Australia: The Land Where Time Began

A biography of the Australian continent 

Matworld – the Biogeochemical Effects of Early Life on Land

Evidence has been growing that life has been on land for billions of years. According to Lenton & Daines microbial mats that were fuelled by oxygenic photosynthesis were probably present in terrestrial habitats from about 3 billion  years ago (Ga) onwards, forming localised ‘oxygen oases’ under an atmosphere that was reducing, which left behind an oxidative weathering signal. Following the Great Oxidation about 2.4 Ga the redox signal was masked by the now oxidising atmosphere, through the mobilisation of phosphorus and other elements by organic acids in weathering profiles. Evidence for ‘greening of the land’ and intensification of weathering in the Neoproterozoic about 0.85-0.54 Ga is equivocal at present. The mid-Palaeozoic about 0.45-0.4 Ga shows, however, global atmospheric changes that are consistent with increased terrestrial productivity and intensified weathering by the first land plants.

It is often assumed that the origin and evolution of life occurred in the oceans, not least because this is where most of the fossil and geochemical record is preserved. However, several lines of evidence have been provided by recent work in support of early land biota (Stueken et al., 2012; Lalonde & Konhauser, 2015; Wellman & Strother, 2015). According to Lenton & Daines the palaeontological evidence has been expertly reviewed recently (Wellman & Strother, 2015) so in this paper the focus is on the biogeochemical evidence for early land biota and corresponding models, which suggests they had significant impacts on the Earth system long before the rise of land plants.

The earliest evidence of life has been found in marine sediments >3.7 Ga which is suggestive of photosynthetic carbon fixation (Ohtomo et al., 2014). The original colonisers of land are believed to probably have been prokaryotic microbial mats in the Archaean Eon (~3.8-2.5 Ga) living under a chemically reducing atmosphere. Lenton & Daines suggest the first mats were conceivably powered by anoxygenic photosynthesis, by using gaseous electron donors such as H₂ or H₂S (or dissolved Fe²⁺ on rocks that were suitably rich in iron). Even with localised recycling, however, their productivity would be limited severely by the supply of electron donors, thereby limiting their potential to leave a signature in the geological record. Lenton & Daines therefore focus on mats that were powered by oxygenic photosynthesis (i.e. with water acting of the electron donor) (Herman & Kump, 2205). Such mats would initially have been dominated by cyanobacteria. They probably gained eukaryotic algae and fungi at some time during the Proterozoic Eon (2.5-0.54 Ga). Eventually, in the Middle Palaeozoic (~0.47-0.45 Ga), nonvascular plants (e.g. mosses, liverworts) were added.

At present a mixture of cyanobacteria, algae, fungi, lichens and non-vascular plants are found in terrestrial mats, that are often termed ‘biological soil crusts’ or ‘cryptogamic cover’(Elbert et al., 2012). Typically, the constituents of such mats have the ecophysiological features that they are poikilohydric and can tolerate desiccation (which is not possible for vascular plants), though they have a limited vertical extent, with the result that they can be outcompeted for light by taller vascular plants. The niche for such mats in the modern world is therefore in arid environment where vascular plants that are homiohydric struggle to maintain internal water pressure. At the other end of the hydrological scale, however, benthic microbial mats colonise environments where grazing is inhibited, which include marine intertidal zones and hypersaline lakes.

In this paper, Lenton & Daines first outline the effects of the ecosystem engineering of mats of the present. Then they consider the biogeochemical effects of ancient terrestrial mats in the 3 key phases in the geological record:

1)    Under a reducing atmosphere in the Archean >2.4 Ga;

2)    Following the Great Oxidation Event ~2.4-0.85 Ga;

3)    In the transition from the Neoproterozoic to the Palaeozoic in the modern world ~0.85-0.4 Ga.

Mats as ecosystem engineers

Cryptogamic cover of the present contributes an estimated 7% of global terrestrial net primary productivity (NPP, ~3.9 PgC/yr), provides almost half of the terrestrial nitrogen fixation (49 TgN/yr; Elbert et al., 2012), emits significant fluxes of nitrous oxide and methane (Lenhart et al., 2015), and has considerable potential for weathering (Porada et al., 2014), all this in spite of being restricted to marginal environments. Extensive vascular plant cover of the present results in the absence of good modern analogues of terrestrial mats in less extreme hydrological environments (i.e., the majority of the land surface). Still, the effects of ecosystem engineering of extant mats provide important clues to the signatures they could have left in the geological record.

Sediments are stabilised by mats and they alter surface properties. The surface albedo can be lowered by cyanobacteria, as well as increasing the temperatures by up to 10^oC through the production of ‘sunscreen’ pigments such as scytomenin (Couradeau et al., 2016). Soil aggregates are formed by cyanobacteria through the production of extracellular polysaccharides (Mager & Thomas, 2011), and filaments that may self-assemble into supracellular ropes (Garcia-Pichel & Wojciechowski, 2009), to form an effective seal that alters the topography of the surface, suppresses wind erosion and water erosion and is expected to enhance weathering. Therefore, mats can leave distinctive structures in the sedimentary record, notable the ‘Roll up’ of eroded mats (Beraldi-Campesi et al., 2014).

Considerable local redox gradients are produced by mats. Typically, the active photosynthetic zone is below the ground, which makes NPP difficult to assess by remote sensing (Raanan et al., 2016), with a mix of sand grains and biological compounds which provide surface protection from ultraviolet (UV) radiation (Rastogi et al., 2014). Cyanobacteria display a dynamic (about 1 hour) response to intermittent precipitation events, with photosynthesis and gas exchange that is limited in the active water saturated state that is comparable to that in aquatic benthic mats (Rajeev et al., 2013). At depths of about 1 mm within such mats supersaturated oxygen oases (≥500 μM) can build up during the day, which drives an O₂ flux to the overlying atmosphere (or water column) and downwards to respire organic carbon that has accumulated (Raanan et al.,  2016). At night a corresponding excess of reductants can be converted to, e.g., methane or organic acids and (in a modern high oxygen environment) consumed locally.

Mats access nutrients that are rock-bound vie enhancement of weathering. Bacteria (Styriaková et al., 2012), lichens (Jackson, 2015), mosses (Lenton et al., 2015) have all been measured to enhance weathering. Only a modest flux of organic carbon in soils was downwards as dissolved organic matter and subsequent to respiration, is needed to support an appreciable concentration of CO₂ in the vadose (aeriated) zone beneath the mat cover, which enhances weathering by carbonic acid (Keller & Wood, 1993). This introduction of carbon that is isotopically light can have an isotope signature that is stable where secondary carbonate forms in karst environments (Kenny & Knauth, 2001). A characteristic geochemical signature of mobilisation of P, Fe and Cu in palaeosols (ancient soils), can be left by the production of organic acids and chelating agents, with the release of Fe and Cu also sensitive to underlying lithology (basalt or granite basement rocks) and _PO₂ values (Neaman et al., 2005a,b).

According to Lenton & Daines it is probable that the formation of soil and the cycling of terrestrial nutrients functioned very differently before vascular plants evolved. It was demonstrated by field measurements that a high level of recycling of P (phosphorus) in modern ecosystems (about 50:1 ratio of P recycled flux to P input/output flux at steady state) as a result of deep vascular rooting systems and associated mycorrhiza, though much lower values of (About 5:1 for lichens on rocks and desert crusts (Porada et al., 2016). Early mat-based ecosystems that were confined to an active layer in the upper few mm, could have stabilised soils but become decoupled from a deeper weathering zone over much of the surface of the land, therefore becoming nutrient limited. Later fungal symbiosis, however, including the formation of lichens and mycorrhizas (Field et al., 2015), providing transfer of nutrients over greater distances would have, at least partially, alleviated this (noting the fossil evidence for lichens begins only about 415 Ma; Honegger et al., 2013).

Lenton & Daines suggest global effects of early mats (dependent on their productivity) could have included drawdown of CO₂ and global temperature by enhancing the weathering of silicate rock (Schwartzman & Volk, 1989; Keller & Wood, 1993; Boucot & Gray, 2001), and increasing the atmospheric content of oxygen through the enhancement of its long-term source from the burial and preservation of organic matter in sedimentary rocks (Lenton & Watson, 2004; Kennedy et al., 2006; Lenton et al., 2016), and suppressing its long-term sink, under an oxidising atmosphere, from the oxidation of ancient organic matter in soil profiles (Kump, 2014). Among specific mechanisms for increasing the oxygen source are:

i)                   enhancing the weathering of P, and thus the production of organic carbon (Lenton & Watson, 2004),

ii)                the production of clay particles that preserve organic carbon (Kennedy et al., 2006),

iii)              and increasing the C:P ratio of buried organic matter (Lenton et al., 2016).

Organic carbon that was terrestrially produced could have been preserved in soils or buried in lakes (Spinks et al., 2014), though it is expected that erosional transport and burial in shallow marine environments are expected to have been the major pathway and are difficult to distinguish from local marine production (until the evolution of high C:P recalcitrant structural compounds.

Archaean oxygen oases

The atmosphere was of a chemically reducing type with _PO₂<10^-5 present atmospheric (PAL) and no ozone layer, as is indicated by sedimentary preservation fractionation of sulphur isotopes that are mass independent (MIF-S) that is caused by UV radiation. According to Lenton & Daines there are, perversely, geochemical signals of biogenic oxygen production that have been recorded in marine sediments that were deposited about 3.0 Ga onwards, in the form of oxidative mobilisation of sulphur, selenium and trace metals (Mo, Re, Cr) and corresponding isotope (e.g. δ⁹⁸Mo) fractionation (Lyons et al., 2014).

Such signals cannot readily be explained by oxidative weathering that is driven by atmospheric oxygen because at _PO₂<10^-5 PAL it would require very slow transit through weathering environments and down rivers in order to achieve any significant mobilisation of redox-sensitive species (Johnson et al., 2014). It may, instead, be explained best by the presence of cyanobacterial mats on the land surface, or in freshwaters, forming ‘oxygen oases’ – that is disequilibrium high concentrations of O₂, as a result of restricted exchange rates with the atmosphere (Herman & Kump, 2005; Stueken et al., 2012; Lalonde & Konhauser, 2015).

It has been suggested by molecular phylogenetic studies that a freshwater and benthic origin for cyanobacteria (Blank & Sánchez-Baracaldo, 2009), which is consistent with microbial mats in a 2.72 Ga lake (Buick, 1992). It is conceivable that such mats in freshwater, or shelf seas, could produce putative signatures of oxidative weathering. The question then becomes: is there any evidence in support of subaerial mats? It has been assumed in many studies that UV levels were prohibitive to subaerial life in the absence of an ozone layer (instead requiring about 10 m of water depth in order to attenuate UV (Berkner & Marshall, 1965). Prokaryotes have, however, many effective strategies for screening themselves from UV (Rastogi et al., 2014). Also, in the chemically reducing atmosphere there would be a photochemical haze of organic compounds that would have provided some degree of UV shielding of the surface (Zerkle et al., 2012).

The presence of subaerial mats is suggested by a series of palaeosols dating to about 3.0 Ga to the Great Oxidation about 2.4 Ga. Palaeosol indicators of oxidative weathering are often attributed to elevated atmospheric oxygen partial pressures (Crowe et al., 2013; Mukhopadhyay et al., 2014), though they are more consistently explained by localised oxygen oases that are formed within mats (Lalonde & Konhauser, 2015). P and Fe depletion are often shown by palaeosols from 2.76 Ga onwards, which is consistent with the presence of organic acids and ligands (Neaman et al., 2005a; Driese & Medaris, 2008; Driese et al., 2011). Also, very isotopically light organic carbon in a palaeosol that dates to about 2.76 Ga indicates the presence of metanotrophs (which consume methane and oxygen) (Rye & Holland, 2000). A palaeosol that is rich in organic carbon with the isotopic signature of photosynthesis directly indicates local NPP (Watanabe et al., 2000).

Proterozoic evidence

The atmosphere became chemically oxidising after the Great Oxidation of about 2.4 Ga with atmospheric oxygen concentration of about 10^-2- 1^-1 PAL, and thereby masking the localised redox signature of terrestrial mats by supporting oxidation of the surface everywhere. The chemical weathering signature of terrestrial mats can still be seen, nevertheless, and is augmented by sedimentological and microfossil evidence.

From about 2.4 Ga onwards palaeosols continue to show surface depletion of phosphorus, and sometimes deep deposition of phosphorus, which is consistent with the dissolution of apatite by organic acids (Prasad & Roscoe, 1996; Driese & Medaris, 2008). With the atmosphere that was now oxidising the redox signature of these palaeosols changed somewhat, now being able to retain iron, at least in palaeosols that are based on basalt basement rock (Neaman et al., 2005,a b). This atmospheric effect competed with the mobilisation of iron that was caused by organic acids, sometimes producing depletion of Fe near the surface of the soil and retention of Fe below (Beukes et al., 2002). It is remarkable that weathering was so intense that laterites sometimes formed, e.g. in the Hekpoort palaeosol from about 2.2 Ga, which suggests the presence of substantial surface biological productivity in tropical climates (Beukes et al., 2002).

Extremely rounded, intensely weathered fine sandstone sediments (quartz arenites), that are found in the Proterozoic (and early Palaeozoic) were interpreted as being a result of surface stabilisation by mats (Dott, 2003). The Makabeng Formation, a dryland system from 2.06-1.88 Ga, shows ‘roll-up’ and other sedimentological structures that are indicative of microbial mats (Eriksson et al., 2000; Simpson et al., 2013). Sedimentary structures that were induced by microbes by 1.2-1.0 Ga, are preserved from a wet environment that was subject to periodic drying (Torridonian succession) (Prave, 2002; Strother & Wellman, 2016). Eukaryotic fossils, which include multicellular forms, have also been yielded by this succession from freshwater and habitats that are subaerially exposed (Strother et al., 2011). Meanwhile, filamentous microfossils from 1.1 to 0.8 Ga were interpreted as terrestrial cyanobacteria (Horodyski & Knauth, 1994). Therefore, eukaryotes were probably present alongside cyanobacteria in terrestrial mats; though it is not clear whether these were algae.

Neoproterozoic greening?

A greening of the land surface during the Neoproterozoic Era (1,0 Ga-54 Ma), has been invoked by several studies particularly as a mechanism to explain the possible rise of oxygen at this time. Yet there were already terrestrial ecosystem that were sufficiently productive to produce organic-rich lacustrine shales about 1.2-1.0 Ga (Spinks et al., 2014), and isotopically light secondary terrestrial carbonates (palaeokarst) 1.1 and 0.8 Ga (Kenny & Knauth, 2001).

Lenton & Daines originally hypothesised that both planetary cooling and a rise in atmospheric oxygen during the Neoproterozoic could be explained by an increase in productivity that drove silicate and phosphorus weathering (Lenton & Watson, 2004). There is some support for increased weathering of phosphorus (Planavsky et al., 2010). It is argued by a variant hypothesis that an increase that is biologically driven in the intensity of chemical weathering that is reflected in the production of pedogenic clay minerals, which provide microsites for the burial of organic carbon in marine sediments, thereby driving a rise in atmospheric oxygen (Kennedy et al., 2006). Careful analysis of the clay content of organic carbon-rich shales, however, shows a sharp decrease (rather than an increase) in the pedogenic clay content in the Neoproterozoic (Tosca et al., 2010). A separate argument for land greening is that isotopically light marine carbonate from about 850 Ma onwards reflect the input of respired terrestrial photosynthetic carbon in groundwater (Knauth & Kennedy, 2009; Kump, 2014). At least some carbonates from the Neoproterozoic record, however, a primary marine (i.e. not groundwater-altered) δ¹³C signature (Lu et al., 2013).

Palaeosols from the Neoproterozoic and Cambrian continue to show element loss profiles that are consistent with intense weathering by organic acids (Retallack & Mindszentry, 1994; Driese et al., 2007). It is intriguing that a palaeosol from the Cambrian has shown patterns of apatite removal and the formation of clay in the top 30 cm that was interpreted as an indication of mycorrhizal fungi (Horodyskyj et al., 2012). 

Evidence for a major phase of land colonisation in the Neoproterozoic is currently equivocal, whereas the cryptospore record of the rise in the Middle to Late Ordovician of vascular plants has been widely accepted (Wellman & Struthers, 2015). It is suggested by ecophysiology modelling of global cryptogamic cover that under the higher concentration of atmospheric CO₂ and associated climate simulated for the Late Ordovician, global NPP could have been near about 25% to 30% of plant cover dominated by vascular plants (Lenton et al., 2016; Porada et al., 2016). It is further predicted by modelling of long-term cycling of carbon that this would have lowered atmospheric _PCO₂, which is consistent with observed global cooling (Lenton et al., 2012) and ocean-atmosphere oxygenation (Lenton et al., 2016), which included the appearance of fossil charcoal, implying that _pCO₂>15% by volume from about 420-400 Ma (Glasspool & Scott, 2010).

Conclusions

Though all the phases of land colonisation cannot yet be tied down, a mechanistic framework has been provided by recent work – informed by ecophysiology and ecosystem engineering effects of extant terrestrial mats – which can make predictions that are testable for the effects on the sedimentary record of early land biota. A consistent picture is beginning to be yielded by combining Earth system modelling and evidence from diverse geochemical indicators of both local and global biogeochemical effects of early life on land. From about 3.0 Ga onwards it is strongly suggested by oxidative weathering, acid weathering and carbon isotope signals in ancient soils, that subaerial microbial mats that were fuelled by oxygenic photosynthesis were present. Following the Great Oxidation event that occurred about 2.4 Ga the redox signal of terrestrial mats was masked by the oxidising atmosphere that was present at this time, though their organic acid weathering signal remains in ancient soil profiles. By about 2.0 Ga mats in drylands are revealed by sedimentary structures. There is microfossil evidence by 1.0 Ga of eukaryotes entering terrestrial ecosystems, which were producing and burying significant amounts of organic carbon. There is only mixed support from the available evidence for hypotheses of an increase in terrestrial productivity and associated weathering during the Neoproterozoic about 0.85-0.54Ga. The well-established rise in the Middle Palaeozoic of the first land plants, in contrast, is accompanied by global atmospheric changes which are consistent with terrestrial productivity and intensified weathering. Lenton & Daines suggest that future work can progress from establishing the existence of early terrestrial biota – which is now considered incontrovertible – to trying to quantify better their productivity and the nature, timing and scale of their biogeochemical effects.

Sources & Further reading

Lenton, T. M. and S. J. Daines (2017). "Matworld – the biogeochemical effects of early life on land." New Phytologist 215(2): 531-537.