Information Circular 45. Utility of Elemental Geochemical Data in Correlation and provenance Studies of Pleistocene Materials: A Case Study in Stearns county, Central Minnesota

Geochemical attributes, especially minor, trace, and rare-earth elements, are commonly used to characterize various kinds of sedimentary rocks and to elucidate their provenance. Similar techniques have been applied to glacial materials with varying degrees of success. In Minnesota, for example, Martin and others (1989, 1991) concluded that few if any geochemical elements were useful for correlating tills across large areas. In contrast, Gowan (1998), in a study of six till units from central Minnesota, concluded that geochemical attributes were useful in delineating stratigraphic units and their provenance. Results of a similar study in southwestern Minnesota (Patterson and others, 1995) were inconclusive. The geochemical studies undertaken to date in Minnesota have utilized the silt and clay-size fraction as a sample medium and Atomic Adsorption spectrometry or Inductively Coupled Plasma emission spectrometry as the principal analytical technique. Unfortunately, both techniques require that the sample medium be dissolved prior to analysis, which can limit the value of the resulting data. For example, relative to the composition of the sample matrix and the particulates, certain minerals and elements may be selectively leached or incompletely dissolved, or spectral signals from some concentrated elements may interfere with weaker signals from less abundant elements. Such problems are best addressed by the use of replicates and standards developed from materials similar to those being analyzed. To our knowledge, sets of material standards that could be used to calibrate the various analytical methods have not been developed for Pleistocene sediments in Minnesota. In this study we avoided problems associated with partial dissolution and interference by using Instrumental Neutron Activation Analysis (INAA) techniques (XRAL Activation Services Inc., Ann Arbor, Michigan). We analyzed 123 subsurface samples from five drill sites in Stearns County, east-central Minnesota, for 32 elements. The results in this report are not directly comparable with those of Martin and others (1988), Gowan (1998), and Patterson and others (1995), because they were obtained on whole-rock samples from which only clasts of pebble or larger size had been removed. Nonetheless, the data provide insight into the utility of geochemical techniques to investigate provenance and transport patterns of glacial materials. The complete file of analytical data is summarized in the Appendix