INTRODUCTION

The magnetic field of the earth changes on a vast spectrum of time scales ranging from seconds to millions of years. Rapid changes during magnetic storms may be so severe that surveying is impossible. Slower changes such as the secular variation pose problems for compiling survey data from different years. You’ve all learned about the magnetic reversal time scale, magnetic stripes of the ocean floor and continental drift. Today we’ll learn more about some of these phenomena, starting with the high frequency ones.

MAGNETIC STORMS

Galileo was the first modern scientist to study sunspots and to explain them in terms of a 27 day rotation of the sun. Associated with the sunspots are vigorous emissions of charged particles which travel to the earth in two or three days. These particles interact with the magnetic field of the earth according to the Lorenz Law and produce a "ring current" that flows westward around the equator in the ionosphere.

The net result is that, after an small initial increase, the field observed at the surface of the earth is reduced by hundreds of nanoTeslas for several days. During storms, which occur synchronously all over the earth, magnetic surveying is not possible.

Based on observations of the sun and its known rotation period, geomagnetic observatories issue predictions of magnetic field behavior. We can use these predictions to schedule field work during magnetically quiet times. You will find predictions at the web sites of the British Geological Survey, the Geological Survey of Canada and the United States Geological Survey.

DIURNAL VARIATION

Every day the field magnitude goes up and down by a few tens of nanoTeslas. The pattern of change is different for each field component and at each latitude but the pattern is roughly the same every day. The variations occur at the same local sun time everyday. The changes are stronger during the day than during the night and stronger during the summer than during the winter. These correlations imply that the basic cause is solar. There are also weak 12 and 12.4 hour variations that imply solar and lunar tidal influences.

These phenomena can be explained by Stewart’s "tidal dynamo theory". The basic idea is that the solar radiation produces a more intense ionosphere on the sunny side of the earth (more charge per unit volume). The ions are pulled into tidal bulges by gravitational attraction of the sun and the moon. Guided by the magnetic field of the earth the ions form two big horizontal current gyres on the sunny side of the earth. The northern hemisphere currents form a counter-clockwise gyre whereas the southern hemisphere one is clockwise. It is amazing that Stewart proposed this theory in 1882 long before the discovery of the ionosphere!

The current gyres are, in effect, two big dipoles locked in place above the sunny side of the earth. As time goes on the earth rotates under the fixed gyres and, according to the Biot Savart Law, the magnetic field changes with a 24 hour period.

The easiest way to correct for diurnal changes is to use a magnetic gradiometer. Both sensors experience the same diurnal change which is cancelled out when we subtract the readings of one sensor from the other.

Given the small amplitude of the diurnal variation and its slow rate of change, we can readily adjust our survey values to some common time datum. In practice this means making repeat "base station" readings at short time intervals (say half an hour) or using a recording base station. In some parts of the world permanent magnetic observatories are close enough to monitor the diurnal variations.

For very large aerial surveys we may be too far from base stations or observatories to use them. In such cases it’s possible to synthesize a diurnal correction curve from the field values at places where the flight lines cross.

Deep-sea surveys may be far from base stations and the anomalies of interest (magnetic stripes) may be smaller than the diurnal variation. In such cases we might be better off using a gradiometer.

Satellite surveys are conducted using orbits hundreds of kilometers above the earth. Here the anomalies are much, much less than the diurnal variation. To reduce the diurnal effect we use only quiet night-time data, use only low orbit data and stack data for many repeat paths.

SECULAR VARIATION

This term refers to changes over years, decades and centuries. These changes can be quite large; here in Durham the total field magnitude is decreasing about 100 nT per year. On the one hand the secular variation is a nuisance. Survey data taken at different times must be adjusted to some common data and, as you know, the difference between magnetic and true north changes too. On the plus side we can use the secular change preserved in sediments as a dating tool.

Observatories in Europe have recorded field directions since the 1500’s. These show changes of tens of degrees. The changes are more or less similar and synchronous throughout Europe but differ from patterns observed in other parts of the world. That is, the secular variation patterns are regional in scope.

Secular variation dating is done by setting up a local time series of inclination and declination for datable material. Organic rich lake sediments datable by 14C are commonly used. In addition we can use speleothems, lava flows or any other datable material with a stable NRM. Once we have a "master curve" for a given region, dating is just a matter of correlating our data series to the master curve.