Temperature plays a fundamental role in controlling many of the physical properties and macroscopic processes of the Earth's mantle, such as viscosity, density, electrical conductivity, convection, plume dynamics and mantle melting. Thermal profiles of the upper mantle are well constrained from geothermometry, heat-flow measurements and the temperature of the olivine-spinel-post spinel transitions. Lower mantle temperatures, on the other hand, are not well constrained with estimates of temperature at 700 km varying by as much as 1500 K (eg; Yagi et al., 1996). Such widely varying models have vastly different implications for Earth processes and properties. Electrical conductivity measurements could, in principle, be used to distinguish between different lower mantle thermal models, provided a) geophysically measured response functions have a sufficiently good depth resolution and b) high quality laboratory conductivity measurements for the relevant minerals exist. Recent conductivity measurements at the high temperatures of the lower mantle (Dobson et al., 1997; Katsura et al., 1998; Xu et al., 1998) show that lower mantle phases have activation energies higher than previously thought, at around 1 eV, sufficient to discriminate between candidate lower mantle geotherms. We present results from forward modelling calculations of the response function based these electrical conductivity measurements. By comparing predicted response functions generated from candidate geotherms with the measured geomagnetic response of the Earth, we are able to discriminate between lower mantle geotherms. We find the data are only consistent with a relatively cold shallow lower mantle.
Dobson DP, Richmond NC & Brodholt JB, Science, 275, 1179-1801, (1997).
Katsura T, Sato K & Ito E, Nature, 395, 493, (1998).
Yagi T & Funamori N, Phil. Trans. R. Soc. London, A354, 1371-1384, (1996).
Xu Y, McCammon & Poe BT, Science, 282, 922-924, (1998).
The seismic anisotropy of a mylonitic peridotite from the Oman Ophiolite has been measured at room pressure and room temperature. Detailed microstructural, mineralogical and chemical analyses have been undertaken together with numerical modelling in order to constrain the relative significance of various microstructural features in controlling the seismic anisotropy of mantle rocks.
The microstructure of the peridotite shows high temperature deformation. It is composed of 57% olivine (Fo91) and 39% orthopyroxene (En88) and small amounts of spinel and clinopyroxene. A mineral lineation is defined by elongated spinel crystals and elongated olivine and orthopyroxene porphyroclasts. A pronounced foliation is defined by flattened crystals and ultramylonitic layers. In the ultramylonitic layers, the grain size is reduced to µ m from the mm-sized porphyroclasts.
P and S wave velocities have been measured at 10 degree intervals over 360 degree on 3 orthogonal planes (2 in the foliation plane, 1 perpendicular to it), and on a fourth plane inclined to the foliation. The measured velocities change as a function of orientation with the greatest velocities being in the foliation plane parallel to the lineation.
Numerical modelling of the velocity results is being used to determine the relative contribution of the spatial distribution of the mineral phases, their crystallographic preferred orientation, and grain boundary structure in producing the observed velocity anisotropy. In this way, we seek to constrain the influence of microstructural parameters not typically accounted for describing the seimic velocity properties of mantle rocks.
We report on an experimental investigation of the viscosity of liquids in the Fe-S system at high pressures. The typically used high pressure Stokes viscometry 'quench and probe' technique is not suitable for this system because of the low viscosities and the lack of requisite control of the rise/fall time of the sphere. Instead, we employ the 'electro-detection method' which was developed for in-situ sphere speedometry measurement (LeBlanc and Secco, 1995). The method is based on the detection of sphere position from an electrical resistance anomaly produced by an insulating sphere passing through a pair of electrodes located in the conducting liquid sample. The experiments are carried out in a 1000 ton multi-anvil press with heating rates as high as 1500oC/min using LaCrO3 furnaces and W/Re thermocouples. Spheres of ruby, 0.5-0.7 mm diameter, are used in BN contained samples of 2.5 mm diameter and 2.5-3.0 mm height. Ruby is inert in Fe-S liquids and provides the requisite electrical conductivity contrast (~107) with the sample. Using Fe electrodes, the measured sphere velocity incorporates the electrical size of the sphere, determined from previous potential field modelling experiments (Secco et al, 1998), which is ~2.6 times the physical size. At pressures of 6 and 8 GPa and for temperatures in the range 1100-1350oC, the viscosities of Fe-27wt%S are in the range 1-6 Pa-s, or ~3 orders of magnitude higher than pure liquid Fe at 1 atm. These values are in agreement with extrapolations of previous viscosity measurements on the same composition up to 5 GPa and 1300oC in a cubic anvil press (LeBlanc and Secco, 1996). Using the same configuration and sample volume, we expect to extend the pressure range of viscosity measurement to 15 GPa in a 5000 ton press.
LeBlanc GE & Secco RA, Rev. Sci. Instrum, 66, 5015-5018, (1995).
Secco RA, LeBlanc GE, Yang H & Seibel JN, Properties of Earth and Planetary Materials at High Pressure and Temperature, AGU Monograph, 101, 495-505, (1998).
LeBlanc GE & Secco RA, Geophys. Res. Lett, 23, 213-216, (1996).
One of the objectives of the Seismic Experiment in Patagonia-Antarctica (SEPA) is to explore the Alvarez (1982) idea of mantle flow around South America, supported by shear wave splitting observations around the continent Russo and Silver (1994). This addresses the larger issue of how asthenospheric flow interacts with continental roots: are the roots dragged by the plate through the mantle, or does mantle flow push the roots? The portable stations sited in Patagonia and the Antarctic Peninsula are ideally placed to detect the shear wave splitting signature of mantle flow, in combination with permanent Antarctic and South American stations. Results from the permanent IRIS/GSN stations on S. Georgia, the Falkland Islands, southern Argentina and the Antarctic peninsula yield typical splitting results except for Palmer Station. The fast polarization direction <phi> and delay time t values are (-59,0.95) at HOPE (S. Georgia; poorly constrained), (64,0.89) at PLCA (SW Argentina), (61,1.0) at EFI (Falkland Islands; possible two-layer complexity) and (78,2.04) at PMSA (Palmer Station, Antarctica). These values are typical of ~1 second continental splitting times except for PMSA, which is among the largest recorded. The preliminary (<phi>,t) values are (30,0.92) at MILO and (2,0.45) at SALM, both in northwestern Patagonia, (50,1.92) at FELL in eastern Patagonia and (-82,0.55) at VTDF in southern Patagonia. The reliable results from the portable stations on the Antarctic Peninsula are also NE-SW, with (<phi>,t) = (61,1.2) at PRAT. These directions neither parallel the Andean orocline nor the Drake Passage. Thus the splitting patterns anticipated by the Russo and Silver, (1994) flow model are not as expected in southern South America. The directions we observe are mostly oblique to the E-W directions found in Venezuela by Russo et al. (1996). Mantle flow around the continent either a) lacks symmetry north and south, b) is not "laminar'' in the south, or c) is impeded through the Drake Passage. All these outcomes are problematic for the development of the S. Sandwich arc, thought to be analogous to the Lesser Antilles, which exists as a consequence of the flow around northern S. America (Russo et al., 1996). If the agent responsible for westward closure of the Pacific by S. America is convective traction on its continental root, this same flow may stagnate in the Drake Passage where it meets the flow due to Pacific closure or be strongly oblique to the western South America margin as suggested by flow modelling by Silver et al. (1998).
Alvarez, W., J. Geophys. Res., 87, 6697-6710, (1982).
Russo, RM & Silver, PG, Science, 263, 1105-1111, (1994).
Russo, RM & Silver, PG, Geology, 24, 511-514, (1996).
Silver, PG, Russo, RM & Lithgow-Bertelloni, C, Science, 279, 60-63, (1998).
Precise determination of the elastic properties of the mantle minerals at pressure and temperature conditions of the Earth's interior is very important for understanding the composition of the mantle. In our laboratory, we have developed techniques to measure elasticity of mantle minerals up to 8 GPa and 1500K using simultaneous ultrasonic velocity measurements and X-ray diffraction studies in a DIA-type, cubic anvil high pressure apparatus (SAM85) installed at beamline X17B at NSLS in Brookhaven National Laboratory (Liebermann, et al., 1998). Ultrasonic measurement is implemented by mounting an acoustic transducer at the back of the WC anvil and including alumina as buffer rod inside the cubic boron epoxy pressure medium. X-ray diffraction spectra from both the sample and NaCl were recorded at elevated pressures and temperatures from which the unit cell volumes of the sample and cell pressures were retrieved. Using this technique, compressional (VP) and shear (Vs) wave velocities measurements and equation of state (P-V-T) study for forsterite have been conducted on a polycrystalline specimen hot-pressed at 5 GPa and 1200°C; the bulk density is 99.3% of the single crystal X-ray value. Compressional and shear wave velocities at ambient P and T agree with single crystal data within 1%. Complete P-V-T and Vp and Vs data have been collected up to 8 GPa and 1273 K along a few decompression /heating /cooling cycles. Analyses from present P-V-T and acoustic measurements provide independent determination of elastic moduli and their pressure and temperature derivatives for this upper mantle phase. Comparison with previous P-V-T studies indicates a good agreement in sample volume at high pressure and high temperature. Results on pressure and temperature dependence of the bulk and shear moduli and comparison with previous ultrasonic measurements at high-pressure room temperature and high temperature room pressure will be presented.
Liebermann, R. C., G. Chen, B. Li, G. D. Gwanmesia, J. Chen, M. Vaughan, and D. J. Weidner, Rev. High Pressure Sci. Technol., 7, 75-78, (1998).
The origin and evolution of geological structures could bea clue for understanding of crust-mantle interaction.For simulation of geological structures evolutionin connection with deep mantle movements the thermomechanical models of different rheology were used. The next problems were considered: forming and evolutionof sedimentary basins, geothermal evolution of sedimentary cover, interaction of changeble sedimentary cover with crust and mantle lithosphere, reconstruction of deep mantle motions by movements of basement surface, simulation of back-arc spreading, geodynamics of collision zones of lithospheric plates, geodynamics of rifts, P-T parameters distribution in sedimentary cover, crust and mantle lithosphere. The results of modelling were considered on the examples of Pre-Caspian Depression and geological structures of Alpine and Pacific belts.The main results of modelling were: 1. Deep depression can arise above upwelling mantle diapir under some relationships between parameters of model. In the process of evolution the structure of swell is changed by structure of depression above raising mantle diapir. 2. Surface grad T depends on basins evolution history and thickness of the lithosphere layers. 3. Continental crust is involved and sinks in the subduction zone during the collision process. 4. Form of downwelling plate in subduction zone depends on correlation between densities and viscosities of lithosphere plates and asthenosphere. Stretching and compressing stresses in subduction plate depend on correlation of gravity and viscosity strengths in media. 5. It is possible to evaluate the lithosphere thickness and the mantle diapir upwelling by original gravity-geothermal model. The results of modelling give good agreement with geological-geophysical data.
The Sveconorwegian orogen in south-west Sweden west of the Protogine zone is generally characterised by an amphibolite to granulite facies metamorphism and penetrative tectonic reworking of older rocks. East of this zone Sveconorwegian structures and metamorphism are mainly observed in narrow, greenschist facies shear zones which eastwards gradually change to more brittle structures but with a consistent kinematic pattern.The coarse porphyritic Askersund granite in south-central Sweden (within the Transscandinavian Igneous Belt and its western parts partly transsected by Sveconorwegian shearzones) has previously been studied with the AMS technique in order to investigate the transition between magmatic and metamorphic fabric within it (Wikström et al 1997). The previously collected samples in two radial traverses from central granite to peripherical gneiss have been further analysed. The natural remanent magnetisation increase towards the periphery in both traverses. The intensity is generally higher in samples from one of the two traverses. Two ancient components of NRM were isolated during the thermal and alternating field demagnetisations. The most stable one gives the palaeomagnetic pole that indicates its Sveconorwegian age. The less stable component can possibly be Vendian. However the Vendian part of the Fennoscandian APWP is still vague, so we cannot be completely sure. Remagnetisation has thus occurred in the Askersund granite-gneiss complex. The blocking conditions have changed, possibly as an effect from burial of the complex. This thermal event in the granite has previously been undetected. The magnetic fabric has been preserved during this processindicating a minor tectonic impact upon the rocks.
The result of a common Sveconorwegian magnetic overprinting in the actual area has implications for the discussion of how far eastwards Sveconorwegian geological processes can be detected.
Wikström A, Sjöberg B, Kapicka A, GFF, 119, 285-290, (1997).
The Magallanes Fault System (MFS) is interpreted as the westward prosecution of the present-day strike-slip boundary between the South America and Scotia plates, extending for over than 600 km across the Tierra del Fuego in both the Argentinean and Chilean territories. Sparse and poorly documented geological evidences of transcurrent fault and associated thrusting found onshore have not yet allowed a precise mapping of this fault system, and prevailing geometrical extrapolations (onshore westwards prosecution of the South Scotia Ridge) have been considered for tracing the MFS from the Atlantic coast of the Tierra del Fuego to the southern Chile Trench. The precise age of the deformation associated to the presence of the MFS, and the distribution of relative movement between the two plates along its length, are also poorly known. In particular, it is not clear the role that the MFS had in the more general context of the Drake Passage opening and the successive development of the western Scotia plate during Oligocene time.
A geophysical and geological investigation have been designed in order to clarify these aspects, and two campaigns were carried out in February-April 1998 and October-November 1998 in the Tierra del Fuego Island, with the collection of GPS-fixed gravimetric and magnetic data points, the execution of field structural geology transects and petrologic samplings, in an area 30 km x 60 km wide, located just on the east of the Fagnano Lake. This research project, called TESAC (Tectonic Evolution of the South America-Scotia plate boundary during the Cenozoic), is part of a scientific collaboration between Argentina and Italy for the study of the Antarctic region and adjacent seas. Main aim of this program is to analyze the geological structure of the segment of the MFS in the Tierra del Fuego region both onshore and offshore, and to reconstruct the principal aspects and timing of the strike-slip activity occurred in the area during the Cenozoic. The offshore part of the survey is planned for the Austral summer 1999-2000.
Preliminary magnetic and gravity maps of the studied area have been produced, and a deep structural model was constructed across the supposed location of the MFS, along the eastward prosecution of the Fagnano Lake, in some parts deeper than 500 m, where the South America-Scotia plate boundary is supposed to be located. The topographic correction for the Bouguer anomaly map has been computed using the digital elevation model derived from interferometry of a pair of Synthetic Aperture Radar (SAR) images acquired by ERS-1 and ERS-2 (Earth Resource Satellite), because of the unavailability of extensive and precise altitude information. Analyses performed on the acquired data furnished important indications on the presence of a main tectonic lineation, accompanied by sub-parallel structures, elongated in an ENE-WSW direction, prosecuting eastward on the Atlantic Ocean in correspondence of a regional gravity minima, as seen on satellite-derived free-air gravity data. This evidence seems to confirm that the MFS effectively represents the onshore prosecution of the structures associated to the South America-Scotia plate boundary, even if significant aspects related to its role in the tectonic development of the western Scotia Sea region are not yet well understood.
The SG4 superdeep borehole is sited on the Tagil volcanic arc in the Middle Urals. It currently extends to a depth of 5.4 km through a sequence of andestic to basaltic lavas, pyroclastic flows and volcaniclastic sedimentary rocks. A programme of seismic experiments, conducted over the past five years, has been used to investigate the crustal structure in the vicinity of the borehole. The purpose of this poster is to present detailed gravity and magnetic data from the area and show how these have contributed to the integrated interpretation of the three-dimensional geological structure at this key experimental site.The gravity data are available as a grid of residual Bouguer anomalies covering an area of 30x50 km. They define a very clear pattern of residual highs and lows which closely parallel geological strike and are interpreted to be due to lithological variations within the volcanic sequence. Modelling indicates that the observed anomaly pattern can be reconciled with the measured densities of rocks intersected in the borehole if an appropriate eastward dip is assumed. A dense, lava unit intersected at the base of the borehole provides a particularly useful marker. The magnetic data cover a smaller (12x14 km) area but offer significantly higher resolution. They were acquired along ground traverses which were 100 m apart close to the borehole and 250 m apart elsewhere. The anomaly pattern is again dominated by a set of features which parallel geological strike and it is possible to correlate magnetisation contrasts measured in the borehole with the surface magnetic anomaly pattern. This correlation is compatible with that made on the basis of the gravity/density data. Key magnetic horizons are the lavas at the base of the borehole and a boundary intersected at a depth of 3 km between a relatively magnetic pyroclastic sequence and underlying non-magnetic volcaniclastic rocks. The anomalies due to lithological contrasts are intersected by magnetic lineaments with a variety of trends which are interpreted as late faults.The potential field data thus respond primarily to bulk, lithologically-related changes in rock density and magnetisation and provide a means of tracing the surface expression of units identified in the borehole. This is in contrast to the seismic reflection method, where the principal features are interpreted to be generated by the acoustic impedance contrasts associated with shear zones.
Hot-Dry-Rock (HDR) is a concept for using Earth's heat as an energy resource. An artificial underground heat exchanger at a depth of 3.0 to 3.5 km has been created in the granitic basement of the Rhine Graben, close to the town of Soultz in France, by hydraulically connecting two deep boreholes over a horizontal distance of 500 m with the hydrofrac-technique. During a 4-month circulation test, over 240,000 m3 of water have been injected and produced at flow rates of 20 to 25 l/s and with outflow temperatures above 140°C. The net output of thermal power exceeded 10 MW. The experiments indicate that the circulation system in the underground is hydraulically open.Numerical models of coupled heat- and fluid-flow help to understand the observations. At the regional scale, previous models suggested deep fluid circulation from East to West through a sandstone aquifer across the Rhine Graben causing a heat flow anomaly at Soultz. This is in contradiction to recent geochemical analyses of pore fluids. Models of the regional flow-system that agree with the pore-fluid chemistry include deep flow through the granitic basement. These models show that in the area of the HDR heat exchanger fluids generally move upwards. The results of the regional simulations are incorporated as boundary conditions in a local 3-D model of the underground heat exchanger at Soultz. This model is based on flow of two kinds: (1) Darcy-flow in the artificially fractured (stimulated) areas of the granite and (2) channeled flow on natural faults. The stimulated volume is a heat exchanger in the classical HDR concept: it is the hydraulic connection between injection and production hole and it provides the surface for heat absorption of the fluid flowing through the fractured rock. Additionally, the hydraulic fracturing of the granite created a connection with the regional fault system of the graben. Although not part of the HDR concept, this extension of the involved hydraulic system improves the long-term heat extraction process because the regional fault system acts as a buffer for temperature and pressure. The numerical models are used to predict the performance of a pilot plant that is planned to utilize an underground heat exchanger at 200°C.
Index of EUG 10 Volume
Further EUG 10 Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page
Last Updated on Thursday, March 18, 1999.
© 1997 Cambridge Publications