A reconstruction of Archean biological diversity based on molecular fossils from the 2.78 to 2.45 billion-year-old Mount Bruce Supergroup, Hamersley Basin, Western Australia

Abstract

Bitumens extracted from 2.7 to 2.5 billion-year-old (Ga) shales of the Fortescue and Hamersley Groups in the Pilbara Craton, Western Australia, contain traces of molecular fossils. Based on a combination of molecular characteristics typical of many Precambrian bitumens, their consistently and unusually high thermal maturities, and their widespread distribution throughout the Hamersley Basin, the bitumens can be characterized as ‘probably of Archean age’. Accepting this interpretation, the biomarkers open a new window on Archean biodiversity. The presence of hopanes in the Archean rocks confirms the antiquity of the domain Bacteria, and high relative concentrations of 2α-methylhopanes indicate that cyanobacteria were important primary producers. Oxygenic photosynthesis therefore evolved > 2.7 Ga ago, and well before independent evidence suggests significant levels of oxygen accumulated in the atmosphere. Moreover, the abundance of cyanobacterial biomarkers in shales interbedded with oxide-facies banded iron formations (BIF) indicates that although some Archean BIF might have been formed by abiotic photochemical processes or anoxygenic phototrophic bacteria, those in the Hamersley Group formed as a direct consequence of biological oxygen production. Biomarkers of the 3β-methylhopane series suggest that microaerophilic heterotrophic bacteria, probably methanotrophs or methylotrophs, were active in late Archean environments. The presence of steranes in a wide range of structures with relative abundances like those from late Paleoproterozoic to Phanerozoic sediments is convincing evidence for the existence of eukaryotes in the late Archean, 900 Ma before visible fossil evidence indicates that the lineage arose. Sterol biosynthesis in extant eukaryotes requires molecular oxygen. The presence of steranes together with biomarkers of oxygenic photosynthetic cyanobacteria suggests that the concentration of dissolved oxygen in some regions of the upper water column was equivalent to at least ∼1% of the present atmospheric level (PAL) and may have been sufficient to support aerobic respiration.

Introduction

Microfossils, stromatolites and isotopes of sedimentary carbon and sulfur all indicate that microorganisms inhabited Earth during the Archean, the time before 2.5 billion years ago (Ga) (Fig. 1). The fossils and isotopes also provide circumstantial evidence for the early evolution of some physiological attributes and metabolic pathways. Sedimentary organic matter strongly depleted in the carbon isotope ¹³C implies the presence of archaeal methanogens and bacterial methanotrophs (Hayes, 1983), fractionated sulfide isotopes in sulfate crystals suggest the existence of mesophilic bacterial sulfate-reducers (Shen et al., 2001), and palimpsest filament tufts in lacustrine stromatolites are consistent with phototropic cyanobacteria (Buick, 1992). However, these lines of evidence depend on chains of inference with concomitant compounding uncertainties. The oldest fossils well enough preserved to be recognized as a certain member of an extant clade only appear in the mid-Paleoproterozoic, providing diagnostic and unequivocal evidence for organisms of the phylum Cyanobacteria at 2.15 Ga (Hofmann, 1976). The oldest remains of possible eukaryotes were discovered in iron deposits 1.87 Ga old Han and Runnegar 1992, Schneider et al 2002, while morphologic evidence for the domain Archaea is not known from any part of the fossil record. Therefore, the phylogenetic position of life in the Archean remains largely obscure. Potentially, a better insight into early Precambrian microbial diversity can be obtained from molecular fossils (biomarkers) preserved in sedimentary rocks.

This paper discusses the paleobiological and paleoenvironmental significance of biomarkers detected in late Archean sedimentary rocks from the Pilbara region of Western Australia Brocks et al 1999, Brocks et al 2003. Biomarkers, their distribution in the Archean samples, their occurrence in other Precambrian strata and their potential biologic source will be discussed according to hydrocarbon class in the following order: normal and branched alkanes, cyclohexylalkanes, adamantanes, tricyclic terpanes, hopanes and the diverse range of steranes and steroids.

A companion paper in this issue (Brocks et al., 2003) gives a detailed assessment of syngeneity of the biomarkers and addressing the possibility that the Archean host rocks were contaminated during drilling and storage, or were adulterated by migrating fluids during post-Archean history. It is concluded that adamantanes and polyaromatic hydrocarbons of the Hamersley Group are ‘certainly syngenetic’, and aliphatic hydrocarbons and polycyclic biomarkers of the Hamersley and Fortescue Group ‘probably syngenetic’. Therefore, the age of the molecules that contain the biologic information is not fully resolved. Hence, paleobiological interpretation presented in this paper should be cited cautiously and with reference to the remaining uncertainty of syngeneity (Brocks et al., 2003).

The interpretation of Archean and Proterozoic biomarkers is complicated by two factors. One is the fragmentary knowledge of biomarker distributions across the great diversity of extant organisms and the second is the problem of how reasonable it is to extrapolate modern biomarker relationships back in time over several billion years. The biological interpretation of geolipids is almost exclusively based on the distribution of biolipids in extant organisms. However, the full biolipid repertoire is only known for a small percentage of microorganisms that have been cultured (Volkman et al., 1993). It is possible, therefore, that many lipids might have a broader biological distribution and more or less taxonomic value than currently known. More problematic, however, are the uncertainties associated with attributing biological sources to biomarkers from rocks several hundred million to billions of years old. Particular pathways of lipid biosynthesis, such as the modification of sterol side-chains, might have evolved independently in different lineages or could have been passed by lateral gene transfer between lineages. So lipids believed to be diagnostic of extant taxonomic groups could also have been prevalent in extinct clades.