University of Texas at Austin

"U-Th Dating of Ostrich Eggshell: Adapting Established Techniques for a New Application"

September 21, 2018 9:00 AM
Staci Lowey

Dr. Staci Loewy
Radiogenic Isotope Geochemist,  UT DoGS         

The newly-amalgamated, JSG Radiogenic Isotope Facility, includes a thermal ionization mass spectrometer (TIMS) and new multicollector inductively coupled plasma mass spectrometer (MC-ICP-MS) and 193 nm laser, two clean labs for sample dissolution and isolation of elements of interest and two prep labs for cutting, crushing or microdrilling samples and picking mineral grains.  It is co-managed by two PhD’d geoscientists with passions for analytical science and solid backgrounds in geologic applications of isotopes, and supported by a team of student assistants. We have a host of well-established analytical capabilities (Sr, Nd, Pb, Hf and Os isotopes in waters, carbonates and silicates) but we are poised to develop new techniques and adapt procedures for new applications. In this talk I will present an example where we adapt U-Th dating procedures, commonly applied to speleothems and coral, to generate reliable ages in ancient ostrich eggshell (OES) found in an early human archeological site by Dr. John Kappelman (UTDGS). At the request of other JSG scientists, we are establishing analytical techniques for analysis of Li isotopes in carbonates, silicates and waters, Rb-Sr dating of high pressure metamorphic rocks, Sm-Nd dating of fluorite-rich veins, U isotopes in marine carbonates and zircon Hf isotopes by laser ablation. We welcome the challenge to develop new capabilities to enhance JSG research goals.

Obtaining more accurate dates for fossils and archaeological sites from the Middle Stone Age (MSA) is critical to studies of human evolution because this time period witnessed the origin of modern Homo sapiens and its dispersal out of Africa and across the rest of the Old World. OES is commonly found in MSA sites and its age can be determined by radiocarbon dating but this approach can only be applied to the youngest end of this time interval (<50 ka).  A second approach is to use U-Th dating which, in contrast to radiocarbon, is reliable across the entire time range of the Middle Stone Age. However, both techniques can be compromised by the precipitation of young calcite into the eggshell’s microscopic openings between the crystals and inside the pores, potentially incorporating detrital U, Th, and C from the soil into the OES, thus complicating age interpretations.

Two-dimensional element mapping by quadrupole LA-ICP-MS of ancient OES identifies higher concentrations of common detrital indicator elements (Al, Mg, Ti) within the inner and outer layers of OES and within the calcite-filled pore clusters. Analyses of extracted pore cluster infill yield significantly higher and more variable detrital Th and U concentrations. After mechanical removal the cone (inner) and crystal (outer) layers and pore cluster infill, U-Th analyses of the internal palisade layer yields lower detrital Th concentrations and more uniform U concentrations. 

Based on these results, a new preparation technique that completely removes the cone and crystal layers along with the pore cluster infill was used to determine ages of the OES palisade layer from an MSA archaeological site. U-Th analyses of palisade layers yield a mean age of ~76 ka (within analytical uncertainty) for all samples. When the pore cluster infill was not removed, a younger age was obtained (53 ka). The technique was further evaluated with AMS 14C dating. In a split of the U-Th sample, radiocarbon infinite ages (~50 ka) were obtained for OES with palisade layers only; OES that retained the pore cluster infill usually yielded younger ages (27-41 ka). The U-rich authigenic pore cluster calcite infill shifts both U-Th and 14C analyses to younger ages. The results of this analytical study support the addition of mechanical removal of the pore cluster infill to the already standard removal of the cone and crystal layers for both U-Th and C dating of OES in order to more fully account for the incorporation of authigenic Th, U, and C into the microscopic OES structure.