Sharpening the U-Th chronometer through technical developments and community implementation

Lead Research Organisation: University of Oxford
Department Name: Earth Sciences

Abstract

The radioactive decay of uranium incorporated in natural carbonates (e.g. corals and stalagmites) provides a powerful way of dating these materials. Such U-Th techniques extend to about half a million years ago and provide the major way in which we can learn about the timing of past climate and environmental change. Over the past 15 years there has been considerable improvement in our ability to measure U and Th isotope ratios and concentrations resulting in a reduction of U-Th age uncertainties by an order of magnitude. Uncertainties are now as low as 0.1%, or 100 years in the age of fossil coral or speleothem that is 100,000 years old. This increase in precision has enabled a wide and expanding range of questions to be answered and is critical to our understanding of the mechanisms of Pleistocene climate and sea-level change. But it has also exposed a problem. Calibration of the tracer solutions used to make U and Th measurements is performed independently in each laboratory using differing techniques and it has become abundantly clear that resulting U-Th ages, while impressively precise, do not agree at this level of precision from one lab to another. There is now inter-laboratory bias at a level that exceeds typical quoted age uncertainty. The cause of this inter-laboratory uncertainty is due to a lack of suitable materials for both calibration purposes and for long-term assessment inter-laboratory agreement. One of the most widely used materials for such calibration, HU-1, has recently been demonstrated to vary, by up to 0.5% between the solutions used in different laboratories. And there exists no widely distributed and well characterised 'age standards' that could be analysed by all of the U-Th laboratories to facilitate quantification of inter-laboratory agreement. We propose a series of actions to address these short fallings in the international U-Th chronology community. We will develop a series of U-Th calibration solutions who's composition is known from first principles metrology (i.e. from the weighing and dissolution of high-purity U and Th metal). We will use these solutions to calibrate the tracers used in three UK and one overseas laboratories - each with a well established reputation for U-Th work. We will then proceed to do develop four different 'age solutions' by taking the U and Th isotopes and mixing them together in proportions that mimic typical compositions analysed by the community. The composition of these 'age solutions' will be measured using the newly and precisely calibrated tracers, so that all compositions will be known and traceable to basic measurements of mass. These 'age solutions' will also be made freely available to all U-Th laboratories who request them worldwide, and we will produce enough of the solutions so that they will last the community 20 years. We will co-ordinate an inter-laboratory comparison exercise so that, for the first time, we will be able to quantify the level to which dates produced in different laboratories agree. As a community we will want to ensure that the level of inter-laboratory variation is minimised, so if labs find their results to be inaccurate they will be able to use the age-solutions, whose compositions are well known, to improve the accuracy of their results. There is very widespread support for this effort in the international U-Th community and we have letters of support from 33 laboratories - the vast majority of all such laboratories worldwide. There are also limitations with the mathematical treatment of U-Th data used to produce dates. Our proposed analytical efforts will be made in concert with the development of these new data reduction template. Overall, these activities will provide dramatic and permanent increase in the reliability of U-Th dates of carbonates - the dates on which so much of our knowledge of Pleistocene climate is based.

Publications

10 25 50