V12B-01
Are there impact-formed zircons in the Hadean record?
Detrital Hadean zircons from the Jack Hills, Western Australia, show a remarkable cluster of crystallization temperatures at 680±25°C. This is particularly surprising as a simple model relating rock composition and Zr concentration predicts that a very broad spectrum of crystallization temperatures (ca. 650°C to 1000°C) with a median value of 780°C, would result from impact melting of the Earth's surface. Magmatic fractionation would tend to increase the aforementioned values. Given the predicted high rate of impacts during the Hadean, the absence of such a population in the Jack Hills zircons could signal a profound sampling problem, a hint of a history much different than previously supposed, or our lack of understanding of zircon formation due to impact related processes. We have begun to examine the latter issue by investigating the crystallization temperatures of zircons formed in melt sheets preserved in the geologic record. The Sudbury Igneous Complex, formed at 1850±3 Ma within the second largest impact crater on Earth, includes two igneous units termed the Black and Felsic Norites. Examination of zircons from each by SIMS confirms their crystallization age at 1847.3±2.2 Ma and yields Ti-in-zircon temperatures of 720°C and 750°C, respectively. This is consistent with that predicted from zircon saturation systematics. A statistical test indicates that the combined norite population is distinct from the Hadean temperature distribution. Thus the question arises: where are the Hadean zircons expected to have formed at >780°C via impact processes? Similar analysis is being pursued for zircons from the Vredefort Impact Structure, South Africa, which should provide further information on impact-formed zircon temperature spectra.
V12B-02
Detrital Shocked Minerals: A New Tool for Identifying Eroded Impacts in the Sedimentary Record
Impact processes dominated the Early Earth and are believed to have culminated in a Late Heavy Bombardment at ca. 3.9-3.8 Ga. No direct geologic evidence of this stage of planetary evolution has been found, however Hadean zircons up to 4.4 Ga have been identified. A new method for identifying impact evidence in siliciclastic sediments has recently been reported (Cavosie et al., 2008; Quintero et al., 2008 this volume), based on the discovery of detrital grains of shocked quartz and zircons eroded from the 2.02 Ga Vredefort Dome in channel sediments of the Vaal River. Here we describe grain scale characteristics of these shocked minerals, and discuss aspects of their preservation. In grain mount, detrital grains of shocked quartz range from sub-angular to round, and characteristically preserve a single set of parallel decorated planar deformation features (PDFs) with spacing of less than 5 microns, analogous to shocked quartz reported from Vredefort impactites. While the PDFs are continuous across most grains, numerous examples were found where a single orientation of PDFs is heterogeneously distributed in an optically continuous grain, such as only being preserved within 15 microns from a grain margin. Where sub-grains are present, PDFs can occur in multiple orientations, or in single isolated sub-grains as small as 100 microns that are surrounded by PDF-free sub-grains. Well-preserved PDFs are found in grains as small as 100 microns. In addition to quartz, detrital shocked zircons were also found. The shocked zircons are subhedral to anhedral, and show evidence of sedimentary abrasion. Planar fractures (PFs) were observed parallel to (010), (100), and in two {hkl} orientations. Up to three PF orientations are present in single zircons, with variable spacing that usually ranges from 5-20 microns. Despite the complex history of the shocked grains from the Vredefort Dome, including post-impact granulite facies metamorphism, exhumation, erosion, and fluvial transport, detrital quartz and zircon provide a robust record of the Vredefort impact in modern sediments. The recognition that shocked quartz and zircon can survive erosion and fluvial transport in siliciclastic sediments opens up new avenues in searching for impact evidence from the Hadean, as shocked grains from the early Earth may be preserved in Hadean detritus. Given the abundance of evidence supporting an origin for the Hadean detrital zircons in quartz-saturated granitoids, large volumes of shocked quartz may have been produced, eroded, and deposited in younger sediments. This is particularly promising, given that all known populations of Hadean detrital zircons occur in quartz-rich siliciclastic rocks. Cavosie et al., 2008. Geology, in review Quintero et al., 2008. AGU, this volume Cavosie et al., 2007. Earth's Oldest Rocks.
V12B-03
Accessory Mineral Records of Early Earth Crust-Mantle Systematics: an Example From West Greenland
Conditions for the formation and the nature of Earth's early crust are enigmatic due to poor preservation. Before c.4 Ga the only archives are detrital minerals eroded from earlier crust, such as the Jack Hills zircons in western Australia, or extinct isotope systematics. Zircons are particularly powerful since they retain precise records of their ages of crystallisation, and the Lu-Hf radiogenic isotope and O stable isotope systematics of the reservoir from which they crystallised. In principle, this allows insight into the nature of the crust, the mantle reservoir from which the melt was extracted and any reworked material incorporated into that melt. We have used in situ methods to measure U-Pb, O and Lu-Hf within single zircon crystals from tonalitic gneisses from West Greenland in the vicinity of the Isua Supracrustal Belt. They have little disturbed ages of c.3.8 Ga, mantle-like O isotope signatures and Lu-Hf isotope signatures that lie on the CHUR evolution line at 3.8 Ga. These samples have previously been subjected to Pb isotope feldspar and 142Nd whole rock analysis and have helped constrain models in which early differentiation of a proto-crust must have occurred. The CHUR-like Lu-Hf signature, along with mantle-like O signature from these zircons suggests juvenile melt production at 3.8 Ga from undifferentiated mantle, yet the other isotope systems preclude this possibility. Alternatively, this is further strong evidence for a heterogeneous mantle in the early Earth. Whilst zircons afford insight into the nature of the early crust and mantle, it is through the Sm-Nd system that the mantle has traditionally been viewed. Titanite often contains several thousand ppm Nd, making it amenable to precise analysis, and is a common accessory phase. It has a reasonably high closure temperature for Pb and O, and it can retain cores with older ages and distinct REE chemistry. It is often the main accessory phase alongside zircon, and it is the main carrier of Nd within the whole rock such that Nd isotope analysis of titanite may be able to see-through later alteration that may have partially reset the whole rock system. We present new in-situ U-Pb, O and Sm-Nd and high-precision U-Pb ID-TIMS and Sm-Nd MC- ICPMS data from individual or fragmented titanite grains. We discuss how these data complement the zircon data and may help to resolve long-standing debates in ancient gneiss terranes, with utility to the nature and formation of crust on the early Earth.
V12B-04
Paleomagnetism of the Astrobiology Drilling Project 8 drill core, Pilbara, Western Australia: implications for the early geodynamo and Archean tectonics
Paleomagnetic measurements from the Archean Pilbara craton have recently been used to argue for the presence of a substantial magnetic field at 3.2 Ga (Tarduno et al., 2007), as well as for extremely fast plate motions or true polar wander (Strik et al., 2003, Suganuma et al., 2006). Paleomagnetic records in the Archean are fundamentally limited by the scarcity of well-preserved, low metamorphic grade Archean rocks. Where such rocks are exposed, paleomagnetic sampling is often difficult or impossible due to pervasive lightning remagnetization and deep weathering of the cratonic surface. More pristine samples can potentially be obtained from shallow drill cores like those obtained by the Astrobiology Drilling Project (ABDP). We present a paleomagnetic analysis of the ~350 m deep ABDP-8 drill core, which was drilled in the East Strelley greenstone belt and which penetrated the Double Bar Formation of the Warrawoona Group, as well as the unconformably overlying Euro Basalt and Strelley Pool Chert units of the Kelly Group. Full sample orientation (declination and inclination) was achieved through the use of a Ballmark™ orientation system. A strong drilling overprint was removed for most samples by alternating field demagnetization to 20 mT. Subsequent thermal demagnetization revealed single-polarity magnetic directions within the Euro Basalt and Double Bar Formation carried by magnetite. The directions from these two Formations are statistically different to >95% confidence, which constitutes a positive unconformity test and indicates that the Euro Basalt direction is primary. Upon tilt correction, the ~3.34-3.37 Ga Euro Basalt direction is indistinguishable from the tilt-corrected direction found previously in the ~3.46 Ga Duffer Formation of the Warrawoona Group (McElhinny and Senanayake, 1980). The Euro Basalt direction, if taken at face value, implies small relative motion of the Pilbara Craton from ~3.46 Ga to ~3.34 Ga. This is inconsistent with the apparent polar wander path presented for the ~3.46 Ga Marble Bar Chert Member of the Towers Formation (Suganuma et al., 2006). The lack of reversals in the sequence is consistent with a low reversal frequency in early Earth history, as has been suggested by dynamo models for the Earth with a small inner core (Coe and Glatzmaier, 2006).
V12B-05
Episodic Earth Evolution: a Mantle Geodynamic Model
Three major regimes of crust-mantle evolution are recorded in U-Pb ages of zircons in granites and large rivers. Plate tectonics operated in the first stage, from ~4.4 to 2.7 Ga; huge peaks of crustal growth separated by long troughs dominated the second stage, from 2.7 to 1.8 Ga, and semi-continuous growth punctuated by large peaks characterized the last stage, from 1.8 to 0 Ga. Individual peaks in the second stage, at 2.7, 2.5, 2.1 and 1.8 Ga, open with massive mafic-ultramafic volcanism and climax 30 m.y. later with intrusion of voluminous granitoids: each peak was initiated by enhanced mantle plume activity, culminated with accelerated plate tectonics that produced large amounts of granitoid crust, and was followed by a long period of diminished tectonic activity. New fluid-mechanics experiments show that this second regime could have resulted from destabilization of a hot dense layer at the bottom of the lower mantle. Domes rising from this layer partially melt to form voluminous mafic magmas, and also trigger enhanced subduction. The first pulse of thermochemical instabilities was synchronous over the whole mantle, like the major 2.7 Ga crustal- growth peak, and later pulses were more disorganized, like the later peaks. During each peak, enhanced subduction removes quickly the upper cold thermal boundary layer: it is therefore followed by an inter-peak period of diminished activity, during which the cold boundary layer is growing again. This episodic evolution closely links continental growth and the extraction of heat from the core. In particular, the onset of the second regim could have been decisive for the timing of the inner core crystallization and the establishment of today's core dynamo regim.
V12B-06
An Archean Terrestrial Fractionation Line for Oxygen Isotopes
The Terrestrial Fractionation Line (TFL) for oxygen isotopes is defined by 17O/16O and 18O/16O analyses of meteoric waters, seawater, sedimentary, metamorphic, and igneous rocks and constituent minerals. Interlaboratory measurements of the slope of the TFL on a plot of d18O vs. d17O revealed eclogitic garnets with a slope of 0.526 and hydrothermal quartz of 0.524 from rocks younger than 0.8 Ga (Giga years before present). New measurements show Archean metamorphic rocks and minerals from Barberton, (3.2 Ga, S. Africa), Isua (3.8 Ga, Greenland), and Acasta (4.0 Ga, Canada) have a slope of 0.524 +/- 0.002 (95% confidence, MSWD = 0.66). Analysis of Ag3PO4 prepared from apatite mineral separates from Isua meta-sediments gives a slope of 0.509 +/- 0.022 (95% confidence, MSWD = 0.59). Taken at face value, steeper slopes on a d17O vs. d18O diagram indicate an approach towards isotope exchange equilibrium. Lower slopes are expected when isotope fractionation is kinetically controlled. The lower slope of 0.509 for Isua apatite suggests that the formation of orthophosphate was kinetically controlled. Kinetic fractionations are known to occur during catalysis of reactions by enzymes secreted by microbes. Enzymatic catalysis confers an advantage on organisms because energy-producing reactions may be induced to occur at lower temperature conditions more accessible to the organism. May it be definitively concluded that enzymatic catalysis was responsible for the measured 0.509 slope? No, abiotic kinetic fractionation cannot be disproven with existing data. The preparation of Ag3PO4 from apatite may have introduced kinetic fractionation as an analytical artifact. Conclusions fully supported by the data suggest: (1) Mixing accompanying the violent birth of the Earth- Moon system had already succeeded in establishing Earth's current oxygen isotope composition by 4.0 Ga; and (2) No trace of an episode of late heavy meteorite bombardment remains in the oxygen isotope compositions of Earth's oldest rocks.
V12B-07
In search of the noble gas 3.52 Ga atmospheric signatures
The isotopic signatures of noble gases in the Present-day mantle and in the atmosphere permit exceptional insight into the evolution of these reservoirs through time ([1]). However, related exchange models are under- constrained and would require direct measurements of the atmospheric composition long ago, e.g., in the Archaean. Drilling in the the 3.52 Ga chert-barite ([2]) of the Dresser formation(Pilbara Drilling Project) , North Pole, Pilbara craton (Western Australia), led to recovery of exceptionally fresh samples preserving primary fluid inclusions unaffected by surface weathering. The whole formation is considered to be an already established basin when hydrothermal processes started. The chemical composition of primary fluid inclusions trapped in hydrothermal quartz from vacuolar komatiitic basalt from 110 m depth were determined by synchrotron X-ray microfluorescence (ESRF, Grenoble,France). Data show that fluids are relatively homogenous, consisting of a Ba-rich fluid and a Fe (+Ba)-rich fluid of hydrothermal origin as concluded by Foriel et al.([3]). The isotopic compositions of xenon and argon trapped in these fluids were measured by mass spectrometry following vacuum crushing. The three argon isotopes show a homogeneous signature quite different from present-day Earth atmosphere but we cannot exclude the possibility that secondary nuclear reactions produced these anomalies. Despite this, the Xe isotopic trends indicate a less radiogenic signature than the Present-day atmosphere, and probably represent a remnant of the Archaean atmosphere. If this xenon composition is primitive then it implies that there is no cosmogenic production through time. However, argon ratios could be explained by cosmogenic production which implies significant surface exposure times. Cosmogenic production will produce correlated argon and xenon isotope signatures. Therefore it is necessary to differentiate primary from secondary composition. To investigate the effects of these nuclear reactions on Xe isotope production, barite from 30m shallower depth in the same core were analyzed. Variable excesses can be linked to spallogenic and cosmogenic reactions ([4] [5] [6]) which allow the primitive Xe isotopic signature to be isolated from subsequent secondary production. Models of the archaean atmospheric noble gas signature can thereby be compared with different theories on primitive atmospheric composition. [1] Staudacher T. Allègre C.J. (1982) EPSL 60, p 389-406 [2] Van Kranendonk MJ., Hickman A.H., Williams I.R. and Nijman W. (2001) Rec.-Geol. Surv. West. Aust. 2001/9, 134 [3] Foriel J., Philippot P., Rey P., Somogyi A., Banks D. and Ménez B. (2004) EPSL, 228, 451-463 [4]Srinivasan B. (1976) EPSL, 31, 129-141 [5]Charalambus S. (1971) Nuclear Physics, A166, 145 [6]Meshik A. P., Hohenberg C. M., Pravdivtseva O. V. and Kapusta Y. (2001) Phys. Rev., C 64, 035205-1 035205-6
V12B-08
Archean Surface Environments and Komatiitic Volcanism: Evidence From the Barberton Greenstone Belt, South Africa
The Barberton Greenstone Belt (BGB) includes a stratigraphic succession 10-15 km thick with at least three major intervals dominated by komatiitic rocks totaling about five kilometers in aggregrate thickness that formed over an interval of approximately 200 million years (3.48 to 3.28 Ga). Significant aspects of the mid- Archean surface environment must have been controlled by the localized build-up of komatiitic landforms, perhaps unique tectonic styles associated with komatiitic volcanism, as well as the chemically highly reactive Mg-rich mineral and glass phases as they were deposited in contact with the Archean atmosphere and hydrosphere. The 3.48 Ga Komati Formation is composed of komatiitic and minor basaltic flow rocks, 3.5 km thick, without any sedimentary interbeds or stratiform zones of alteration that might suggest pauses in volcanism. We suggest this may represent a time interval as short as 105 years. These lavas were likely deposited in submarine lava fields of large dimensions and perhaps well away from the high heat flows associated with vent areas. There is no compelling evidence to support either a mid-oceanic rift or subduction zone tectonic setting. However, it is clear that this was a geologic environment removed from the input of continent-derived sediment. The 3.29 Ga Weltevreden Formation, over 3 km in stratigraphic thickness in places, also appears to have formed over a very short time span, with little or no deposition of sediment. Unlike the Komati Formation, abundant komatiitic tuffs are interbedded with the lavas and these may document environments somewhat more proximal to volcanic vents for the Weltevreden Formation. By contrast, the 0.6-1.0 km thick Mendon Formation is composed of six or more distinctive members that each include komatiitic lava flows capped by carbonate and silica-enriched alteration zones and sedimentary cherts. The base of the formation is 3.33 Ga, several units in the interior are 3.30 Ga, and conformable cherts of the immediately overlying Fig Tree are 3.27 Ga. Both geology and geochronology are consistent with 107 or more years for this formation – requiring up to 3-4 orders of magnitude lower rates of local volcanic deposition compared to other komatiitic units of the BGB. Immobile incompatible element ratios suggest that each member is from a discrete and separate mantle-derived magma batch, and like the Komati Formation, probably erupted from distant vents. However, significant changes in stratigraphy from south to north, and across a series of regional faults, do support an opening rift-basin model for the Mendon Formation. The geology of these komatiitic formations does not support models for vast, dynamic, and focused hydrothermal systems. Major cross-cutting zones of alteration do not exist; silica and carbonate alteration is stratiform and contemporary with sedimentation at each pause in volcanism. Komatiitic rocks only meters below the alteration zones typically contain fresh igneous minerals, commonly pyroxenes and rarely even olivine, and the rocks are generally only modified by hydration and minor losses of sodium and calcium. Our preferred model for the development of the Mendon units, as well as similar komatiitic units in the Hooggenoeg and Kromberg Formations, includes deposition of flows, shallow-marine to subaerial exposure and surface-driven hydration, silicification and carbonation contemporaneous with accumulation of organic sediments and volcanic ash from distal sources.