A triaxial fluxgate magnetometer for use on the sea floor has been built and tested. This magnetometer, which is small and easy to handle, is housed in a pressure-tight glass sphere. The instrument is equipped with a timed release which enables free-fall installation and automatic recovery. Maximum period of measurement is 60 days with 3-minute samplings and the accuracy of the measurement is ±0.8nT. This corresponds to the error of ±1 least significant bit unavoidable in digital conversion using a 16-bit AD converter.

View full abstract

A remote sensing technique has been developed for high resolution mapping of subsurface structures using ratio-wave transmissions through the ground. A transmitter and receiver are lowered in boreholes drilled on opposite sides of the underground region to be imaged. The measured attenuation and velocity of the electromagnetic waves propagated between these boreholes is determined by the electromagnetic properties of the intervening media. Maps showing the distribution of these electromagnetic properties between the boreholes can be produced by tomographic reconstruction-a technique used for several years in medical X-ray imaging. In some cases these maps can be directly related to geologic structure. This remote sensing technique has a variety of possible applications including mapping underground fracture systems; locating tunnels or water-bearing strata; mapping burn fronts during in situ coal gasification and monitoring in situ oil shale retorts.

View full abstract

A number of electromagnetic instruments with fixed frequency transmitter-receiver has been developed by Centre de Recherches Geophysiques (Garchy, France) in order to measure magnetic susceptibility and electrical resistivity of ground.
A dipole-dipole induction tool (SH3) was built for some meter penetration from the surface (spacing=1.5m, frequency=8kHz) and has allowed electrical resistivity and magnetic susceptibility mapping to be applied to archaeological exploration.
A “ROMULUS” probe (spacing 0.85m, frequency=4kHz) has followed as a borehole version (diameter 0.04m) for mineral exploration by Bureau de Recherches Géologiques et Minières (Orleans, France).
Another borehole instrument called “ERIC” was then designed on the same principle (frequency=1kHz) but with variable transmitter-receiver spacing (5m, 10m, 20m) for ten meter scale investigation of conductive bodies around the hole.
The method, its field of application and design of equipments is described. Several examples of shallow investigation surface maps and borehole induction logs are presented.

View full abstract

In the summer of 1981, a scalar audio-magnetotelluric survey was completed in four locations in Northern Ontario. Apparent resistivities, for frequencies from 13Hz to 8, 570Hz, were measured at 101 stations.
A survey in Moody Township near Lake Abitibi clearly outlines the clay properties and thickness, and also shows that, at the lower frequencies, we are able to map the electrical resistivity of the basement beneath the clay and/or esker cover. Over a region, tentatively identified as metasediments, the bedrock is seen to be strongly anisotropic, reflecting anisotropy of the sediments themselves.
In a second region, Marter Township near Engelhart, two survey profiles mapped a region of thickening clay-rich overburden as well as located high resistivities over a large esker. These profiles were taken in an area where the overburden had been drilled in a program of the Ontario Geological Survey. Our results can be interpreted to be in general agreement. A similar example reported from eastern Manitoba shows that the near-surface apparent resistivity is typically as low as 10ohm-m reflecting the conducting clay. The resistivity then rises sharply to values of 1, 000ohm-m to 10, 000ohm-m reflecting the Precambrian bedrock.
Surveys in Bowman Township near Matheson, were conducted on a two-dimensional grid and the data consistently show the presence of four layers, a thin resistive surface layer a few meters thick, followed by a very conductive layer of clay. Bedrock is at a depth of 50-100m and is very resistive. This survey gives a map of the resistivity and the thickness of the clay-rich glaciolacustrine sediments and clearly outlines near surface features such as an esker.
In addition to the AMT survey, we have collected clay samples from sites in Larder Lake and in Marter Township to study their electrical properties. Laboratory measurements show that in the Kirkland Lake area, the clays have resistivities on the order of 20ohm-m in agreement with the high frequency AMT data over clay-rich overburden.

View full abstract

Two major factors hinder mineral exploration at this present time: (1) a conventional resistivity borehole log is often more indicative of the resistivity of the pore-filled fractured rock in the close locale (<1m) of the borehole than the actual resistivity of the layer in which the probe is located; and (2) ground-based em techniques, both natural and controlled source, are often unable to locate a mineralized zone beneath another mineralized zone.
In this paper, the theory is presented for the basis of a conceptually new type of borehole technique based on the ratio of the measurement of the natural horizontal magnetic field variation to its gradient with depth, down the hole, viz. V_{x, d}(ω)=H_{x, d}(ω)/(∂H_{x, d}(ω)/∂z). Defining the “downhole apparent resistivity”, ρ_a(ω, d) by
ρ_a(ω, d)=ωμ₀|V_{x, d}(ω)|²
it is shown that, for a 1D earth structure, as ω tends to infinity, then ρ_a(ω, d) tends to the actual resistivity of the layer in which the probe is located. Also, ρ_a(ω, d) is independent, in the 1D case, of any structure above it, and weakly dependent in the 2D case. The technique has the benefit of most borehole methods of being far superior at resolving structure at depth below it, e.g., a second good-conducting zone, than equivalent ground-based methods (e.g., MT and/or GDS).
Application of the technique to some theoretical 1D and 2D structures is presented, as well as a discussion of the feasibility of constructing the necessary sensor for the proposed technique.

View full abstract

The principles of electromagnetic induction in the earth by ocean currents are reviewed with an emphasis on their interaction with the conducting earth and their modal structure. This is illustrated theoretically by examining a Kelvin wave off of the California coast and experimentally by separating seafloor measurements into ionospheric and oceanic parts. An attempt is made to estimate the electrical conductivity of the oceanic lithosphere from the polarization of the magnetotelluric fields near coastlines, yielding an upper limit of 0.001S/m. Possible interpretations of this value in terms of local current channeling are discussed.

View full abstract

A comparison of results from field stations and a laboratory analogue model shows that the electromagnetic response of the Queen Charlotte Islands region is primarily related to the distribution of land and seawater. Almost all of the induction arrows from the simple land-sea model, both in phase and quadrature, have similar directions to those available from the five field sites over the period range of 4 to 120 minutes. The few arrows which differ in direction are influenced by conductive crustal features, not included in the simple model. In contrast to the excellent agreement in direction, the amplitudes of the coastal induction arrows differ between the model and field results: the model amplitudes are too large at short periods (4min) and too small at long periods (120min). Thus a more complex structure is required for the lithosphere and upper asthenosphere for the oceanic and land regions than the uniform layer of the present laboratory model. The effect of deflection of the induced electric currents is observed in the model near all capes and bays along the coastline and is very large near Rose Point at the northeast corner of the islands. Contour plots and three-dimensional views of the magnetic and electric field components clearly demonstrate the difficulty in locating a representative reference station for correcting marine or land magnetic surveys in this type of region.

View full abstract

Geomagnetic variation data have been measured along a profile from central Vancouver Island, across the continental shelf to the deep seafloor on the Juan de Fuca plate. As expected, the maximum response in the geomagnetic coast effect occurs near the shelf-edge and the fall-off inland and seaward is observed. The direction and magnitude of the real part of the induction arrows vary slightly with period between 30 and 103 minutes. The induction arrows are co-linear at all sites and are aligned approximately normal to the shelf-break trend. At the seafloor sites, the imaginary part of the induction arrows has a large variation with period, reversing its direction at periods less than 35 minutes. At the land sites, the imaginary part increases in magnitude with increasing period.
The responses of several numerical models were calculated using the finite difference algorithm of Brewitt-Taylor and Weaver. The comparison of the calculated and observed responses shows that a good conductor is required at sub-lithospheric depths beneath the profile. The addition of a conductive wedge of sediments beneath the continental shelf improves the fit. The best-fitting model incorporates this wedge and a conductive, shallow-dipping slab representing the Juan de Fuca plate subducting under Vancouver Island.

View full abstract

The electromagnetic response of the Assistance Bay region (N. W. T., Canada) is studied using laboratory analogue model and field station measurements. Magnetic field components for the frequency range 0.1-15Hz, which includes the well known lightning induced Schumann Resonances, were measured at 8 sites along a 6km line through the central region of Assistance Bay and roughly perpendicular to the Cornwallis Island coastline. Two of the stations were on land and six were off-shore on the ice. Although the field data were examined in the three bandwidths 0.1-6Hz, 6-10Hz, and 10-15Hz, the present work deals only with the 8Hz Schumann Resonance component within the 6-10Hz range. Calculated average amplitudes and phases of the three magnetic field components show the relative response of the region at stations along the 6km line. The response is discussed in terms of the effect of the on shore topography, the coastline, and the bathymetry.
A laboratory analogue model of the Assistance Bay region was constructed and model measurements carried out for a simulated frequency of 8Hz for E-Polarization (electric field in the east-west direction) and H-Polarization (electric field in the north-south direction) of an overhead uniform horizontal inducing field. The magnetic field amplitudes H_x, H_y, H_z for both polarizations for 10 traverses perpendicular to the coastline are presented and the anomalous fields attributed to the response of the coastline bay, capes, bathymetry and source field polarization.
The analogue model and the field station coast effects (vertical to horizontal field ratio) are compared for the model traverse coinciding with the 6km line of field stations. The analogue model results show a maximum enhancement of approximately 1.3 directly over the coastline with rapidly decreasing values seaward followed by a second enhancement over the sharp ocean depth gradient beyond the 55m depth contour. The field station results do not show maximum at the coastline, but rather an increasing value seaward, reaching a value of 1.7 at the station over the sharp ocean depth gradient near the 55m depth contour, indicating the effective coastline to be at the deep water edge. This difference may in part be attributed to the shallow water conductivity being smaller than simulated in the laboratory model. The effect of the 2m layer of ice, as well as the accumulation of fresh water from streams in the Assistance Bay region, may lead to a significantly lower effective conductivity than assumed for the analogue model. An effective decrease in the shallow water conductivity would lead to decreased enhancement at the coastline and relative increased enhancement off-shore at the deep water edge. Other factors possibly contributing to the seemingly anomalous seaward coast effect could be subsurface geologic and permafrost structure, and the nature of the source field. The laboratory overhead inducing source is adequate for simulating ionospheric currents of large extent but may not be suitable for the Schumann source field associated with distant lightning strokes due to lack of a significant vertical electric field component. This source field effect requires further study.

View full abstract

Magnetometer array studies have been undertaken in Finland since June 1981 with 31 modified Gough-Reitzel magnetometers, which are on loan from the University of Münster to the University of Oulu. Two arrays were operated during 1981. They were situated in the central part of the Fennoscandian Shield. The first array was operated for about two months in summer. Half of the array was then moved, and the second array operated for two months in autumn.
Two events from the second array have been digitized and analyzed. Magnetograms, polarization ellipses and induction vectors have been drawn for these events. The qualitative interpretation shows that the area under investigation is electrically inhomogenous, and three anomalous zones have been detected.

View full abstract

Magnetotelluric and geomagnetic depth-sounding surveys were made in the Harz Mountains, Germany, to investigate the conductivity structure across its northern edge as well as the extension of crystalline intrusions inside the mountains. Magnetic and electric pulsations were recorded along a north-south-profile of 80km length. Because the statioins were recording simultaneously, transfer functions for anomalous horizontal and vertical magnetic variations could be obtained with the north end station defined as normal. Their amplitude and phase indicate a large-scale discontinuity in conductivity.
This model explains well the smooth behaviour of the magnetic transfer functions but not the telluric impedances and their distinct changes from site to site. For this reason a special study of the effects of local anomalous conductivity zones on the telluric field was carried out. The impedances found there are influenced by strong direct-current distortions. Furthermore, a vertical electric field was detected inside the uppermost layer. Because it is also associated with a local anomaly, it can be used to determine the orientation and magnitude of that anomaly.

View full abstract

Employing the finite difference technique the concept of a secondary inducing source field at an altitude between 180 to 350km in south western Nigeria has been examined. The study compared numerical results with field results obtained by ONI and AGUNLOYE (1976) at Ibadan, Okitipupa and Igbanke. At both Ibadan and Okitipupa the numerical results tend to support the presence of a secondary inducing source field at 210 km and 320km respectively, while the secondary inducing source height at Igbanke was found to be closer to that of the equatorial electrojet.

View full abstract

In order to investigate the electrical conductivity structure of a subduction zone, and to obtain the vertical profile of the conductivity beneath an old oceanic plate near the subduction zone, we observed electric and magnetic variations on the sea floor of the Northwest Pacific ocean over two months in the summer of 1981. Magnetometers and electric field recorders were installed at six sites across the Japan trench. The outermost station is located about 600km from the coast of Japan and 450km from the trench axis.
Short period variations in the geomagnetic field are greatly attenuated on the deep sea floor outside the trench, while they are significant in the records on the continental slope. Clear correlation is seen between the variations in the vertical and the east component on the slope, indicating that the geomagnetic variations are severely affected by the coast effect. Unusually large magnitudes of Parkinson vectors that exceed unity were obtained with their directions normal to the trench axis.
A three layer structure was tentatively assumed for inversion of magnetotelluric soundings. The result suggests the existence of an electrically conducting layer at depths of 150-170km beneath the oceanic plate of about 125 my of age. In comparison with the existing conductivity profiles in the Pacific, this seems to support the hypothesis that the conducting layer becomes deeper with the plate age, reflecting the evolution of the oceanic lithosphere.

View full abstract

Observations of geomagnetic variations were made at 14 sites in the Northern Honsyû of Japan in 1981. At many sites variations in the electric field were simultaneously recorded. The objective was to investigate the electrical resistivity structure of the island arc, since the Northern Honsyû has various features of a typical island arc.
Variations in the magnetic and the electric field at frequencies from about 10^-4Hz to 17kHz were observed at some of the stations by use of fluxgate magnetometers as well as induction type ones, although at most of the sites observation covered the frequency range of 10^-4 to 10^-1Hz. Examining the transfer function of the geomagnetic variation, it has been revealed that the coast effect is predominant, having unignorable influence on the geomagnetic variations even at such a remote station as 150km from the coast of the Pacific. Magnetotelluric soundings suggest that the lower part of the crust is of low resistivity. On the Japan Sea side, a low resistivity layer of 60ohm·m appears at a depth of about 20km, which coincides well with the depth of a seismological boundary, the Conrad disconituity, between the upper and the lower crust.

View full abstract

A 1200km discontinuous M-T profile from Buenos Aires to San Miguel de Tucuman, gives a great complexity of results concerning the structure of the crust, the lithosphere and the upper mantle. At the north-west end of the profile appears a new geodynamic process associated with the Nazca subduction plate, generating one important geothermal area of about 200km diameter.

View full abstract

The channelling effect is studied with the aid of a numerical model in which two oceans are represented by conductive thin sheets, one occupying a half-plane and the other a quarter-plane, situated on the surface of and in electrical contact with a conducting half-space. The oceans are connected by a narrow conductive channel through which induced currents can be channelled. The electromagnetic response of this model to a regional magnetic field which is periodic in time and uniform in space, oriented either perpendicular or parallel to the channel, is calculated using the algorithm developed by Dawson and Weaver. The limitations of the numerical method restrict the study to periods between 1.5min and 10min, and only the latter period is allowed if the thin conductive sheets are to be interpreted as oceans. It is found that for a field of period 10min both horizontal deflection (into the channel) and vertical leakage (into the underlying medium) of the induced currents are significant near the channel entrances. The actual paths of the anomalous currents in the surface plane are displayed by plots of the anomalous surface current vectors. These show that the quadrature currents are very definitely channelled from one ocean to the other even when the regional flow is perpendicular to the channel. The effect is less marked with the in-phase currents but channelling is still present. At a period of 1.5min, however, the channelling effect has all but disappeared. Local induction appears to predominate when the period is small. Difference magnetic field vectors and induction vectors in the region of the channel are also shown, and comparisons are made with the corresponding vectors calculated for the same model but with the channel removed. It is concluded that the presence of the channel can be clearly recognized from the behaviour of these vectors.

View full abstract

We showed in the past how the anomalous magnetic fields in the Rhine Graben can be interpreted by induction in elongated sedimentary conductive layers (DUPIS and THERA, 1982); these layers existence was confirmed by magnetotelluric (M. T.) soundings and by electrical logs. The problem is different for the Northern Pyrénées because of a lack of petroleum logs and of a significant number of M. T. soundings. The model developed in this paper is the result of a geological interpretation from a geological map, complemented by electrical soundings, M. T. soundings and DGS (differential geomagnetic sounding) information. It shows that the results obtained for the anomalous magnetic field (BABOUR and MOSNIER, 1979) and first interpreted as due to “channelling currents”, are perfectly described by the effect of induction in local structures.

View full abstract

A two-dimensional (2-D) structural transition from a flat valley floor to a mountain slope, with a presumed reverse fault at its base, was investigated by the audio-magnetotelluric sounding method. By calling on previously available geological information, it was possible to reformulate the problem so that only two basic questions remained to be answered: (1) does the structure effectively involve a fault, and (2) if so what is the dip angle of this fault? Since the soundings away from the mountain furnished the 1-D valley structure, it was possible to construct a 2-D model in which the only specified parameter was the unknown dip angle. Even the absence of a fault could be simulated in the model by giving this parameter a definite value. The model responses were computed numerically by the technique of finite differences and compared with the measured response curves, yielding quite clear answers in favor of a reverse fault dipping at an angle of less than 30° under the mountain. It was also possible to secure some information concerning details of the valley structure.

View full abstract

A comparison of induction arrows calculated from analogue model and field station measurements is reported for fifteen locations in the Newfoundland region of Atlantic Canada. The comparison is made at four periods 100s, 300s, 900s, and 1800s. The differences between the total induction response as represented by the field results and part of the induction response due to the conductive ocean alone as represented by the model results are presented in the form of difference induction arrows. These difference arrows are representative of the anomalous conductivity associated with the tectonic features of the region. The major tectonic structures identified as contributing to the anomalous response in Newfoundland are the Carboniferous sedimentary basins both onshore and in the Gulf of St. Lawrence and a conductive feature in eastern Newfoundland probably associated with the ancient margin of the proto-Atlantic Iapetus Ocean. There is also a suggestion that the crustal conductive zone underlying the Nova Scotian Shelf extends to the Grand Banks off Newfoundland.

View full abstract

The behaviour of the electromagnetic fields over the Hainan Island region of the South China Sea is studied using a scaled laboratory analogue model. The model source frequencies used simulate periods of 5min. to 30min. in the geophysical scale. To examine the electromagnetic field variations due to the irregular coastlines, the deep ocean, a narrow ocean channel, and an atoll (modelled as a seamount), field components for individual traverses over the model are presented. The coast effect for the bay coastline is visible only for short periods at which the depth of the bay is at least. 1 skin depths. The response of the seamount, overlain by a 500m depth ocean, differs greatly from the response of an island. Electric current flow over the seamount, confined to this thin layer, produces a quadrature field directly opposite to that for an island. Buried features such as the seamount and depth profile of the deep ocean show the largest in-phase response at the lower frequency where the overlying ocean is much less than. 1 skin depth thick.

View full abstract

This paper gives a summary of magnetic and magnetotelluric measurements in central Germany. Except for the results concerning the conductivity structure in this area, criteria for the rotation of the anomalous horizontal magnetic field and the elimination of DC-distortions of the electric field are described.

View full abstract

Electromagnetic response data for the Earth covering a wide frequency span are analysed using linearised inverse theory. The response measure adopted is the natural logarithm of the complex apparent resistivity. Model finding and interpretation takes place in plane geometry, a simple transformation converts the solutions to spherical geometry for global problems. Thus the methods will apply to local problems, where the combination of AMT and MT data provides several decades of frequency, and to global problems of which this paper is an example. The global data used here are drawn from four sources and cover more than 6 decades in frequency (3.5×10^-9 to 8.7×10^-3Hz).

View full abstract

The geomagnetic storm-time variation Dst allows information concerning the global distribution of electrical conductivity in the upper mantle to be obtained. Such storm events can be analysed for periods up to 10 days following the storm commencement. The penetration depth of the associated induced currents within the Earth is over 1000km. Global response estimates were derived from storm data analysed by MARSHALL (1980) covering the frequency range 0.07 to 2cpd. The response data were inverted to give conductivity profiles within the upper mantle using the algorithms due to BAILEY (1970a, 1970b), PARKER (1980) and PARKER and WHALER (1981), and FISCHER et al. (1981a). The results obtained from these inversions are consistent, and reveal two major discontinuities in conductivity of several orders of magnitude at depths between 650-750km and 950-1050km. It is also suggested by the results that the uppermost region of the mantle is poorly conducting.

View full abstract

The method of the transient electromagnetic field migration based on the “reverse” continuation of the field within the conducting Earth is developed. The properties of the migrated field and the possibilities of electromagnetic migration method application for the geoelectric inverse problems solution are studied in detail.

View full abstract

Laboratory measurements have been made of the impedance of saturated samples of Berea Sandstone and Spirit River Formation sandstone throughout the frequency range of 5Hz to 13MHz. When the data from a saturated sample are presented as a complex impedance plot the frequency dispersion produces a “non-ideal” distribution of points in the form of a depressed and slightly distorted semicircle and an inclined straight line. It is possible to separate the low frequency electrode response from the bulk sample response, thereby isolating the effect of polarization at the sample/electrode interface. The frequency response of the sample can be modelled by an equivalent circuit proposed by RAISTRICK et al. (1976) which includes a frequency dependent admittance term. Different samples and varying levels of water saturation cause changes in the complex impedance plots which can be related to this admittance term. The form of this term and its significance with respect to rock properties needs further investigation.

View full abstract

Kinetics of the oxidation process of titanomagnetites are described with the aid of topochemical equations using measured values of the activation energy, the order of the reaction, and the frequency factor. The kinetics were used to derive the changes of physical properties.

View full abstract

From the Urach 3 deep drilling project and the Hot Dry Rock project in Falkenberg, both in West Germany, we have investigated rocks taken from cores and determined the porosity, electrical conductivity, and permeability. The porosity was measured with a Hg-porosimeter, which allows differential determination of the pore radius distribution from 35Å to 75000Å. Measurements were done on 13 samples from 1635 to 3257m depth from Urach 3, and on 7 samples from 25m to 300m depth from the Falkenberg site. Depending on the rock condition, whether fresh or hydrothermally detiorated rocks, the porosity values vary from 0.6 to 1.9 vol % for the Urach gneiss samples. This can be correlated to the mineral composition, mineral grain size, mineral alteration, and also to the measured borehole resistivities. The more homogeneous Falkenberg granite samples differ in porosity only from 1.4 to 1.7 vol %. The only variable is the number of cracks in the samples. In both cases an inverse correlation of the porosity with the V_p velocity could be found, e. g. high porosity is correlated to low V_p velocities.
Measurements of the electrical conductivity on those samples as a function of pressure, temperature, pore space filling and pore pressure show that the dominating factor for the conductivity is the percentage of pore space filling. The presence of ore minerals, such as pyrite, makes no significant contribution to the conductivity. The borehole resistivity log data could be reproduced in the laboratory. The observed large differences in the rock resistivities (50 to 2000Ohm·m) could be interpreted by different pore space fillings.
Correlation of the resistivity with porosity allows calculation of the permeability.

View full abstract

The electrical structure of the crust and upper mantle beneath three regions of Scandinavia has been delineated by the magnetotelluric (MT) and the horizontal spatial gradient (HSG) techniques. The analyses were applied to data recorded by the Scandinavian IMS magnetometer array complimented by telluric observations.
Models compatible with the response functions observed in northern Sweden and northeastern Norway/northern Finland are distinctive by exhibiting: (i) a negligibly small resistivity contrast across the seismic Moho; and (ii) the unequivocal existence of an electrical asthenosphere beneath both regions. In definite contrast, the response function observed in southern Finland demands a highly conducting layer (of resistivity around 10-50Ωm) in the lower crust, and an order of magnitude increase in resistivity on entering the mantle. This increase is at a depth compatible with the known seismic Moho for the region.
It is not possible to make a quantitative estimate of the depth to the electrical asthenosphere beneath southern Finland, due to lack of long period information, but a qualitative measure indicates that the asthenosphere depth increases with increasing distance towards the centre of the north European craton.

View full abstract

The problem of estimating the viscosity of the upper mantle is investigated, based on the approximate relation between the conductivity and the viscosity with due regard to the activation energies. The results of combined analysis of postglacial uplifts and deep electromagnetic soundings on the Baltic Shield are compared.

View full abstract

Geomagnetic induction studies of the electrical conductivity of the sublithospheric mantle are currently underway at the University of Washington. This work is based upon the analysis of magnetic time variations in the frequency range 0.01 cpd to 0.2 cpd. The data are obtained from a set of 76 magnetic observatories distributed as uniformly as possible over the earth's surface.
Pioneering work by Banks and others has shown that the P₁⁰ spherical harmonic adequately describes the major part of magnetic field variations in the above range of frequencies. The dominance of the zonal ring current term has been exploited by a number of workers in attempts to determine the global distribution of electrical conductivity. The magnetovariational transfer function, i. e. the vertical over horizontal ratio of the Fourier components of the field variations, is estimated. This function can be related to the magnetotelluric impedance function, and hence can be inverted to yield conductivity structures.
Application of one-dimensional Backus-Gilbert inverse theory to data obtained at Tucson and Honolulu has shown a resolvable difference in conductivity structures at the two sites. This work serves as the basis for the current effort to delineate lateral mantle conductivity variations using the world-wide distribution of magnetic observatory data.
The data for this work consists of 1358 station years. Each station-year contains up to 366 daily mean values for each of the three magnetic components H, D and Z. Much of these data exist as simultaneous timeseries from sites distributed around the globe. The simultaneity of data will allow us to test the validity of the P₁0 assumption and identify and remove noise. A statistical basis for the treatment of spurious data will also be discussed.

View full abstract