Harve S. Waff

Intergranular basaltic melt is distributed in thin, elogated inclusions

We describe a method to analyze the melt distribution in experimentally produced ultramafic partial melts. It is shown that the melt inclusions can be approximated by ellipses in two dimensions and by penny-shaped ellipsoids in three dimensions. The aspect ratios of these ellipses (the ratio of the minor to the major axis) can in turn be used to calculate bulk physical properties of partial melts. We apply this method to two olivine-basalt samples with 3.2% and 0.75% melt fraction. In the samples analyzed approximately 75% of the melt is contained in inclusions with much smaller aspect ratios than triple junction tubules. The reduction of P-wave velocities calculated for this melt distribution is twice as large as for melt distributed solely in triple junction tubules.

Effects of crystalline anisotropy on fluid distribution in ultramafic partial melts

The textures of experimentally produced ultramafic partial melts show consistent and significant deviations from the morphology predicted by isotropic equilibrium theory. Flat crystalline interfaces are pervasive in these systems and they coexist with smoothly curved boundaries which are predicted by the isotropic theory. Long-duration experimental runs on olivine-basalt mixtures held at pressures between 1.0 and 2.0 GPa and temperatures from 1350°C to 1400°C were evaluated. Scanning electron microscope images of samples with 2.5 or more volume percent melt showed at least 20% of observable grain boundaries to be wetted by the melt. In addition, approximately 60% of the melt tubules occurring along triple junctions in this sample were found to have at least one flat interface, an effect which increases the ratio of permeability to porosity. Both the experimental evidence and theoretical considerations indicate that the flat faces are stable equilibrium or steady state features in these partial melts, and that they are crystallographically controlled. The crystal-melt morphology is influenced by the crystalline equilibrium habit (obtained from minimization of surface energies of individual crystals) under the constraints of polycrystalline aggregates. A dependence of the style of melt distribution on melt fraction was observed in these runs. At very low melt fractions (less than 1 vol %) the texture is dominated by melt-filled triple junctions and mostly dry grain boundaries, whereas at higher melt fractions (but below 5 vol %) more melt pockets and melt films along grain boundaries appear. An interpretation of the observed texture is made, applying established crystal growth and interface theories to steady state partially molten systems. The extensive occurrence of flat or faceted crystallographic faces in partial melts requires major changes in the modeling of their permeabilities, as well as bulk elastic, anelastic, and electrical properties, from existing models of melt distribution. In regions of the upper mantle where olivine lattice preferred orientation is expected (e.g., in the vicinity of mid-ocean ridges) the presence of faceted faces and associated changes in melt distribution will produce anisotropic permeabilities and changes in seismic attenuation.

Electrical conductivity of magmatic liquids: effects of temperature, oxygen fugacity and composition

Abstract The effects of temperature, f O2 and composition on the electrical conductivity of silicate liquids have been experimentally determined from 1200 to 1550°C under a range of f O2 conditions sufficient to change the oxidation state of Fe from predominantly Fe 2+ to Fe 3+ . Oxidation of ferrous to ferric iron in the melt has no measurable effect on the conductivity of melts with relatively low ratios of divalent to univalent cations. Under strongly oxidizing conditions a minor decrease of conductivity is detected inth high ΣM 2 /ΣM + ratios. It is concluded that for purposes of estimating the conductivity of magmatic liquids, f O2 may be ignored to a first approximation. Both univalent and divalent cation transport is involved in electrical conduction. Melts relying heavily on divalent cations for conduction, i.e. melts with relatively large ΣM 2+ /ΣM + ratios, show strong departures from Arrheenius temperature dependence with the apparent activation energies decreasing steadily as the temperature increases. Conductivities dominated by the univalent cations, in melts with relatively small ΣM 2+ /ΣM + ratios, show classical Arrhenius temperature dependence. These observations are discussed in terms of the general characteristics of the melt structure. Compositional variations within the magmatic range account for much less than an order of magnitude variation in electrical conductivity at a fixed temperature. This observation, combined with previous measurements of the conductivity of olivine (A. Duba, H.C. Heard and R. Schock, 1974) make it possible to state with reasonable confidence that melts occurring within the mantle will be more conductive by 3–4 orders of magnitude than their refractory residues. Potential applications to geothermometry are discussed.

Olivine crystallographic control and anisotropic melt distribution in ultramafic partial melts

In the upper mantle, physical properties (permeability, electrical conductivity, and seismic velocity) are highly dependent on the melt geometry. Both flat and curved melt interfaces exist in olivine-basalt partial melts. To further understand anisotropic wetting behavior of partial melting within the mantle, crystal orientation studies have been conducted to determine the Miller indices of flat interfaces, wetted by the melt in olivine-basalt partial melt samples. The samples were equilibrated at pressures of 1-2 GPa, and temperatures of 1350–1400°C for 142–549 hours. Crystal orientation studies show that lower Miller indices of crystals are associated with wetted-flat melt interfaces and melt films. Specifically, (010), (110), and (021) olivine faces in these studies were dominantly wetted, indicating that they may be potentially large contributors to permeability and seismic anisotropy in the upper mantle where partial melts occur with high degrees of preferred crystallographic alignment.

Results of a magnetotelluric traverse across western Oregon: crustal resistivity structure and the subduction of the Juan de Fuca plate

Abstract As part of project EMSLAB, we have collected and analysed wideband magnetotelluric data along an east-west transect in western Oregon. Preliminary modelling of the data using one-dimensional inversions based upon rotationally-invariant earth response functions was followed by finite-element two-dimensional modelling. The models produced indicate the presence of an electrical conductor beneath the Oregon Coast Range dipping eastward at 12–18° from a depth of 23–32 km. We believe that this conductor includes the thrust surface of the subducting Juan de Fuca plate and/or adjacent water-saturated rocks. Its high conductance (about 200 S) is thought to be due to one or more of the following mechanisms: (1) sediments subducted atop and with the Juan de Fuca plate, (2) saline fluids produced by dehydration of the former, or (3) seawater contained within subducted oceanic basalts. There is a distinct possibility that the high conductivity is due primarily to the presence of subducted sediments, in contrast with the notion that the subduction of young, buoyant lithosphere retards sediment subduction at this convergent margin. The conductive layer is overlain by relatively resistive rocks presumed to be accreted oceanic lithosphere. Model-determined resistivities for the upper part of the Coast Range section are in good agreement with deep well-log data. A strong electrical contrast appears in the determinant phase pseudosection between the Coast Range and the Willamette Valley suggesting a structural boundary between the two provinces. A surficial conductor is present in the valley to depths of 1–2 km and is due to alluvial fill. Induction arrow data show the geomagnetic coast effect and a smaller effect by the Willamette Valley alluvial fill.

Verification of five magnetotelluric systems in the mini‐EMSLAB experiment

Given the degree of complexity of modern magnetotelluric (MT) instrumentation, comparison of the total performance for two or more systems is an important verification test. This paper compares the processed data from five MT systems which were designed and constructed separately, and which employ different electrode types, electrode separations, magnetometers, and methods of signal processing. The comparison shows that there is a high degree of agreement among the data from the different systems. The study also demonstrates the compatibility and reliability of the MT systems employed as part of EMSLAB Juan de Fuca (Electromagnetic Sounding of the Lithosphere and Asthenosphere Beneath the Juan de Fuca Plate). This project, proposed by a consortium of institutions, involves not only magnetotellurics studies but also studies of magnetic variation, on land and on the sea bottom. The project calls for the real-time MT systems to occupy stations along segments of a profile in Oregon. A composite profile will be created from the segments. Prior to commencing the main MT profiling phase, one week was set aside in August, 1984, for all groups to record and process MT data sequentially at six sites in diverse geologic terrains; this experiment was called mini-EMSLAB.

Relations of electrical conductivity to physical conditions within the asthenosphere

This paper examines the constraint placed by electrical conductivity on physical conditions existing within the mantle. Determination of bulk electrical conductivities of multiphase materials from knowledge of individual phase conductivities and their relative fractions is discussed in the light of recent studies of the equilibrium phase geometries of partial melts. It is concluded that existing models which are based on assumed geometries offer little refinement over Hashin-Shtrikman variational bounds. The possible effects of the gravitational field on phase geometry in partial melts and the resulting conductivity anisotropy are considered. At this time it appears that minimum melt fractions can be safely estimated from a knowledge of Earth conductivity combined with laboratory data on melt and crystalline conductivities and the Hashin-Shtrikman upper bound. The question of whether or not the asthenosphere corresponds to a zone of partial melting is also addressed.