GEOLOGY, December 2007 1139 INTRODUCTION The Paleocene–Eocene Thermal Maximum (PETM) was a period of extreme global warming ca. 55 Ma characterized by widespread nega- tive oxygen and carbon isotope values in carbonate and organic matter. These isotope anomalies indicate that isotopically light carbon was rap- idly (within a few millennia) emitted into the ocean-atmosphere system, causing global warming of 5–9 °C (e.g., review in Bowen et al., 2006), and widespread carbonate dissolution in the oceans (e.g., Zachos et al., 2005). Recovery took ~170 k.y. (Röhl et al., 2006). There is no agreement on the source of the isotopically light car- bon: one explanation for the carbon isotope excursion (CIE) is the release of ~2000–2500 Gt of isotopically light (~–60‰) carbon from oceanic methane clathrates (Dickens et al., 1995). Arguments against this hypothesis include discrepancies between the size of the Paleocene gas hydrate reservoir, the amount of oceanic carbonate dissolution, and discordance between the temperature increase recorded and amount of greenhouse gas needed (Pagani et al., 2006a). Various alternate sources of isotopically light carbon have been proposed (e.g., Pagani et al., 2006b; review in Thomas, 2007). We lack not only understanding of the cause of this global warm- ing event, but also its ending. Increased biological productivity on land (Beerling, 2000) or in the oceans (Bains et al., 2000) could have caused CO 2 drawdown. There is strong evidence for increased productivity on shelves, in epicontinental basins, and along some continental margins (e.g., Gibbs et al., 2006; Thomas, 2007). However, in the open ocean the productivity record remains problematic, with conflicting interpreta- tion based on different proxies for productivity, even for the same sites (Stoll et al., 2007; Bains et al., 2000). Barite formation in the water column and its accumulation in sedi- ments are closely related to export production (e.g., Paytan and Griffith, 2007). Barite precipitates continuously in seawater, and is a closed sys- tem after burial. Barite is not affected by diagenesis in oxic sediments (Paytan et al., 1998), but its preservation in the sediment depends on seawater saturation levels with respect to barite (Dymond et al., 1992; Paytan and Griffith, 2007). Thus, knowledge of barite saturation in the water column is required for quantitative interpretation of barite accu- mulation rate records. The Sr/Ba ratio in barite has been suggested as a proxy for the degree of barite saturation (Van Beek et al., 2003). Marine barite crystals forming in the water column have a wide range of Sr concentrations (Bertram and Cowen, 1997), and high-Sr barite is significantly more soluble than low-Sr barite (Moninn and Cividini, 2006; Rushdi et al., 2000). Barite crystals with high Sr/Ba values thus dissolve preferentially in the water column and at the sediment-water interface, and the degree of preferential dissolu- tion strongly depends on the saturation state of the ocean with respect to barite (Monnin and Cividini, 2006; Averyt and Paytan, 2003). The present average Sr/Ba of barite in core top sediments is 36.6 mmol/mol (Van Beek et al., 2003; Averyt and Paytan, 2003, 2007), ranging from ~25 to 45 mmol/mol. This range reflects variability in sea- water and microenvironment Sr concentration, the effect of temperature on partition coefficients, spatial variability within a crystal, the presence of accessory minerals in the barite sample separates, and the saturation Geology, December 2007; v. 35; no. 12; p. 1139–1142; doi: 10.1130/G24162A.1; 2 figures; Data Repository item 2007279. © 2007 The Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or editing@geosociety.org. Barite accumulation, ocean productivity, and Sr/Ba in barite across the Paleocene–Eocene Thermal Maximum A. Paytan Institute of Marine Science, University of California–Santa Cruz, Santa Cruz, California 95064, USA K. Averyt Intergovernmental Panel on Climate Change, Working Group I Support Unit, 325 Broadway, Boulder, Colorado 80305, USA K. Faul Environmental Sciences, Mills College, 5000 MacArthur Boulevard, Oakland, California 94613, USA E. Gray Geological and Environmental Science Department, Stanford University, Stanford, California 94305-2115, USA E. Thomas Department of Geology & Geophysics, Yale University, P.O. Box 208109, New Haven, Connecticut 06520-8109, USA, and Department of Earth and Environmental Sciences, Wesleyan University, Middletown, Connecticut 06459, USA ABSTRACT The Paleocene–Eocene Thermal Maximum (PETM), ca. 55 Ma, was a period of extreme global warming caused by rapid emission of greenhouse gases. It is unknown what ended this episode of greenhouse warming, but high oceanic export productivity over thousands of years (as indicated by high accumulation rates of barium, Ba) may have been a factor in ending this warm period by carbon sequestration. However, Ba has a short oceanic residence time (~10 k.y.), so a prolonged global increase in Ba accumulation rates requires an increase in input of Ba to the ocean, increasing barite saturation. We use a novel proxy for barite satura- tion (Sr/Ba in marine barite) to demonstrate that the seawater saturation state with respect to barite did not change across the PETM. The observations of increased barite burial, no change in saturation, and the short residence time can be reconciled if Ba burial decreased at continental margin and shelf sites due to widespread occurrence of suboxic conditions, leading to Ba release into the water column, combined with increased biological export production at some pelagic sites, resulting in Ba sink reorganization. Keywords: Paleocene–Eocene Thermal Maximum, barite, paleoproductivity, carbon sequestration.