The ability to provenance geologic materials such as chert to their point of origin is immensely useful to archaeologists and other investigators of prehistoric cultures. While simple visual classification and analysis can be useful among separate chert formations, it is inadequate to deal with an intra-formational study such as the Prairie du Chien. In the latter case, geochemical analyses must be employed in order to begin to uncover coherent regional variations which may facilitate the provenance study.

The large-scale sampling and Neutron Activation Analysis of cherts throughout much of the Prairie du Chien group has revealed a number of interesting results. It appears that the Shakopee and Oneota formations behave geochemically as essentially a single formation, with no coherent differences between the two. There do not seem to be any other coherent variations in the vertical dimension, either. This may be either fortunate or unfortunate for those interested in the provenance of Prairie du Chien chert; although it maybe would be useful to make such vertical distinctions in some cases, it also would likely contribute more complications to the issue of horizontal geochemical variances.

There does not appear to be any strong alteration of chemical concentrations in weathered (streambed) deposits. As expected, there are detectable shifts in the concentration of some elements between outcrop and streambed cherts, but the magnitude of the differences is generally within the broad variations of the Prairie du Chien composition.

Some of this homogeneity seems to also extend through the horizontal extent of the Prairie du Chien. Coherent changes in overall chemistry can be detected across the formation as a whole, but there is a great deal of overlap between the ranges of element concentrations found at each end-member. The basic geochemical trend seems to be quite similar to the Model #4 outlined by Luedtke and Meyers (1984): a gradual increase of elemental concentrations in a single direction (Figure 4). It should carefully be noted, however, that the axis of this concentration gradient (northwest-southeast) lies parallel to the direction of maximum geographic spread in the sample locations. Is this a coincidence, or does the distribution of sample locations have an influence on the identified geochemical trends? This is an untestable hypothesis; the only means of verification requires geologic sampling in the northeast-southwest directions, where there are no Prairie du Chien outcrops.

In geochemical provenance studies, the most useful tool for data analysis has traditionally been Discriminant Analysis. It can be very effective for distinguishing between materials from different formations. In this study, this statistical tool was found to be able to distinguish effectively between individual far-separated sources at opposite ends of the field area. However, the gradient of elemental concentrations in the Prairie du Chien is so gradual, and has so much overlap throughout its range, that Discriminant Analysis has proven to be ineffective for distinguishing between regions of the overall formation. In this sense, the methodology of inter-formational provenance techniques do not transfer effectively to an intra-formational study.

The potential for archaeological provenance of Prairie du Chien chert is not impossible, but certainly not very easy. The regional trends identified throughout the formation provide a tantalizing possibility for provenance identifications, but the large overlap of geochemical ranges makes confident source determinations difficult. The results are certainly not discouraging, however, for those who might be interested in applying such techniques to other formations. Definite coherent geochemical trends exist and are quantifiable within geologic formations. Perhaps other formations will exhibit stronger variations and be more amenable to provenance investigations. The potential for intra-formational provenance seems particularly promising for situations in which prehistoric people utilized only a few discrete sources for raw material rather than a broad procurement pattern throughout the entire outcrop area.

     NEXT     PREVIOUS     TABLE OF CONTENTS

Intro and Background     Fieldwork     Sample Prep     Data Analysis     PCA     Correspondence Analysis    Stepwise DA   

Discriminant Analysis     More PCA     Element Trends     Conclusions     Bibliography     Appendix A: Part 1 Part 2 Part 3 Part 4 Part 5     Appendix B     Appendix C

The views and opinions expressed in this page are strictly those of the page author.
The contents of this page have not been reviewed or approved by the University of Minnesota.