2014 GSA Annual Meeting in Vancouver, British Columbia (19–22 October 2014)

Paper No. 63-5
Presentation Time: 2:20 PM


SCHOENE, Blair, Department of Geosciences, Princeton University, 208 Guyot Hall, Washington Road, Princeton, NJ 08544-1003, bschoene@Princeton.EDU

Establishing causality between mass extinctions and their drivers requires resolving power of better than 100,000 years on depositional ages for ash beds that bracket important events, and also on potential causes for extinction events including large igneous province eruptions or bolide impacts. In rocks < 100 Ma, this level of precision is achievable on single zircon dates using modern CA-ID-TIMS U-Pb geochronology, but becomes more difficult earlier in the Phanerozoic. Typically, achieving higher precision involves calculating weighted-means on multiple analyses. However, increasing evidence shows that the majority of zircon bearing igneous rocks contain zircon that grew over tens to hundreds of thousands of years or more, implying uncertainties can mask real age dispersion, and weighted means could be biased too old by pre-eruptive zircon growth. Similarly, Pb-loss not remediated by chemical abrasion can bias dates younger.

Using data from mass extinctions as an example, this talk explores available tools to bear on the problem. 1) Geochemistry on the same zircons dated by ID-TIMS can help test the assumption of cogenetic zircon growth required for robust weighted-mean dates, because zircons having crystallized in the same batch of magma immediately before eruption should have the same trace element signature. 2) A Bayesian approach to building age models for stratigraphic sections that incorporates the law of superposition and individual zircon date uncertainty distributions.

By releasing the assumption of zircon age homogeneity required for weighted means and increasing the importance of single analyses, tests of laboratory reproducibility are essential. Efforts through the EARTHTIME initiative over the last decade, such as the mixing, calibration, and distribution of tracer solutions, have improved interlaboratory agreement in U-Pb ID-TIMS geochronology to better than 0.1%. This work must continue as we demand more precise interpretations of the synchroneity between mass extinctions, their drivers, and associated environmental change.