Webb, Gregory E.1
1The University of Queensland, Brisbane, Australia
Trace element geochemistry is useful increasingly in ancient and not so ancient carbonate rocks where it provides the basis for radiometric dating and many palaeoenvironmental proxies. High precision U-Th and U-Pb analyses in carbonate samples are converging to close the ‘undatable’ window past 500 ka. A variety of palaeothermometers are used commonly (e.g., Sr/Ca, Mg/Ca, U/Ca, etc.) and rare earth elements (REEs) inform the source of ancient water masses, water quality and redox states. Other elements (e.g., Ba, Mn, V, etc.) provide proxies for biological processes and palaeoproductivity as well as terrestrial processes, including firing. Redox sensitive elements (e.g., Mo, V, U, etc.) inform complex oxygenation scenarios in Precambrian seas. Trace elements are providing an ever increasing tool kit for sedimentologists, stratigraphers and palaeoenvironmentalists.
However, a variety of pitfalls accompany the expanding use of trace element geochemistry in carbonate rocks. As carbonate minerals are metastable at the Earth’s surface, sedimentologists are highly attuned to the problem of diagenetic alteration. Effective sample vetting is crucial, but new core scanning technologies and SEM approaches are easing sample selection. Regardless, many samples are complex mixtures of sources with differing elemental concentrations and distributions that require detailed understanding of the sample, which elements are being targeted for analysis and the reservoir for which they are meant to serve as proxies. For example, trace elements in marine precipitates are sourced from ambient seawater, but depending on the sample, elements also may reflect siliciclastic detritus (contamination) and organic components, which may or may not cause fractionations or enrichments. Additionally, analysed element distributions may record local microenvironments rather than the ambient water masses in which they occur. Where ambient water chemistry is targeted, elemental contributions of all other sources must be identified and removed. For bulk samples (dissolution ICP-MS), contaminants can be removed post-analysis using mixing lines to quantify contamination in each sample. Alternately, sequential etching may attempt to analyse separate sources independently during dissolution. Increasingly, laser ablation (LA) ICP-MS combined with LA mapping can be used to identify and sample increasingly small, specific targets while avoiding contaminants. Other elemental mapping approaches (e.g., synchrotron based x-ray fluorescence, SEM electron dispersive spectroscopy) also aid sample vetting and targeting of appropriate sources. The source of the proxy elements is critical.
Although ICP-MS is increasingly common, significant technical issues remain past adequately low blanks and high count rates. As sample size decreases (e.g., LA analysis), low sample volume exacerbates low element concentrations leading to poor data quality. However, even apparently ‘low quality’ data for some elements, like the REEs, provide useful information owing to their self-normalising behaviour. Cohesive REE data suggest adequate precision to carry information, regardless of calculated detection limits and groups of less cohesive data can be analysed statistically to provide some information.
Overall, successful interpretation of trace element proxies requires more than a good geochemical laboratory; it requires knowledge of the finite relationship between the targeted proxy elements and the sample being analysed (i.e., context) along with application of the most appropriate technique for the job.
Gregory Webb is a palaeontologist and carbonate sedimentologist who specialises in ancient and modern corals, coral reefs and microbialites. He uses petrographic and geochemical means to understand how organisms make rocks and how those rocks record environmental information in their geochemistry and morphology.