Attribution: You must attribute the work in the manner specified by the author or licensor (but no in any way that suggests that they endorse you or your use of the work).Noncommercial ‒ you may not use this work for commercial purpose.No Derivative works ‒ You may not alter, transform, or build upon this work.Sharing ‒ Individual scientists are hereby granted permission, without fees or further requests to GSA, to use a single figure, a single table, and/or a brief paragraph of text in other subsequent works and to make unlimited photocopies of items in this journal for noncommercial use in classrooms to further education and science.

The oxygen content of the oceans is intimately connected with the geochemical cycles of carbon and other elements in a complex network of feedbacks. Hence, biogeochemists are interested in reconstructing how redox conditions in the oceans—and in the broader environment—have changed with time. In the wake of recent analytical advances, this decade has seen the invention of several new paleoredox proxies, as well as the refinement of existing ones, catalyzing a renaissance of research on this topic. Much of this effort has focused on the Precambrian, an era that apparently witnessed at least two very large, stepwise increases in environmental oxygenation (Canfield, 2005). In this issue, Pearce et al. (2008, p. 231) demonstrate the promise of a new tool, molybdenum (Mo) stable isotopes, to understand the smaller and more rapid redox perturbations that have occurred in Phanerozoic oceans.

Pearce et al. (2008) aim to quantify the extent of seafloor anoxia during the Early Jurassic (Toarcian) Ocean Anoxic Event (OAE), a time when carbon release rates from methane hydrate dissociation were similar to those from modern fossil fuel burning (Hesselbo et al., 2000; Cohen et al., 2007). Mo stable isotopes are useful in this effort because there is a marked contrast in the extent of Mo isotope fractionation during removal of Mo from the oceans to at least two sedimentary sinks. Specifically, light Mo isotopes adsorb preferentially to ferro-manganese oxyhydroxides with a fractionation factor of ∼1 ‰/amu (Barling and Anbar, 2004), while there is little fractionation in H2S-rich, “euxinic” settings where Mo is quantitatively transferred from seawater to sediments. Therefore, it has been proposed that δ98/95Mo in the oceans reflects the relative importance of oxic versus euxinic Mo sinks, and hence the areal extent of oxic versus euxinic seafloor (Barling et al., 2001; Siebert et al., 2003). δ98/95Mo in black shales and other organic-rich sediments might therefore be used to reconstruct the seawater δ98/95Mo record (Barling et al., 2001). The power of this proxy is that it potentially yields information about the extent of basin-wide, or even global, anoxia; most other proxies provide specific information about the local redox environment.

While this view of the Mo isotope budget in the oceans is essentially correct, a number of factors complicate quantitative interpretations of Mo isotope variations in marine sediments. Most important is the growing realization that Mo isotopes are also fractionated during removal in intermediate redox environments where O2 is scarce but H2S is not abundant. In these so-called “suboxic” or “anoxic” settings, Mo accumulates alongside organic carbon in sediments, but the seawater-to-sediment transfer is not quantitative. The result is an expressed isotope fractionation that is intermediate in magnitude between those in euxinic and oxic settings (Poulson et al., 2006; Siebert et al., 2006). This phenomenon creates two problems. First, it is necessary to demonstrate that the local redox conditions during deposition of any particular sediment being analyzed were euxinic, with H2S > 100 µM (Arnold et al., 2004; Williams et al., 2004). Otherwise, there is the danger that variations in δ98/95Mo in sediments reflect changes in local environmental conditions rather than changes in seawater δ98/95Mo. Second, the importance of such sediments to the global Mo ocean budget is unclear but potentially large (McManus et al., 2006), and so there is an extra sink in the Mo isotope mass balance equation that is poorly constrained. Expansion of suboxic seafloor at the expense of oxic seafloor would leave the Mo isotope budget relatively unperturbed, due to the similar magnitude and direction of fractionation in both these sinks (Anbar et al., 2005).

To cope with the first of these problems, Pearce et al. (2008) employ a multiple proxy approach (Sageman et al., 2003; Brumsack, 2006). They use new measurements of total organic carbon (TOC) and Re/Mo, and prior measurements of the degree of pyritization (DOP) from nearby in the Cleveland Basin (Raiswell et al., 1993), to argue that local conditions were persistently euxinic at their sampling site during the Toarcian OAE (Interval 2). DOP is of particular value because it relates directly to the H2S content of local bottom waters, unlike TOC (Lyons and Severmann, 2006). Additionally, the interpretation of DOP (and other sedimentary Fe proxies, like Fe/Al) is not tied to assumptions about the relative ocean inventories of Re and Mo—assumptions that may not be valid during extended periods of widespread anoxia that could alter the ocean budgets of one or both of these elements. The data indicate that Interval 2 is a good candidate for application of the Mo isotope proxy.

The picture that emerges from the Mo isotope measurements is that ocean oxygenation during the Toarcian OAE was profoundly depressed. The importance of non-fractionating, as opposed to fractionating, sinks for Mo from the oceans increased by about an order of magnitude, which may translate into a tenfold expansion of euxinic seafloor. As the authors note, this estimate is probably a lower limit because the Mo budget in the oceans probably did not reach a new steady-state during Interval 2, which spans ∼200,000 yr (25% to 50% of the modern Mo ocean residence time). Also, because the relative importance of the suboxic sedimentary sink is unconstrained, Interval 2 could represent not only a tenfold expansion of euxinic seafloor but, additionally, a great expansion of suboxic seafloor at the expense of oxic seafloor. To resolve this ambiguity, it would be useful to have proxies that are particularly sensitive to the extent of suboxic sedimentation. Such proxies may emerge in the near future from the development of the U and Re stable isotope systems (Miller et al., 2007; Stirling et al., 2007; Weyer et al., 2008).

Pearce et al. (2008) logically propose that fine-scale variations of δ98/95Mo and Mo concentration within Interval 2 are a consequence of a shortened Mo ocean residence time in Mo-depleted oceans. If so, they argue, these proxies record oscillations in ocean anoxia that may be linked to episodes of methane release and enhanced continental weathering, possibly driven by orbital precession. The details of the linkages between methane release (as recorded in δ13C) and Mo proxies are speculative, but this intriguing interpretation highlights the need for a more nuanced understanding of the Mo geochemical cycle.

Consistent with the Mo isotope message, the Mo content of these organic-rich sediments is remarkably low, as might be expected from enhanced rates of Mo removal to euxinic and suboxic ocean sediments. The Mo concentration in contemporaneous deep seawater can be inferred from the ratio Mo/TOC, which compensates for variations in the burial flux of organic carbon that likely carries Mo to sediments (Algeo and Lyons, 2006). Mo/TOC in Interval 2 is < 1 ppm/wt%, a value unprecedented in the Phanerozoic, and reminiscent of Archean black shales (Scott et al., 2008; Anbar et al., 2007). A Mo concentration lower even than that in the deep waters of the modern Black Sea—some 100 times lower than in today's open oceans—is implied (Algeo and Lyons, 2006). At such concentrations, Mo scarcity in upwelling regions of the oceans could have affected the biological nitrogen cycle because of the importance of Mo for nitrogen fixation and nitrate assimilation. If the ocean Mo inventory was as depleted as the data imply, then such suggestions, previously confined to the Precambrian (Anbar and Knoll, 2002), may be relevant to unraveling the bio-geochemistry of the Toarcian and other OAEs.

We thank S. Severmann for insightful discussions.