Photic Zone Euxinia During the Permian-Triassic Superanoxic Event

See allHide authors and affiliations

Science  04 Feb 2005:
Vol. 307, Issue 5710, pp. 706-709
DOI: 10.1126/science.1104323


Carbon and sulfur isotopic data, together with biomarker and iron speciation analyses of the Hovea-3 core that was drilled in the Perth Basin, Western Australia, indicate that euxinic conditions prevailed in the paleowater column during the Permian-Triassic superanoxic event. Biomarkers diagnostic for anoxygenic photosynthesis by Chlorobiaceae are particularly abundant at the boundary and into the Early Triassic. Similar conditions prevailed in the contemporaneous seas off South China. Our evidence for widespread photiczone euxinic conditions suggests that sulfide toxicity was a driver of the extinction and a factor in the protracted recovery.

The most severe extinction of the past 500 million years occurred in the Late Permian (1, 2). The biotic crisis was accompanied by an oceanic anoxic event (OAE) that may have lasted up to 8 million years. Although different authors report various anoxic intervals, the most severe conditions persisted during the first 1 to 3 million years (3, 4). Anoxia has been proposed to have had a major role in driving the extinction (5, 6); surface outcropping of sulfidic waters and emissions of hydrogen sulfide to the atmosphere provide a kill mechanism that might account for the terrestrial and marine extinctions (7).

In anoxic zones of modern-day stratified lakes or restricted marine environments (e.g., the Black Sea and Antarctic fjords), conditions are favorable for bacterial reduction of sulfate to sulfide (e.g., 8). Chlorobiaceae (green sulfur bacteria) are typical of these environments in which hydrogen sulfide extends into the photic zone, where it serves as the electron donor required for anoxygenic photosynthesis. Chlorobiaceae use a distinct assemblage of light-harvesting pigments comprising bacteriochlorophylls c, d, and e and the carotenoids isorenieratene and chlorobactene. Identification of these compounds, or their diagenetic alteration products, in sediments provides unequivocal evidence for photic zone euxinic (PZE) conditions in the past (e.g., 912).

Here, we use carbon and sulfur isotopic data and biomarker and iron speciation analyses in a drill core (Hovea-3) from the onshore Perth Basin, Western Australia (13), to establish the redox conditions in the water column of the southern Tethys Ocean during the Permian-Triassic (P-T) superanoxic event. Biomarkers diagnostic for anoxygenic photosynthesis by Chlorobiaceae were identified in P-T boundary sediments of the organic-matter (OM)–rich Hovea-3 core and in coeval samples from the OM-lean Meishan-1, a new core drilled at the type section of Meishan, South China (fig. S1); these biomarkers demonstrate that waters of the Tethys Ocean were periodically euxinic in the photic zone during and after the extinction event.

Changhsingian and Griesbachian sediment samples of Hovea-3 (1960- to 1995-m depth) contain C18 and C19 aryl isoprenoids (Figs. 1A and 2A), and the Griesbachian sediments contain isorenieratane, the C40 parent hydrocarbon (Figs. 1A and 2B and fig. S2). Highly specific bacteriochlorophylls can also give rise to distinctive maleimides (9). Methyl iso-butyl maleimide was identified in the polar fractions [see (13) for separation of maleimides]. The highest concentrations of all these pigment derivatives are preserved in the Griesbachian, reflecting high green sulfur bacterial activity and, thus, PZE conditions. Isorenieratane and aryl isoprenoids (including low pristane/phytane ratios) also occur in the latest Changhsingian and earliest Induan (Griesbachian) sediments (beds 22 to 27) of the global boundary strato-type section and point (GSSP) at Meishan, South China, which suggests that PZE conditions were widespread (Fig. 1B and Fig. 3).

Fig. 1.

Typical gas chromatography–mass spectrometry selected-ion chromatogram for the 133-Da fragment ion that is diagnostic for aryl isoprenoids of an aromatic hydrocarbon fraction [see (14) for fractionation of extracts]. (A) From 1979.9 m (Hovea-3). This trace shows pseudohomologous series C12-C31 compounds identified as aryl isoprenoids with a 2,3,6-trimethyl substitution pattern and the C40 biomarker isorenieratane. These hydrocarbons were identified by comparison with a reference sample (fig. S2). (B) From 107.94 m (Meishan-1). This trace for Meishan Bed 24-6 shows an assemblage of Chlorobiaceae biomarkers similar to those seen in Hovea-3. The enhanced relative abundance of isorenieratane reflects a lower degree of thermal maturity in the sediments from South China.

Fig. 2.

Composite plot of Hovea-3 organic and inorganic data. (A) Aryl isoprenoid concentrations (μg/g of total organic carbon). (B) Isorenieratane concentrations (μg/g of total organic carbon). (C) Concentrations of vanadyl and nickel porphyrins (μg/g of total organic carbon) [see (14) for analysis of porphyrins]. (D) FeHR/FeT ratios. FeHR, highly reactive iron; FeT, total iron; FeP, pyritic iron [see (14) for isolation of iron species]. (E) δ34S of pyrite relative to Vienna Canyon Diablo troilite (VCDT) standard in ‰ [see (14) for sulfur isotope analyses]. (F) δ13C of pristane and phytane relative to Vienna Pee Dee belemnite (VPDB) standard in ‰ [see (14) for 5A molecular sieving and compound-specific isotope analyses].

Fig. 3.

Composite plot of Meishan-1 organic geochemical, biomarker, and bulk carbon isotopic data. Pristane/phytane ratios show their lowest values just below and just above the extreme negative carbon-isotopic excursion. Maxima of aryl isoprenoid and isorenieratane abundances in beds 24 to 27 bracket the extinction horizon. Notably, there are additional maxima of these markers at beds 34, 35, and 37, which indicates that there was a waxing and waning of PZE conditions that continued well into the Triassic (fig. S1).

Although the above data provide evidence for PZE during the P-T transition, the presence of benthic epifaunal macroinvertebrates such as Claraia and spirorbids demonstrates that, because these animals would have required some oxygen, the euxinia was episodic (14).

Independent evidence for euxinic conditions in the Lower Triassic seas off Australia is provided by high amounts of metalloporphyrins, in particular vanadyl (VO) porphyrins (Fig. 2C) (15). These samples also contained high concentrations of Ni (II) porphyrins.

We also measured iron species, dithionite-extractable (FeD), pyrite (FeP), and total iron (FeT) (16) in the Hovea-3 sediments. (FeD FeP)/FeT values up to 0.7 indicate euxinic + conditions in the basal-most Triassic (Fig. 2D), comparable to values reported from the modern euxinic Black Sea (17). Variations in the ratio may be due to changes in the oxycline position, the area of sediment covered by anoxic waters, and changes in weathering rates.

The isotopic composition of the oceanic sulfur reservoir is partly controlled by the balance of bacterial sulfate reduction and sulfide oxidation. These processes lead to 34S-depleted sulfide and related pyrite (18) and may result in sulfate relatively enriched in 34S. Burial of pyrite thus removes isotopically light sulfur from the seawater pool. Sedimentary sulfides in the Hovea-3 samples (Fig. 2E) are depleted in 34S compared with contemporaneous seawater sulfate (19, 20) and follow a trend toward heavier values approaching the P-T transition, as seen in carbonateassociated sulfate in other P-T samples from northern Italy and Iran (19, 20).

Sulfur-isotope fractionations between sedimentary pyrite and contemporary sulfate (19, 20) up to about 50 to 60 per mil (‰) indicate that reservoir effects did not substantially influence the isotopic signatures. The 34S/32S ratios for the sulfides in the Basal Triassic of the Perth Basin are consistent with euxinic conditions as found in the modern Black Sea and the Pliocene Mediterranean Basin (21). The change in isotope discrimination implies that there was a perturbation in the sulfur cycle in the Upper Permian and a relative increase in the fraction of sulfur buried as pyrite in the Lower Triassic compared with the Permian (22) or that there was a change in other factors (e.g., quality of OM) influencing overall sulfur isotope discrimination. Isotope shifts in reduced sulfur across the P-T boundary indicate changes in the sulfur cycle similar to those reported in sections from Japan (22).

Stable carbon isotopic data from the bulk kerogen fraction of Hovea-3 record an abrupt 7.5‰ negative shift from the Upper Permian to the Lower Triassic consistent with a localized palynofacies change from charcoal-wood dominated OM to algal-amorphous OM, respectively (23). In contrast, δ13C values of the molecular fossils pristane and phytane vary gradually across the P-T transition (Fig. 2F), representing in part a change in OM inputs or an increase in stratification toward the Triassic that caused enhanced recycling of 13C-depleted CO2. However, recent δ13C data on higher plant and phytoplankton biomarkers show similar isotopic changes across the P-T transition, which indicates a global disruption of the carbon cycle (24).

The δ13C values of pristane and phytane, derived from phytol side chains of the chlorophylls of algae and cyanobacteria, are robust proxies for the isotopic composition of phytoplankton (e.g., 25, 26). In contrast, C14-C18 n-alkyl carbon chains have many inputs, comprising primary producers and heterotrophs, and their δ13C values represent a weighted average of these. If derived from primary sources such as algae and cyanobacteria, they should be depleted in 13C compared with the co-occurring isoprenoids by 1.5‰ (e.g., 25, 26). This pattern is observed in the Triassic data (Fig. 4), in which there is independent evidence, such as the abundance of porphyrins and algal microfossils, for high primary productivity. Alternatively, isotopic enrichment of C14-C18 n-alkyl carbon chains can occur through heterotrophic processing of primary photosynthate or dominant inputs of isotopically heavy bacterial biomass. This pattern of anomalous isotopic ordering is seen in the Permian sediments. Figure 4 demonstrates this phenomenon, which suggests that there was a shift in the mode of carbon cycling at the P-T transition in Hovea-3. A similar reversal in the expected C-isotopic relation between polymethylenic and polyisoprenoid lipids has been reported for the Proterozoic-Cambrian transition (27). Moreover, this transition in C-isotopic relations is also evident in data from the P-T transition of Western Slovenia (28), where the isotopic trends of inorganic carbon and kerogen are similar to those seen at Meishan and the Southern Alps. Together, these data provide further evidence for a switch in the mode or extent of organic carbon remineralization at the P-T transition.

Fig. 4.

Δδ = (average δ13C of n-C17 and n-C18) – (average δ13C of pristane and phytane) ‰ (relative to VPDB standard in ‰) with depth (meters). Measurements determined by isotope-ratio-monitoring gas chromatography–mass spectrometry (IsoPrime, Micromass, Manchester, UK) of saturate, branched/cyclic, and n-alkane fractions [see (14) for 5A molecular sieving and compound-specific isotope analyses].

The data show that PZE conditions occurred during the P-T superanoxic event and document a major disruption of the carbon and sulfur cycles. The onset of PZE in Hovea-3 coincides with a sharp facies change, reflecting rapid transgression (e.g., 23). The association of oxygen-poor water and rapid transgression is key for the hypothesis that anoxia caused the extinction event (5). Given that similar conditions prevailed elsewhere in the Tethys Ocean during the P-T event, we propose that sulfide toxicity in the ocean and emissions of hydrogen sulfide to the atmosphere were important drivers of the largest mass extinction in the past 500 million years and may have also been a factor in the protracted recovery (7).

The local association of PZE and high-sedimentary total organic carbon contents (TOC) at Hovea-3 are similar to those of other Mesozoic OAEs (29). However, the conditions creating the record here were not repeated elsewhere, even within Western Australia, and must have been very localized. Because PZE also occurred in the contemporaneous seas off South China, localized high algal productivity probably played a key role in the formation of a petroleum-rich source rock in the Perth Basin.

Supporting Online Material

Materials and Methods

Figs. S1 and S2


References and Notes

View Abstract

Stay Connected to Science

Navigate This Article