| This work shows that uranium-lead dating of carefully sampled and petrographically screened Cenozoic sedimentary carbonates can provide high-precision ages that can be used to constrain and correlate terrestrial successions. Lacustrine and paleosol calcite from Plio-Pleistocene Olduvai Gorge, Tanzania were selected for initial study. Samples have a range in U/Pb ratios and relatively high U concentrations, but U-Pb ages have high uncertainties. Lacustrine samples incorporate volcaniclastic debris and contain authigenic clays, which inhibit complete dissolution and complicate U-Pb systematics. Small samples of roots yielded analytical data with large uncertainties.; Lacustrine tufas from the Miocene Barstow Formation, southern California are the focus of the second part of this dissertation. Two techniques are developed to characterize samples with the greatest potential for providing accurate and precise ages. Phosphor imaging is used to map the distribution of U and Th in polished slabs of rocks and minerals. This non-destructive technique provides moderate to high resolution even in samples of relatively low U concentration (∼10 ppm) and can be used to increase the range in U/Pb in aliquots for dating. Combined petrography, X-ray diffraction, phosphor imaging, and synchrotron X-ray absorption spectroscopy are used to understand the conditions that formed the Barstow tufas. Tufas contain elevated concentrations of Mn, Fe, Sr, and U. X-ray absorption near edge spectroscopy shows that U is U(IV). Tufa calcite formed at the interface between alkaline lake waters with a high U/Ca ratio and groundwater or springs bearing Ca and trace elements such as Sr. Rapid, possibly microbially mediated, precipitation of calcite allowed for incorporation of high concentrations of trace metals as structural substitutes or fine-scale inclusions.; Five samples of U-rich Barstow Formation tufa calcite yield U-Pb ages (2σ) of 14.83 ± 0.39 Ma, 15.28 ± 0.29 Ma, 15.39 ± 0.17 Ma, 15.83 ± 0.96 Ma, and 16.25 ± 0.22 Ma. These ages agree with pre-existing Ar geochronology. The most precise age has a 1% uncertainty, which is comparable to uncertainties for K-Ar ages from the same time period.; New screening tools and the great success in dating Tertiary lacustrine tufa suggests that, with careful sampling, it may be possible to obtain precise and accurate ages of analogous samples from Pliocene and Pleistocene paleoanthropological sites. |
