Oceanic anoxia and extinction in the latest Ordovician

Abstract

The Late Ordovician (Hirnantian) mass extinction (LOME) was marked by two discrete pulses of high species turnover rates attributed to glacial cooling (LOME-1) and subsequent expansion of anoxic marine conditions (LOME-2). However, the mechanisms and extent of global marine anoxia remain controversial. In this study, we present uranium isotope (δ238U) data from a new Ordovician-Silurian (O-S) boundary carbonate section in the Southwest China to explore the extent/duration of the global marine anoxia, and links to the LOME. This section was found to continuously record the characteristic Hirnantian paired carbon isotope excursion (HICE) and complete benthic faunal turnover across the O-S boundary based on detailed stratigraphic constraint. The poor correlations between δ238Ucarb data, local redox proxy (Ce anomalies), and diagenetic indicators (e.g., Mn and Sr contents, Mn/Sr ratio, δ18O), as well as microscopic petrographic examinations of O-S carbonates suggest that most of the δ238U data are not produced by either local redox conditions or post-depositional diagenetic processes. By coupling our new uranium isotope data with a stochastic U isotope mass balance model, we suggest there were two episodes of widespread marine anoxia over this time interval revealed by systematic changes of δ238U with two negative shifts punctuated by one positive one in between. The former took place in the late Katian, predating the LOME-1 episode and the pronounced climatic cooling; while the latter occurred synchronously with the LOME-2 in the mid-late Hirnantian, separated by a potential oceanic oxygenation event. This presents a more complicated picture of anoxia—as inferred from U isotope records—as a driver of the end-Ordovician mass extinction. The insignificant biotic response to late Katian anoxia is consistent with the idea that the effects of anoxia on marine ecosystem can be highly variable. On the other hand, the presence of anoxia in mid-late Hirnantian strengthen the connection between the oceanic deoxygenation and fauna turnover in LOME-2.