component of their research. This paper looks at the ways in which this cooperative venture has used GIS to further archaeological research in Jordan, including the detection and elimination of sample bias in an archaeological survey, the construction of environmentally based site probability models, the use of an erosion model to track the introduction of terrace agriculture during the Iron Age, spatial analysis of pottery sherds from the surface of an excavation site, and the production of traditional and non-traditional maps for use as interpretive aids.
Since 1966, State Historic Preservation Offices(SHPOs) have been helping to preserve historic buildings, sites, structures, districts, and landscapes by identifying and recording them. This nation-wide effort has resulted in a inventory of over 4,000,000 sites. Unfortunately, most of this information is accessible only through paper files and maps. A mission of the National Park Service is to provide the technical assistance necessary for SHPOs to automate their inventory. In the case of Virginia, a GIS is being implemented to provide the platform for input, manipulation, analysis, and output of spatial data, while a relational database management system is being used to enter and maintain attribute data. Therefore, it was necessary to develop a GIS application capable of integrating the two. The purpose of this paper/demonstration is to show how a application was developed in ArcView, through Avenue programming, to create a synergy between the GIS data and the RDBMS. The user user-friendly application allows quick, easy access to large amounts of data, thus allowing SHPOs to take a more proactive approach to the monitoring and preservation of cultural resources.
Filling in the space between measurements is a major problem in archaeology, and an obvious niche for GIS. Archaeological data require specialized interpolation techniques because they typically come in gridded or polygonal units of aggregation. The first goal of this paper is to look at ways to use interpolaton to recover some of the information lost in spatial averaging. A second goal of this paper is to make interpolation accessible to archaeologists. This is done through graphic examples that illustrate how the choice of options may affect experimental and interpretative results. In our experiment, we generate an original surface with known properties, evaluate average surface values for each of the pieces, and then reconstruct the surface from the measured averages. We experiment with alternate original surfaces, dissected in various ways, and with using different techniques to reconstruct the original surfaces. We visualize by having the computer display results, intermediate steps, and outputs in the form of shaded surfaces, thus making tangible principles that would be likely to remain obscure if embedded in mathematical formulae or statistical indices. We follow up with comparison of statistical indices, including correlation coefficients, bias, and error. Experiments broadly confirm that interpolation works -- that interpolated reconstructions are in most cases both visually and statistically superior to the polygon or grid average representations on which they are based.