Looking Through Windows Onto The Earliest History Of Life On Earth And Mars


'Department of Earth Sciences, University of Oxford, Parks Road, Oxford, OX' 3PR, UK

2Department of Earth Sciences and Centre for Geobiology, Allegaten 4', Bergen-5007, Norway

Abstract We know that planet Earth is about 4.5 billion years old but what is less clear is when it first became home to life. Locating the first evidence for life on Earth is a question of considerable complexity and controversy. Biogeochemical signals or examples of cellular preservation from the early Archean (greater than ~3 billion years in age) are scarce and vigorously debated. Understanding the relationship between a specific signature in the terrestrial rock record and a specific organism and/or environment is a key issue that astrobiologists must address in order to succeed in any search for life, extinct or extant, on other planets. We here present an overview of putative biogenic signals described from some of Earth's oldest rocks and highlight the most promising areas for future research.

Keywords Abiogenic, Akilia, Archean, Astrobiology/Astrobiologist, Barberton, Biogenic, Carbon, Earth, Greenland, Isotope, Isua, Life, Mars, Microfossil, Microtube, Pilbara, Stromatolite, Sulfur

1. Introduction

The early fossil record here on planet Earth is essential to both designing the search strategy for seeking life elsewhere in our universe and to ratifying the evidence collected in this search. There are two fundamental challenges that face geologists and astrobiologists here. The first involves locating and recognizing preservational windows, that is to say, the types of rocks and fossil evidence needed to uncover the early history of life. The second involves the scientific community reaching a consensus on this evidence for the earliest appearance of life in the rock record. This terrestrial "ground truthing" process as reviewed below should proceed hand-in-hand with the exploration of other planetary bodies.

The search for the earliest life on Earth relies on finding ancient rocks where biosignals may still be preserved - these are our 'windows'. Unfortunately, suitable minimally deformed rocks are relatively rare compared to the Phanerozoic rock record. Although several Archean greenstone belts have been studied worldwide, only two - the Pilbara of Western Australia and the Barberton Mountain Land in southern Africa contain intact stratigraphic piles of early Archean age. The rocks from these two regions are as old as ~3,500 Ma and neither region is highly metamorphosed. Older rock successions are known, some possibly as old as 3,850 Ma, like those of Greenland and Labrador. But these are of higher meta-morphic grade, which makes identification of any putative biogenic signals even more difficult and controversial.

Efforts to find the earliest life in these rocks have been focused on meta-sedimentary rocks that are predominantly chemical in origin, for example cherts and banded-iron-formations and, more recently, on hydrothermal and even volcanic rocks. Detrital meta-sediments, such as sandstones and mudstones that are composed of fragments of older rocks, are less common in the Archean rock record, but may be some of the best preserved windows for investigation, where post-mortem processes which act to destroy biological information may have been minimized.

This chapter aims to provide an overview of putative biogenic structures and chemical traces from the three localities on Earth which preserve meta-sedimen-tary rocks in excess of ~3 billion years old: the Isua greenstone belt and Akilia Island, Greenland; the Pilbara craton of Western Australia; and Barberton mountain land, South Africa/Swaziland.

Two main questions need to be asked of any such reports. First, are the candidate structures or chemical traces truly ancient? In other words, can it be demonstrated beyond doubt that they are both indigenous and syngenetic to the rock in which they are found? Second, is the morphology of the candidate structures, along with associated chemical traces, indicative of biology? For example, do they show evidence of tiers of metabolic processing, and do they occur in a context that is conducive for life? At the same time, can abiogenic mechanisms be discounted for their formation?

Throughout the following discussion we use these and other criteria to evaluate putative biogenic structures or traces and to test whether they can provide reliable evidence for early life on Earth. We aim to present a balanced account of the currently available evidence for early Archean life and to also provide suitable notes of caution for interpreting the earliest candidate biosignatures. (Alternative recent reviews of this topic can be found in Schopf et al., 2007 and Van Kranendonk et al., 2007, the latter focuses more broadly on the early rock record.)

2. Greenland: The Earth's Oldest Supracrustal Rocks

The earliest rock record on Earth is somewhat fragmentary and although minerals have been dated from as far back as 3,910-4,270 Ma, these are single zircons contained within much younger c.3,100 Ma Jack Hills quartzites of Western Australia

(Wilde et al., 2001). We must therefore turn to the Isua Greenstone Belt and Akilia Island in southwest Greenland (Fig. 1) to find the oldest intact supracrus-tal, i.e. volcanic and sedimentary rocks on our planet. These have a minimum age of ~3,700 Ma in Isua (Moorbath et al., 1973; Nutman et al., 1997a) and could be as old as ~3,850 Ma on Akilia Island (Nutman et al., 1997b; Mojzsis and Harrison, 2000). Unfortunately these rocks have been subjected to intense meta-morphism, so any fossilized morphological remains of Earth's earliest biosphere, if any ever existed at that time, have been destroyed by heat and pressure. Instead we must rely on chemical signatures within these rocks that may give clues to the former existence of life. The most widely used chemical signatures are isotopic ratios, in particular carbon isotope fractionations (reported as 513C PDB values) that are widely believed to record the signals of ancient metabolic activity and biological processing (see for example Schidlowski, 2001).

There has been much debate and controversy surrounding the value of the Greenland successions to provide the oldest possible window into earliest life (see for example a recent review by Whitehouse and Fedo, 2007). One such controversy centres on the tiny island of Akilia, just off the coast of southwestern Greenland (Figs. 1c-e) and is here used to highlight the importance of understanding the geological context of these ancient rocks. It begins with a report by Mojzsis et al. (1996) of evidence for life in a >3,850 Ma banded quartz pyroxene rock on Akilia. These authors analysed the carbon isotope signatures of graphite inclusions within ~10 |im3 sized grains of apatite from a small outcrop that had previously been interpreted as a sedimentary banded-iron-formation (BIF) (Nutman et al., 1996). They found a strongly negative mean 513C value of -37 +/- 3% o for this graphitic carbon within the apatite grains. Since organisms prefer to use the light isotope of carbon, Mojzsis et al. went on to state that this "provides evidence for the emergence of life on Earth by at least 3,800 Myr before present'. The key to this statement relies upon two now widely scrutinized assumptions. First, is their interpretation of this rock unit as a primary sedimentary BIF deposit. Crucially, if the rock is indeed a BIF, then the carbonaceous material could preserve a primary biological signal that was shielded from later metamorphism by being trapped within the apatite grains. Second is their interpretation of negative 513C values as an unambiguous fingerprint for metabolic activity in the early biosphere.

Challenges to these assumptions are now legion. A second team of geologists (Fedo and Whitehouse, 2002) examined the Akilia site and determined that the outcrop, which is only a few tens of square meters in total area (Figs. 1d-e), is not a meta-sediment. A re-examination of the field relationships of the outcrop showed the banding to comprise discontinuous boudinage tails caused by multiple, intense deformation events. Furthermore, detailed geochemical data (major elements, trace elements and rare earth elements) from this team have pointed to an ultramafic igneous protolith. An igneous protolith for this rock means that the carbon would have little or no biological relevance; Fedo and Whitehouse (2002) therefore advanced two abiogenic scenarios that could also

Figure 1. (a) Turbidite sedimentary rocks from the -3,700 Ma ISB, West Greenland (Rosing, 1999); (b) photomicrograph of carbon grains within (a) (Rosing, 1999); (c) locality map for the Isua and Akilia localities in West Greenland discussed in the text; (d) sketch from aerial photograph of the location of the quartz pyroxene rock (purported BIF) on the southwest peninsula of Akilia Island; (e) field exposure of the purported -3,850 Ma Akilia Island BIF in contact with surrounding ultramafic rocks. Figure 1b and c are reprinted with permission from Science, Volume 283. Rosing, M. T. 13C Depleted carbon microparticles in >3,700-Ma sea-floor sedimentary rocks from West Greenland, pp. 674-676. Copyright (1999) AAAS.

Figure 1. (a) Turbidite sedimentary rocks from the -3,700 Ma ISB, West Greenland (Rosing, 1999); (b) photomicrograph of carbon grains within (a) (Rosing, 1999); (c) locality map for the Isua and Akilia localities in West Greenland discussed in the text; (d) sketch from aerial photograph of the location of the quartz pyroxene rock (purported BIF) on the southwest peninsula of Akilia Island; (e) field exposure of the purported -3,850 Ma Akilia Island BIF in contact with surrounding ultramafic rocks. Figure 1b and c are reprinted with permission from Science, Volume 283. Rosing, M. T. 13C Depleted carbon microparticles in >3,700-Ma sea-floor sedimentary rocks from West Greenland, pp. 674-676. Copyright (1999) AAAS.

explain the strongly negative 513C signal recorded in this outcrop. One involves the decarbonation of metasomatized ultramafic rock, and the other serpentiniza-tion of an olivine-bearing ultramafic protolith followed by metamorphic decar-bonation.

The story took a further twist when Dauphas et al. (2004) came out in support of a sedimentary BIF origin for this deposit. They used a new approach involving iron isotopes and found that the Akilia rocks were enriched in the heavy iron isotope when compared to igneous rocks (see also a later paper that again uses Fe isotopes coupled to a greater array of additional geochemical traces, Dauphas et al., 2007). This, they argued, was best interpreted as evidence for the transport, oxidation and precipitation of ferrous iron from hydrothermal vents and thus consistent with these rocks indeed being the oldest water lain sedimentary deposit preserved on Earth. However, this finding was overtaken by yet another twist in this controversy when Lepland et al. (2005) re-examined 17 apatite samples from Akilia, including the sample originally used in the Mojzsis et al. study. These authors and another independent team (Nutman and Friend, 2006) failed to find graphite inclusions in any of the apatite grains raising the possibility that the original findings of Mojzsis et al. (1996) were an artefact.

The final convolution in this discussion to date came when some of the authors of the original studies replied to their critics in a pair of papers (Manning et al., 2006; McKeegan et al., 2007). In the first of these, new mapping, geo-chronology and geochemistry data are presented which support the originally reported >3,850 Ma age for these rocks, although ambiguity in the cross-cutting relationships due to the complex deformation history of the Akilia outcrops still leaves some doubts over these age constraints in some quarters (e.g. Eiler, 2007). The second paper uses laser Raman micro-spectroscopy to confirm the presence of graphitic inclusions in the apatite grains together with ion microprobe analysis to corroborate the original negative 513C measurement. It remains to be explained why two independent teams did not find these inclusions (Lepland et al., 2005; Nutman and Friend, 2006), and perhaps it is the case that these inclusions are much less abundant than was originally contended. Notwithstanding, these two recent studies have renewed hopes in some quarters that the Akilia rocks may after all hold evidence pertinent to the origins of life on Earth. However, considerable ambiguities do remain, especially regarding the exact nature of the proto-lith and the complex metamorphic and metasomatic history of the Akilia rocks, as explained in a recent review paper by Eiler (2007).

A more promising location to search for ancient biomarkers appears to be a low strain domain in the north east of the Isua greenstone belt where primary igneous and sedimentary fabrics are in part preserved. Researchers have for sometime now argued for a biogenic component to graphitic carbon in this area (e.g. Hayes et al., 1983). In addition to the Akilia Island sample, Mojzsis et al. (1996) also investigated apatite grains at a second locality, in an iron carbonate-rich rock from this low strain area. They again argued that these apatite grains contained carbonaceous inclusions whose isotopic values of 513C = -30 ± 3%c are consistent with a biogenic origin. Once again, however, more recent work has cast doubt upon their claims. Lepland et al. (2002) investigated apatite grains from eight different meta-sedimentary samples and discovered that they were entirely free from graphite inclusions. They did, however, find abundant graphite inclusions in associated meta-carbonate rocks, formed by metasomatic processes; detailed petrography and rare earth element geochemistry led them to re-interpret the original Mojzsis et al. sample as metasomatic in origin. If the graphite in these metacarbonate rocks was formed via the reduction of carbonate ions during thermal decomposition of iron-carbonate as proposed by Van Zuilen et al. (2002, 2003), then it cannot be biogenic.

A third promising claim for life in these Greenland rocks comes from graphite globules in a meta-sedimentary unit within the low strain area of Isua (Figs. 1a-c; Rosing, 1999; Ueno et al., 2002). Trails of graphitic globules within some garnet and biotite metamorphic porphyroblasts suggest a primary origin for this graphite, while a 513C value of ~ -19%o led Rosing (1999) to conclude that the carbon is biogenic in origin. As ever, extreme care must be taken to exclude the possibility of more recent contamination, especially given that modern endolithic coccoids have been found in cracks within BIFs of the Isua greenstone belt (Westall and Folk, 2003), and experiments have indicated post-metamorphic addition of organic matter to some Isua samples (Van Zuilen et al., 2002). The Rosing (1999) turbidite sample, however, was shown by Raman spectroscopy to contain well crystallized graphite that is markedly different from post-metamorphic contamination (Van Zuilen, 2005, Fig. 5b), and thus the original claim still stands.

The debate surrounding the ~3.8 billion year old rocks from Greenland is ongoing and, despite the great challenges associated with the Akilia and Isua areas, these rocks are unique in representing the period in Earth history where geological processes as we know them today may first be recognizable and where conditions for the emergence of life may have become tolerable. The fact that unambiguous signals for life have not been forthcoming, as yet, should not deter further detailed study into this earliest window into potential life on Earth.

3. East Pilbara Granite-Greenstone Terrane

The Pilbara craton of Western Australia (Fig. 2a) is composed of three ancient granite greenstone terranes; East Pilbara, West Pilbara and Kurrana. The oldest rocks are exposed in the East Pilbara where the lowermost group is the Warrawoona Group (Fig. 2b), deposited from 3,515-3,420 Ma. This consists mostly of mafic volcanic rocks of the Double Bar Formation, Table Top Formation, North Star Basalt, Mt Ada Basalt, and Apex Basalt.

These are interspersed with thin chert horizons and felsic volcanics of the 3,515-3,500 Ma Coucal Formation, the 3,472-3,465 Ma Duffer Formation, the ~3,460 Ma 'Apex chert' and the 3,458-3,427 Ma Panorama Formation (Van Kranendonk, 2006). The Kelly Group lies unconformably above these, consisting of the >3,350 but <3,427 Ma Strelley Pool Chert, the 3,350-3,325 Ma Euro Basalt and the 3,325-3,315 Ma Wyman Formations. This pile, in turn, is unconformably overlain by the ~3,240 Ma Sulphur Springs Group.

These lower three groups of the Pilbara Supergroup house some of the Earth's oldest purported stromatolites, microfossils and microtubes (Fig. 2), as outlined below. (A recent alternative review of these rocks is available in Van Kranendonk, 2007.)

Was this article helpful?

0 0

Post a comment