U-Pb zircon geochronology and depositional age models for the Upper Triassic Chinle Formation (Petrified Forest National Park, Arizona, USA): Implications for Late Triassic paleoecological and paleoenvironmental change

The Upper Triassic Chinle Formation is a critical non-marine archive of low-paleolatitude biotic and environmental change in southwestern North America. The well-studied and highly fossiliferous Chinle strata at Petrified Forest National Park (PFNP), Arizona, preserve a biotic turnover event recorded by vertebrate and palynomorph fossils, which has been alternatively hypothesized to coincide with tectonically driven climate change or with the Manicouagan impact event at ca. 215.5 Ma. Previous outcrop-based geochronologic age constraints are difficult to put in an accurate stratigraphic framework because lateral facies changes and discontinuous outcrops allow for multiple interpretations. A major goal of the Colorado Plateau Coring Project (CPCP) was to retrieve a continuous record in unambiguous superposition designed to remedy this situation. We sampled the 520-m-long core 1A of the CPCP to develop an accurate age model in unquestionable superposition by combining U-Pb zircon ages and magnetostratigraphy. From 13 horizons of volcanic detritus-rich siltstone and sandstone, we screened up to ∼300 zircon crystals per sample using laser ablation−inductively coupled plasma−mass spectrometry and subsequently analyzed up to 19 crystals of the youngest age population using the chemical abrasion−isotope dilution−thermal ionization mass (CA-ID-TIMS) spectrometry method. These data provide new maximum depositional ages for the top of the Moenkopi Formation (ca. 241 Ma), the lower Blue Mesa Member (ca. 222 Ma), and the lower (ca. 218 to 217 Ma) and upper (ca. 213.5 Ma) Sonsela Member. The maximum depositional ages obtained for the upper Chinle Formation fall well within previously proposed age constraints, whereas the maximum depositional ages for the lower Chinle Formation are relatively younger than previously proposed ages from outcrop; however, core to outcrop stratigraphic correlations remain uncertain. By correlating our new ages with the magnetostratigraphy of the core, two feasible age model solutions can be proposed. Model 1 assumes that the youngest, coherent U-Pb age clusters of each sample are representative of the maximum depositional ages and are close to (<1 Ma difference) the true time of deposition throughout the Sonsela Member. This model suggests a significant decrease in average sediment accumulation rate in the mid-Sonsela Member. Hence, the biotic turnover preserved in the mid-Sonsela Member at PFNP is also middle Norian in age, but may, at least partially, be an artifact of a condensed section. Model 2 following the magnetostratigraphic-based age model for the CPCP core 1A suggests instead that the ages from the lower and middle Sonsela Member are inherited populations of zircon crystals that are 1−3 Ma older than the true depositional age of the strata. This results in a model in which no sudden decrease in sediment accumulation rate is necessary and implies that the base of the Sonsela Member is no older than ca. 216 Ma. Independent of these alternatives, both age models agree that none of the preserved Chinle Formation in PFNP is Carnian (>227 Ma) in age, and hence the biotic turnover event cannot be correlated to the Carnian−Norian boundary but is rather a mid-Norian event. Our age models demonstrate the powers, but also the challenges, of integrating detrital CA-ID-TIMS ages with magnetostratigraphic data to properly interpret complex sedimentary sequences.