Parameters in the use of pXRF for archaeological site prospection: a case study at the Reaume Fort Site, Central Minnesota

Abstract

Field portable/hand-held x-ray fluorescence (pXRF) analyzers have been characterized as potentially useful for archaeological site prospection, but little has been published on the parameters of their use in this manner. The purpose of this study is to explore whether the variability of surface geochemistry as characterized with a pXRF analyzer corresponds with subsurface archaeological features at a site subsequently excavated, and what conditions influence the success of this endeavor, including feature depths, soil moisture, and sample processing. A 520 m2 within-site area was systematically surveyed on a 2 m interval, within which several types of archaeological features were excavated (chimney bases, wall trenches, and a bonebed of faunal waste), taking readings in situ and collecting samples for ex situ testing (undried, dried but not powdered, and dried/powdered). The four different tests of each grid location, analyzed through univariate and multivariate tests, showed that the pXRF surface data does correspond with some types of subsurface features when those features are very shallow (within 5 cm of surface level) and are associated with clayey fills. Further, the data from the subsurface samples provides excellent distinction of feature fills from other sediments, regardless of sample preparation. In situ surface survey with pXRF analyzers may however be adequate for sites with a thorough baseline geochemical database.