USE OF ZIRCON U-PB AGES AND BULK TRACE ELEMENT AND ND AND HF ISOTOPE COMPOSITIONS TO DETERMINE SOIL PROVENANCE IN A LIMESTONE TERRANE, MIDDLE TN, USA

This study aimed to assess the potential for using zircon U-Pb geochronology to identify the provenance of soil and the siliciclastic fraction of limestone, and to test the hypothesis that soil in middle TN formed through in-situ chemical weathering of "dirty" limestone. Samples of soil and underlying limestone bedrock were collected from two sites in the drainage basin of the Harpeth River. Samples of potential exotic sources included alluvium from the same basin, and loess from west TN. Limestone samples were dissolved in acid, and zircon was separated from the residuum. Bulk samples were analyzed for mineralogy, major and trace element concentrations, and 143Nd/144Nd and 176Hf/177Hf isotopic compositions for use as additional provenance tests.

Zircon age spectra indicate that bedrock sediments were derived from rocks formed during the Grenville and Taconic orogenies in the Appalachians. All provenance indicators showed that a soil composite from the A and B horizons of an alfisol at one site was formed by chemical weathering of the underlying limestone and from the now missing overlying limestone unit. In contrast, soil samples of the B1 and B2 horizons of an ultisol collected from a terrace 35 m above the river consist of alluvium or loess. While zircon U-Pb age spectra and Hf isotopes link the B1 ultisol to bedrock formations eroding upstream, trace element, Nd isotopic, and mineralogical compositions do not because sediments derived from the eroding bedrock formations are altered by sedimentary sorting and possibly loess deposition. Results suggest that zircon U-Pb age spectra are a robust tool for determining the provenance of carbonate lithologies, and for tracing inputs into soils, including from bedrock units removed by erosion, but should be used with other provenance indicators.