The present World Magnetic Survey is described and discussed in relation to defining the earth's main field and secular change. Some effects dependent on geomagnetic noise levels affecting the survey are indicated. Difficulties incident to the extrapolation of faulty surface-field data to the surface of the core are noted. The extrapolated results of previous surveys are used to indicate the possible principal descriptive features of surface fluid motions that are derived for the outer core.

View full abstract

Mobile recording magnetometers were developed during and after World War II. These new instruments have permitted the recording of long magnetic profile lines. These long magnetic profiles have clearly shown that two distinctly different types of anomalies occur. Many local anomalies are found with amplitudes up to a few thousand gammas and widths up to a few hundred kms. Very large regional anomalies are found with amplitudes of tens of thousands of gammas and widths of several thousands of kms. There are no noticeable anomalies between there two types.
It is suggested that world magnetic charts made to a scale of less than 1/30, 000, 000 should be smoothed so as to eliminate the effect of all local anomalies on the isoline shapes. The characteristics of local anomalies could be indicated on the chart by a color coded scheme which in effect would indicate a chart error which might be expected because of local anomalies in the area.

View full abstract

All available surface and airborne observations of magnetic declination from 1955.0 and of vector force components from 1945.0 up to mid-June 1964 have been brought up to the epoch 1965.0. The reduced data have been subjected to a series of four spherical harmonic analyses in such a way that the apparent external part has been reduced whilst retaining a reasonable fit to the observations. Three similar analyses have been performed on the extrapolated time derivatives of X, Y and Z at 113 magnetic observatories. Results are given in terms of Schmidt quasi-normal coefficients up to the 8th order and degree for the main field and up to the 6th order and degree for its rate of change.

View full abstract

Geomagnetic local anomalies measured in a horizontal plane square can be represented by a finite two-dimensional Fourier's series. The series is finite according to the finite number of information given by a measurement.
This representation of the anomalous field allows the direct calculation of the surface relief of a magnetized layer capable to produce the measured anomalies. If an average depth of the relief and the magnetization of the layer are given, the shape of the relief is unique.
A consideration of the spectrum of this representation and of the corresponding spectrum of the anomalies measured in a profile offers the possibility to give an interpretation of irregular spatial fluctuations of the anomalous field in profiles by a description of the character of the magnetization inhomogeneity in the underground. Although the anomalies are measured in profiles the magnetization distribution used for interpretation is three-dimensional without a preferred direction.

View full abstract

The critieron for separating the earth's main field from magnetic anomalies of crustal origin is reviewed. A concise resume of the nature of magnetic anomalies over the Mid-Atlantic Ridge, the anomalies caused by great elongated but unidentified bodies and the magnetic anomalies over seamounts is presented.

View full abstract

Studies were made of accurate, total field aeromagnetic data of excellent uniformity over an area in N. W. Ontario extending from 48°45′N to 53°N latitude and from 86°W to 94°W longitude. The rocks in this area are of early Precambrian age (Archaean era). The regional effects from the data were removed. The residual values were then analytically filtered to obtain the total field stripped of its near-surface, high amplitude components. Eighty-five individual anomalies were selected and analyzed to determine the horizontal dimensions and the depths to the tops and bottoms of the causative bodies and the magnitude and direction of total polarization vectors associated with the bodies. The bodies were assumed to be vertical rectangular prisms of arbitrary polarization. A brief summary of the method used is presented.
The horizontal dimensions of the bodies vary from 0.8km (.5mile) to 5.1kms (3.2miles). The depths to the top of the bodies in the filtered aeromagnetic map are found to be in the range 4.8kms (3miles) to 8kms (5miles) from the surface of the ground. The depths to the bottom of the bodies are between 17.7kms (11miles) and 24kms (15miles) for most of the anomalies, with a mean depth of 20kms (12.4 miles), which is interpreted as the mean depth to the Curie point geotherm. The horizontal vectors of the polarizations are subparallel to the magnetic lineation of the filtered aeromagnetic map. An interesting and surprising result of this analysis is the detection of the presence of alternatingly normal and reverse polarization roughly in the East-West direction. The mean inclination and declination for normal polarizations are 67.6° and 258.7° respectively whereas those for reverse polarizations are -67.9° and 81.3° respectively. The consistency in the directions of magnetization in the two clusters for normal and reverse polarization indicates strongly the absence of any dominant component of magnetization parallel to the present field. On the basis that the magnetizations are permanent and the reversals are due to field or selfreversal, the calculated pole position for the rocks of Archaean era is found to be at a latitude of 31.5°N and a longitude of 136.9°W. The intensity of magnetization of the bodies lies within a wide range from .0006 to .349cgs emu with the average of .055cgs emu.

View full abstract

The general problem of analysing the depth of near surface magnetic sources which contribute to the magnetic anomaly spectrum has been analysed in terms of crossings per unit interval. Assuming a random distribution of either magnetic poles or magnetic dipoles aligned in any of three orthogonal directions, explicit functions have been derived expressing the crossings per unit interval for three orthogonal field components directly in terms of either the depth to a thin layer or depth to the top of a thick layer.
The statistical weakness of the method is explained: it appears to be most useful in supplementing other statistical techniques. The speed and simplicity of the method, however, makes it an attractive possibility in extracting preliminary information from the large number of World Magnetic Survey aeromagnetic three-component profiles now being acquired.

View full abstract

The secular variation in the geomagnetic field revealed from geomagnetic data during the latest Century may be characterized by the following five major characters; i. e. (a) decrease of the moment of geomagnetic dipole with rate of 0.05%/year, (b) westward precessional rotation of the dipole with speed of 0.05 degree/year, (c) northward sifting of the dipole with speed of 2km/year, (d) westward drift of the non-dipole field with speed of 0.2 degree/year, and (e) growth and decay of the non-dipole field with annual rate of 10γ/year in the order of magnitude.
Based on the concept of the hydromagnetic dynamo of the earth's core, a possibility of theoretical interpretations of the five main characteristics is discussed. (a) and (b) characteristics are hardly possible to be quantitatively interpreted at present. (c) characteristic may be interpreted as due to a simple harmonic oscillation of S₂⁰ component. (d) characteristic has been interpreted by Bullard et al as due to the electromagnetic coupling between the conductive liquid core and the less conductive solid mantle. (e) characteristic may be interpreted as due to growth of magnetic fields caused by upwelling of the strong toroidal field (T₂⁰) by local convection motions and their decay.

View full abstract

Analyses of geomagnetic variations, having a period of about 11 years, have been made using the data from 27 observatories for the last 55 years. Using a particular differentiation process together with a numerical high-pass filter, in-phase variations of about 11-year period have been extracted for all the observatories against a background of the gradually changing large scale secular variation. Fourier analyses of these modified mean annual values revealed that the amplitude of the 11-year variation in the north component varies from about 2γ to 17γ according to the observatory. Approximating the 11-year variation field by a P₁ field, the coefficient of the first spherical harmonic was determined by a least square calculation as 8.5γ for the north component and 9.1γ for the vertical component. The ratio of the coefficient of internal origin i to that of external origin e was obtained as 0.16±0.08. From this, the electrical conductivity in the lower mantle has been estimated as 6×10^-10 e. m. u. at about 1600km depth.
It has also been shown that the effect of the 11-year variations on analyses of the rates of change in the field are negligible, and that the variation is likely to be caused by the growth or decay of the equatorial ring current.

View full abstract

Since the magnetocrystalline anisotropy constants (K₁, K₂) and the magnetostriction coefficients (λ₁₀₀, λ₁₁₁) of titanomagnetites change appreciably with temperature in the low temperature range, the magnetic susceptibility and remanent magnetization of these minerals depending on these parameters also change markedly with temperature. In particular, the magnetite-rich titanomagnetite has the isotropic point, at which K₁=0, at a low temperature. The marked reduction of magnetocrystalline anisotropy energy around the isotropic point results in some remarkable magnetic phenomena. These phenomena are: -
(a) On cooling, IRM almost vanishes or is sharply reduced at the isotropic temperature, and on subsequent heating, once diminished magnetization is recovered to a certain extent above the isotropic temperature;
(b) By heating from a temperature below the isotropic point to the room temperature in a magnetic field (H), the sample acquires remanent magnetization;
(c) As IRM almost vanishes at the isotropic temperature on cooling and its recovery on heating is of a small amount (<1/5) while TRM's recovery rate is 80%-90%, the procedure of cooling down a rock sample to a temperature below the isotropic point can work as an effective method of cleaning IRM and VRM of magnetites and titanomagnetites.

View full abstract

The reversible susceptibility of large single crystals of hematite carrying different intensities of various types of remanent magnetization, i. e., IRM, TRM and ARM has been measured in a small A C magnetic field (H≈3œ) using a transformer bridge. In a stoichiometric αFe₂O₃ specimen, reversible susceptibility has also been measured before and after cooling and heating through the Morin transition. The results of these experiments have revealed that reversible susceptibility decreases steeply with increasing IRM but that it is independent of the magnitude of TRM, ARM and memory of any types of remanence except when TRM and ARM are almost saturated. It may be concluded that the blocking mechanism of TRM and ARM is similar to the mechanism of memory of remanence but completely different from that of acquisition of IRM. It may be inferred that stable TRM, ARM and memorized remanence are carried as magnetically hard particular regions in the crystals while soft remanence is due to the more usual process of the domain wall movement.

View full abstract

The results of a deliberate search for stable secondary magnetization in Precambrian igneous rocks of St. Francois Mountains, Southeast Missouri, U. S. A., are described. These rocks, at least one billion years old, possess stable components of natural remanent magnetization, which are not always of thermoremanent origin. The results of thermal demagnetization and of some subsidiary experiments show that rhyolite porphyries of Flint Hill have been remagnetized by strong fields due to lightning and are of little use for paleomagnetic investigations. Similar experiments conducted on felsites and bedded tuffs of Johnson Shut-Ins area reveal the presence of stable and significant magnetization of a direction similar to that found by Scharon, Hayes, and Anderson for the whole of St. Francois Mountains. The granite porphyry from the same area carries a stable component of natural remanent magnetization, which is of secondary origin, but is of little significance for the paleomagnetic studies of the whole system of rock units.
It is concluded that the rock units investigated carry no stable secondary magnetization which masks their original magnetization and leads to wrong conclusions about the direction of Precambrian geomagnetic field. From this it is inferred that the natural remanent magnetization of all other rock units of St. Francois Mountains is unlikely to be masked by stable secondary components.

View full abstract

The problems in the crystallographic structure of maghemite is first critically reviewed and magnetic evidences for a superstructure presented. A nondestructive method was used to find maghemite's Curie point and the latter shown to be coincident with its crystallographic transition temperature. The implications of this discovery for rockmagnetism are discussed, including the possibility of a pressure induced remagnetizaiton.

View full abstract

The first part of this paper describes the investigation of thermomagnetic effects in three basalt samples. It was found that when a sample was heated in air the ferrimagnetic mineral component changed from a homogeneous member of the Ulvöspinell-Magnetite-series into two components. One of these had the same O/Fe-ratio, the other one the same Fe/Ti-ratio as the original mineral. No changes took place when the sample was heated in a vacuum of 10^-5 Torr.
The second part deals with the observation of domain patterns in natural magnetites and the changes of these patterns under the influence of external magnetic fields. With the help of a new polishing technique powder patterns could be observed on magnetite grains in a non-conducting rock matrix.
In the third part the magnetization of the Suevites-that are tuffaceous rocks formed in connection with the formation of the Ries crater-is discussed. At all locations investigated the magnetization is nearly anti-parallel to the present direction of the earth's magnetic field. The small scatter among the directions from different location is taken as an indication for simultaneous magnetization of all Suevites.

View full abstract

The measurement of the magnetic moment of rock samples with a fluxgate magnetometer is discussed. A sample-fluxgate-configuration is proposed for which the influence of inhomogeneities of the sample and geometric inaccuracies is a minimum. Special attention is paid to samples of cubical shape.

View full abstract

The direction and intensity of remanent magnetism has been measured on a series of iron meteorite specimens and their magnetic domain patterns have been studied on polished surfaces. Specimens were carefully collected in the field and their geographic orientation noted. From these specimens, the magnetic effects of the geomagnetic field could be estimated and only those specimens which lacked such effects were used for further study. Specimens obtained from museums or similar sources are not considered satisfactory for these magnetic studies.
The intensity of remanent magnetism for octahedries from Arizona Crater averages 11.6×10^-3emu/cm³, whereas that for octahedrites from odessa Crater averages 6.6×10^-3emu/cm³. Domain patterns for schreibersite have a chain-like cross-section and are rod shaped in longitudinal section. Kamacite has a rectangular pattern when carefully prepared or under other conditions, a maze pattern.

View full abstract

Some preliminary data on secular variations of the geomagnetic field are given for the time 600 to 1500 A. D. as determined by archaeomagnetic methods. The inclination of the field has varied from a low of approximately 40 degrees to a maximum of some 60 degrees. Values of declination vary approximately from 10 degrees east of true north to 15 degrees west.

View full abstract

Intensity of natural remanent magnetization of potteries from Peru, Bolivia and Mexico together with igneous rocks (lavas and pyroclastic rocks) from Mexico has been measured. The specimens cover the ages between about 0 A. D. and 1400 A. D. Most of the potteries and rocks have been found to contain magnetite or maghemite with titanium which is stable against heating in air. The intensity of NRM and its demagnetization by heating have been compared with the artificial partial TRM in laboratory so that the secular variation in the geomagnetic total force at the localities could be estimated. The results seem to imply that the intensity of the geomagnetic total force in the South and Central America at about 500 A. D. was almost twice as large as that at present but that it was only 1.2 times larger than the present one at about 0 A. D. More work will be required to find whether this change is a worldwide tendency or due to a local anomaly.

View full abstract

The Thelliers' method of estimating the ancient geomagnetic field intensity was undertaken with some of the samples whose direction of NRM was reported in Part 1 of the paper. The results indicate that the intensity of geomagnetic field about 2, 000 years ago had the maximum value, being 1.5 times as large as that at present. The results also show a fairly good agreement with those reported by several authors such as Thelliers, Burlatskaya, and Nagata et al.

View full abstract

Magnetic measurements have been made on small discs prepared from a series of prehistoric Indian pottery pieces collected from Southwestern United States. The specimens ranged in age from 600 to 1400 A. D. Thermal demagnetization experiments allowed the assessment of the potential use of these materials to estimate the paleointensity of the geomagnetic field. It is concluded that they are satisfactory for the purpose provided suitable methods are used and tests are made to indicate mineralogic changes.

View full abstract

The Lovejoy formation consists of undeformed basalt flows distributed across the northern Sierra Nevada, California, and has been dated by the potassium-argon method as early Miocene. Measurements of remanent magnetization of 200 specimens collected at 16 widely separated sites have provided 13 independent determinations of paleomagnetic field direction during the relatively short period of eruption. Of these, 5 are not greatly different from the field of a reversed axial dipole, while 8 correspond to virtual geomagnetic north poles ranging in latitude from 60° north to 8° south. Stability of magnetization was established by a. c. and thermal demagnetization. Viscous magnetization, with intensity of the order of 50% of the natural thermoremanent component, was completely removed by peak alternating fields of 75 to 200 oersteds. The mean of the 13 virtual geomagnetic north poles lies at 76° north latitude and 74° east longitude, but the precision parameter k for these directions is only 4. Attempts to relate the degree of divergence of individual poles from the mean (or from the geographic pole) to the apparent relative intensity of the ancient field were unsuccessful because observed irreversible increases up to five-fold in the susceptibilities of specimens heated to 600°C invalidate any comparison of natural T. R. M. to artificial T. R. M. It is concluded that these basalts were erupted during a period of time, about 24 million years ago, in which the earth's field was predominantly reversed but varied in direction to an unusual degree.

View full abstract

A method of estimating the grain volume-coercive force distribution n(H_co, v) for rocks which can be treated by the Néel theory of non-interacting single domain grains is described. The method involves the analysis of a. c. demagnetization curves of partial TRM's produced in various temperature ranges. Experimental curves reported by Everitt (1961) for a sample containing single domain grains of maghemite have been analysed and a partial outline of n(H_co, v) for this sample found. The accuracy of the method was tested by re-solving for the original a. c. demagnetization curves and was found to be reasonably good. Prediction of results of other experiments carried out on the same sample, however, gave only semiquantitative agreement with experiment. Some possible explanations of the observed discrepancies are considered.
When grain interactions are important, the coercive force of a grain in increasing fields, a, becomes different from the coercive force in decreasing fields, b. It is then useful to find the grain distribution n(a, b). A method of deducing n(a, b) from n(H_co, v) is described and applied to the sample used above. The results are found to resemble actual distributions found by Bate (1962) for synthetic samples containing single domain maghemite particles.

View full abstract

Recent findings of geomagnetic variation anomalies in East Europe and Japan are reviewed. It is now concluded that mantle conductivity anomalies are fairly common all over the earth and that some of them are very local.
In order to see to what extent a large scale mantle conductivity anomaly affects geomagnetic variations, a theory of electromagnetic induction in a conductor which is not exactly a sphere is briefly described. The meaning of the depth of the conducting part of mantle as determined on the assumption of spherical symmetry is discussed.

View full abstract

Anomalies in geomagnetic variations in the Arctic Archipelago are first discussed. The gross characteristics of the Alert (Ellesmere Island) anomaly can be explained by a number of models involving a 100km upheaval of the 1400-1500°C isotherm to within 25-30km of the surface. The solution is not unique and may be in error by as much as 60%. However the response parameters most useful have been clearly defined. Magnetotelluric investigations lead to unusually high conductivities and a large anisotropy as expected. However a detailed synthesis of the magnetic variation results with the magneto-telluric analysis has not proved possible. At the present time, no self consistent explanation is known, but recent aeromagnetic and other data do not support any compositional explanation.
The suppression of the vertical field fluctuations at Mould Bay, Prince Patrick Island has been investigated further and the finite areal extent of the anomaly to the east has been demonstrated. Negative heat flow experiments and seismic investigations are described and their consequences outlined. A summary of the experimental variation and magnetotelluric data, now being examined, is given. The latter again qualitatively confirms the existence of a gross conductor, but at the present time it is not known if this is thermal or compositional in origin.
Experiments by the University of Toronto and the Dominion Observatory in southeastern Canada are described, which show that, within a confidence level of 20%, there are no gross anomalies in the region of Southern Ontario from London to Ottawa, or across the Logan fault in Western Quebec. However, work by the University of British Columbia indicates appreciable conductivity anomalies of a varied nature from Crescent Valley, in southern British Columbia to Lethbridge, in southwestern Alberta. Current investigations of a coast effect off Vancouver Island are also outlined.
It appears clear that considerable theoretical advances are necessary to clarify which, if any, of the unusual features found are related to abnormal temperatures in the upper mantle. Where an integrated geophysical approach has been attempted in Canada, no self consistent explanation has yet been found.

View full abstract

The orientations of the vector changes ΔF in the total magnetic field for time intervals of twenty minutes are plotted on polar diagrams for three magnetic stations in the Soviet Arctic. For two of these stations, Tikhaya Bay and Cape Chelyuskin, which are in the auroral zone the vector changes are oriented along a line. This suggests that the sources of the field are linear current elements. At the third station, Tixie Bay located south of the auroral zone, the vector changes lay on a plane whose orientation appears to be related to the contact between a Precambrian platform and Mesozoic folding. A large, discrete magnetotelluric event is described which occurs at Cape Chelyuskin one hour and 54 minutes before it occurs at Tikhaya Bay.

View full abstract

Impedance relations for non-uniform conductors with plane and spherical boundaries are derived which show the similarity between the magnetic potential method and the magnetotelluric method. The impedance relations are so presented that they can be used for any number of layers inside the earth, each representing a different conductivity distribution. Calculations of impedance and phase values for conductors with plane boundaries show that the horizontal wave length of the source can be assumed to be infinite when determining moderate resistivity distributions at shallow depths. Similar calculations for conductors with plane and spherical boundaries for a few conductivity distributions show that the effect of the earth's sphericity can be neglected for periods less than a day when determining conductivity distribution by the magnetotelluric method.

View full abstract

The convective flow conducting heat from the inner core of the earth to the mantle may follow a jet-stream-like pattern in which trapped segments of fluid form cylindrical rotating columns—as observed in laboratory studies in cylindrical rotating vessels. This suggests a very simple hydromagnetic version of the dynamo theory in which bundles of magnetic lines of force twisted by the rotation of the outer columns play a predominant role. The westward drift of the convection pattern is a reaction to the eastward drift which the jet stream imparts to the inner core. This drift bends the bundles of force lines in such a way that the twisting provides the curl H (near the equator) which supports the earth's main field. The dynamo action tends to produce a main dipole moment parallel to the axis of rotation and effects that are intrinsically fairly small are proposed to account for the present inclination between them. One such effect is inhomogeniety of the temperature distribution and conductivity of the mantle due to convection related to continental configuration. Another arises from irregular kinking of the almost axial bundles of lines of force threading through the solid core and twisted by its rotation relative to the mantle. It is thus plausible that magnetic and rotational axes have remained nearly parallel so that paleomagnetism approximately traces the course of the kinematic poles. Field reversal and continental drift are also briefly discussed.

View full abstract

The magneto-hydrodynamic theories of the earth's main magnetic field rest on the process of simple induction. Since e_magn>>e_mech, this is only possible, if the stronger magnetic field become force-free or pressure-compensated. It is supposed, that the period of≈50 years, that is to be find in the secular variations of the magnetic field vector, in the amplitudes of the Chandler-Wobble and in the lenght of a day are caused by a magneto-hydrodynamic oscillation. The origin of the dipolefield of the earth is to be find in the boundary-layer of the core.

View full abstract