GSA Connects 2022 meeting in Denver, Colorado

Paper No. 223-10
Presentation Time: 10:30 AM

BAYESIAN INTEGRATION OF ASTROCHRONOLOGY AND RADIOISOTOPE GEOCHRONOLOGY


TRAYLER, Robin, Department of Geosciences, Boise State University, Boise, ID 83725; Department of Life and Environmental Sciences, University of California, Merced, 5200 N Lake Road, Merced, CA 95343, SCHMITZ, Mark, Department of Geosciences, Boise State University, 1910 University Drive, Boise, ID 83725 and MEYERS, Stephen R., Department of Geoscience, The University of Wisconsin - Madison, Madison, WI 53706

Age-depth models that relate stratigraphic position to time play an important role in interpreting the rate and timing of environmental change throughout Earth history. Astrochronology­—using the geologic record of rhythmic astronomical oscillations to measure the passage of time—has proven a valuable technique for generating age-depth models and durations of time in rock sequences. However, in the absence of temporal anchoring information, many deep time astrochronologies float in ‘absolute’ numerical time. Alternatively, radioisotopic geochronology (e.g., U-Pb, 40Ar/39Ar) produces point-estimates of numerical age, usually dispersed randomly throughout stratigraphy, which can be used to anchor floating age-depth models.

In this study we present a Bayesian approach for integrating radioisotopic geochronology and astrochronology into age-depth models. Most existing Bayesian accumulation models use a stochastic random walk to approximate the variability and uncertainty of sedimentation. Integration of the astrochronologic record and radioisotopic dates allows reduction of uncertainties related to interpolation between dated horizons and captures subtle changes in sedimentation rate recorded by astrochronology. The method simultaneously tunes astrochronologic records to astronomical periods and anchors to radioisotopic dates, while incorporating prior information about sedimentation rate, superposition, and the presence of major hiatuses. Resulting anchored age depth models preserve both the continuity of floating astrochronologies and the precision and accuracy of modern high precision radioisotopic geochronology.

We have developed and tested our methods on synthetic data and applied it to a published record from the Ediacaran Jibalah Group (Saudi Arabia). Model uncertainties are relatively constant with depth, primarily controlled by the precision of radioisotopic ages, and significantly reduced relative to stochastic random walk models between dated horizons; the uncertainty in accumulation rate was improved by a factor of three. Furthermore, since the resulting age-depth models combine both astrochronology and radioisotopic geochronology in a single inversion, they leverage the strengths of, and naturally resolve ambiguities between, the two timekeepers.