South American Triassic geochronology: Constraints and uncertainties for the tempo of Gondwanan non-marine vertebrate evolution

Abstract

Gondwanan sedimentary deposits preserve a rich archive of Triassic non-marine vertebrate evolution. This fossil record is integral to understanding early Mesozoic global change events, including the end-Permian and end-Triassic mass extinctions, Carnian Pluvial Episode, and macroevolutionary events such as the origin of dinosaurs. Until very recently, almost all of these fossil assemblages were dated by exclusively biostratigraphic means, which made robust correlation to the GSSP-defined timescale difficult. Furthermore, recent advances in radioisotopic dating and magnetostratigraphy have demonstrated that many of these biostratigraphic schemes were imprecise and that key index taxa have different first and last appearances across geographic space. Thankfully, over the past ten years, new radioisotopic and magnetostratigraphic age constraints from fossiliferous sequences in South America have allowed the revision of the absolute ages and relative correlation of key Gondwanan vertebrate assemblages. Here, we review these geochronologic age constraints from South America, describe and revise their accuracy and uncertainties, present new U–Pb zircon age data for a key section in Venezuela, infer preliminary age models for these successions, and discuss what they mean for the correlation of fossiliferous Triassic units in Gondwana. This synthesis suggests that although radioisotopic age data are often numerous, the geological uncertainties associated with U–Pb zircon dates using micro-beam techniques (LA-ICPMS and SIMS) mean that the age of most sedimentary units cannot be constrained better than a precision of ± 3–5 Ma. Although CA-TIMS U–Pb zircon ages and 40Ar/39Ar ages can be more precise and accurate, they only result in well-constrained age models when multiple ages are available throughout the section (e.g., Ischigualasto-Villa Unión Basin of northwest Argentina), and even then, issues with lateral correlation within basins remain. Nonetheless, these data demonstrate that South America has high potential for developing a precise and accurate Triassic non-marine numerical timescale for Gondwanan vertebrate evolution.