Sterane biomarkers as indicators of palaeozoic algal evolution and extinction events

https://doi.org/10.1016/j.palaeo.2006.03.050Get rights and content

Abstract

Algae are among the oldest organisms known to inhabit the Earth's biosphere with a record dating back some 2.7 Ga. Only few marine algae produce hard parts that can be preserved over geologic time and consequently their fossil record is incomplete. Molecular fossils or biomarkers add complementary information to the fossil palynomorph record. Steroids are important constituents of eukaryotic cell-membranes and are preserved in sediments as steranes. C28- and C29-steranes are indicators for the presence of green and C27-steranes for the presence of red algae, respectively. The relative abundance of steranes allows the investigation of the fossil record for Palaeozoic algal diversification and evolution.

In this study, 500 Palaeozoic rock samples, representing a broad facies variety from the Upper Ordovician to the P/T-boundary were investigated for sterane compositions. In addition three profiles covering the Hirnantian, the F/F and the Hangenberg-Event were analyzed for possible shifts in the sterane distribution associated with extinction events. A sharp increase of the C28/C29-sterane ratio from < 0.55 to > 0.70 at the Devonian/Carboniferous boundary implies a fundamental change in the green algae assemblage from more primitive, mainly C29-sterane-producing algae, to modern C28-sterane-producing algae. For all investigated extinction events a pronounced but short-lived rise in the C28-sterane content occurs that is attributed to an episodic increase in prasinophytes. The gradual radiation of algae may have been triggered by frequent mass extinctions in the Upper Devonian culminating with the massive decline of acritarchs at the D/C-boundary. The coeval rise in the C28/C29-sterane ratio indicates the presence of an non-encysting algal group and coincides with the global augmentation of numerous filamentous Codiacea (Siphonales) and the rise of euspondyle and metaspondyle Dasycladales.

Introduction

Palaeozoic phytoplankton consist mainly of three major groups: Cyanophytes, acritarchs and algae. Steroids represent biomarkers for algal eukaryotes, whereas hopanes are typical biomarkers for prokaryotic cyanophytes (Ourisson et al., 1979, Brocks et al., 1999, Summons et al., 1999). Acritarchs are organic walled microfossils of uncertain affinities (Evitt, 1963), although a relationship to dinoflagellate cysts has been invoked by numerous investigations (Tappan, 1980, Lipps, 1993, Colbath and Greenfell, 1995, Strothers, 1996, Moldowan and Talyzina, 1998, Moldowan et al., 2001, Armstrong and Brasier, 2005).

At present a great variety of organisms including algae, land plants, fungi, and animalia produce steroids (Volkman, 1986, Patterson, 1994, Nes and Venkatramesh, 1994, Volkman et al., 1998, Volkman, 2003, Volkman, 2005). For Palaeozoic marine depositional environments, algae are assumed to have been the main producers of steranes, with increasing land–plant derived contributions starting not before the Upper Devonian. Studies of recent algae commonly indicate the presence of a variety of sterols with usually one or two sterols predominating (Volkman, 1986, Patterson, 1994, Nes and Venkatramesh, 1994, Volkman et al., 1994, Volkman et al., 1998, Volkman, 2003, Volkman, 2005). The sterol composition may vary depending on environmental factors including growth rate, temperature, illumination, nutrient supply and others (Volkman, 2003, Volkman, 2005). Although sterol composition is often crucial for identification of Recent algae, unambiguous identification of individual algal species based exclusively on steroid composition is not feasible (Volkman, 1986, Patterson, 1994, Nes and Venkatramesh, 1994, Volkman et al., 1998, Volkman, 2003, Volkman, 2005). Phylogenic relationships based on the predominance of specific steroids, however, are well established (Patterson, 1994). Huang and Meinschein (1979) summarized that C27-steranes are biomarker for red algae while green algae preferentially produce C28- and C29-steranes. This generalized pattern was used by Grantham and Wakefield (1988) in their sterane analysis of 400 Phanerozoic oils of postulated marine origin. In their study of averaged 50 Ma intervals a constant, nearly exponential rise of the C28/C29-sterane ratio throughout geological time was observed, reflecting the progressive evolution of algae. The content of C27-steranes exhibits no systematic trend but on average remains stable throughout the Palaeozoic, corresponding to the palaeontological record of red algae (Tappan and Leoblich, 1973, Tappan, 1980).

Due to the well-known dependency of sterol compositions on environmental factors and the presence of suites of different steroids in individual species any application of steroid composition towards evolutionary interpretations has to depend on large well averaged data sets. This is already evident from the pioneering work of Grantham and Wakefield (1988) in which steroid distribution over time is contemplated separately for different lithofacies. Their 50 Ma averaged time step for the Jurassic reveals a C28/C29-sterane ratio of 0.35 for carbonate-sourced and 0.90 for siliciclastic-sourced oils. The large deviations presumably result from sampling of oils derived from extreme environments and from low sample density in this time interval.

The study of Grantham and Wakefield (1988) covered the entire Phanerozoic and the albeit large number of 400 oils corresponds to 600 Ma of evolution. On average this results in 1 sample per 1.5 Ma years and each 50 Ma time step is represented by 75 oil samples. However, this sample distribution is biased because the 400 oil samples derive from approximately 10 different petroleum source rocks only and those source rocks will represent exceptional environmental conditions of sedimentation, which are not representative for a larger stratigraphic interval.

Higher frequency oscillations in the marine C28/C29-sterane ratio have to be assumed during the Phanerozoic and especially during the Palaeozoic due to the occurrence of severe mass extinction events (Raupp and Sepkoski, 1982, Raupp and Sepkoski, 1984, Sepkoski, 1993, Erwin, 1998, House, 2002, Bambach et al., 2002, Bambach et al., 2004), major glaciations (Crowell, 1978, Frakes et al., 1992, Martini, 1997), and the evolution of land plants and associated changes in soil biochemistry and nutrient flux to the oceans (Algeo et al., 2001, Martin, 2003, Falkowski et al., 2004). Due to the large averaging steps of 50 Ma used by Grantham and Wakefield (1988) the effect of these catastrophic events on marine algal evolution and sterane composition can not be deciphered.

This work examines sterane composition and algal evolution during the Palaeozoic following higher temporal/stratigraphic resolution in order to address higher frequency variability in sterane distribution due to evolutionary radiation and catastrophic mass extinctions. In contrast to Grantham and Wakefield (1988), we decided not to analyze reservoir oils but to investigate sedimentary rock extracts.

Both approaches have advantages as well as disadvantages. Oil samples are almost exclusively allochthonous and therefore, a degree of uncertainty results from the oil-source rock correlation, which is needed to derive the stratigraphic age of the formation that generated the oil. Furthermore, oils in commercial quantities are almost exclusively generated from organic matter rich rocks, mainly black shales, deposited under restricted water circulation and reducing conditions (Peters et al., 2005). Thus a strong bias will be exerted on sterane composition due to environmental conditions of sedimentation. For a given time interval of several million years or a sediment succession of several hundred meters, source rocks will be of very limited extent but contribute all the oil generated from the whole interval, again leading to a severe bias. Oils do, however, provide good average values for a larger drainage area. They are usually not affected by secondary alteration or weathering and biomarkers always occur in larger quantities thus simplifying analysis and enhancing signal to noise ratios (Peters et al., 2005).

Rock samples offer advantages over oils in that they are stratigraphically better resolved. They allow analysis of intervals lean in organic matter and, thus, devoid of oil source rocks. Rock samples are available from very different lithologies and facies regimes and thus may much better represent the whole range of environmental conditions occupied by fossil algae. Rock samples are, however, subject to secondary alteration or weathering and may be impregnated with migrating bitumen of different stratigraphic origin thus obscuring the precision of the stratigraphic assignment. In many cases extract and biomarker yields of rocks, especially non-source rocks are lean, consequently leading to a reduction in analytical precision.

In this study we analyzed bitumen extracted from sedimentary rock samples in order to facilitate high resolution sampling over short intervals representing extinction events, an approach not feasible when investigating oils. We provide data at two different temporal resolutions, one data set of time-averaged sterane compositions to follow general trends in algal evolution and for comparison with previously published data by Grantham and Wakefield (1988). A second data set covers three Palaeozoic Events: The Hirnantian Event from a section on Anticosti Island, Canada, the F/F-Event from a section in Kowala, Holy Cross Mountains, Poland and the Hangenberg-Event from a section in the Appalachian Basin, USA. Supplementary data on lithofacies and stratigraphy is available from the PhD thesis by P. Empt accessible at: http://kups.ub.uni-koeln.de/volltexte/2004/1264/.

Section snippets

Methods

500 rock samples of different stratigraphy, lithology, palaeoclimate conditions and palaeogeographic locations as shown in Fig.1 were investigated. A screening analysis was carried out, to ensure sufficient biomarker concentrations and to exclude samples that received terrigenous organic matter input or were impregnated by migrating fluids. The screening procedures involved solvent extraction and analysis of whole extracts by gas chromatography with flame ionization detector (GC-FID) and

Samples and palaeogeographic settings

To minimize the bias of depositional environment on sterane distributions, samples were selected from a wide range of marine environments and in dense stratigraphic succession. As shown in Fig 1, the Ordovician, Devonian and Carboniferous samples represent a wide range of palaeogeographical positions, the Silurian and Permian samples lack such a diversity. Besides formations or locations represented by single samples only for a few key sections extended sample suites were collected. The most

Secular palaeozoic sterane evolution

The sterane composition of 500 dated rock extracts were investigated by GC/MS and the C28/C29-sterane ratio was calculated. Individual data points were averaged in 50 ma steps, (Fig. 2) to facilitate a direct comparison with sterane evolution composed by Grantham and Wakefield (1988). Sterane composition was then averaged in 25 ma steps (Fig. 2) to reveal higher frequency variations. One reason for the steady increase of the C28/C29-sterane ratio in Grantham and Wakefield (1988) work results

Conclusion

The C28/C29-sterane ratio shows systematic variations during the Palaeozoic that can be attributed to marine algal evolution and extinction events. In general, from the Ordovician to Upper Devonian, the C28/C29-sterane ratios lie below a threshold value of 0.55. Exceptions from this trend are restricted exclusively to the Kacak, Upper and Lower Kellwasser extinction events. These short-lived events show a reversible increase in C28/C29-sterane ratios, attributed to a episodic change in the

Acknowledgements

We thank numerous colleagues who provided the sample material accompanied by stratigraphic and facies information for this study. Reviewers R. Pancost and L. Stasiuk provided thoughtful comments that helped to improve the manuscript. The German Research Foundation is acknowledged for providing financial support under grant Schw554/7. This study is part of the Priority Programme “Evolution of System Earth” and benefited from the support by other programme participants.

References (51)

  • J.K. Volkman et al.

    Sterol biomarkers for microalgae from the green algal class Prasinophyceae

    Organic Geochemistry

    (1994)
  • J.K. Volkman et al.

    Microalgal biomarkers: a review of recent research developments

    Organic Geochemistry

    (1998)
  • T.J. Algeo et al.

    Effects of the middle to late Devonian spread of vascular land plants on weathering regimes, marine biotas, and global climate

  • H.A. Armstrong et al.

    Microfossils

    (2005)
  • R.K. Bambach et al.

    Anatomical and ecological constraints on Phanerozoic animal diversity in the marine realm

    Proceedings of the National Academy of Sciences of the United States of America

    (2002)
  • R.K. Bambach et al.

    Origination, extinction, and mass depletions of marine diversity

    Paleobiology

    (2004)
  • J.J. Brocks et al.

    Archean molecular fossils and the early rise of eukaryotes

    Science

    (1999)
  • P. Copper

    Evolution, radiations, and extinctions in proterozoic to middle palaeozoic reefs

  • P. Copper

    Reefs during the multiple crisis towards the Ordovician–Silurian boundary: Anticosti Island, eastern Canada, and worldwide

    Canadian Journal of Earth Sciences

    (2001)
  • P. Copper

    Reef development at the Frasnian/Famennian mass extinction boundary

    Palaeogeography, Palaeoclimatology, Palaeoecology

    (2002)
  • J.C. Crowell

    Gondwana glaciation, cyclothems, continental positioning and climate change

    American Journal of Science

    (1978)
  • C. Downie

    Acritarchs in British stratigraphy

    Geological Society of London, Special Report

    (1984)
  • W.R. Evitt

    A discussion and proposals concerning fossil dinoflagellates, hystrichospheres, and acritarchs, I and II

    Proceedings of the National Academy of Sciences of the United States

    (1963)
  • P.G. Falkowski et al.

    The evolution of modern eukaryotic phytoplankton

    Science

    (2004)
  • L.A. Frakes et al.

    Climate Modes for the Phanerozoic

    (1992)
  • Cited by (158)

    View all citing articles on Scopus
    View full text