Sequencing and correlation of the fossil record
Correlation of ancient ash beds
Well calibrated biosequences are essentially time-lines of macroevolutionary history and form the basis for high-precision geologic time scales. Geologists are concerned with developing tools that will accurately and efficiently arrange biostratigraphic data into such time lines. This seemingly simple goal is actually an enormously complex one. The complexity results, in part, because such large biostratigraphic databases must be efficiently integrated, which requires sequencing huge data sets into manageable pieces, then combining the results. Adding to this difficulty is the handling of fossil biotas that have the same age, but differ as a result of facies and preservation. And finally, there is the problem of providing independent means of correlating strata that do not contain overlapping biota. We have received NSF funding to pursue this intriguing problem using mid-Ordovician sections in the eastern and central USA. The goal is combining conodont biostratigraphic data from carbonates in the west with graptolite-dominated strata in eastern black shales. The key to making these linkages are the abundant and widespread volcanic ash layers (now altered to K-bentonites) that occur throughout the region. By applying novel tephrochronological techniques to Ordovician bentonites we are developing a chronologically and chemostratigraphically independent framework in which to correlate the rich conodont and graptolite biostratigraphic record. Because the original glass in the volcanic ash has long since been diagenetically altered, we cannot rely on either whole-rock or glass chemical compositions, the techniques normally used on Pleistocene – Recent ash beds. Instead, we have developed techniques to use both the minor element composition and isotopic composition of apatite phenocrysts still preserved within the K-bentonite layers. Previous work has shown that initial Sr isotopic compositions of apatite can be powerful ‘fingerprints’ of ancient ash beds (Samson et al., 1995b), and that in many cases Nd isotopic compositions are unique (Samson, 1996). We are thus exploring the use of these isotopic systems to help us correlate individual ash beds over large (~1,000 km) distances. Successful correlation of closely-spaced ash beds would establish time planes on the 103 to 104 years scale, providing sequencing resolution at an astonishing level for ~ 450 Ma strata.
Chronometry of ancient ash beds
Although long-distance bentonite correlation is enormously helpful in sequencing studies it does not provide numerical time constraints. Luckily, the majority of the bentonites contain primary zircon phenocrysts in addition to apatite phenocrysts. We will determine U-Pb zircon dates with the highest possible level of precision for the bentonites that can be chemically correlated over the longest distances. We will thus benefit from the significant advantage of the chemical abrasion-TIMS technique pioneered by Mattinson (2005). This process involves annealing zircon at ~ 900oC, followed by HF-leaching to dissolve areas of zircon that experienced Pb loss, all prior to normal spiking and dissolution. Previous time-scale work that has used this technique has achieved extremely precise ages (e.g., 206Pb*/238U dates with uncertainties of just a few hundred thousand years: Bowring and Schmitz, 2003; Mundil et al., 2004; Ramezani et al., 2007). Because of the importance of these tephrochronolgoical studies to refining the Paleozoic Time Scale, we have joined the EARTHTIME initiative and plan to use the new, internationally calibrated, quadruple (202Pb-205Pb-233U-235U) spike for all of our U-Pb measurements. This tracer solution is being distributed to isotope laboratories working on geological Time Scale issues that have demonstrated the ability to determine high precision U-Pb zircon dates with Pb blank levels of < 2 picograms. The beauty of using a ‘dual’ double spike is the ability to directly determine mass fractionation during the analysis, not otherwise possible for U or Pb isotope ratio measurement. The importance of the distribution of a widely distributed calibrated spike is that U-Pb ages determined from different laboratories can be compared directly. Thus uncertainties in individual spike calibrations no longer need to be propagated, resulting in a Geological Time scale of greater accuracy.