Chris J. Hawkesworth

Episodic growth of the Gondwana supercontinent from hafnium and oxygen isotopes in zircon

It is thought that continental crust existed as early as 150 million years after planetary accretion, but assessing the rates and processes of subsequent crustal growth requires linking the apparently contradictory information from the igneous and sedimentary rock records. For example, the striking global peaks in juvenile igneous activity 2.7, 1.9 and 1.2 Gyr ago imply rapid crustal generation in response to the emplacement of mantle ‘super-plumes’, rather than by the continuous process of subduction. Yet uncertainties persist over whether these age peaks are artefacts of selective preservation, and over how to reconcile episodic crust formation with the smooth crustal evolution curves inferred from neodymium isotope variations of sedimentary rocks. Detrital zircons encapsulate a more representative record of igneous events than the exposed geology and their hafnium isotope ratios reflect the time since the source of the parental magmas separated from the mantle. These ‘model’ ages are only meaningful if the host magma lacked a mixed or sedimentary source component, but the latter can be diagnosed by oxygen isotopes, which are strongly fractionated by rock-hydrosphere interactions. Here we report the first study that integrates hafnium and oxygen isotopes, all measured in situ on the same, precisely dated detrital zircon grains. The data reveal that crust generation in part of Gondwana was limited to major pulses at 1.9 and 3.3 Gyr ago, and that the zircons crystallized during repeated reworking of crust formed at these times. The implication is that the mechanisms of crust formation differed from those of crustal differentiation in ancient orogenic belts.

Rare earth element geochemistry of oceanic ferromanganese nodules and associated sediments

Abstract Analyses have been made of REE contents of a well-characterized suite of deep-sea (> 4000 m.) principally todorokite-bearing ferromanganese nodules and associated sediments from the Pacific Ocean. REE in nodules and their sediments are closely related: nodules with the largest positive Ce anomalies are found on sediments with the smallest negative Ce anomalies; in contrast, nodules with the highest contents of other rare earths (3 + REE) are found on sediments with the lowest 3 + REE contents and vice versa. 143 Nd 144 Nd ratios in the nodules (∼0.51244) point to an original seawater source but an identical ratio for sediments in combination with the REE patterns suggests that diagenetic reactions may transfer elements into the nodules. Analysis of biogenic phases shows that the direct contribution of plankton and carbonate and siliceous skeletal materials to REE contents of nodules and sediments is negligible. Inter-element relationships and leaching tests suggest that REE contents are controlled by a P-rich phase with a REE pattern similar to that for biogenous apatite and an Fe-rich phase with a pattern the mirror image of that for sea water. It is proposed that 3 + REE concentrations are controlled by the surface chemistry of these phases during diagenetic reactions which vary with sediment accumulation rate. Processes which favour the enrichment of transition metals in equatorial Pacific nodules favour the depletion of 3 + REE in nodules and enrichment of 3 + REE in associated sediments. In contrast, Ce appears to be added both to nodules and sediments directly from seawater and is not involved in diagenetic reactions.

The generation and evolution of the continental crust

Abstract: The continental crust is the archive of the geological history of the Earth. Only 7% of the crust is older than 2.5 Ga, and yet significantly more crust was generated before 2.5 Ga than subsequently. Zircons offer robust records of the magmatic and crust-forming events preserved in the continental crust. They yield marked peaks of ages of crystallization and of crust formation. The latter might reflect periods of high rates of crust generation, and as such be due to magmatism associated with deep-seated mantle plumes. Alternatively the peaks are artefacts of preservation, they mark the times of supercontinent formation, and magmas generated in some tectonic settings may be preferentially preserved. There is increasing evidence that depletion of the upper mantle was in response to early planetary differentiation events. Arguments in favour of large volumes of continental crust before the end of the Archaean, and the thickness of felsic and mafic crust, therefore rely on thermal models for the progressively cooling Earth. They are consistent with recent estimates that the rates of crust generation and destruction along modern subduction zones are strikingly similar. The implication is that the present volume of continental crust was established 2–3 Ga ago.

Detrital zircon record and tectonic setting

ABSTRACTDetrital zircon spectra refl ect the tectonic setting of the basin in which they are deposited. Convergent plate margins are charac-terized by a large proportion of zircon ages close to the depositional age of the sediment, whereas sediments in collisional, extensional and intracratonic settings contain greater proportions with older ages that refl ect the history of the underlying basement. These differences can be resolved by plotting the distribution of the difference between the measured crystallization ages (CA) of individual zircon grains present in the sediment and the depositional age (DA) of the sedi-ment. Application of this approach to successions where the original nature of the basin and/or the link to source are no longer preserved constrains the tectonic setting in which the sediment was deposited.INTRODUCTION Detrital zircons are a minor constituent of clastic sedimentary rocks, yet their physiochemical resilience and high concentrations of certain key trace elements means that they have become an important phase in sedimentary provenance analysis and in crustal evolution studies (e.g., Cawood et al., 2007b; Hawkesworth et al., 2010). Large numbers of in situ, high precision analyses of both igneous and detrital zircons are now available, and a striking feature of the zircon record is that it clusters into peaks of crystallization ages (Condie et al., 2009). Compilations of crys-tallization ages for detrital and igneous zircons show remarkably similar patterns of peaks and troughs, although with some variation in the rela-tive amplitude of the peaks (Condie et al., 2009). This coincidence sug-gests that the sedimentary record is a valid representation of the magmatic record (Hawkesworth et al., 2010).We establish that detrital zircon spectra have distinctive age distribu-tion patterns that refl ect the tectonic setting of the basin in which they are deposited. These patterns are principally controlled by (i) the volumes of magma generated in each tectonic setting and their preservation poten-tial, (ii) the ease with which magmatic and detrital zircons of various ages and origins become incorporated into the sedimentary record, and (iii) the record of old zircons incorporated into the sediment. These in turn provide a framework that can be used to constrain the tectonic setting of sedimen-tary packages. This approach distinguishes between three tectonic settings (i.e., convergent, collisional, and extensional), and it is most sensitive when the depositional age of the sediment investigated is well constrained. Basin setting will evolve with tectonic regime; for example, arc-continent or continent-continent collision will result in the evolution of convergent and extensional basins into collisional foreland basins. Hence the three settings distinguished herein are end-members, and the zircon age patterns associ-ated with each show a spectrum of distributions that merge and overlap rather than defi ne discrete fi elds. Discriminant plots developed for igneous rock geochemistry (e.g., Pearce and Cann, 1973) or sediment framework modes (e.g., Dickinson and Suczek, 1979) often have diffuse boundaries or overlap between fi elds, but remain important approaches in understanding and constraining tectonic setting. Equally important, exceptions to simple end-member classifi cations can provide insight into subtleties of tectonic process, such as outlined below for Avalonia in eastern North America.

Magmatism and continental break-up in the South Atlantic: high precision40Ar-39Ar geochronology

A detailed ArAr study of the Parana-Etendeka continental flood basalts (CFB) has been undertaken using laser spot analysis. Data provide information not only on the age of the samples but also on the variability of the non-radiogenic Ar component and state of alteration. The results indicate the Parana-Etendeka CFB were erupted over 10 million years between 137 and 127 Ma providing a minimum overall eruption rate of ∼ 0.1 km3 yr−1. This is an order of magnitude less than that previously proposed for this, the Deccan and Siberian CFB provinces but it is similar to estimates for Hawaii, Iceland and the Columbia River CFB. In detail, the new ArAr analyses indicate that magmatism within the Parana-Etendeka province migrated from NW to SE prior to and during the opening of the South Atlantic, providing an explanation for the asymmetry of the CFB lavas about the South Atlantic. Moreover, chemically defined magma types were erupted at different times in different places, and so within the Paranathey may not be used as reliable chronostratigraphic units. Rather, such magmatic units may reflect the extent of compositionally distinct source regions in the uppermost mantle, and indicate that partial melting took place over a wide area under the ParanaBasin. The NW-SE migration of the onset of magmatism might be interpreted as a plume trace, but the inferred rate of movement is 3 times faster than that inferred from subsequent magmatism on the Rio Grande Rise. It is argued that the onset of magmatism reflected extensional tectonics normal to the Ponta Grossa dyke swarm. Given that magmatism preceded the main phase of rifting, and that extension across the Ponta Grossa was moderate, current plume models have difficulty in predicting sufficient melting within the asthenosphere.

Controls on trace element (Sr-Mg) compositions of carbonate cave waters: implications for speleothem climatic records

Abstract At two caves (Clamouse, S France and Ernesto, NE Italy), cave drip and pool waters were collected and sampled at intervals over a 2–3 year period. Mg/Ca and Sr/Ca concentration ratios, corrected for marine aerosols, are compared with those of bedrocks and, in some cases, aqueous leachates of soils and weathered bedrocks. Cave waters do not lie along mixing lines between calcite and dolomite of bedrock carbonate, but typically show enhanced and covarying Mg/Ca and Sr/Ca. Four factors are considered as controlling processes. (1) The much faster dissolution rate of calcite than dolomite allows for the possibility of increase of Mg/Ca if water–rock contact times are increased during drier conditions. A theoretical model is shown to be comparable to experimental leachates. (2) Prior calcite precipitation along a flow path is a powerful mechanism for generating enhanced and covarying Mg/Ca and Sr/Ca ratios. This mechanism requires the solution to lose CO 2 into pores or caverns. (3) Incongruent dolomite dissolution has only limited potential and is best regarded as two separate processes of dolomite dissolution and calcite precipitation. (4) selective leaching of Mg and Sr with respect to Ca is shown to be important in leachates from Ernesto where it appears to be a phenomenon of calcite dissolution. In general selective leaching can occur whenever Ca is sequestered into precipitates due to freezing or drying of soils, or if there is derivation of excess Sr and Mg from non-carbonate species. The Ernesto cave has abundant water supply which in the main chamber is derived from a reservoir with year-round constant P CO 2 of around 10 −2.4 and no evidence of calcite precipitation in the karst above the cave. Two distinct, but overlying trends of enhanced and covarying Mg/Ca and Sr/Ca away from the locus of bedrock compositions are due to calcite precipitation within the cave and, at a variable drip site, due to enhanced selective leaching at slow drip rates. Mg-enhancement in the first chamber is due to a more dolomitic bedrock and longer residence times. The Clamouse site has a less abundant water supply and presents geochemical evidence of prior calcite precipitation, both in the cave and in overlying porous dolomite/dedolomitized limestone bedrock. Initial P CO 2 values as high as 10 −1 are inferred. Experimental incubations of Clamouse soils which generated enhanced P CO 2 and precipitated CaCO 3 had compositions similar to the karst waters. Calcite precipitation is inferred to be enhanced in drier conditions. Hydrological controls on cave water chemistry imply that the trace element chemistry of speleothems may be interpretable in palaeohydrological terms. Drier conditions tends to promote not only longer mean residence times (enhancing dolomite dissolution and hence Mg/Ca), but also enhances degassing and calcite precipitation leading to increased Mg/Ca and Sr/Ca.

Evolution of the continental crust

The continental crust covers nearly a third of the Earth’s surface. It is buoyant—being less dense than the crust under the surrounding oceans—and is compositionally evolved, dominating the Earth’s budget for those elements that preferentially partition into silicate liquid during mantle melting. Models for the differentiation of the continental crust can provide insights into how and when it was formed, and can be used to show that the composition of the basaltic protolith to the continental crust is similar to that of the average lower crust. From the late Archaean to late Proterozoic eras (some 3–1 billion years ago), much of the continental crust appears to have been generated in pulses of relatively rapid growth. Reconciling the sedimentary and igneous records for crustal evolution indicates that it may take up to one billion years for new crust to dominate the sedimentary record. Combining models for the differentiation of the crust and the residence time of elements in the upper crust indicates that the average rate of crust formation is some 2–3 times higher than most previous estimates.

A Change in the Geodynamics of Continental Growth 3 Billion Years Ago

Continental Growth Spurts The appearance and persistence of continents through geologic time has influenced most processes on Earth, from the evolution of new species to the climate. The relative proportion of newly formed crust compared to reworked, or destroyed, older crust reveals which processes controlled continental growth. Based on the combined analyses of Hf-Pb and O isotopes in zircon minerals, Dhuime et al. (p. 1334) measured continuous but variable rates of new crustal production throughout Earths history. Increased rates of crustal destruction starting around 3 billion years ago coincide with the onset of subduction-drive plate tectonics, slowing down the overall rate of crustal growth. Isotopic analysis of zircons reveals the proportion of crust formed and destroyed on continents throughout Earth’s history. Models for the growth of continental crust rely on knowing the balance between the generation of new crust and the reworking of old crust throughout Earth’s history. The oxygen isotopic composition of zircons, for which uranium-lead and hafnium isotopic data provide age constraints, is a key archive of crustal reworking. We identified systematic variations in hafnium and oxygen isotopes in zircons of different ages that reveal the relative proportions of reworked crust and of new crust through time. Growth of continental crust appears to have been a continuous process, albeit at variable rates. A marked decrease in the rate of crustal growth at ~3 billion years ago may be linked to the onset of subduction-driven plate tectonics.

Crustal contamination versus enriched mantle: 143Nd/144Nd and 87Sr/86Sr evidence from the Italian volcanics

Abstract143Nd/144Nd, 87Sr/86Sr, and REE analyses are presented on a wide variety of Pliocene-Recent volcanic rocks from central Italy. 143Nd/144Nd varies from 0.51214–0.51289 and 87Sr/86Sr from 0.7255-0.7036; while the rare earth elements are characterised by light RE enrichment and a significant negative Eu anomaly. These Italian volcanics are tentatively subdivided into three zones: (1) N. Tuscany where the magmas are believed to reflect crustal anatexis. (2) A central zone in which hybrid (crust/ mantle) rocks have been recognised. (3) A southern zone, south of Rome, where mantle-derived magmas are identified which have been largely unaffected by interaction with continental crust. At Roccamonfina, in zone 3, Rb/Sr and Sm/Nd pseudo isochrons are observed but since the calculated ages are 0.5 and 2.0 b.y. respectively it is argued that a simple isochron model is not applicable and that the data are most easily explained by a recent mixing event within the upper mantle. It is envisaged that this occurred during metasomatism of the upper mantle source region by a fluid that had high 87Sr/86Sr and low 143Nd/144Nd and was enriched in K, Rb, and LREEs but relatively depleted in Sr2+ and Eu2+.

The petrogenesis of Mesozoic Gondwana low-Ti flood basalts

Abstract New major, trace element and isotopic data for Jurassic basalts from SE Australia indicate that they are strikingly similar to the Jurassic tholeiitic rocks of Tasmania and the Transantarctic Mountains. These rocks are all characterised by low TiO2, P2O5, Na2O, Fe2O3, Ti/Zr, Ti/Y and eNd, and high SiO2, Rb/Ba, Rb/Sr, 87Sr/86Sr and 207Pb/204Pb, relative to oceanic basalts. They therefore comprise a major province, termed the Ferrar magmatic province, which extended for 3000–4000 km across the Gondwana supercontinent. A review of the other Mesozoic low-Ti CFBs suggests that the Ferrar rocks are an extreme example of these magma types. It is striking that both the major and trace element compositions are different from oceanic basalts, which suggests that these features are linked, and it is argued that they were derived from distinctive source regions in the sub-continental mantle. Such source regions were variably depleted in major and minor elements, and then relatively enriched in highly incompatible elements and Sr and Pb isotopes, which is best explained by the introduction of a small amount of subducted sediment. The tectonic setting of the Ferrar magmatism is poorly constrained, but at present there is no clear geochemical evidence for the involvement of asthenospheric plume material in the petrogenesis of these low-Ti CFBs.