Abstract

Photic zone euxinia (PZE) has proven important for elucidating biogeochemical changes that occur during oceanic anoxic events, including mass extinction and conditions associated with unique fossil preservation. Organic geochemical analyses of a 380 Ma invertebrate fossil, which included well-preserved soft tissues, from the Gogo Formation (Canning Basin, Western Australia) showed biomarkers and stable isotopic values characteristic of PZE and a consortium of sulfate-reducing bacteria, which lead to exceptional fossil and biomarker preservation. The carbonate concretion contained phytoplankton, green sulfur bacteria (Chlorobi), and sulfate-reducing bacteria biomarkers with an increasing concentration toward the nucleus where the fossil is preserved. The spatial distribution of cholestane unequivocally associated with the fossilized tissue and its high relative abundance to the total steranes suggest that the fossil is a crustacean. The presence of an active sulfur cycle in this Devonian system, including sulfate reduction and the resulting PZE, played a pivotal role in the preservation of soft tissue from the fossil and its associated low-maturity biomarker ratios.

INTRODUCTION

Photic zone euxinia (PZE) is defined as the presence of H2S at the chemocline in water columns where oxygen is absent and light is available for photosynthesis by, for example, green sulfur bacteria (Chlorobi). Chlorobi play a prominent role in the sulfur cycle at the chemocline of stratified water columns (e.g., extant Black Sea) and their biomarkers provide the most direct molecular and isotopic evidence for the onset of anoxygenic photosynthesis in ancient water columns. PZE has proven important for elucidating biogeochemical changes that occur during oceanic anoxic events, especially those conditions associated with the largest mass extinction event during the Phanerozoic (Grice et al., 2005). In addition, PZE has been reported in sediments hosting concretions with exceptionally preserved fossils (Lagerstätte) in which soft tissue is identified (e.g., Heimhofer et al., 2008, 2010; Schwark et al., 2009). The Upper Devonian Gogo Formation Lagerstätte of Australia preserves a unique reef fauna. The formation comprises a sequence of shales with thin beds of limestone and bedded micritic limestone concretions representing the basinal facies of the reef complex. The Gogo Formation interfingers with, and is equivalent to, the Sadler Formation marginal slope facies, and overlies the platform facies of the Pillara Formation (e.g., Playford et al., 2009). A late Givetian age, Stichopus hermanni zone is suggested for the lower part of the section; however, the fossil-bearing nodules from the upper part of the section are well constrained to conodont zones 3–5 (e.g., Playford et al., 2009). The exceptional three-dimensional preservation of the skeletal fossils has long been recognized; recently, soft tissues replaced by apatite have been described (Briggs et al., 2011; Long and Trinajstic, 2010). We investigate biogeochemical cycles leading to this exceptional preservation.

METHODS

A fossilized invertebrate (of unknown affinity) within a carbonate concretion was separated into two parts. A thin slice (5 mm thick) comprising the fossilized tissue was separated from the matrix. Additional samples were taken within the matrix, from concentric layers away from the nucleus (Fig. 1). All the samples were ground, organically extracted, and separated by liquid chromatography into saturated, aromatics and polars. Semiquantitative analyses were performed on the saturated fraction. Polar fractions were treated with Raney nickel to release the C-S bound compounds. Solid residues were treated with HCl to release the carbonate associated biomarkers. Gas chromatography–mass spectrometry and gas chromatography–isotope ratio–mass spectrometry were performed on the saturated and aromatic fractions. More details and methods can be found in the GSA Data Repository1.

RESULTS AND DISCUSSION

The n-alkanes, the regular isoprenoids pristane (Pr) and phytane (Ph), steranes, hopanes, and components derived from Chlorobi carotenoids were the main compounds identified, showing qualitative and quantitative (Fig. 1) differences in molecular composition as well as δ13C between the innermost (fossil) and the outermost samples of the concretion (matrix).

The C-20 isomerization of steranes in the fossil sample is dominated by the thermally unstable biological epimer (20R), rather than the geologically favored 20S configuration (Peters et al., 2005). The C29 5α, 14α, 17α(H) sterane has a 20R/20R + 20S of 0.15. These data, in addition to an average ratio for 22,29,30-Trisnorneohopane (Ts)/Ts + 22,29,30-Trisnorhopane (Tm) of 0.26 and the absence of triaromatic steroids, are consistent with relatively low thermal maturity of the organic matter (OM) within the fossil. Similarly low thermal maturities have been reported in concretions from different geological ages and depositional settings (e.g., De Craen et al., 1999; Heimhofer et al., 2010; Pearson et al., 2005), all with relatively low concentrations of clay minerals. Isomerization reactions are reported to be catalyzed by clays (e.g., Peters et al., 2005; Nabbefeld et al., 2010), explaining the low thermal maturities reflected by biomarkers in the carbonate concretions.

PALEOENVIRONMENTAL CONDITIONS

The average Pr/Ph of 0.5 from the fossil is consistent with anoxia during the time of fossil preservation and with the redox conditions in the Devonian paleowater of the Gogo Formation (Playford et al., 2009). However, evidence of PZE, accounting for the fossil preservation, is provided by the identification of a suite of monoaryl, diaryl, and triaryl isoprenoids with a 2,3,6/3,4,5-trimethyl substitution pattern (Fig. 2A), largely attributed to carotenoids of Chlorobi. The intact biomarkers isorenieratane, renieratane, and the Devonian marker paleorenieratane (Fig. 2; Fig. DR1 in the Data Repository) were also identified, representing direct evidence for PZE associated with deposition of OM-rich sediments (Summons and Powell, 1986; Brown and Kenig, 2004; Grice et al., 2005; Marynowski et al., 2011). The extent of the PZE that accompanied the fossil preservation was evaluated based on the aryl isoprenoid ratio (AIR) proposed by Schwark and Frimmel (2004), and compared with oils and their potential source rocks from the same basin (i.e., anoxic muds of the Gogo Formation) (Maslen et al., 2009; S. Tulipani, 2012, personal commun.). Figure 2C suggests anoxic conditions for all the samples evaluated; however, the lowest AIR was found in the fossil, indicating persistent PZE at the time of fossil preservation and episodic PZE conditions for the deposition of the muds.

It has been suggested that derivatives of Chlorobi carotenoids, in particular aryl isoprenoids, can be generally formed via thermally induced electrocyclic reactions (Xinke et al., 1990). However, crocetane (Greenwood and Summons, 2003), reported to be a thermal product of isorenieratane in Devonian samples from the Gogo Formation, is absent in the concretion. Desulfurized aryl isoprenoids were obtained from the polar fraction of the fossil and were dominated by a C32 triaryl isoprenoid and a limited distribution of mono- and diaryl isoprenoids (Fig. 2B). Sulfurization of biomarkers occurs at early stages of diagenesis and, given the low maturity of the sample, these Chlorobi-derived biomarkers may be formed by a radical reaction at the time of deposition and/or within the chemocline, being produced by electrocyclic reactions initiated by light and not only associated with subthermal processes (Grice et al., 1996).

Preservation of organic matter throughout sulfurization is an important pathway under anoxic conditions caused by reactions of H2S produced by sulfate-reducing bacteria at early stages of diagenesis, especially when there are low iron levels (Sinninghe Damste and De Leeuw, 1990). The desulfurized saturated fraction showed a typical distribution for a paleoenvironment rich in H2S, showing abundant n-alkanes (C15 to C32) with an even-over-odd carbon predominance related with microbial sources from a highly saline, carbonate-rich environment (Dembicki et al., 1976; Summons et al., 2010). In addition, the Pr/Ph (0.13) is consistent with the prevailing reduction of phytol to dehydrophytol under euxinic conditions at the time of sulfur incorporation of the functionalized biolipids.

CHEMOTAXONOMY

Quantitative analyses of biomarkers performed on all samples (Fig. 1) provide a sense of spatial variation within the concretion, showing a greater abundance of biomarkers toward the fossil nucleus consistent with a high phytoplanktonic input at the early stages of the concretion formation. The isotopic and molecular differences found in the saturated biomarkers of the concretion arise from sinking OM derived from phytoplankton, and in situ decomposition products of secondary OM recycled by microbial activity of sulfate-reducing bacteria building the concretion; more remarkably, biomarkers associated with the invertebrate are preserved within the fossilized soft tissue.

Steranes are biomarkers derived from sterols present in the lipid membrane of eukaryotes (Peters et al., 2005). In the fossil layer, where the invertebrate tissue is preserved, cholestane is the dominant sterane (75% of steranes) and its concentration is as much as 10 times more abundant compared to the external layers of the concretion (Figs. 1 and 3A; Fig DR2). Comparable predominance of the C27 sterane was found in the S-bound fraction (Fig. 3B) as well as in the bitumen released after decarbonation of the fossil. The dominance of cholestane in this concretion is unique and suggests that it is indigenous to the fossil biomass, making it suitable as a marker for taxonomic identification. However, the contribution of sterols from algae and cyanobacteria cannot be excluded; among extant taxa, similar predominance of the C27 sterol has only been reported in crustaceans and some molluscs (e.g., Kanazawa, 2001). The absence of the shell in the sample precludes this animal being from the Mollusca. We here place the unidentified invertebrate within the subphylum Crustacea due to the presence of segmented muscle bands, preserved as apatite and calcite (Figs. DR3 and DR4), and the dominance of cholestane in the fossil as a diagenetic product of cholesterol.

Cholesterol in a crustacean is obtained from an external source and/or from dealkylation of sterols present in their algal diet, as they cannot biosynthesize it de novo (Kanazawa, 2001). Herbivory studies (Grice et al., 1998) show that the δ13C of cholesterol serves as a conservative marker of dietary sterols at this trophic level. For example, the δ13C value of cholestane (−30.5‰) in the fossil (Fig. 3A) represents an average of the δ13C value of the sterols in the phytoplanktonic community. Cholestane is enriched by 4.0‰ and 3.5‰ compared with the average of the C16 to C19n-alkanes (−34.8‰) and Ph (−34.0‰), respectively (Fig. 3A). These isotopic disparities are consistent with the different biosynthetic pathways for fatty acids and isoprenoids as well as their location of synthesis in phytoplanktonic cells (Schouten et al., 1998).

In contrast with the steranes, the long-chain n-alkanes (>n-C22) in the sample are widespread through the concretion and dominate the saturate fraction in both fossil and matrix (Fig. 1) released after acid treatment, similar to results reported in bacterially mediated ooids (Summons et al., 2010). Another source is proposed here for these long-chain n-alkanes, which are relatively depleted in δ13C (average −40‰) (Figs. 3A, 3C). In addition, aliphatic compounds are important in the diagenetic products of the biopolymer chitin in modern and fossil cuticles (e.g., Gupta et al., 2009), but the δ values reported here are inconsistent with a chitin source (Schimmelman et al., 1986). However, the strong affinity of these 13C depleted n-alkanes with the carbonate minerals may be the result of autolithification of bacterial cell walls during bacterially mediated soft tissue mineralization (Briggs, 2003). Sulfate-reducing bacteria have been recognized to be strongly involved in preservation of soft tissue by promoting authigenic mineralization in anoxic sediments (e.g., Briggs, 2003) and during precipitation of carbonates by decomposition of OM (Berner, 1969). Calcite, apatite, and dolomite along with pyrite (3–4 μm to 20 μm in diameter) were identified in the fossil (see the Data Repository). Based on the evidence of anoxic conditions, mineral association, and 13C-depleted long-chain n-alkanes, sulfate-reducing bacteria at the sediment-water interface are here proposed to play a pivotal role in the mineralization involved in tissue preservation, and in the precipitation of the carbonates under alkaline conditions (e.g., Londry et al., 2004; Marynowski et al., 2011; Ladygina et al., 2006; Briggs, 2003). δ13C of the carbonate (–7.1‰) supports a mixture of sources (sulfate-reducing bacteria and diagenetic carbonates) (Coleman, 1993).

IMPLICATIONS AND CONCLUSIONS

The exceptional preservation of a set of biomarkers (phytoplanktonic, Chlorobi derived, and sulfate-reducing-bacteria related) in the fossil points to rapid encasement of the crustacean enhanced by sulfate-reducing bacteria under PZE, preventing the crustacean tissue and sinking OM from further decomposition. This work provides the first evidence of PZE playing a vital role in fossil (including soft tissue) and biomarker preservation. The presence of organosulfur compounds suggests that preservation of the Crustacean occurred within the water column and/or at the chemocline under persistent PZE. The sulfate-reducing bacteria recycled the OM anaerobically, decreasing the alkalinity and leading to conditions that allow the carbonate concretion to accumulate (Berner, 1969). δ13C differences between cholestane, phytane, and n-alkanes (<C23) all support high phytoplankton productivity (Schouten et al., 1998; Grice et al., 2005). It is likely that a phytoplankton bloom along with limited water circulation promoted the anoxic conditions and eventually led to the development of PZE in a stratified Devonian paleowater column.

Grice and Melendez thank the Australian Research Council for a Queen Elizabeth II Discovery grant (awarded to Grice; “Characteristics of OM formed in toxic, sulfide-rich modern and ancient sediments”), and for Melendez’s Ph.D. stipend. Melendez and Ladjavardi acknowledge Curtin University for fee waivers, and Trinajstic acknowledges a Curtin Fellowship. Greenwood thanks the John de Laeter Centre for a Research Fellowship. We also thank Geoff Chidlow, Stephen Clayton (WA-OIGC) and David Adams (Centre for Microscopy, Characterisation & Analysis, The University of Western Australia) for technical support, Tim Senden (Australian National University) for providing the invertebrate nodule, and anonymous reviewers for highly constructive comments.

1GSA Data Repository item 2013030, detailed methods, sample information, is available online at www.geosociety.org/pubs/ft2013.htm, or on request from editing@geosociety.org or Documents Secretary, GSA, P.O. Box 9140, Boulder, CO 80301, USA.