GEOLOGY | Volume 44 | Number 2 | www.gsapubs.org 1 Global-ocean redox variation during the middle-late Permian through Early Triassic based on uranium isotope and Th/U trends of marine carbonates Maya Elrick 1 , Victor Polyak 1 , Thomas J. Algeo 2,3,4 , Stephen Romaniello 5 , Yemane Asmerom 1 , Achim D. Herrmann 6 , Ariel D. Anbar 5 , Laishi Zhao 3 , and Zhong-Qiang Chen 4 1 Department of Earth and Planetary Sciences, University of New Mexico, Albuquerque, New Mexico 87131, USA 2 Department of Geology, University of Cincinnati, Cincinnati, Ohio 45221-0013, USA 3 State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Wuhan, Hubei 430074, China 4 State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan, Hubei 430074, China 5 School of Earth and Space Exploration and Department of Chemistry and Biochemistry, Arizona State University, Tempe, Arizona 85287, USA 6 Coastal Studies Institute and Department of Geology and Geophysics, Louisiana State University, Baton Rouge, Louisiana 70803, USA ABSTRACT Uranium isotopes ( 238 U/ 235 U) in carbonates, a proxy for global-ocean redox conditions owing to their redox sensitivity and long residence time in seawater, exhibit substantial variability in the Daxiakou section of south China from the upper-middle Permian through the mid-lower Triassic (~9 m.y.). Middle and late Permian ocean redox conditions were similar to that of the modern ocean and were characterized by improving oxygenation in the ~2 m.y. prior to the latest Permian mass extinction (LPME), countering earlier interpretations of sustained or gradually expanding anoxia during this interval. The LPME coincided with an abrupt negative shift of >0.5‰ in d 238 U that signifies a rapid expansion of oceanic anoxia. Intensely anoxic conditions persisted for at least ~700 k.y. (Griesbachian), lessening somewhat during the Dienerian. Th/U concentration ratios vary inversely with d 238 U during the Early Triassic, with higher ratios reflecting reduced U concentrations in global seawater as a consequence of large-scale removal to anoxic facies. Modeling suggests that 70%–100% of marine U was removed to anoxic sinks during the Early Triassic, resulting in seawater U concentrations of <5% that of the modern ocean. Rapid intensification of anoxia concurrent with the LPME implies that ocean redox changes played an important role in the largest mass extinction event in Earth history. INTRODUCTION The latest Permian mass extinction (LPME) was the largest biocrisis of the Phanerozoic, resulting in a loss of >90% of species globally and the nearly total restructuring of marine and terrestrial ecosystems (Chen and Benton, 2012). A broad consensus has emerged that the LPME was triggered by the massive eruptions of the Siberian Traps through injection of large quan- tities of CO 2 and CH 4 to the atmosphere, lead- ing to extreme global warming and widespread marine anoxia (Korte and Kozur, 2010; Payne and Clapham, 2012). However, the timing of onset, the duration, and the extent of oceanic anoxia are debated, leaving the causal relation- ship between deoxygenation and the LPME uncertain. Arguments have been made for wide- spread anoxia starting as early as 10 m.y. prior to the LPME (Isozaki, 1997; Kato et al., 2002; Wignall and Twitchett, 2002), although more recent studies have concluded that ocean redox conditions were spatially heterogeneous (Bond and Wignall, 2010) and that anoxia was confined mainly to intermediate-depth oceanic oxygen- minimum zones (OMZs; Algeo et al., 2010, 2011; Winguth and Winguth, 2012). However, all of the proxy-based studies to date have relied on proxies that evaluate local rather than global water mass redox conditions. The ratio of 238 U/ 235 U (expressed as d 238 U in units of per mil [‰] deviation from a standard) is a recently developed proxy for globally inte- grated paleocean redox conditions (Weyer et al., 2008; Romaniello et al., 2013; Dahl et al., 2014). It was applied to the Permian–Triassic interval by Brennecka et al. (2011) and Lau et al. (2016); both studies documented an abrupt ~0.3‰ negative d 238 U shift at the LPME in sec- tions in south China and Turkey, corresponding to an ~6× increase in the global area of anoxic sedimentation. The Brennecka et al. (2011) study examined only a narrow, ~200-k.y.-long interval around the LPME, whereas the Lau et al. (2016) study reported U-based ocean redox trends for an ~20-m.y.-long interval spanning the latest Perm- ian through Middle Triassic. To date, U isotopes in marine carbonates have not been examined for the full upper Permian to test ideas concern- ing global ocean redox changes leading into the LPME event. We present d 238 U and Th/U trends from upper-middle Permian through lower Trias- sic marine carbonates at the Daxiakou section in south China (Hubei Province) to evaluate global seawater redox conditions in an ~8-m.y.-long interval prior to the LPME and <1 m.y. after the event. BACKGROUND Geologic Setting Daxiakou is located in Hubei Province in south-central China, an area close to the paleo- equator during the late Permian to Early Trias- sic (Fig. DR1 in the GSA Data Repository 1 ). The study section was deposited on a carbonate ramp, at water depths of >200 m, on the northern (paleowestern) margin of the Yangtze platform (Zhao et al., 2005). The section comprises the middle and upper Permian Maokou, Wujiaping, and Dalong Formations and the lower Triassic Daye Formation (Fig. 1) that consist of lime- stones and argillaceous limestones (Table DR1). Precise age control is provided by the dozens of ash layers near the Permian-Triassic boundary, conodont biostratigraphy, and carbonate d 13 C stratigraphy (Wang and Xia, 2004; Tong et al., 2007; Shen et al., 2012; Zhao et al., 2013). U Isotope and Th/U Systematics The main source for U to the global ocean is weathering of continental crust, and the main sinks are suboxic and anoxic continental margin 1 GSA Data Repository item 2017041, supplemen- tary information, figures, and data tables, is available online at www.geosociety.org/pubs/ft2017.htm, or on request from editing@geosociety.org. GEOLOGY, February 2016; v. 44; no. 2; p. 1–4 | Data Repository item 2017041 | doi:10.1130/G38585.1 | Published online XX Month 2016 © 2016 Geological Society of America. For permission to copy, contact editing@geosociety.org.