Success and failure in Cenozoic global correlations using golden spikes: A geochemical and magnetostratigraphic perspective

Abstract

We review examples of placement of select Cenozoic Global Boundary Stratotype Sections and Points (GSSP = "golden spikes") from a geochemical and magnetostratigraphic perspective. Though biostratigraphy is the sine qua non for placing modern chronologic tools (high-resolution radiometric dates, magnetochronology, and astrochronology) into proper stratigraphic context, its use as the primary correlation tool for GSSP is less desirable than using magnetostratigraphic or chemostratigraphic markers. Here, we advocate for defining of GSSPs at lithologic levels that are globally correlatable using magnetic reversals and/or abrupt geochemical changes, and the avoidance of biostratigraphic datum levels (that are necessarily biogeographically restricted) as primary criterion for correlation. We provide two examples where placement of golden spikes in association with global geochemical events has been successful: the Cretaceous/Paleogene (K/ Pg) and Paleocene/Eocene boundaries. We discuss three boundaries where placement of the GSSP is unfortunately not at distinct isotopic or magnetostratigraphic events (the Eocene/Oligocene, Oligocene/Miocene, and Pliocene/ Pleistocene (= Neogene/Quaternary) boundaries. Lectostratotypes for isotopic reference are proposed for the Eocene/Oligocene boundary at South Atlantic Site 522 and for the Oligocene/Miocene boundary at Southern Ocean Site 1090.