Kenneth L. Kvamme

Geophysical Surveys as Landscape Archaeology

Recent advances in technology and practice allow geophysical surveys in archaeology to produce maps of subsurface features over large areas and in potentially great detail. It is shown through a series of case studies from two regions in North America that archaeo-geophysical surveys can produce primary information suitable for the study of site content, structure and organization, for examining spatial patterns and relationships, and for directly confronting specific questions about a site and the past. Because large buried cultural landscapes can now be revealed, it is argued that an alternative perspective on regional or landscape archaeology may be possible because space can be viewed in terms of tens of hectares as opposed to the tens of square meters typical of archaeological excavations. Moreover, by placing focus on such buried features as dwellings, storage facilities, public structures, middens, fortifications, trails, or garden spaces that are not commonly revealed through most contemporary surface inspection methods, a richer view of archaeology, the past, and cultural landscapes can be achieved. Archaeo-geophysical surveys can also play an important role in Cultural Resource Management (CRM) contexts as feature discovery tools for focusing expensive excavations, thereby reducing the amount needed and lowering costs. Their utility is weighed against shovel test pits as a primitive and costly form of prospecting.

Southwestern Pottery Standardization: An Ethnoarchaeological View from the Philippines

ABSTRACTArchaeologists assume that products of specialists will be standardized and that standardized pottery reveals the presence of full-time potters. To test these assumptions, two populations of contemporary Philippine cooking pots are studied: those made on a household basis by the Kalinga, a “tribal” society, and those from Paradijon, a neighborhood of full-time potters. The data indicate that pots from Paradijon are much more standardized than those from Kalinga. Furthermore, the degree of standardization among prehistoric cooking pots from Grasshopper Ruin, in east-central Arizona, is examined in this context. We conclude that without knowledge of ceramic classes that were meaningful to the pottery makers, it is difficult to make reliable assessments of standardization in prehistoric assemblages.

An experiment in archaeological site location: modeling in the Netherlands using GIS techniques

Abstract Regional models portraying archaeological expectation or sensitivity based on knowledge of the current archaeological situation can be of great use to archaeologists in the Netherlands. If past work can be used to guide future efforts to the most archaeologically sensitive regions, benefits in efficiency, cost‐reduction, preservation, etc. can be realized. A GIS‐based approach to archaeological modeling is described that summarizes archaeological expectation along multiple environmental dimensions. The approach is based on a mix of objective data together with archaeological experience and expertise. An application to the Regge Valley region of the Netherlands shows excellent results in terms of archaeological expectation and performance on known site samples. These results are placed within the context of the many problems and difficulties that modeling research must address in this area of Europe, and a variety of suggested improvements are offered.

One-Sample Tests in Regional Archaeological Analysis: New Possibilities through Computer Technology

Archaeologists commonly employ two-sample statistical tests in regional locational analyses that compare environmental measurements obtained at site locations against measurements taken at random locations from the background environment. One-sample tests that compare a site sample against a background standard are conceptually and statistically superior, but have been difficult to implement for continuous data types. This situation now is changed owing to a relatively new computer technology known as Geographic Information Systems (GIS). GIS can provide a complete description of the nature of the background environment of entire regions for categorical and continuous data types, thereby allowing the ready application of one-sample testing strategies. Examples of several GIS-based one-sample tests are given using data from east-central Arizona. Such conventional tests only should be applied, however, when the observations can be shown to be statistically independent through tests for spatial autocorrelation.

Spatial Autocorrelation and the Classic Maya Collapse revisited: Refined techniques and new conclusions

Abstract A recent paper in this journal [Whitley & Clark (1985) Spatial Autocorrelation Tests and the Classic Maya Collapse: Methods and Inferences. Journal of Archaeological Science 12, 377–395.] presented an introduction to the topic of spatial autocorrelation. That paper focused particularly on a spatial autocorrelation test for a variable distributed in areal units (e.g. counties or countries). The test was employed to analyse the terminal distribution of dated monuments at lowland Classic Maya sites which yielded the conclusion of a lack of geographic pattern in the dates and, by inference, the Maya collapse itself. This result is particularly surprising because pattern clearly is indicated in a simple plot of the data and a host of archaeologists have perceived, or have shown with other techniques, pattern in the distribution of these dates. Whitley & Clarks analysis is shown to be seriously flawed, however, because their use of an area-based statistic forced them to severely compromise the quality of the point-pattern data set they analysed. A revised spatial autocorrelation test for point-distributed data is presented that may generally be more appropriate for archaeological data. This test is used to re-analyse the Maya data and it is demonstrated that significant geographic pattern does indeed exist, thus agreeing with the perceptions and findings of Mayanists. Owing to the contradictory findings obtained by two very similar analyses, the need for a careful and thoughtful approach to the analysis of data is emphasized as is the importance of matching appropriate inference methods to the problem at hand.

Current Practices in Archaeogeophysics

As the cost of conducting archaeological research continues to rise, so has the need to define site structure while minimizing excavation efforts. At the same time there is a growing ethic for site conservation and the use of noninvasive procedures for site exploration. Near surface geophysical methods are increasingly seen as useful and cost-effective tools for archaeological exploration because they provide a means to remotely sense what lies beneath the earth. By raising the probability of encountering sought-for features, data recovery efficiency through excavation is increased and costs are reduced. Yet, the very success of geophysical methods has led to a misconception that they are suited only for discovery purposes. A growing perspective is that they are useful, in themselves, for acquiring primary data of relevance to a variety of research questions (Dalan, 1993; Summers et al., 1996). In some contexts, for example, and depending on research goals, the need for excavation may be precluded as when geophysical surveys provide accurate plans of buried architectural remains that offer sufficient information for cultural resource management purposes. Indeed, the simple mapping of architectural and other anthropogenic features (e.g., trenches, ditches, pits, middens, pathways) through geophysics now constitutes a principal means of site inventory in the United Kingdom and elsewhere (David, 1995; Payne, 1996).

Permutation Technique for the Spatial Analysis of the Distribution of Artifacts into Classes

A permutation test for assessing intrasite patterning of artifact distributions in archaeological space is presented. This test requires none of the usual assumptions about the data; permits either exact locational or counts/grid data in one-, two-, or three-dimensional space; and provides an exact method for testing any hypothesis concerning the nonrandom allocation of artifacts into classes within a site.

The nineteenth-century Mandan/Arikara village at Fort Clark: Results from a multi-instrument remote sensing investigation

The nineteenth century earthlodge village at the Fort Clark State Historic Site (32ME2) in North Dakota was occupied by two different groups between 1822 and 1861. The village was initially constructed by the Mandans who later left it in 1837 following a smallpox epidemic. It was reoccupied by the Arikaras in 1838. Both occupations are evidenced by large circular lodge depressions, an open plaza area, trails, a fortification ditch, and hundreds of smaller features visible on the ground surface. Additionally, two well-known fur trade posts were constructed adjacent to the Mandan/Arikara village. The Upper Missouri Outfit’s or American Fur Company’s Fort Clark was in operation from 1830 to 1860. An opposition post, Fort Primeau, established by the St. Louis Fur Company operated between 1846 and 1861. Fort Clark was a center of trade between indigenous groups and Euroamericans during the early to mid nineteenth century (Wood 1993; Wood et al. 2011). A vast record of Native and Euroamerican life at Fort Clark was left by its numerous visitors, artists like George Catlin, Karl Bodmer and John James Audubon, the naturalist-explorer Prince Maximilian of Wied, as well as the early anthropologist Lewis Henry Morgan. Moreover, the head clerk Francis Chardon (1997), left a daily diary of Fort life. With the exception of railroad construction and cultivation along its margins, the site has suffered little disturbance since acquisition by the State Historical Society of North Dakota (SHSND) in 1931. Although intermittent archaeological investigations occurred since the early twentieth century (Wood et al. 2011), systematic work was initiated only in the mid-1980s, when more than 2200 features visible on the surface were mapped (Wood 1993:Figure 3). In 2000 and 2001 a multi-instrument geophysical survey of a linear transect 20 × 400 m east-west across the Native plains anthropologist, Vol. 61 No. 240, November, 2016, 449–468

Magnetic investigations of nomadic group encampments at Fort Clark State Historic Site, North Dakota

Abstract Extensive geophysical investigations were performed at Fort Clark State Historic Site, North Dakota, in 2011 and 2012. One objective of this project was to investigate potential nomadic group camping locations associated with the Mandan-Arikara village and the Fort Clark trading post that are only briefly mentioned in historic accounts. We anticipated that over time nomadic group camp activities at Fort Clark would leave characteristic signatures of magnetic anomaly clusters and increased magnetism isolated from the village and the trading post itself. To evaluate this assertion, a series of magnetic susceptibility and magnetometry surveys were undertaken in the open spaces south of the village and west of the trading post, in the area referred to by Prince Maximilian when he visited the site in 1833. Although our results are not conclusive, magnetic findings indicate the village periphery, and even areas closer to Chardon Creek, were actively used as evidenced by enhanced susceptibility and gradiometry anomalies.