Earth-logs

From an empirical standpoint the mass extinction at the close of the Palaeozoic, 251 Ma ago, links closely with eruption of the largest known flood basalt pile in Siberia, and there is no known extraterrestrial impact that tallies. So it seems likely that the P-T event was generated by the influence of a mighty mantle plume on surface conditions. Careful statistical analysis of the marine faunas that preceded and followed the event give some clues to geochemical conditions associated with the extinctions and slow Triassic recovery of animal diversity (Bottjjer, D.J. et al. 2008. Understanding mechanisms for the end-Permian mass extinction and the protracted Early Triassic aftermath and recovery. GSA Today, v. 18, September 2008 issue, p.4-10). Brachiopods, bivalves and bryozoans, in terms of their respective diversities and abundance relative to one another, changed markedly. On the late Permian seafloor, brachiopods and bryozoa fell in both measures, whereas bivalves exploded in numbers but became dominated by just 4 genera. This ecological lop-sidedness continued in the early Triassic. Such an oddity in itself suggests that some kind of geochemical stress was present in marine environments for a protracted period of time. The most likely stressful agents are increased CO₂ and H₂S, and decreased oxygen. The faunal review goes on to discuss the need for experimental manipulation of oxygen, carbon dioxide and hydrogen sulfide concentrations to see the effects on modern organisms.

Another approach to the issue of the P-T event is to model the conditions that may have led to anoxia linked with increased CO₂ and H₂S (Meyer, K.M. et al. 2008. Biochemical controls on photic-zone euxinia during the end-Permian mass extinction. Geology, v. 36, p. 747-750). The authors use an Earth system model of ocean circulation coupled with one for the distribution of atmospheric moisture when the continents were assembled into the Pangaea supercontinent. Chemical constraints were 12 times the current carbon dioxide content for the atmosphere and about one fifth of its present oxygen content (see New twist for end-Permian extinctions in the May 2005 issue of EPN), roughly those accepted for the time. The supply of phosphate to the oceans was varied up to 10 times present values. Specifically, the model examined the likely effects of such conditions on the likelihood of hydrogen sulfide production in the oceans and its transfer to the uppermost ocean water. Increasing supply of phosphate inexorably drives global near-surface conditions towards anoxia and H₂S – rich conditions. Even adding sulfide-oxidising bacteria to the surface waters doesn’t prevent runaway toxicity, including export of hydrogen sulfide to the atmosphere that would drive many land animals to extinction. It is hard to think of a more pervasive and effective ‘kill mechanism’, nor one that would have lingered for longer, thereby satisfying the evidence, including the extremely long biota recovery time during the Triassic. The two accounts taken together cast doubt on a determining role for the Siberian flood basalts, which were relatively short-lived, although volcanic emissions of CO.₂ and SO₂ may have placed a chemical ‘last straw’ on already stressed organisms.

Plant evolution summarised

Papers on palaeobotany, especially the evolution of plants are a lot less frequent than those on many other broad geoscience topics. So to see a review is welcome (O’Donoghue, J. 2008. Petal Power. New Scientist, v. 200 1 November 2008 issue, p. 36-39). O’Donoghue summarises recent publications on the rise of the angiosperms – Darwin’s “abominable mystery” – since the Jurassic. Accepted wisdom has long been that the earliest flowering plants were akin to modern magnolias that seem anatomically primitive. That assumption has much to answer for, because palaeobotanists sought evidence for big, simple flowers. It was a piece of pure luck that resolved the issue, in the form of fossilised debris from an 83 Ma wildfire found in Sweden. The carbon-rich clay contained masses of flowers only a few mm across, which resemble those of walnuts, plane trees and saxifrages. A shift in focus to minute blooms enabled Chinese geologists working on evidence for the habitat of the famous feathered dinosaurs to find the earliest flowers yet in an early Cretaceous (125 Ma) lagerstätte. They are humble indeed, resembling duckweed. Pushing back further the time of separation of the angiosperms from a presumed gymnosperm (cycads, ginkgoes and conifers) has depended on molecular evidence from living primitive flowering plants, and came up with a humble shrub from New Caledonia (Amborella), the srat anise plant (a member of the Austrobaileyales group) and water lilies. A molecular-clock approach suggests an evolutionary jump from gymnosperms took place as far back as the early Jurassic. The peculiar means of sexual reproduction evolved by angiosperms – giving animals a ‘free lunch’ with the perk to plants of their carrying pollen – cut the amount of energy involved in reproduction by massive pollen release for wind fertilisation and production of seeds without guaranteed fertility used by gymnosperms. In turn is resulted in a massive adaptive radiation by insects in particular, seeing the bees evolve from predatory wasps in early Cretaceous times. Now, to a very large extent, angiosperms are dependent on the humble bee. An alarming fact since bee diseases and parasites are currently getting the upper hand, while we become ever dependent on food sources from a  dominantly angiosperm crops.

The strange case of the line-dancing arthropods

Lagerstätten – sites of extraordinarily good fossil preservation – generally throw up surprises and oddities, and those of Cambrian age in China are no exception. Cambrian arthropods, notably the trilobites but also shrimp-like creatures, are not uncommon in them. But any animals that appear to have been engaged in communal activities are cause for both a double-take and a short communication (Hou, X-G. et al. 2008. Collective behaviour in an Early Cambrian arthropod. Science, v. 322, p. 224). About 22 groupings of shrimp-like fossils show individuals linked in ‘nose-to-tail’ chains, the tail (telson) of one in front being lodged in the carapace of that behind. Not only that, but the chains are meandering. ‘Follow-my-leader’ behaviour is seen in modern lobsters bent on migration; perhaps the inspiration for Lewis Carroll’s Lobster Quadrille in Alice in Wonderland. Since no modern arthropods link in such chains for reproductive purposes, and mouth-clenching a partner’s tail is not good evidence for feeding behaviour, the authors’ conclusion is that indeed the diminutive and very ancient creatures were probably hooked-up to go somewhere more conducive to their habits.

Evidence for earliest photosynthesisers takes a knock

The first tangible and isotopic evidence for the permanent presence of oxygen in the Earth’s atmosphere appears in sedimentary rocks dated at about 2.4 Ga. From that we can surmise that some organisms had previously evolved the photosynthetic trick of breaking the hydrogen-oxygen bonds in water: nothing else is known in nature to produce free oxygen on a planetary scale. Frustratingly, the earliest undisputed fossils of such organisms – blue-green bacteria – are a lot younger; around 2 Ga. Structures in sedimentary rocks back to 3.5 Ga, such as stromatolites, which do look a lot like products of living cyanobacteria and may have a biogenic origin, do not contain cellular structures that would constitute proof. So a report in the late 1990s of organic-chemical evidence for cyanobacteria from 2.7 Ga old sediments was greeted with some relief. These oldest biomarkers also included compounds characteristic of eukaryotes; an even more astonishing outcome, given that the oldest undoubted eukaryote fossils are from 1.5 Ga sediments. The ancient biomarkers have been much celebrated, but there is a problem: if cyanobacteria were around at 2.7 Ga in sufficient amounts for their biomarkers to be preserved, how come it took 300 Ma for oxygen to build up in the atmosphere? A novel technology for geochemists has been applied to resolve the issue of the Archaean biomarkers (Rassmussen, B. et al. 2008. Reassessing the first appearance of eukaryotes and cyanobacteria. Nature, v. 455, p. 1101-1104). One of the co-authors, Jochen Brocks of the Australian National University, was an originator of the study on biomarkers, so clearly the new technology has thrown matters into considerable disarray. The oily biomarkers accompany solid kerogen in the late Archaean sediments, in microscopic amounts. Ion-probe mass spectrometry with a 50 nm resolution has provided carbon-isotope measurements of minute samples of several kinds of hydrocarbon in thin sections. These show, with little room for doubt, that the organic compounds thought to have been biomarkers for cyanobacteria and eukaryotes formed by ‘cracking’ of kerogen during thermal metamorphism at about 2.2 Ga. Any other claims based on supposedly specific biomarkers are likely to be ‘tarred with the same brush’. How annoying: complex life clearly was around before 2.4 Ga, some of capable of photosynthesis, but that conjecture cannot be proven!

Much ingenuity has been harnessed to design robotic geochemistry that will be aimed at the popular topic of ‘Life on Mars’ in the coming decades. It would be no surprise if biomarkers are targeted. Yet it is entirely possible that hydrocarbons of inorganic origin can yield such compounds, given some geothermal heating…

See also: Fischer, W.W. 2008. Life before the rise of oxygen. Nature, v. 455, p. 1051-2.

The Ediacaran fauna of the late Neoproterozoic (occurring between 575-543 Ma) marks the first clear sign of animal life, although the affinities of many of the taxa are obscure. ‘Molecular clocks’ based on differences between the DNA of living organisms seems to suggest a last common ancestor of all animals somewhat earlier than the Ediacaran period, perhaps as early as 1000 Ma. Whatever that first animal was, its emergence and that of the Ediacarans took place in climatically and chemically peculiar times. The Neoproterozoic was marked by at least three glacial epochs that left traces at palaeolatitudes as low as the tropics: so-called ‘Snowball Earth’ events. It also contains the most erratic swings in carbon isotopes that are known from the geological record, which have something to do with ups and downs of life at the time, probably variations in global biomass and/or the rate at which organic carbon was buried in seafloor sediments. Among Neoproterozoic sediments two are outstanding: graphitic and sulfidic mudrocks; banded iron formations (BIFs) which are sulfur-poor. BIFs of that age have been an enigma, the most massive and long-lived being those in the Palaeoproterozoic (before 1.8 Ga) and the Archaean. Neoproterozoic BIFs seem to mark the return after a billion years of most peculiar ocean chemistry, when soluble iron(II) ions were abundant at all depths in the ocean yet were oxidised to insoluble iron(III) at the sites where Fe₂O₃ was deposited in huge amounts. In the earlier BIF period that had to have been where oxygen was being locally emitted by primitive blue-green bacterial photosynthesisers, i.e. in shallow water. We must surmise that occurred again in the Neoproterozoic, although the source of oxygen would then have included more advanced oxygenic photosynthesisers. But that is not the puzzle. How did ocean-wide conditions return to allow the abundance of dissolved iron(II) ions and why did those conditions not prevail in the BIF-less billion years?

Donald Canfield of the University of Southern Denmark has long been immersed in issues of ocean-chemistry evolution in relation to atmospheric oxygen levels, and offered an answer to the second question that has largely replaced the once accepted wisdom that ocean water became oxygenated throughout after 1.8 Ga thereby allowing iron to enter oxidised minerals immediately it emerged in ocean-floor basalts magmas. Instead, he suggested that the deep ocean, at least, contained abundant hydrogen sulfide as witnessed by sulfur isotope patterns in marine sediments. In other words, oceanic Fe(II) was efficiently precipitated through the Mesoproterozoic in the form of sulfides. The H₂S was probably generated by bacterial reduction of sulfate ions, themselves derived by oxidation of on-land exposures of sulfidic rocks because of low but increasing atmospheric oxygen. Canfield and a rich variety of international colleagues once again has an authoritative say, this time as regards the Neoproterozoic iron formations (Canfield, D.E. et al. 2008. Ferruginous conditions dominated later Neoproterozoic deep-water chemistry. Science, v. 321, p. 949-952).

If the supply of sulfate from the continents waned, then bacterial production of sulfides would follow suit in sulfur-poor oceans. Provided deep-ocean oxygen levels remained very low, iron(II) derived from continually generated ocean-floor basalts and their hydrothermal alteration could once again pervade the oceans. Oxygen in shallow water would again encourage precipitation of hematite and BIFs. This hypothesis does not need a special explanation for fully oxygenated Precambrian oceans reverting back to anoxia in the Neoproterozoic and then back and forth in their oxygen concentrations to explain short BIF episodes, merely variations in the supply of sulfate from weathered continental surfaces. Canfield et al. tested this hypothesis by examining the proportions of total iron in 800-530Ma sediments contained by minerals able to react easily with their environment, such as sulfides and carbonates, and the proportions of such reactive iron in sulfide minerals. In modern oxygenated waters the proportion of such reactive iron in sediments does not rise above about 40%, and is often lower. In the Neoproterozoic samples, shallow marine rocks obeyed the modern <40% rule, but those from intermediate to deep-water settings (below storm-wave base) sometimes show far higher values. That is a clear signature of anoxic waters, and it persists into the Cambrian. Interestingly, many deep-water sediments from the Ediacaran Period do show signs of oxygenation, while others were anoxic. Among the sediments deposited under anoxic conditions none have iron sulfide proportions as high as those produced in modern euxinic basins such as the Black Sea, thereby signalling a dearth of bacterially generated H₂S and low sulfate supply to the oceans as predicted. But why did the supply dry up? One possibility is that chemical weathering on the continents plummeted during ‘Snowball Earth’ episodes. Yet the evidence for anoxic, high iron(II) conditions in the oceans persisted well beyond the times of the known glacial epochs. Another plausible explanation is pyrite burial, analogous to that of carbon, and subduction of sulfide-rich sediments that progressively completely stripped the oceans of sulfate. What of the effect on early animal life? Iron is an essential micronutrient, much touted today as a means of encouraging phytoplankton blooms in ocean surface water. Together with rising shallow-water oxygen levels, perhaps an explosion in food supply enabled large early animals, such as the Ediacarans, to develop and thrive, instead of much smaller precursors whose survival as fossils would be less likely.

The next big step was also one of geochemistry, when animals became able to secrete calcium-rich skeletons by extracting that element from seawater. It took place around 543 Ma at the start of the Cambrian, while iron-rich deep waters were also common. Was there somehow a connection between the two chemical highlights of the late Precambrian? Calcium is very interesting metabolically: too little and cells do not function properly; too much and they die. The ‘window’ of metabolically tolerable calcium concentrations is narrow. One possible means whereby calcium-rich hard parts may have developed among animals is that their outer cells were harnessed by evolution to rid the body of excess calcium in an organised way, creating the opportunity for both armour and armaments. Would elevated iron enhance the solubility of calcium in ocean water?

See also: Lyons, T.W 2008. Ironing out ocean chemistry at the dawn of animal life. Science, v. 321, p.923-924.

The Great Ordovician Diversification

Geologists in general learn that the tangible fossils first appeared at the start of the Cambrian Period. So they did, but we refer to that event as the Cambrian Explosion, but it was hardly explosive as there were very few fossil taxa of Lower Cambrian age. Indeed, by the end of the Cambrian only 500 or so genera are known. Fossils truly exploded in the later Ordovician, reaching 1600 genera, which number wasn’t exceeded until the start of the Cretaceous, 300 Ma later. Sudden rises in diversity, like mass extinctions, demand an explanation, but few have been offered for the late Ordovician explosive diversification, unlike the mass extinction at its close, which halved the number of genera living at the time. That has been attributed to the widespread glaciation of Gondwanaland, the fall in sea levels drastically reducing ecological niches (a wilder scenario is that the extinction was caused instantaneously by a gamma-ray burst from a nearby supernova, but there is little evidence for such an event).

The Ordovician has been assumed to have been a period that experienced ‘supergreenhouse’ conditions because of a far greater proportion of CO₂ in the atmosphere in the early Palaeozoic. Advances in stable-isotope analyses of small samples allow that idea to be tested (Trotter, J.A. et al. 2008. Did cooling oceans trigger Ordovician biodiversification? Evidence from conodont thermometry. Science, v. 321, p. 550-554). Julie Trotter of the Australian National University and her French and Canadian colleagues show that oxygen isotopes in conodonts that range in age from Lower Ordovician to Lower Silurian changed steadily with time. Assuming the conodont animals were planktonic, the increase in the proportion of ¹⁸O represents decreasing sea-surface temperatures, from around 40ºC (truly supergreenhouse) to levels very similar to those that prevail in today’s tropical ocean, around 30ºC, to even more temperate levels (24ºC) by the close of the Ordovician. So it seems as if cooling encouraged rapid evolution of new organisms at that time.