Age calibration of geomagnetic polarity reversals around the Cretaceous-Paleogene boundary

Abstract:

Testing different Cretaceous-Paleogene boundary (KPB) extinction hypotheses is inhibited by insufficient geochronology, exemplified in the geomagnetic polarity time scale (GPTS). The GPTS is used for age control in studies lacking means for high-precision age determination, ranging from studies on climate change to the evolution of life across the KPB. If well-calibrated, the GPTS would provide a powerful tool for probing deeper into the events around the KPB extinctions. Current calibration of the GPTS (GTS2012; Ogg, 2012) across circum KPB chrons (C30n-C28n) draws heavily on the age results of Swisher et al. (1993) for the KPB, which has been shown to be ~ 200 ka too old per reanalysis by Renne et al. (2013). Further, GPTS estimations rely heavily on astronomical tuning of ODP cores and land-based records (Zumaia) (Husson et al., 2011; Thibault et al., 2012; Kuiper et al., 2008; Westerhold et al., 2008), which fail to account for possible biasing effects of climate signals other than those due to orbital forcing, such as those associated with the KPB events. Moreover, complex sedimentation in marine sections following the mass extinction can obscure orbital signals and complicate cyclostratigraphic interpretation (Westerhold et al., 2008).

Terrestrial deposits in the Hell Creek region of NE Montana (USA) provide an opportunity to refine the ages of polarity reversals near the KPB (C30n-C28n), and test the accuracy of orbitally tuned chronologies e.g. (Ogg, 2012). These strata are interbedded with abundant sanidine-bearing ashes, which have yielded ⁴⁰Ar/³⁹Ar ages with resolution as good as ± 11 ka and absolute accuracy in the range of ± 40 ka (Renne et al, 2013; Sprain et al., 2014). Further, these sections have relatively uniform sediment accumulation rates, which support the use of interpolation to calculate reversal ages from bounding tephra layers (Sprain et al., 2014).

Preliminary results using new ⁴⁰Ar/³⁹Ar ages and magnetostratigraphic data suggests a duration of ~ 200 ka for the Pg. portion of C29r, roughly two 100 ka eccentricity cycles shorter than the orbitally tuned estimate of ~ 400 ka of Westerhold et al. (2008). Combining this new result with maximum estimates for the Cret. portion of C29r (Sprain et al., 2014), the duration of C29r is ~ 450 ka, significantly less than the 710 ka estimate from GTS2012 (Ogg, 2012).