Burned features are often readily detectable in magnetic surveys(1).
precise targeting can make it possible to recover samples of datable materials efficiently and with minimal disturbance .
Magnetic prospection for datable materials
This page is based on:
Jones, Geoffrey (2014). Magnetic prospection for datable materials. Presented at the 2014 Plains Anthropological Society Conference, Fayetteville AR, October 28-November 1, 2014.
Thermal features, such as hearths and burned structures, are often detectable in magnetic surveys. These features are rich sources of archaeological data. In particular, thermal features are sought as sources of datable material. Among the dating methods that require or can be performed on burned or thermally-altered materials are:
Magnetic surveys can map many features across a site, and these can be targeted specifically for datable materials. This allows more samples to be collected efficiently and with minimal disturbance.
Radiocarbon dating May generally be considered the most useful and cost-effective physical dating method for archaeological investigation.The preservation advantage of charcoal over unburned wood makes it a common source of datable carbon. The hearth, as a locus of food preparation may also offer other datable materials such as bone or maize cobs. The magnetic enhancement of burned features makes it possible to map and target them as sources of datable carbon.
The Eagle Tree site (48CO2920)(2) in northeastern Wyoming offers an example of the use of magnetic survey for targeting datable features. Eagle Tree was a spatially extensive prehistoric campsite with a very sparse distribution of features and artifacts. It was the subject of archaeological mitigation performed by GCM Services, Inc. prior to mining. Magnetic survey identified a number of subsurface features, and allowed efficient recovery of far more data than might have been possible with conventional archaeological techniques alone.
Magnetic survey results from the Eagle Tree Site. Prehistoric hearth features were very clearly detected. Shovel testing or other conventional techniques would be very unlikely to detect any of these sparsely scattered features. The dates of the hearths in themselves are very informative. They indicate that instead of a single occupation, the site was occupied repeatedly, probably by very small groups. Most of the dates occur at intervals of some decades through a span of roughly 500 years in the late prehistoric. This might suggest a cultural tradition of use, or a period where the site was more favored by environmental conditions. There are two outlying dates (underlined), one preceding the others by a few hundred years, and the other by many thousands. It may be noted that the very early date does not have a magnetic expression in the survey data above. It was detected magnetically, but at a greater depth.
Mitigation at Eagle Tree also involved the search for older components in the deeply stratified sediments. The magnetometer is limited to approximately one meter in its range of detection of prehistoric features. At Eagle tree it was used in conjunction with mechanical excavation to investigate at greater depths. Successive iterations of mechanical stripping and magnetic survey of the stripped surface were performed. The aim of this approach was to open a large area for investigation, while being able to detect features before they were impacted. Features and associated occupation surfaces could then be excavated by hand (this approach has been successful elsewhere).
Gene Munson examines a Paleo-Indan hearth discovered at approximately eight meters below the modern surface. It was detected magnetically and then exposed by hand excavation.
Magnetic survey results from the of the mechanically excavated surface. The hearth feature shown at left appears as a distinct magnetic high.
Archaeomagnetism, paleointensity, and thermoluminescence are dating methods that may be used where datable carbon is absent. Although based on different properties, all of these methods are performed on samples that have been heated. Archaeomagnetism and paleo-intensity are magnetic methods, and associated analyses may useful information, such as the number of times a sample might have been heated. Potassium-argon dating also estimates the date that a sample was last heated, but is most useful in contexts too old to be relevant to New World archaeology.
The fill of the hearth contains charcoal that could be (and was) dated more accurately and cheaply than could be done with these other methods, but this carries with it an opportunity. Archaeomagnetism and paleointensity are only useful if samples can be compared to regionally specific curves tracking the history of the geomagnetic field. These curves can only be built by correlating data from magnetic samples against independent techniques such as radiocarbon and dendrochronology. Such curves do not yet exist for the region where this hearth occurs, but would be very useful for dating features that lack charcoal. Paleointensity also offers the potential to date unoriented magnetic samples, including ceramics.
Dendrochronology relies on having a known local history of tree growth patterns. Although it does not require that a sample be burned, in many environments, samples are unlikely to survive without the preservation advantage of charcoal over unburned wood. The magnetic effects of burning can also offer a means for locating samples through magnetic survey, through locating features such as burned houses of palisades. Of course, burned timbers are likely to be datable by radiometric means, but if a local dendrochronology has been established, it is a more precise dating method (with the understanding that both date the death of the tree, rather than the use or burning of the wood). Conversely, radiocarbon dates may be of value in anchoring dendrochronologies. Dendrochronology has also been important in establishing archaeomagnetic curves and calibrating radiocarbon dates. Aside from its value for dating, tree-ring data may be of interest for its implicit climate information.
The Eastview site(4) is shown here (at left) as an example of burned timbers found in a structure identified with magnetic survey. The house at Eastview was excavated as part of a field school conducted by Courson Archaeological Research. Although dendrochronology was not performed at the site, it illustrates how datable timber samples might be directly targeted.
The hearth has played a central role on many human activities. It is often a focus of food processing, shelter, crafts, and social life. Burned structures can offer superior preservation and a snapshot of daily life.
The spatial dimension inherent in magnetic survey results can contextualize temporal data, especially on multicomponent sites. Beyond a narrow focus on dating, geophysical survey (magnetic and otherwise) can be used to search materials for useful for a wide variety of analyses, including botanical, faunal, lipids, soil chemistry, etc...
While magnetometry is not effective on every site, where it can be applied it has the potential to broaden the range of temporal questions that may be asked, and to help answer them more cheaply and in greater detail.
1. Tomasic, John (2013). 2012 Phase IV Excavations of 14my388 at the Eastep Riverbank Stabilization Project, an Nrcs Undertaking In Montgomery County, Kansas. With contributions by: Rolfe D. Mandel; Edwin J. Miller; Andrew Wyatt; Rebecca Freidel; Whitney Cleaver. Kansas State Historical Society.
Jones, Geoffrey (2013). Magnetic survey at Eagle Tree Site (48CO2920). Archaeo-Physics LLC Report of investigation No. 199. Prepared for GCM Services, Inc. (Butte Montana). Archaeo-Physics, LLC. Minneapolis, MN.
3. Munson, Gene (2013). Excavation Of Holmes Draw Site (48CA2834): A Paleoindian, Early Plains Archaic, Middle Plains Archaic And Late Prehistoric Campsite. With Contributions by: Geoffrey Jones; John Albanese; David McKee; Viktor Kujawa; PaleoResearch Institute; Beta Analytic, Inc. Prepared for Peabody Powder River Operations, LLC, Gillette, Wyoming by GCM Services, Inc., Butte, Montana.